MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

matching transistors - DIY
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author matching transistors - DIY
delayed
Hi,
I am looking into this and see Ian Fritz, October 2010 article and Ray's MFOS article, Transistor Matching 101 under the Synth-DIY 101 tab within his website. The MFOS shows a little circuit and a larger GO no GO circuit.

Has anyone made up one of these? If using Ian's how do you swap the emitter leads easily. Setting up a switch perhaps?

Does anyone have some images of their testing bread board setup? Or better suggestions than these two examples?

I would like to see what other people are doing to match their transistors.

Thanks,
roglok
I built both on stripboard, the simple MFOS and Ian's. Both work fine, but Ian's is a lot simpler to build.

I used DIP sockets for testing the transistors. Left is Ray's, right Ian's:



Ray's is missing the power connector - was short of those and harvested it for a module oops
e-grad
Do these sockets work?

I had built a noise module with a socket for the noise source (transistor) some time ago. However, the transistor didn't made sufficent contact to make it feasible.
roglok
e-grad wrote:
Do these sockets work?

I had built a noise module with a socket for the noise source (transistor) some time ago. However, the transistor didn't made sufficent contact to make it feasible.


The machined sockets do work - not sure about the cheap ones. They actually sit in there quite firmly... Also keep in mind that this is only a temporary circuit...
qfactor
Roglok: so you used the 8-pin IC sockets for the transistors? How do the E-C-B pins of a transistor plug in?
roglok
qfactor wrote:
Roglok: so you used the 8-pin IC sockets for the transistors? How do the E-C-B pins of a transistor plug in?


You have to stretch them apart a little bit. I used a fine tipped permanent marker to designate the pinout directly on the stripboard. I can post a detail pic for you later...
roglok


I've inserted one transistor as illustration. If you look closely, you can make out the C B E markings next to the sockets. The lower two pins on each sockets are not used. You can see that one end of each 100K resistor has been bent upwards. These are the ends connected to the emitters and where I attach my DMM. Of course you could substituted this with something more robust, but since I don't match transistors every day it has worked just fine so far...
emmaker
You might check to see if the Ray Wilson circuit has the same flaw that the Moog schematic has. That is there needs to be a 10K resistor between the collector and emitter of the transistors in the PNP tester like the NPN one has.

Also keep in mind those sockets don't have a lot of insertion cycles. Might want to stack them 2-3 high and replace them as they wear out. That's a lot easier than having to unsolder and replace the one on the board.
J3RK
I have Ian's circuit permanently assigned to a small bread-board. Seems to do the trick quite nicely. I now only use monolithic matched pairs on rare occasions. I typically match them into groups, (by writing the values down and then grouping them by that) so I end up with maybe 5-6 good pairs after maybe 15-20 minutes of testing. That goes a fairly long way. thumbs up
emmaker
Did a quick look for the Moog flaw and found this:

http://www.dragonflyalley.com/constructionTransistorMatching.htm
fonik
i made myself a professional one-off PCB, but anyways, i use machined SIL sockets (single row, cut to the correct length. and they work fine. actually a little bit too good. i have to insert the transistors carefully in order not to bend the legs...
roglok
emmaker wrote:
Did a quick look for the Moog flaw and found this:

http://www.dragonflyalley.com/constructionTransistorMatching.htm


Bill and Will seem to be happy with DIP sockets, too grin

On a sidenote: I have previosuly matched pairs of 2N3904 and 2N3906 from the same batches - the VBE of almost all units tested was already within +/- 2mV of their counterpart without selection...

I guess this means that you can often just pop in any pair of modern transistors from the same batch and get satisfactory results...
kvitekp
Here's another version of MFOS Vbe matching tool: http://midisizer.com/vbe-matching, I still have a couple of spare pcbs.
delayed
kvitekp wrote:
Here's another version of MFOS Vbe matching tool: http://midisizer.com/vbe-matching, I still have a couple of spare pcbs.


how much would you want for 1 or 2 of the pcbs? feel free to pm me. thanks,
jonbstevens
I breadboarded Ian's when i wanted matched transistors for for CGS DUSG's. I get close to 2 octaves in tune once i've got them trimmed correctly.

I'm not sure what sort of tuning to expect with unmatched transistors, but i'm pretty happy.
fonik
okay, while the MFOS circuit is called 'simple', Ian Fritz' circuit really IS simple. and there is no need for a precise voltage supply or matched resistors.
just a set of three resistors, a diode, and a DPDT on-on switch for each NPN and PNP transistors:


Ian Fritz' Transistor Matcher by fonitronik, on Flickr
delayed
fonik wrote:
okay, while the MFOS circuit is called 'simple', Ian Fritz' circuit really IS simple. and there is no need for a precise voltage supply or matched resistors.
just a set of three resistors, a diode, and a DPDT on-on switch for each NPN and PNP transistors:


Ian Fritz' Transistor Matcher by fonitronik, on Flickr




Would you like to share your layout?
fonik
the eagle files, you mean? i would/could ask ian.
delayed
Yes, that would be good. Or just the SCH. I am guessing you set up the switch to swap the "e" and would like to see that part. A toner layout would even be better but I see that is asking a bit ++.
roglok
very handsome, matthias!
fonik
roglok wrote:
very handsome, matthias!

and small !!! Will have to mount it to a panel .
J3RK
fonik wrote:
okay, while the MFOS circuit is called 'simple', Ian Fritz' circuit really IS simple. and there is no need for a precise voltage supply or matched resistors.
just a set of three resistors, a diode, and a DPDT on-on switch for each NPN and PNP transistors:


Ian Fritz' Transistor Matcher by fonitronik, on Flickr


That is nice! thumbs up I'd much prefer to use that over my breadboarded version. Mr. Green

It would be cool to make a "Builder's Utility Module" that contained this, a tuning calibrator for 1V/Oct and maybe 1.2V/Oct, and maybe some peak level indicators or something along those lines.
frijitz
fonik wrote:
the eagle files, you mean? i would/could ask ian.

Of course, no problem. It's not like some clever invention, after all, basically just a Wheatstone bridge.

Ian
diablojoy
Quote:
That is nice! I'd much prefer to use that over my breadboarded version.

It would be cool to make a "Builder's Utility Module" that contained this, a tuning calibrator for 1V/Oct and maybe 1.2V/Oct, and maybe some peak level indicators or something along those lines.

seconded
an accurate note indicator would be very useful
qfactor
diablojoy wrote:
Quote:
That is nice! I'd much prefer to use that over my breadboarded version.

It would be cool to make a "Builder's Utility Module" that contained this, a tuning calibrator for 1V/Oct and maybe 1.2V/Oct, and maybe some peak level indicators or something along those lines.

seconded
an accurate note indicator would be very useful


An A440 reference tone in this module would be good too! thumbs up
elmegil
This was one of my early projects. Taught me the lesson of "read the specs for the switch, don't just look at it in a picture with no size reference." :-)

It's all wires, resistors and diodes, and a 9V battery on the inside. And one of those switches is a 4PDT :-) It's Ian's circuit....

Edit: having trouble remembering attachments lately, apparently.
qfactor
elmegil wrote:

It's all wires, resistors and diodes, and a 9V battery on the inside.


So this circuit works with a +9V and gnd? No -9V?
elmegil
I misspoke, there are two 9V batteries inside.

I also revised it at one point and put in the trimmer to zero out the difference.

Here are the guts:
marvkaye
I built my transistor matcher from Ian Fritz's schematic, included trimmers to get the resistances matching and then added switches to handle the emitter swaps. After having used the unit for awhile (most recently matching for the expo converters on several Thomas Henry/Fonik VCO555 PCBs) I find that I prefer to install one trannie as a reference and then match the DUTs by recording their variations from the reference part and putting those with the same readings in the same bin. It's actually faster than swapping both trannies and swapping emitters, although I can do it either way. Live and learn. The next update is to replace the DIP sockets with a nice ZIF socket to make putting the transistors in and out a bit easier. With the DIP sockets I have to handle each transistor... there's no way to push the pins in using a pair of tweezers, simply too much force is required.

The circuit is built on stripboard and the DIP sockets have really long pins (wire-wrap type) so they're installed standing tall enough for me to be able to use 3/4" standoffs to secure the board (the 4 black screws surrounding the sockets.)

Anyway, it works really well... I found that in about an hour I was able to match up 4 pairs to +/- 50uV (yep, that's 50 micro volts).... that was before I realized that most people consider their transistors matched as long as they are within 1.5-2mV. Duhhh... Since all my trannies came out of the same batch I probably could have used them without matching and met that spec... I've got leftovers from the process that match to 0.5mV already. Live and learn.



Here's the trans matcher alongside my MFOS V/Oct calibrator... another great project.



Thanks for looking.....

<marv>
Altitude909
perfect timing for this thread, need to match me some 3904s
oldenjon
fonik wrote:
and there is no need for a precise voltage supply or matched resistors.

Wait, Ian's method calls for a matched pair of 100k resistors. What's more is that they should be matched using a 4.5 digit DMM, and a meter with those specs isn't exactly cheap.... Did I miss something?
Altitude909
oldenjon wrote:
fonik wrote:
and there is no need for a precise voltage supply or matched resistors.

Wait, Ian's method calls for a matched pair of 100k resistors. What's more is that they should be matched using a 4.5 digit DMM, and a meter with those specs isn't exactly cheap.... Did I miss something?


You can buy 0.1% resistors and a reasonable price (0.75$ ea last I looked). a few bucks of those is better than sorting through a big pile of 1%.

If you're in the market for a good DMM, I just grabbed one of these: http://www.ebay.com/itm/370769141376?ssPageName=STRK:MEWNX:IT&_trksid= p3984.m1497.l2649

5.5 digit and the one I got was laboratory clean
frijitz
oldenjon wrote:
Wait, Ian's method calls for a matched pair of 100k resistors.

I recommend matched resistors or adding a small cermet trimmer as described in the writeup. If the resistors aren't matched, then you have an offset voltage to subtract and you can end up trying to determine a small difference between two large numbers. The method still works, it's just that you loose resolution, which makes a difference when you get into the uV regime.

Ian
fonik
here are the eagle files.
delayed
Super cool of you to post these. Thank you.
delayed
getting late/early now. wiring tomorrow. kinda think i have this flipped, but since it is a mirror of itself i don't know if that matters. will investigate later.
megaohm
oldenjon wrote:
fonik wrote:
and there is no need for a precise voltage supply or matched resistors.

Wait, Ian's method calls for a matched pair of 100k resistors. What's more is that they should be matched using a 4.5 digit DMM, and a meter with those specs isn't exactly cheap.... Did I miss something?


Check out page four "What Current?".
Make the -V variable.

With my cheap meter measuring 100K resistors I only get:
100.?.
So, one decimal place.
If I measure 10K resistors my meter gives me two decimal places: 10.??.

In this case, I'd replace the matched 100K resistors with 10K resistors (I would replace them with 30K, actually, because my meter still gives two decimal places with that value - 30.??K).

So, by dropping to a lower value resistor you can get better resolution when matching the resistors. Depends on your meter.

Plus, it seems a good idea to have adjustable current because then you can set it according to the circuit you will be using the matched pairs in.
frijitz
megaohm wrote:
So, by dropping to a lower value resistor you can get better resolution when matching the resistors. Depends on your meter.

Percentage-wise 100.x and 10.xx are the same resolution.

Ian
doobedy
jonbstevens wrote:
I breadboarded Ian's when i wanted matched transistors for for CGS DUSG's. I get close to 2 octaves in tune once i've got them trimmed correctly.

I'm not sure what sort of tuning to expect with unmatched transistors, but i'm pretty happy.


Two things:

1) Matching Vbe gets you temperature stability, and has nothing to do with tracking performance as far as I can tell.

2) Does anybody (Ian?) know which transistors have good log conformance, which should get us better tracking.

I just did a couple hours of matching for two DUSG boards, using 2n3904 and 2n3906s, for 8 pairs in total. I haven't wired them up yet but I'm guessing a couple octaves is probably pretty good judging by past experiences.

As far as the rest, I matched the 100k resistors to one decimal point, but I don't think it matters too much as long as you you leave Q1 in it's spot, and compare a succession of Q2s. Like Ian says in his PDF, as long as the difference is the same, they match. No need for supermatched resistors.

I got all my matches down to +/- .1mv according to my cheap meter, but it took a good 5 minutes for temps to really settle down during tests, a small fan might help, but whatever. It is kind of fun. Like the world's worst slot machine.
daverj
Seeing the mention of 9 volt batteries above got me thinking that Ian's circuit could be simplified to run off a single battery instead of two. So I came up with a simple change and PM'd Ian about it. He came back with an even simpler modification to do it. We went back and forth a couple of times and ended up with the following schematics.

I feel guilty even adding my name to the drawings since in the end all I did was prod him into coming up with the changes.

A couple of notes:

-- I did three drawings. The first one uses a pair of DPDT switches and is the most universal. It's the best one to build. The second drawing is for people wanting to build something extremely cheap, without the switches. But it's different for NPN and PNP, plus requires you to manually swap the meter probes during matching. The third one is a variation on the second, combining the two with a single battery but separate sockets and probe points.

-- The sockets are marked EBC but the B and C are connected together so no need to twist the leads around if you have transistors that are in an ECB package instead of EBC.

-- With 1% 100K resistors the trimpot could theoretically be 2K. But since trimpot tolerance is not as good, it should be slightly larger. I picked 5K since it's the next larger standard size. If you can get a 2.5K or 3K trimpot that would be even better. You adjust the trimpot by switching the meter probes back and forth, turning the trimpot until you get the same reading with just the +/- sign changing.

-- All three drawings show the same simplified circuit. Just without the switches in two of them.

-- The same simplified circuit could also be connected to a regular power supply too instead of a battery. It just needs one power rail plus ground for that.





qfactor
daverj wrote:
Seeing the mention of 9 volt batteries above got me thinking that Ian's circuit could be simplified to run off a single battery instead of two. So I came up with a simple change and PM'd Ian about it. He came back with an even simpler modification to do it. We went back and forth a couple of times and ended up with the following schematics.


Thank you all for this!! applause we're not worthy
gaetano
Interesting about the Moog 'flaw', which I didn't notice, as I have only used the Moog version for matching NPN's
Dave Kendall
Many thanks Ian and Dave - a veryuseful circuit thumbs up

Quote:
2) Does anybody (Ian?) know which transistors have good log conformance, which should get us better tracking.


Is there a way we can measure for that?
Would it be something like varying the current (voltage?) to the DUT in set increments and comparing readings with an "ideal" set of values for a log curve?

How important is log conformance for VCOs and VCAs respectively?

Also, how useful/important is the hFE reading in the scheme of things - do the two trannies need to match tightly, and does the hFE value need compensating for in a VCA/VCO circuit in some way?

Sorry for all the questions, but I've been wondering about these things for a long time, and haven't yet found anywhere that discusses them.
A way to avoid expensive matched pairs and get *really* high performance would be truly magical. Well worth the time matching IMO. Ideal if you have time and not money..... wink

cheers,
Dave
ejr27233
As I've only got a 3.5 digit multimeter is there any point in using a x100 gain op amp in the circuit to measure the offset mV?.
frijitz
Dave Kendall wrote:

Quote:
2) Does anybody (Ian?) know which transistors have good log conformance, which should get us better tracking.

Is there a way we can measure for that?

The best log transistors are the special super-transistors that are sold as hi-conformance devices. The best are the MAT0x devices. The LM394 and SSM2010 are also very good.

Log conformance is usually characterized by the spurious series resistance in the emitter path. You will see this given different names. The standard measurement of this resistance is the slope of the collector to emitter voltage vs base current, with the collector open. The conformance is not usually specified for most transistors, except for the super-transistors. I measured a number of other devices a while back, as did someone else on the s-diy list.

Quote:
How important is log conformance for VCOs and VCAs respectively?

Quite important for VCOs. A high-frequency tracking adjustment does a pretty good job of compensating for errors, but it still works best to start with a good device. Most VCAs are linear so log characteristics don't enter in.

Quote:
Also, how useful/important is the hFE reading in the scheme of things - do the two trannies need to match tightly, and does the hFE value need compensating for in a VCA/VCO circuit in some way?

Usually that is not considered. High hFE is required for the ideal equations to be accurate, though. (See Analog Devices Nonlinear Handbook for this.)

Quote:
A way to avoid expensive matched pairs and get *really* high performance would be truly magical. Well worth the time matching IMO.

Once again, matching is for temperature compensation, not for tracking.

Ian
frijitz
daverj wrote:
...So I came up with a simple change...

Not a big deal, but I just noticed that Dave has reversed the position of the reversing switch and the DVM leads compared to my original. This means that the mismatch is now the average of the two readings and the error is half the difference. Probably this is even easier to use than the original.

Ian
Dave Kendall
Hi Ian.

Thanks for the reply - more pieces of the puzzle. smile
I found some posts on [sdiy] from you, putting MAT02/03 at the top, followed by LM394 and SSM2210 in terms of good log conformance.
Googling and so on, I couldn't find any other write-ups or data about how much worse other transistors perform compared to them.

It would be good to know how far off (or not) regular transistors are compared to the above.
I had a look at 3 datasheets for the BC847BS, a dual NPN surface mount matched pair that I've seen mentioned as possible substitutes, but couldn't find any mention of Rb or log conformance data.

Is it possible to measure this with a typical set of SDIYer skills/tools and a little patience?


cheers,
Dave
elmegil
While we're on the topic.....

I've been trying to think about how to pack relatively simple VCAs into a tight space, and this circuit:

http://urekarm.tripod.com/synth/jx_vca.pdf

Seems like it could be promising. I could build 6 of those on a 40 x 50 hole stripboard.

What I'm not sure about is the "Selected" for the two NPN transistors.

Would you expect this is the "normal" matching we've been discussing here, or some other means of selecting the pair? The site that is on doesn't have any direct links to the schematic, BTW, I just found it via google, and I don't see any reference to or discussion of it.

Unfortunately attempting to email the account listed at the site gets bounced so I have been unable to ask directly to the designer.
daverj
frijitz wrote:
daverj wrote:
...So I came up with a simple change...

Not a big deal, but I just noticed that Dave has reversed the position of the reversing switch and the DVM leads compared to my original. This means that the mismatch is now the average of the two readings and the error is half the difference. Probably this is even easier to use than the original.

Ian


Oops. I didn't even notice. If it's easier to use then it's a happy accident. If people want to use it the old way, simply move the meter connection down below the switch.
KnobHell
I used a zif socket on my tester. The ZIF socket makes it really easy to swap transistors. Would work with any design.

frijitz
Dave Kendall wrote:
It would be good to know how far off (or not) regular transistors are compared to the above.

I dug up my old notes and found results for 2N3904 (1.6R), CA3083 (1.3R) and CA3046 (5.0R).

Quote:
I had a look at 3 datasheets for the BC847BS, a dual NPN surface mount matched pair that I've seen mentioned as possible substitutes, but couldn't find any mention of Rb or log conformance data.

Right, just what I said above.

Quote:
Is it possible to measure this with a typical set of SDIYer skills/tools and a little patience?

Yes, it's not difficult.

Ian
qfactor
Just tried my tranny matcher based on what was discussed here, and was testing a bunch of 3906.
Funny thing was there were 3 of these trannies that displayed a negative voltage (between -6.2mV to -12.2mV). Is something amiss here?
The rest of my tests returned with trannies measuring 0.4mV to 22.2mV. All positive readings.
I'm a noob in matching trannies smile
xpmtl
KnobHell wrote:
I used a zif socket on my tester. The ZIF socket makes it really easy to swap transistors. Would work with any design.


Great idea. thumbs up
bartleby
why do you guys all build two separate circuits for npn and pnp? seriously, i just don't get it

the only difference i see between ian fritz's testing circuits for npn and pnp transistors is the polarites of the diode and of the overall power connection. so why not just put the diode in a socket as well, so you can adjust its polarity depending on which type of transistor you want to test?

that's what i did, anyway. it's not a pretty build (i'm not a stripboard virtuoso, and was in a bit of a hurry when i built this), but it works fine for pnp, and i can't see why it wouldn't work for npn when i reverse diode and power polarity...

elmegil
I used a 4PDT switch to switch the polarities myself.
LetterBeacon
Apologies, but this might be a stupid question:

Ian specifies in his pdf file that the multimeter must be 4.5 digits.

My DMM specifies this in the manual:

Range = 200mV
Resolution = 0.1mV
Accuracy ±(5%+1)

Is my DMM accurate enough? I can't work it out!
Altitude909
LetterBeacon wrote:
Apologies, but this might be a stupid question:

Ian specifies in his pdf file that the multimeter must be 4.5 digits.

My DMM specifies this in the manual:

Range = 200mV
Resolution = 0.1mV
Accuracy ±(5%+1)

Is my DMM accurate enough? I can't work it out!


How many digits after the decimal place?

This is what a 5.5 looks like
fonik
LetterBeacon wrote:
Apologies, but this might be a stupid question:

Ian specifies in his pdf file that the multimeter must be 4.5 digits.

My DMM specifies this in the manual:

Range = 200mV
Resolution = 0.1mV
Accuracy ±(5%+1)

Is my DMM accurate enough? I can't work it out!


DMM specs explained
http://www.brianhoskins.co.uk/website/digital-multimeter-accuracy/
LetterBeacon
Thanks for the replies.

Turns out my DMM isn't quite good enough. Do you need to spend £100+ on a DMM to be able to accurately match transistors then?
EATyourGUITAR
LetterBeacon wrote:
Thanks for the replies.

Turns out my DMM isn't quite good enough. Do you need to spend £100+ on a DMM to be able to accurately match transistors then?


that depends on your requirements. if you require perfectly matched transistors than yes. as stated in the thread, some people have had ok results by randomly picking two transistors from the same batch without any testing. it depends on the circuit, the batch of transistors, and your requirements with respect to sound and temperature stability. if you only need matched transistors for one project, you can use any of the integrated circuits mentioned in the thread. they range from $0.40 to $12 depending on specifications, package, product life cycle, rarity etc..
LetterBeacon
Trouble is not all PCB layouts allow you to use ICs when they call for matched transistors.

I reckon there's a market there for one of the SDIY stores (Thonk, Synthcube?) to make available matched pairs of 2n3904s etc. I would happily pay for them.
EATyourGUITAR
LetterBeacon wrote:
Trouble is not all PCB layouts allow you to use ICs when they call for matched transistors.

I reckon there's a market there for one of the SDIY stores (Thonk, Synthcube?) to make available matched pairs of 2n3904s etc. I would happily pay for them.


I can install pins in a breakout board with teflon tubing so that when I bend the pins to fit the pcb, nothing gets shorted. if I add up the cost of the breakout pcb + matched 3906 in sot-363 + teflon tubing I still spend less than $1. you can also design the breakout board to fit a specific pinout if you like and the cost is about the same. no pin bending required.
LetterBeacon
All spare PCBs have now been accounted for!

Thanks very much and I'll get them posted off tomorrow.


*******************************************************

Because I'm a bit lazy, I thought I'd get Fonik's transistor matching PCB fabricated. The minimum order was 10, so I now have 9 spare.

If anyone wants one of those PCBs I'm selling them at £2 each inc. postage (£2.50 WW). I'm not making a profit on these, the £2 is just to cover my costs, and I thought these might be useful for the community. I only have 9 so it's first come first served. Please PM me if you want one.

DISCLAIMER: All I've done is take Fonik's Eagle files, generated the Gerber files and then sent those files off to be fabricated. I haven't had a chance to test the board out and I probably won't until after Christmas. You buy these at your own risk! Saying that, as I haven't touched the board layout at all, there's no reason why they wouldn't work, I'm just covering my arse!

I have both Fonik's and Ian's permission for this.

xpmtl
LetterBeacon, i'll take one, can you PM me your paypal addy?

Thanks
sonicwarrior
LetterBeacon wrote:
I haven't had a chance to test the board out and I probably won't until after Christmas.


I'll wait until it's tested then. oops
aladan
LetterBeacon wrote:
I reckon there's a market there for one of the SDIY stores (Thonk, Synthcube?) to make available matched pairs of 2n3904s etc. I would happily pay for them.


These have been available from MFOS since time immemorial:

http://musicfromouterspace.com/index.php?MAINTAB=SYNTHDIY&VPW=1910&VPH =979

Cheers,
A.
LetterBeacon
sonicwarrior wrote:
LetterBeacon wrote:
I haven't had a chance to test the board out and I probably won't until after Christmas.


I'll wait until it's tested then. oops


Too late they're all gone! I guess by buying one you're taking a bit of a punt, but at £2 a board, it's not much of one!

Quote:
These have been available from MFOS since time immemorial:

http://musicfromouterspace.com/index.php?MAINTAB=SYNTHDIY&VPW=1910&VPH =979

Cheers,
A.


Ah yes I'd forgotten about that! I do remember it taking a good while for stuff to arrive from Ray though...
diablojoy
Quote:
Ah yes I'd forgotten about that! I do remember it taking a good while for stuff to arrive from Ray though...

if this is for the TTSH then i think you will have the time to wait.
dont think i have ever had to wait more then a month fotr rays stuff

2n3904's from the same batch are really close these days as well.
elmegil
diablojoy wrote:
Quote:
Ah yes I'd forgotten about that! I do remember it taking a good while for stuff to arrive from Ray though...

if this is for the TTSH then i think you will have the time to wait.
dont think i have ever had to wait more then a month fotr rays stuff

2n3904's from the same batch are really close these days as well.


I'd be kind of hesitant to pay $12 a pair for 9 pairs, and you're still left with what to do for NPNs....

No disrespect for Ray, I know that matching is a labor intensive process, and he is including tempcos too.... But I can do the labor myself and use parts that are a few cents a pair.
diablojoy
Quote:
I'd be kind of hesitant to pay $12 a pair for 9 pairs, and you're still left with what to do for NPNs....


yeah pretty expensive although i can see where ray is coming from
I will match my own, in any case and i already have tempco's
I think you meant PNP for the 2n3906
elmegil
diablojoy wrote:
Quote:
I'd be kind of hesitant to pay $12 a pair for 9 pairs, and you're still left with what to do for NPNs....


yeah pretty expensive although i can see where ray is coming from
I will match my own, in any case and i already have tempco's
I think you meant PNP for the 2n3906


Uh, yeah... d'oh!
Jop
Thanks Matt, pcb arrived today.

Great to have this tiny circuit on pcb instead of breadboard.

Happy matching nanners
fonik
from my experience:
use a low impedance path from the PCB to your DMM, don't use cheap aligator clip cables. best attach your test probes directly and firm to the PCB (be creative)
i put a socket from a socket wrench upon the transistors to cover them. keeps them at a constant temp.

usually i still have to wait about half a minute until the readings settled.
J3RK
fonik wrote:
i put a socket from a socket wrench upon the transistors to cover them. keeps them at a constant temp.


Nice idea! Heat isn't going to easily transfer through that. very frustrating <---Heat trying to get through the socket... hihi
Paradigm X
yes thanks letterbeacon, forgot to mention mine came. Thought it was an empty envelope at first! many thanks

thumbs up
iopop
Just wanted to report back that I successfully built and tested the PNP section of the matcher PCB manufactured by LetterBeacon. Thank you LetterBeacon, fonik and frijitz!

Tested a few random 2n3906s and all were matched at around 0,26 mV.
Cablebasher
Any chance they will be another run of these pcbs?

I missed out last time but would find one really handy.

Thanks.
LetterBeacon
@ iopop -phew! Glad it works! I still haven't gotten round to building mine!

Cablebasher wrote:
Any chance they will be another run of these pcbs?

I missed out last time but would find one really handy.

Thanks.


I won't be doing another one as I have my board now. The Eagle files are earlier in this thread, why don't you get some made at iTead? It's very cheap and I'm sure you'll be able to sell the surplus ones off to other muffwigglers.
flts
^- I'm in for one. I might even make a run if nobody else wants to do it.
regenbot
i'll take one also if someone does a run thumbs up
Reality Checkpoint
I would like one too if somebody could do a run.

I am afraid I do not have the 'mad puter skilz' to do it myself. Dead Banana
EATyourGUITAR
I have the gerber files made from the eagle files fonik posted. I am ready to order the PCB's from itead but I need more than just 3 people. I need 1 more person to make this worth it. there is a 1 month wait for the PCB's to get here from china.

price will be $6 in the USA and $10 international. this includes PCB, packaging, shipping and paypal fee.

fonik if you do not wan't me to sell PCB's from your eagle file, I will take this down. I just want to help the people that refuse to use eagle to make gerber files and order PCB's.
grilojoe
me! I want one!
Broadwave
I agree with Diablojoy… All my NPN/PNP transistors were supplied "on tape" - So probably, they're all from the same batch

I didn't match any transistors when I built my 2600 clone

The only reason I didn't match them is because I was basically "testing" each PCB as I went along, thinking I'd swap for matched pairs at a later date if I needed to… turns out that I didn't need to.

I've not run into any problems, but I found that thermal coupling seemed to be more important (a SMALL dab of thermal compound between the transistors, then tightly wrapped with some self adhesive aluminium tape)
M-Circus
@EATyourGUITAR

I'll have one too pretty please.
flts
I think it'll be easier & cheaper for me to order directly since there's at least one local friend who wanted one. So I'll skip for now but might have some extras later.
steffensen
EATyourGUITAR
Im up for one board buddy!
Reality Checkpoint
steffensen wrote:
EATyourGUITAR
Im up for one board buddy!


Me too!
EATyourGUITAR
I guess both people in finland are out, but I already have 4 new people. I will order PCB's today.
Reality Checkpoint
EATyourGUITAR wrote:
I guess both people in finland are out, but I already have 4 new people. I will order PCB's today.


Thank you!
flts
I ordered some PCBs as well. Just for fun decided to try out TinySine / TinyOS who probably use the same factory as Itead etc. anyway.

If/when the PCBs arrive & turn out to be alright (= provided I didn't mess up anything in Gerber conversion and the PCB service is OK), I can mail extras to regenbot and any other local wigglers at least. Rest can be mailed to EU wigglers as well (I suppose it'd cost more to mail one to US than what ordering 10 pieces yourself would cost), I'll post if/when I have any extras available once I receive the PCBs.
fracinfrucer
EATyourGUITAR wrote:
I guess both people in finland are out, but I already have 4 new people. I will order PCB's today.


If not too late, I would love one as well! thumbs up
jaidee
If there is a spare board going, I will take one (UK).

Thanks!
EATyourGUITAR
there will be 10 PCB's ordered. anyone in the world that wants one, except finland, can get one. 4 of them are still available for anyone coming late to the thread. itead takes a month. I will post when I have the PCB's.
regenbot
Nice, thanks thumbs up

flts wrote:
I ordered some PCBs as well. Just for fun decided to try out TinySine / TinyOS who probably use the same factory as Itead etc. anyway.

If/when the PCBs arrive & turn out to be alright (= provided I didn't mess up anything in Gerber conversion and the PCB service is OK), I can mail extras to regenbot and any other local wigglers at least. Rest can be mailed to EU wigglers as well (I suppose it'd cost more to mail one to US than what ordering 10 pieces yourself would cost), I'll post if/when I have any extras available once I receive the PCBs.
Cablebasher
I'd love one to UK if not too late?

Cheers.
EATyourGUITAR
there are only 10 PCB's for now but if you would like to be added to the waiting list, post in the thread so you can get in on the next batch 2 months from now.

so far I have 9 PCB's reserved for these people. I can sell one more PCB from this first batch. I will PM people my paypal info so you can pre-pay and be done with it or you can pay for it when I have the PCB's.


Cablebasher
Reality Checkpoint
grilojoe
M-Circus
steffensen
fracinfrucer
jaidee
jbdiver
mikecameron
jaidee
Payment sent!
M-Circus
@EATyourGUITAR:

PM received. I'll send you the money tomorrow.
grilojoe
Payment sent!

Thanks!

screaming goo yo
EATyourGUITAR
this batch of 10 is sold out. all 10 people are listed here. if there is interest next month, we can maybe do another batch of 10. I will post here when I receive the boards from itead studio.

Cablebasher
Reality Checkpoint
grilojoe
M-Circus
steffensen
fracinfrucer
jaidee
jbdiver
mikecameron
Synesthesia
Cablebasher
Payment sent!
sammy123
I'd be interested a board.
yan6
count me in for one please
Samuron
I'd like one.
raisinbag
Hey the little triangle on the fonik designed pcb of Ians circuit, is the negative power lead right? Thanks.
Maco
Want one twisted
ctrl+z
One for me, please.
EATyourGUITAR
shipped
Cablebasher
Reality Checkpoint
grilojoe
M-Circus
steffensen
fracinfrucer
jaidee
mikecameron
Synesthesia
jbdiver

waiting list for next batch
sammy123
yan6
Samuron
Maco
ctrl+z
bugfight
trainspotter

please note that I did not design this PCB therefor I can not provide any tech support. all the information you need should be in the thread. if you still have any questions you can ask them here and hopefully someone can help you.
bugfight
+1 please
trainspotter
one for me please
InfraXpert
Put me down for one, please
M-Circus
EATyourGUITAR wrote:
shipped
Cablebasher
Reality Checkpoint
grilojoe
M-Circus
steffensen
fracinfrucer
jaidee
mikecameron
Synesthesia
jbdiver


w00t

That was quick!

Guess I'll have to figure out how to populate this thing then..

Edit:Not the biggest assembly job I see. hihi
Reality Checkpoint
M-Circus wrote:
EATyourGUITAR wrote:
shipped
Cablebasher
Reality Checkpoint
grilojoe
M-Circus
steffensen
fracinfrucer
jaidee
mikecameron
Synesthesia
jbdiver


w00t

That was quick!

Guess I'll have to figure out how to populate this thing then..



Me too!

I missed this shipped notice.

Thank you EATyourGUITAR for doing this. thumbs up
grilojoe
I too missed the shipped notification, but mine arrived today safe & sound. Now, to learn how to use it. Mr. Green
fracinfrucer
grilojoe wrote:
mine arrived today safe & sound. Now, to learn how to use it. Mr. Green


Me too, on both accounts! hihi Thanks EATyourGUITAR!! thumbs up
flts
FWIW, my batch was shipped around two weeks ago - I'm sure they'll get to Finland soon and I'll then send forward to you Finns who asked for one or more. May have some spare ones left for EU wigglers, I'll get back to this once the boards arrive and I verify that they look OK. lol
fracinfrucer
fonik wrote:
okay, while the MFOS circuit is called 'simple', Ian Fritz' circuit really IS simple. and there is no need for a precise voltage supply or matched resistors.
just a set of three resistors, a diode, and a DPDT on-on switch for each NPN and PNP transistors:


Ian Fritz' Transistor Matcher by fonitronik, on Flickr


Anybody know what part Fonik is using for the power connector? It looks kind of specific but I haven't had any luck figuring out what it is. Actually those pins he's using for the DVM posts look better than what I had found too... hihi

By the way, I went with a Fluke 8050A, since I didn't have a 4.5 digit multimeter. $40 shipped on ebay and works perfectly. They seem to be had for dirt cheap pretty regularly.
sammy123
Looks like right angle pin headers. Here are an unshrouded version.

http://www.taydaelectronics.com/connectors-sockets/pin-headers/2x40-pi n-2-54-mm-right-angle-double-row-pin-header.html
Jop
Power connector (sorry no us source):

http://www.reichelt.nl/WSL-10W/3/index.html?&ACTION=3&LA=446&ARTICLE=2 2818&artnr=WSL+10W&SEARCH=WSL+10W

DMM pins:

http://www.reichelt.nl/index.html?ACTION=3;ARTICLE=15321;SEARCH=RTM%20 1-100

Hope this is of any help.
flts
How about the DPDT switches, are they of some standard footprint or will only specific model(s) fit the PCB? I remember someone would have mentioned the exact ones used before in the thread but search fails me...
fracinfrucer
Thanks Sammy123 and Jop! thumbs up I can't seem to find the same one from a US distributor either, so I'll just use the substitutions I was planning on. Unshrouded for the power connectors and the typical pin headers for the DVM posts. It won't look as pretty but should work. hihi


flts wrote:
How about the DPDT switches, are they of some standard footprint or will only specific model(s) fit the PCB? I remember someone would have mentioned the exact ones used before in the thread but search fails me...


I just got the calipers out and these look like the right size if you want to match up the datasheet with a European distributor:

http://www.mouser.com/ProductDetail/Mountain-Switch/108-1MD1T6B1M2QE-E VX/?qs=sGAEpiMZZMvudeGI7i40XCnz47JvH4ug0EsixlwcUaE%3d
Jop
I've used this one:
http://www.reichelt.nl/Switches-Rocker/MS-168/3/index.html?&ACTION=3&L A=2&ARTICLE=13142&GROUPID=3275&artnr=MS+168

But please notice these are with solder lugs instead of PCB connectors, they original did not fit and I used pliers to make the lugs smaller.

Better to order the correct one right away :-)

http://www.reichelt.nl/Switches-Rocker/AS-500FPC/3/index.html?&ACTION= 3&LA=2&ARTICLE=4401&GROUPID=3275&artnr=AS+500FPC
fracinfrucer
Has anyone tried female solder pins for the DVM connectors? It seems like since alligator clips are not a solid enough connection for measurement, that installing female solder pins and sticking the tip of a probe in there, might be a better option.
GrantB
fracinfrucer wrote:
Has anyone tried female solder pins for the DVM connectors? It seems like since alligator clips are not a solid enough connection for measurement, that installing female solder pins and sticking the tip of a probe in there, might be a better option.


I use bananas It's peanut butter jelly time!

it's been said that's what bananas were originally intended for seriously, i just don't get it
sduck
So, I've built up a version of Ian's circuit on breadboard. Seems to work well.

My question is, how close is close enough? And, does my fluke 115 even have enough resolution to tell? I've been measuring 2N3904's for the last hour, all from the same tape, and as far as I can tell they're all really close. I'm always always getting readings like this after letting the things stabilize for a minute or two -



I then pull the trannies and switch the emitters, to measure the difference as per Ian's instructions. For this pair it read -0.001 after switching them and letting it settle.

These would be for a TTSH - are these going to be close enough, and I should stop worrying?
gddfp
sduck wrote:
So, I've built up a version of Ian's circuit on breadboard. Seems to work well.

My question is, how close is close enough? And, does my fluke 115 even have enough resolution to tell? I've been measuring 2N3904's for the last hour, all from the same tape, and as far as I can tell they're all really close. I'm always always getting readings like this after letting the things stabilize for a minute or two -



I then pull the trannies and switch the emitters, to measure the difference as per Ian's instructions. For this pair it read -0.001 after switching them and letting it settle.

These would be for a TTSH - are these going to be close enough, and I should stop worrying?

Set your Fluke to measure mV -- one dial twist to the right. hihi
And yes, if you're measuring trannies from the same batch, they're bound to be very close to each other... certainly close enough for a TTSH.
Still, there's always the rotten apple in the basket (I had one, from the same batch, which was no less then 10-ish mV off!), so it's best to find matching pairs anyway, just to be sure.
sduck
Doh! d'oh! Of course. Thanks.
fonik
sduck wrote:
And, does my fluke 115 even have enough resolution to tell?

IF i read the specs correct your flukes 'resolution' is 1mV max (last digit), with an error of 2 (again in the last digit).
so even though the last digit of your fluke shows i.e. 1mV, it could be 2mV or even 3mV. in other words, if you compare two transistors and both show 2mV there might be a difference of 4mV actually.
sduck
Actually, what I was reading once I figured out how to use the right setting was readings like 1.1 mV or -0.6 mV. So I suspect it's close enough - I was able to find enough pairs that were 0.5 mV apart or less.
flts
Okay, the boards finally arrived from TinyOS, haven't built one and tested yet but they look fine to me.

Edit: all spoken for, going to

1. InfraXpert
2. latigid on
3. STØJ
4. ctrl+z
5. zeta

Shipped on 14th of February 2014.
InfraXpert
I would love one, please
latigid on
I'm in (to CH) if that's okay
flts
Yeah, actually let's say Europe instead of EU since the letter prices are same, so CH is fine.
STØJ
I would also like to buy one pcb if possible? PMed!
flts
STØJ, just send me your postal address details via PM and you'll get one.
flts
Okay, one more left.
latigid on
What is the cost btw? Did you use the same files as previous fabbed already?
flts
all 5 spoken for. i'll pack and send them (along with the ones that wigglers from turku, finland wanted) this week.

the cost to you is nothing since i'll ship via cheap letter post and the cost to me was like 1 euro per board. i now have one of me and two for my friends who needed one which is all i wanted, and i don't want to make paypal any richer for the sake of getting <5 euros per board including shipping.

the boards are made using gerbers generated from fonik's eagle file earlier in the thread. i ordered them via tinyos but probably ends up to the same cheap fab houses that everyone is using for prototyping etc. as far as i see the boards look fine, but i obviously give no guarantee whatsoever lol
latigid on
Many thanks kind sir! thumbs up
flts
The letters are in the mail now. Expect them in a week or two I suppose, unless there are weird delays with weather or sun spots.
STØJ
flts wrote:
The letters are in the mail now. Expect them in a week or two I suppose, unless there are weird delays with weather or sun spots.


Thank you!
baloo
I would like one pcb if still available seriously, i just don't get it ???
flts
Mine are all gone now, better ask EATyourGUITAR if he's going to do more batches I suppose.

FWIW, if you don't want to build the Fritz circuit on a proto board or wait whether someone is going to order more, the Eagle files for Fonik's PCB are earlier in this thread. It took me around 30 mins to get correct Gerber files out of Eagle without having ever done that before (had to read some documentation to figure out how they wanted the files). And a total of 14 or 15 euros - including shipping from HK / China to EU - for a set of 10 PCBs.
STØJ
STØJ wrote:
flts wrote:
The letters are in the mail now. Expect them in a week or two I suppose, unless there are weird delays with weather or sun spots.


Thank you!


Received pcb for 2 days ago. Thank you fits.
ctrl+z
@flts

I got the pcb yesterday.
Thanks a lot!
latigid on
Got mine here too, much appreciated!
Samuron
EATyourGUITAR wrote:

waiting list for next batch
sammy123
yan6
Samuron
Maco
ctrl+z
bugfight
trainspotter


Since there don't seem to be any new additions, please remove me from the list.
EATyourGUITAR
batch #2 ordering today
sammy123
yan6
Maco
bugfight
trainspotter
baloo

Samuron <-- not sure if you are in or out. PM
baloo
nanners
Samuron
EATyourGUITAR wrote:
batch #2 ordering today
sammy123
yan6
Maco
bugfight
trainspotter
baloo

Samuron <-- not sure if you are in or out. PM


Ah; I had given up. Yes, I'll take one, thanks! smile

PM sent
Clemdu
I'd be interested too, is it too late for this current batch ? Thanks ! thumbs up
Heavy Metal Kid
If it's not too late, I'd like to order one as well.
EATyourGUITAR
I PM'd everyone on the list

batch #2
Samuron
sammy123
yan6
Maco
bugfight
trainspotter
baloo
Clemdu
Heavy Metal Kid
kvitekp
Count me in if it's not too late, please.
jbdiver
General question: what's an acceptable tolerance for matching transistors? I know this may depend on the application. I've been trying to match within 0.002v for the TTSH project. Is this reasonable?
Morbid
kvitekp wrote:
Count me in if it's not too late, please.

Me too please oops
cleaninglady
What diode is used for Ian's circuit ? Is it a 4148 or some sort of .6 volt ?
xpmtl
4148
M-Circus
@EATyourGUITAR

Do you think you could possibly be so kind and post the BOM for this PCB? I'm aware there aren't that many parts, but I'm a bit useless at reading schematics... help
tekzon
hey!
if not too late?
i'd be in for 2 pcb's,
Thanks
xpmtl
2 x 3K9
4 x 100K 0.1% or matched 1%
2 x 1N4148
2 x DPDT ON/ON switch
M-Circus
That is indeed simple. ;-) And when taking the time to inspect the PCB properly, I discover it's actually labeled with what goes where... d'oh!

Looking back through this thread, I gather you'll also need some sort of power supply and a few soldering nails, whatever those might be...

http://www.reichelt.nl/WSL-10W/3/index.html?&ACTION=3&LA=446&ARTICLE=2 2818&artnr=WSL+10W&SEARCH=WSL+10W

and

http://www.reichelt.nl/index.html?ACTION=3;ARTICLE=15321;SEARCH=RTM%20 1-100

Which raises a couple of more questions here:

How exactly do you hook that kind of power supply up to the power source? This thing's still supposed to run on a 9V bolt battery, right?

And what's the purpose of the 4 pins that go in the holes marked DVM- and DVM+? Are you supposed to connect anything to them?
latigid on
The PSU is eurorack footprint and DVM is your digital volt meter or multimeter... sorry but this is pretty basic info!
M-Circus
Yeah, I did see that it looked like a euro connector but I hadn't seen any information in this thread that these were now made to run on power from a euro system.

Gotcha on the digital volt meter abbreviation. That's where you measure then.

As for your "this is pretty basic info" comment: I believe I pointed out in my first post today that this field isn't one where I excel at theoretical knowledge. I'm however usually not too scared taking on new tasks, I'm pretty good at reading instructions and all the stuff I've built so far has worked the way it's supposed to. I'm however pretty dependent on being allowed to ask silly questions now and then to those more knowledgeable - like the friendly people on here. Can't really see why you need to beat me over the head with it.

But I got it now. I've figured out how to build and use this thing.
fonik
xpmtl wrote:
4 x 100K 0.1% or matched 1%

when you use the switches, the 100k don't need to be matched. in fact it is the switches purpose to allow measuring the error.
latigid on
M-Circus wrote:
Yeah, I did see that it looked like a euro connector but I hadn't seen any information in this thread that these were now made to run on power from a euro system.

Gotcha on the digital volt meter abbreviation. That's where you measure then.

As for your "this is pretty basic info" comment: I believe I pointed out in my first post today that this field isn't one where I excel at theoretical knowledge. I'm however usually not too scared taking on new tasks, I'm pretty good at reading instructions and all the stuff I've built so far has worked the way it's supposed to. I'm however pretty dependent on being allowed to ask silly questions now and then to those more knowledgeable - like the friendly people on here. Can't really see why you need to beat me over the head with it.

But I got it now. I've figured out how to build and use this thing.


I'm glad you've got it, best of luck with the matching thumbs up
xpmtl
Thanks Matthias.

I've just read Ian's paper again and he calls for matched resistors. In fact that makes sense as you can use the other method to find good pairs and in that case it's better to have them matched.

From what i understand the switch allows to find the mismatch between the resistors. Once you know it you don't have to use the switch anymore theoricaly.

All you do then is dVbe = Vdmv(a) - g

if g = 0 then your calculations are even simpler

Trying to understand all this so don't hesitate to correct me if i'm wrong Mr. Green
frijitz
fonik wrote:
xpmtl wrote:
4 x 100K 0.1% or matched 1%

when you use the switches, the 100k don't need to be matched. in fact it is the switches purpose to allow measuring the error.

True, but it's still best to used resistors matched as well as you can get. Otherwise, your result is a small difference between two large numbers.

Ian
flts
I matched a bunch of 2N3904 from the same batch using Fonik/Fritz PCB yesterday. The biggest differences were something like 3mV (rare, due to one or two outliers) and the closest pairs were around 0.05 mV (50 microvolts) apart which is probably as close as I can hope to measure somehow reliably with my meter. All in all, it seems that modern transistors from the same batch will be close to each other in this sense.

Which kind of brings me to a question that has only briefly been touched in this thread - what kind of matches do you expect / want to get for which particular use to get "adequate" performance?

Now, I understand this is a bit stupid question since it's just one particular parameter that is being matched here, and different types of circuits will have different requirements as well. However, quite often (such as now in TTSH build docs) you tend to see something like "these pairs should be matched" without any additional specification - what do you consider a good enough match by default in that case?

The MFOS site seems to hint that less than 2mV difference would be considered a good pair as far as Vbe goes - would that be a good rule of thumb unless a stricter criteria is explicitly specified or unless one has enough knowledge to analyze how mismatch will affect this particular circuit?
Heavy Metal Kid
Yeah, I matched NPN and PNP transistors for the TTSH project last week, the ones that were delivered by Mouser. All of them were well within the 50µV that Fritz recommends in his paper. I couldn't really find anyone that had a greater difference than 10-20µV or so.
flts
Heavy Metal Kid wrote:
Yeah, I matched NPN and PNP transistors for the TTSH project last week, the ones that were delivered by Mouser. All of them were well within the 50µV that Fritz recommends in his paper. I couldn't really find anyone that had a greater difference than 10-20µV or so.


Looks like your batch was even tighter tolerance than mine then (ordered that bunch from Mouser as well), or I just didn't measure correctly. That 50µV figure Ian gives is for very specific use (making a high performance discrete OTA) and I'm not sure how that applies to different types of synth circuits in general.
LED-man
matching with a microcontrollerbased tester (quartz and ref.voltage source, not chinese transistortester clones without quartz)

i have matched in last year arround 500transistors,
this are from 2 different vendors and comes on roles or in bags.
(in bags are better)

normally i got arround 5 values with a range of 2-5mV for 100 transistors
i match on 1mV.
mangros
flts wrote:
I matched a bunch of 2N3904 from the same batch using Fonik/Fritz PCB yesterday. The biggest differences were something like 3mV (rare, due to one or two outliers) and the closest pairs were around 0.05 mV (50 microvolts) apart which is probably as close as I can hope to measure somehow reliably with my meter. All in all, it seems that modern transistors from the same batch will be close to each other in this sense.


That's about the same experience I'm having, from a bag of 2N3904s from Tayda. Many pairings about 1.3mV, but I've got three pairs with 10-50µV and six pairs around 100µV. I keep meaning to test some THAT300s just to double check my method.

Done with a fan blowing over the test area (including the loose transistors ready for testing), and when I got a decent match I'd press the transistor bodies together with some pliers to see if it improved (usually it did).
xpmtl
what multimeter do you guys use to measure uVolts?
Heavy Metal Kid
I used an Elvis II board at the University lab.
xpmtl
Expensive piece of kit!
Altitude909
xpmtl wrote:
what multimeter do you guys use to measure uVolts?


I grabbed a HP 3478A for $150 on ebay
mangros
UNI-T UT61E.
xpmtl
Altitude909 wrote:
xpmtl wrote:
what multimeter do you guys use to measure uVolts?


I grabbed a HP 3478A for $150 on ebay


That's what i have too. good deals on the bay for that model paid mine 120$ shipped and it's still comparing fine with my 34401A. Thought it would be completely out of cal but not.
flts
I've been meaning to buy a HP 3478A as well since those seem like really good deals (except that the screen is crappy and when the battery goes out you lose calibration and it's a bitch to change on the fly).

As it stands, I have a Brymen BM869 portable multimeter which roughly compares to the higher end Fluke handheld models or cheap end bench devices as far as short term specs is considered at least.

As I interpret the specs (0.02% + 2 digits at the mV range, no separate microvolt measurement) and reviews, it should do mV well, reasonably up to somewhere around 100 µV range, still give a decent (if not super accurate) indication on the 10 µV range and nowhere near enough accuracy for proper µV measurement.
oldenjon
I have my 3478A up in the for sale section. Just sayin
EATyourGUITAR
batch #2

SHIPPING

Samuron
bugfight
Clemdu
yan6
Heavy Metal Kid
kvitekp

PENDING
sammy123
Maco
trainspotter
baloo

check your PM's if you have not paid. I can ship these when you are ready.

tekzon wrote:
hey!
if not too late?
i'd be in for 2 pcb's,
Thanks


if anyone on this list cancels their order I can sell it to you. if not I will add you to the list for batch #3.
tekzon
if anyone on this list cancels their order I can sell it to you. if not I will add you to the list for batch #3.[/quote]

Ok,
Any idea on when this third batch could be?
Thanks!
EATyourGUITAR
I will wait a few more days to see what happened and then I will make a decision. if you are the only person on the list, I can just give you the file so you can order it from china.
flts
I still have my Gerber files generated from the Eagle files as well. As I said, the batch I did at TinyOS / TinySine was like total $15 per 10 PCBs including shipping, and they looked & worked fine (at least the one I built for me did), and it took me like half an hour of learning as a complete Eagle noob to get those router files produced.

I'm not sure what the etiquette is with these things since it's neither my own design nor my own PCB layout, but the Eagle files ARE freely available here and that's where the Gerbers were trivially generated from. So would it be considered a faux pas or a nasty thing for some reason to just attach them to this thread for others to use? They should be ready for TinySine/TinyOS and probably Seeedstudio & some other Chinese suppliers as well.
fonik
i asked ian fritz before uploading my files.
cleahy
EATyourGUITAR wrote:
I will wait a few more days to see what happened and then I will make a decision. if you are the only person on the list, I can just give you the file so you can order it from china.


I'd definitely be interested in a board too please! applause
EATyourGUITAR
batch #2

SHIPPING

Samuron
bugfight
Clemdu
yan6
Heavy Metal Kid
kvitekp
Maco
trainspotter

PENDING
sammy123
baloo

sammy123 awaiting your shipping info

baloo awaiting payment

batch #3
tekzon
tekzon
frozenkore
cleahy
tabularasa
Is there still a place free in batch3?
I also have to match some transistors for a TTSH on the board looks sexy.
EATyourGUITAR
batch #2

SHIPPED

Samuron
bugfight
Clemdu
yan6
Heavy Metal Kid
kvitekp
Maco
trainspotter
sammy123

PENDING
baloo

baloo awaiting payment

batch #3

PAID

tekzon
tekzon
frozenkore
cleahy
oldenjon

PENDING
tabularasa
kvitekp
Thank you for doing this, @EATyourGUITAR
mrand
Hi there, I'm looking for some expertise.

Using I.Fritz's transistor matching method, I've matched some 3904s, as best as I can with my low res DMM and 1% resistors. To my surprise, the difference of hfe within any given "matched pairs" is much greater than within unmatched pairs. I guess I thought Vbe difference would be correlated with hfe difference, but am measuring otherwise. Can anyone speak to this? Hope I'm making sense...
Samuron
Got my board today; thanks!
mad_magician
Can I get some advice? So I built ijfritz's NPN matcher on a breadboard and I am trying to work my way through ~150 2N3904s.

Picked up an HP 3478A on Ebay a few weeks ago and it seems to be pretty solid. However, I never see the number stay stable or lock down. If I measure the incoming voltage, the rails are dead on and stay fairly stable (.01V up/back over 5-10 seconds)

The 100K resistors measure:
99.505K Ohm
99.516K Ohm

Breadboard is in a metal box totally covered, transistors are touching on the flat sides.

I am using the "golden" and TUT method However, I see the measurement drift between values constantly and even after an hour or more it does not settle.

Example of what is on my bench right now:
+02.983mV <-> +03.051mV

So I am at a loss of what to do here. Is this "close enough for government work"? Should I try putting this on a pad per hole board I have handy to see if I get better results?
very frustrating

Thanks

MM
frijitz
mad_magician wrote:
Picked up an HP 3478A on Ebay a few weeks ago and it seems to be pretty solid. However, I never see the number stay stable or lock down. If I measure the incoming voltage, the rails are dead on and stay fairly stable (.01V up/back over 5-10 seconds)

That sounds like a lot of drift for a decent PS. On my diy bench supply with a cheap DVM I measure 12.022 +/- .001 V over time. IOW, only the last digit is changing, which isn't too surprising. Your variation in PS voltage shouldn't affect the matching measurement, but I'm a bit concerned about your metrer's stability. You might want to try measuring a fresh 9 V battery to see if you get the same kind of variation.

Ian
mad_magician
frijitz, thanks for the response.

I tore everything down and looked at my power supply and it looks much less stable than I originally thought. Looks like I will need to find a better one pronto! (It is a 30+ year old "educational" supply) so no surprises there.
Clemdu
Board received, thanks a lot ! thumbs up
Heavy Metal Kid
Got my board the other day!
M-Circus
Got one of these boards from eatyourguitar a while back, and have it assembled here. Considering I've

a) only got a €15 multimeter to measure stuff with

and

b) I've bought transistors from the same batch (from Mouser, on tape),

...I am however wondering how much sense there there will be in attempting to do matches. I'm leaning against just using them, without any further fuzz. Anyone have any thoughts on this?

For the TTSH btw.

Edit: When I think of it, there was a whole discussion on this topic in another thread only days ago, so just ignore me. I'm apparently approaching dementia in my late thirties...
EATyourGUITAR
batch #3

SHIPPED

tekzon
tekzon
frozenkore
cleahy
oldenjon
Tabularasa
emdot_ambient
chrisp
tmeade


I have 2 PCB's available. I will not be ordering more.

price is
$6 USA
$10 international
PM me to buy. I will give you my paypal address.
EATyourGUITAR
I sold them all. I am done selling transistor matching PCB's. if anyone wants the .zip file with the gerbers, you can PM me.
Heavy Metal Kid
Thanks!
StillNotWorking
Looking at the simplified version of Frietz tester and the recommended work procedure where we make notes of the difference on all measured transistors I wonder can I simply use the diode tester on my DMM for this? Doesn't the diode test add voltage and load?
latigid on
The diode tester gives forward voltage drop only as far as I know.
wavydave
EATyourGUITAR wrote:
I sold them all. I am done selling transistor matching PCB's. if anyone wants the .zip file with the gerbers, you can PM me.


Thanks for the 2 PCBs.
indigoid
EATyourGUITAR wrote:
I sold them all. I am done selling transistor matching PCB's. if anyone wants the .zip file with the gerbers, you can PM me.


I, too, am interested in Gerbers. Thanks. Will PM. (but can't PM until my post count is greater than 0)
EATyourGUITAR
I would like the people that were responsible for creating this project to know that I have sent indigoid a zip file containing the gerbers I rendered in eagle.
fonik
yep, i know. it became a topic on the synth DIY mailing list. 8_)
alexg
I would be in for another batch !
EATyourGUITAR
fonik wrote:
yep, i know. it became a topic on the synth DIY mailing list. 8_)


I did not see that. was there a problem?
elmegil
No, folks were asking about matching transistors and I pointed them at this thread. The guy who got the gerbers went back on list and told others that he had them.

I didn't see any problem....
indigoid
Yep, that was me posting to synth-diy about this.

No PCBs left now. The last spare PCB I had was claimed by a Wiggler today and will be in the mail very soon.
Siri
Hi all,

I just built Dave's first drawing (thanks Dave!) (https://www.muffwiggler.com/forum/viewtopic.php?p=1136513#1136513). I've read all the documents, couple of threads, and i'm succesfully lost. help

I bought matched resistors (so i took out the trimmer).
I put power, then two transistors in, wait it stabilize, it does, then what?
Some say if you have matched resistors you don't need to use the switch. If i don't use the switch then what?

I'm just lost about what to do next. I've read Ian's paper, but i don't get it. And English is not my first language.

Any help would be so much appreciated.

Many thanks

Greg

EDIT : i'm trying to be more clear.

In Ian's pdf, in :

Quote:
Bottom Line
OK, OK, so what I actually do?


[...]

(4) Repeat (3) with the emitter leads swapped.
(5) Calculate half the difference of the two readings. That's your mismatch.


And, more down,

Quote:
Multiple devices

[...]

(2) One at a time, plug the other devices into the Q2 position and measure the mismatch
between them and Q1 using the procedure described above. (You do not need to swap
the transistors with this method.)


How can you measure the mismatch without swaping?

help help help
ubiquiphilia
ah nice one. very useful, thanks.
qfactor
Quote:
Multiple devices

[...]

(2) One at a time, plug the other devices into the Q2 position and measure the mismatch
between them and Q1 using the procedure described above. (You do not need to swap
the transistors with this method.)


How can you measure the mismatch without swaping?

help help help[/quote]

I understand this part as the following:

Take 3 transistors, Q1, Q2 and Q3.
Plug Q1 and measure the mismatch with Q2. Let's say, mismatch of Q2 (from Q1) comes to 1.1mV.
Now remove Q2, plug in Q3 and measure mismatch between Q1 and Q3. Suppose the mismatch is also 1.1mV, this would mean that Q2 AND Q3 are matched!! nanners nanners
seriously, i just don't get it
Siri
Quote:
Plug Q1 and measure the mismatch with Q2. Let's say, mismatch of Q2 (from Q1) comes to 1.1mV.


This is this part i don't get.
1.1mV is what you read straight on the DDM?

Because more up in the pdf he says the mismatch is half of the difference of two readings, but he say to not swap (or switch) so how can you get two reading? seriously, i just don't get it
daverj
This:

Siri wrote:

EDIT : i'm trying to be more clear.

In Ian's pdf, in :

Quote:
Bottom Line
OK, OK, so what I actually do?


[...]

(4) Repeat (3) with the emitter leads swapped.
(5) Calculate half the difference of the two readings. That's your mismatch.


and this:

Siri wrote:
And, more down,

Quote:
Multiple devices

[...]

(2) One at a time, plug the other devices into the Q2 position and measure the mismatch
between them and Q1 using the procedure described above. (You do not need to swap
the transistors with this method.)


How can you measure the mismatch without swaping?

help help help


Are two different methods with two different purposes.

The first one is how you measure the difference between two specific transistors.

The second one is how you can sort out a bag full of transistors into groups that match each other. Or to find two transitors from the bag that match.

First lets see if your "matched resistors" truly are matched. Plug in two transistors into the sockets and connect the voltmeter. Power it up and wait a little while for the heat from your fingers to dissipate and the voltage stabilizes. Now write down the voltage.

Now leave the voltmeter alone (don't change it's connection) and pull out the two transistors and swap which socket they go into. Power the circuit and again wait for the temperature and voltages to stabilize. Compare that voltage to the one you wrote down. If the resistors are truly matched then the two voltages will be exactly the same except for the plus or minus sign in front of them. (one will be a positive voltage and one a negative voltage)

If the voltages are exactly the same in both cases then the resistors are close enough to not create an error. If there is a slight difference between them then the resistors are not exactly the same.

If the voltages are exactly the same then from now on you can trust the reading on the voltmeter without having to swap the transistors and do any math. If not, then you still have to do the math Ian talked about to get the true measurement.

Method one that you quoted above is used to find the difference between two specific transistors. If you only have two transistors, or are measuring a dual transistor package, then this method shows you the difference between those transistors.

On the other hand if you have a bag of transistors and want to find two out of the bag that match, to then use together in a circuit, you would use method two that you quoted above.

With method two you are using one transistor as a reference. You will not actually be using that transistor in the circuit you are building. You will use it to find two other transistors to use.

You start by putting a transistor in one socket, and that one stays there for the rest of the test. You then put a second transistor in the other socket and measure it's voltage difference relative to the first. Write that number down.

Now take out that second transistor and put it somewhere so you know it is the one you are trying to match. Now dig through your bag of transistors and stick them one at a time into that second socket and look at the voltages. When you find one that has exactly the same voltage as the one you set aside, you now have two transistors that match. Put that new one with the one you set aside and that is your pair to use in the circuit. You can now take out the first transistor from the first socket and throw it back in your bag of transistors. It was just a reference that you used to find the other two. You don't actually use that first one in your circuit.

This method works because if the difference between the two selected transistors and the reference transistor are exactly the same, then there is no difference between those two. Be sure to pay attention to the plus/minus sign when doing those measurements. The sign must match as well as the voltage number.
Siri


Many thanks Dave, you made it really clear.
qfactor
Dave da man!! applause thumbs up It's motherfucking bacon yo
Poldenstein
No more excuses, I have to face transistor matching or I won't finish my TH 555-VCO..
So I was starting to look at Ian And Dave schematics and asking myself:
Seen that I have a very crappy DMM, couldn't I amplify the voltage to be measured with something like a TL072 (and secondarily, how?).
Please excuse my noobness.
weedywhizz
Any PCBs left or anyone knows a source ?
Synthsense
weedywhizz wrote:
Any PCBs left or anyone knows a source ?


Yes, I'd be happy getting one of those too
airfrankenstein
So with all these transistor matchers built does anyone have 2 pairs of matched 2N3906's to sell? Or if I buy them off the same roll will the values probably be close enough?
Siri
I went back today to use the board I made following daverj first drawing, and the numbers on the multimeter would never settle. They would not move fast, but go up and down every 15sc or so, with a difference (between the up and the down) of about 0.4mV.

Then i hooked it up on a benchtop power supply, and now they don't go up and down anymore, they settle after a minute.

I know my circuit is working because when i touch, or blow on the transistors the numbers start increasing really fast, and it looks like it behaving as it should.

So is it possible that the battery setup is not that reliable? the battery was new.
cleaninglady
fonik wrote:
okay, while the MFOS circuit is called 'simple', Ian Fritz' circuit really IS simple. and there is no need for a precise voltage supply or matched resistors.
just a set of three resistors, a diode, and a DPDT on-on switch for each NPN and PNP transistors:


Ian Fritz' Transistor Matcher by fonitronik, on Flickr


It would be great if Ian would let this one go up on Oshpark , he'd sell and truckload I'm sure !
fuzzbass
I wanted a little rig for this so would not have to set up and tear down the circuit on my breadboard. But I do not like sockets for this, because shoving the little pins into them means handling the transistors. That slows everything down - you have to wait for them to settle back down to room temp. Just a little bit of heat from your fingers will move the DMM significantly.

I took Ray Wilson's suggestion and got a zif socket and built the Ian Fritz unit. I just open the socket, tweeze transistors in, and close it.

It has a switch to flip the readings. I used .1% resistors and value of "G" (error introduce by the resistors) is tiny: 150uv. With this error, I know that any two transistors that within 1mv of each other are a match.

This rig can match NPN or PNP. To change, you flip the socketed diode and reverse the supply power. It accepts PDIP packages too.

Its ugly, I know!

xnor
I added a MOTM power adapter to Fonik's PCB and have posted it to OSH Park

https://oshpark.com/shared_projects/XHWPvGBc

It ended up being 13.50USD for 3 shipped in the US.

I haven't tested this yet as I just submitted the PCB for fabrication.
Crashlander42
Would something like this be acceptable for transistor matching?

http://www.mcmelectronics.com/product/TENMA-72-1012-/72-1012

i currently have -

http://www.amazon.com/gp/product/B00T1AXDUE?psc=1&redirect=true&ref_=o h_aui_detailpage_o04_s00

but I don't exactly trust it... It never seems to give the same reading twice...
Altitude909
^
It should be OK. It doesnt have to be accurate, just repeatable since you are comparing parts not taking absolute readings.

Personally I grabbed an old 5.5 digit HP meter on ebay. There are plenty to be found at a decent price and a 20 year old premium meter will always beat out a brand new low end budget one
Crashlander42
The one on MCM is actually on sale for $79 right now. Which seems like a fairly good deal. Will search around on eBay a bit.

Just picked up a TTSH, so I've got some transistor matching ahead of me...
Crashlander42
So, stupid question time. I apologize if this has been asked before. I built the Ian Fritz circuit, and am trying to match some 3904's. I set the DVM to 200m, and I get 00.9x, and occasionally 01.xx. Usually the second digit after the decimal (00.0x) or the x in my example wanders up and down and does not stay steady, even after a good minute or two.

I assume I'm doing something stupid.
qfactor
Crashlander42 wrote:
So, stupid question time. I apologize if this has been asked before. I built the Ian Fritz circuit, and am trying to match some 3904's. I set the DVM to 200m, and I get 00.9x, and occasionally 01.xx. Usually the second digit after the decimal (00.0x) or the x in my example wanders up and down and does not stay steady, even after a good minute or two.

I assume I'm doing something stupid.


What I do is to leave it to stabilize, and this can take even up to 5 minutes or longer! I would leave this and go do other things (like solder some resistors/caps) in the meanwhile and come back to check if the readings have stabilized.

Though even after 5 minutes or so, sometimes, the readings would still be fluctuating, albeit, very slowly (like up or down after 20-30 seconds).
If that's the case, then I'll just use whatever it is showing within that time frame, and go along with that reading.
Crashlander42
Another question. What digit am I supposed to be matching to. 1mV? .01mV? How much is good enough, say, for the TTSH filter and VCA.
qfactor
Crashlander42 wrote:
Another question. What digit am I supposed to be matching to. 1mV? .01mV? How much is good enough, say, for the TTSH filter and VCA.


I think Altitude909 mentioned it earlier that you'd only need the readings to match.
For example, if out of 5 trannnies you match, 2 of them settles on a reading of 1mV while the other 3 have readings of .01mV, 02mV and 0.12mV, respectively, then the 1st two transistors would be the ones you'd use.

It doesn't matter what readings they end up with except that they are of the same, then they are considered matched! thumbs up
Crashlander42
I phrased that badly. I suppose I mean, what tolerance am I looking for? Should they be matches within 1m/v, .1mV, or .01mV.

If I understand right, with the Ian Fritz method I'm not really matching readings, because the reading it gives me is the difference between the two transistors.
qfactor
Crashlander42 wrote:
I phrased that badly. I suppose I mean, what tolerance am I looking for? Should they be matches within 1m/v, .1mV, or .01mV.

If I understand right, with the Ian Fritz method I'm not really matching readings, because the reading it gives me is the difference between the two transistors.


Yeah, that's the gist of Ian Fritz's matching idea!
As for the tolerance, I don't think you need to match that some more, as each transistor you are matching would have been made to be within a certain tolerance?
Otherwise a 2N3904 would not be a 2N3904, if it's tolerance is not in the range it was manufactured to be, which would mean that it would be a "dud" 2N3904, which would be a different matter from matching transistors! hihi
elmegil
In my experience matching with Ian's circuit you don't really need that many decimal points of precision.

That's what you mean by "tolerance", not like the tolerance of an electronic part.

I have hd the same experience though -- it takes a fair while for them to settle down. Some folks have reported putting a little "hat" over the two transistors can help them settle faster, as part of the issue is that you're measuring at such fine levels that heat from your hands, or stray air currents in the room you're working in can affect the values.
Crashlander42
elmegil wrote:
In my experience matching with Ian's circuit you don't really need that many decimal points of precision.

That's what you mean by "tolerance", not like the tolerance of an electronic part.


Thanks! "How precisely should I match them" is what I should have said.

Me not to good with words and stuff tonight.

I have a plastic container over it in hopes that it will help. I noticed the Moog matching method says to match within 2MV, so I guess I don't need to get too crazy with it. It seems like most my transistors are well within that even with the wandering digits.
elmegil
Crashlander42 wrote:
elmegil wrote:
In my experience matching with Ian's circuit you don't really need that many decimal points of precision.

That's what you mean by "tolerance", not like the tolerance of an electronic part.


Thanks! "How precisely should I match them" is what I should have said.

Me not to good with words and stuff tonight.


You knew what you meant, and I got it, but I figured it wouldn't hurt to make it clear for everyone :-D

And yeah, today's transistors, if you pull them off a string of the same batch, are highly likely to be closely matched already. For a few projects I haven't bothered if I have them on tape (I still do if they're a loose bag though).
spotta
xnor wrote:
I added a MOTM power adapter to Fonik's PCB and have posted it to OSH Park

https://oshpark.com/shared_projects/XHWPvGBc

It ended up being 13.50USD for 3 shipped in the US.

I haven't tested this yet as I just submitted the PCB for fabrication.

I just ordered three as I'm after one myself, 2 going spare (once they arrive) if anyone wants one.
Reality Checkpoint
spotta wrote:
xnor wrote:
I added a MOTM power adapter to Fonik's PCB and have posted it to OSH Park

https://oshpark.com/shared_projects/XHWPvGBc

It ended up being 13.50USD for 3 shipped in the US.

I haven't tested this yet as I just submitted the PCB for fabrication.

I just ordered three as I'm after one myself, 2 going spare (once they arrive) if anyone wants one.


PM incoming!
Grumskiz
spotta wrote:
xnor wrote:
I added a MOTM power adapter to Fonik's PCB and have posted it to OSH Park

https://oshpark.com/shared_projects/XHWPvGBc

It ended up being 13.50USD for 3 shipped in the US.

I haven't tested this yet as I just submitted the PCB for fabrication.

I just ordered three as I'm after one myself, 2 going spare (once they arrive) if anyone wants one.


I would also like one smile
spotta
Both spoken for. thumbs up
hamildad
doh.

anyone got a spare PCB they can sling in an envelope?

I'd be interested in making this ahead of getting my TTSH parts.
whoop_john
daverj wrote:
-- I did three drawings. The first one uses a pair of DPDT switches and is the most universal. It's the best one to build. The second drawing is for people wanting to build something extremely cheap, without the switches. But it's different for NPN and PNP, plus requires you to manually swap the meter probes during matching. The third one is a variation on the second, combining the two with a single battery but separate sockets and probe points.

-- The sockets are marked EBC but the B and C are connected together so no need to twist the leads around if you have transistors that are in an ECB package instead of EBC.

-- With 1% 100K resistors the trimpot could theoretically be 2K. But since trimpot tolerance is not as good, it should be slightly larger. I picked 5K since it's the next larger standard size. If you can get a 2.5K or 3K trimpot that would be even better. You adjust the trimpot by switching the meter probes back and forth, turning the trimpot until you get the same reading with just the +/- sign changing.

-- All three drawings show the same simplified circuit. Just without the switches in two of them.

-- The same simplified circuit could also be connected to a regular power supply too instead of a battery. It just needs one power rail plus ground for that.


I have come up with this layout for Dave & Ian's first circuit, as above. I am thinking of sending off for some PCBs, as I need one myself. Could someone check it over and comment please? I intend to use cheap and ubiquitous 16 pin ZIF sockets, but any 0.1 inch socket could be substituted. Size is 100mm x 26mm.

If there is a demand for these I could post PCBs out to people at modest cost. Speak up if you are interested and I'll do a batch to cover everyone's needs. Failing that I will get about ten PCBs made and sell on 9.

InfraXpert
hamildad wrote:
doh.

anyone got a spare PCB they can sling in an envelope?

I'd be interested in making this ahead of getting my TTSH parts.


I have one, PM me with your address.

:0)
PK808370
whoop_john, I seem to be going backwards in time as I do more DIY, so I will be needing such a device. I would like 1 of these PCBs. Thank you,
InfraXpert
Well, its not like I have a production going just a few extras of the ones mentioned below:
xnor wrote:
I added a MOTM power adapter to Fonik's PCB and have posted it to OSH Park

https://oshpark.com/shared_projects/XHWPvGBc

It ended up being 13.50USD for 3 shipped in the US.


Please note that the added MOTM power adapter is not the commonly used one but smaller version that I do not recognize.

Send me a PM with your address if you want one of my extras.
thresholdpeople
whoop_john wrote:

I have come up with this layout for Dave & Ian's first circuit, as above. I am thinking of sending off for some PCBs, as I need one myself. Could someone check it over and comment please? I intend to use cheap and ubiquitous 16 pin ZIF sockets, but any 0.1 inch socket could be substituted. Size is 100mm x 26mm.

If there is a demand for these I could post PCBs out to people at modest cost. Speak up if you are interested and I'll do a batch to cover everyone's needs. Failing that I will get about ten PCBs made and sell on 9.



Hey whoop_john did this pcb run ever happen? I'm super interested in a board. Did you get my PM?
Chaotic
Bought this little fellow. Quite cheap.
I have no experience with other testers and dont know about the accuracy
http://www.banggood.com/DIY-M12864-Graphics-Version-Transistor-Tester- Kit-LCR-ESR-PWM-With-Case-p-997023.html
whoop_john
thresholdpeople wrote:
whoop_john wrote:

I have come up with this layout for Dave & Ian's first circuit, as above. I am thinking of sending off for some PCBs, as I need one myself. Could someone check it over and comment please? I intend to use cheap and ubiquitous 16 pin ZIF sockets, but any 0.1 inch socket could be substituted. Size is 100mm x 26mm.

If there is a demand for these I could post PCBs out to people at modest cost. Speak up if you are interested and I'll do a batch to cover everyone's needs. Failing that I will get about ten PCBs made and sell on 9.



Hey whoop_john did this pcb run ever happen? I'm super interested in a board. Did you get my PM?


Yes the boards arrived in UK at 7.15am on Christmas Eve and I received your PM.

I am in the process of building a board up and testing it as we speak. I will have ten spare boards, of which, including yourself, two are already spoken for, assuming the PCB and shipping cost is agreeable (I am not out to rob anyone, so it should be a sensible cost).
thresholdpeople
Great. Thank you! It's peanut butter jelly time!
whoop_john


Built up module. I used a 1k preset rather than the suggested 5k, which was fine for balancing the resistance, as my two 1% 100k resistors were well within tolerance.

Banana plugs fit the large plated through holes on the pcb. There are other holes to suit pins or the tip of a probe. Croc clips can be used on the edges of the large hole.



A battery clip can be fitted instead of the Doepfer socket or Synthesisers.com style socket. In this case thread the wires through the holes and tie a knot to prevent the wires breaking off. Note the only power lines used on the synth connectors are 0v and 12v/15v. Don't connect the negative battery connector to -12/15v, connect it to 0v if you make your own connector.

No need to use a ZIF socket, any 0.1" socket will do, but ZIFs are so cheap I saw no reason not to, for ease of use.

Cost? Within UK I can send PCB including postage for £3.50.
Within the EU the cost is €5.5 including postage.
Rest of the world is $6 USD including postage.

I can only accept Paypal, which I will email you. So far I think 4 people are interested, in which case there are 6 left.
elmegil
Beautiful work, very well thought out options. I built my own in a radio shack box ages ago, so I'm not in need, but I would jump at it if I were.
thresholdpeople
Looks great!
hallvard
Thank you for this contribution! I made something on stripboard, but this looks a lot tidier smile.
whoop_john
The ZIF socket I used is:

ZIF SOCKET LINK



Which even for two sockets is cheap enough to throw at the cat and make the baby laugh.

The rest of the bits should be available from Tayda and all the usual suspects.
thresholdpeople
Wow that's a lot cheaper than the zif sockets on mouser!

I guess for the difference in cost it doesn't matter, but does it feel solid like you'll be able to open and close the pins many times over before the handle gives out?
whoop_john
thresholdpeople wrote:
Wow that's a lot cheaper than the zif sockets on mouser!

I guess for the difference in cost it doesn't matter, but does it feel solid like you'll be able to open and close the pins many times over before the handle gives out?


It seems fine to me, really. I have a few component testers with these sockets and they do what it says on the tin. I don't think it is going to fly off the handle any time soon. A couple of bucks inc. shipping and you get a spare thrown in. It is possible to unscrew the assembly and see what any problem might be.

Sure, they are knock-offs of old 3M designs, but who wants to pay the prices of the Minnesota Mining and Manufacturing Company?
spacedog
cool

I hope you still have one for me.

Guinness ftw!
whoop_john
So far the following have expressed an interest:

qfactor
thresholdpeople
hallvard
xahdrez
duff
wallybob
spacedog

…so I shall have three left, assuming all the above still want one and can pay Paypal. I will be messaging everyone individually in due course.

Within UK £3.50 including postage.
Within the EU €5.5 including postage.
Rest of the world is $6 USD including postage.
sonicwarrior
I'd like to have one, too. thumbs up
smcumber
Add me to the list, I'm in the USA
Quadram
also interested, i'm in France
whoop_john
That sounds like all ten gone then. I will get my act together and message you all. Tomorrow I hope, but I am going down with a cold.
sonicwarrior
whoop_john wrote:
Tomorrow I hope, but I am going down with a cold.

No hurry for me. I'm currently recovering from a cold myself. So speedy recovery for you.
whoop_john
OK, I have messaged the following ten people individually:

qfactor
thresholdpeople
hallvard
xahdrez
duff
wallybob
spacedog
sonicwarrior
smcumber
Quadram

If there proves to be further demand, I will get some more of these PCBs made.
thresholdpeople
It's peanut butter jelly time!

Payment sent! Thank you!

Feel better!
qfactor
Sent!
sonicwarrior
Sent, too. we're not worthy
whoop_john
sonicwarrior wrote:
Sent, too. we're not worthy


Many thanks. Sonic, please message me with your postal address.
whoop_john
Of the ten people, nine have paid, many thanks, amazing speed everyone. But I only have addresses for eight people.

Post Office closed today, so they won't go out today. Happy New Year to everyone,

Whoop John

(no it isn't me)
whoop_john
qfactor
thresholdpeople
hallvard
xahdrez
duff
spacedog
sonicwarrior
smcumber
Quadram

Your transistor tester boards are in the post to you today.

wallybob, all attempts to email/message you have had no response. If you still want a board, speak up now, or I will pass the last one I have on to the next person on the list.
PWM
whoop_john
Hi John, do you have a spare PCB left? I'd be very interested.
whoop_john
PWM wrote:
whoop_john
Hi John, do you have a spare PCB left? I'd be very interested.


I am more than happy to get some more or keep a stock of these for people. I would quite like to have comments back from the initial ten sent out, so I can make any suggested improvements if necessary.
qfactor
whoop_john wrote:
PWM wrote:
whoop_john
Hi John, do you have a spare PCB left? I'd be very interested.


I am more than happy to get some more or keep a stock of these for people. I would quite like to have comments back from the initial ten sent out, so I can make any suggested improvements if necessary.


Will do, John. Thanks!
thumbs up
PWM
Great news! No rush, I'll keep an eye on this thread.
sonicwarrior
Forgot to mention this: Got my PCB (on Tuesday AFAIR).
spacedog
Got my board today.
Thanks!

Guinness ftw!
drane2
emmaker wrote:
You might check to see if the Ray Wilson circuit has the same flaw that the Moog schematic has. That is there needs to be a 10K resistor between the collector and emitter of the transistors in the PNP tester like the NPN one has.


Can anyone help me to understand why the 10k is required, or why it's a flaw that it's missing in the Moog PNP tester schematic? I've been trying to wrap my head around it all day, even setting up a spice sim that steps the resistor value from 0 to 10k, which just results in a Vbe variation of about 1uV.[/img]
dirkwiggler
whoop_john wrote:
PWM wrote:
whoop_john
Hi John, do you have a spare PCB left? I'd be very interested.


I am more than happy to get some more or keep a stock of these for people. I would quite like to have comments back from the initial ten sent out, so I can make any suggested improvements if necessary.


Hey John,

Can you add me to the list for the next batch too please?
qfactor
Can someone please explain to me how do you tweak the trimmer to get only a change in + and -?
I don't quite get it! seriously, i just don't get it d'oh!
thresholdpeople
Hey John,

Got my PCB today!

Thanks!
lcf
I was about to ask for a pcb and i thought that I have this "dock" that came with my UNI-T UT61E multimeter.
Could I use it to match transistors please??? hmmm.....
duff
lcf wrote:
I was about to ask for a pcb and i thought that I have this "dock" that came with my UNI-T UT61E multimeter.
Could I use it to match transistors please??? hmmm.....


No, the 61E doesn't have a transistor test mode.
fredp
Hi John,
Could you please add me to the list too? I am located in France
Thanks
lcf
Ok thanks. thumbs up
So i have to take mv mesures on the transistor matcher? I won't need transistor test mode, right??
Altitude909
lcf wrote:
Ok thanks. thumbs up
So i have to take mv mesures on the transistor matcher? I won't need transistor test mode, right??


You are confusing 2 things:

Meters test for Hfe which is DC current gain

You want to match transistors for Vbe which is the voltage drop between the base and emitter of a part. you need a specialized device to accurately measure that
humax5600
@ whoop_john

Hi John,
Can you take me to the waiting list also ?
I am located in Germany.

Happy Easter
lcf
Altitude909 wrote:
you need a specialized device to accurately measure that

What do you mean by specialized device?? A multimeter with transistor test mode for example??
Altitude909
No.

The transistor test mode on a meter measures Hfe, not Vbe

You need something like this: http://www.musicfromouterspace.com/index.php?CATPARTNO=NONE&PROJARG=TR ANSISTORMATCHER%2FTRANSISTORMATCHER.html&MAINTAB=SYNTHDIY&SONGID=NONE& VPW=1266&VPH=747
colorbars
Definitely interested in a pcb if still available!

Just built an Ian Fritz transistor matched on stripboard and tried it out last night, works great too!
Flareless
Hi John,

Could you please add me to the list as well. I would definitely like one of these boards. I'm located in Ontario, Canada.

Thank you!
ehafh
ahoy John,
please add me to the list when you get a chance. :]

located in USA
Musicology of Dreams
Hi John

Me too i like this useful pcb please 1 pcs
location in France smile
Thanks and best regards
delayed
Has anyone ever built the MFOS go/nogo tester?
thetwlo
do these things work for matching?

http://www.taydaelectronics.com/new-arrivals/esr-meter-transistor-test er-lcd-diode-multimeter-capacitance.html

http://www.ebay.com/itm/New-GM328-Transistor-Tester-LCR-ESR-Meter-Squa re-Wave-Generator-MOS-PNP-NPN-WL10-/321942702081

[/img]
Altitude909
thetwlo wrote:
do these things work for matching?


no
Bamboombaps
Altitude909 wrote:
thetwlo wrote:
do these things work for matching?


no


What's the problem with them?
Altitude909
They dont measure Vbe
Bamboombaps


So this bunch of data I collected using one of them( hfe and uf ) is useless for the purpose of matching transistors for vcos etc?
Altitude909
Yes uselss, for the umpteenth time:

"matched transistors" in the synthesizer sense refers to the matching them to within 2 mV of Vbe NOT Hfe (gain) which you can easily measure with a meter.

Two perfectly matched parts for Vbe can have different Hfe. They are two completely different things.

Voltage drop between Base andEmitter which is a measurable number in volts

has little to do with

Hybrid parameter forward current gain, common emitter which is a unitless multiplier

Read this

There very well may be cases where you need to match hFE but the tracking of VCOs and VCFs isnt generally one of them
Bamboombaps
Yikes that's me told lol

Thanks for the info tho im always learning. even when teacher shouts at me The MSN Smack! dizzy
audiohawk
whoop_john wrote:
PWM wrote:
whoop_john
Hi John, do you have a spare PCB left? I'd be very interested.


I am more than happy to get some more or keep a stock of these for people. I would quite like to have comments back from the initial ten sent out, so I can make any suggested improvements if necessary.


If its still going on... 2 PCBs for me please.
Thank you
Altitude909
You can OSHpark some of Ray's design from here:

https://midisizer.com/other/vbe-matching/
ehafh
Altitude909 wrote:
You can OSHpark some of Ray's design from here:

https://midisizer.com/other/vbe-matching/


this is my first time using those file types / pcb manufacturing sites etc.
is it normal for this to pop up? i could only get the *.sch file to upload to oshpark
"Warning: Your project doesn't contain a bottom silk screen."

thanks
Altitude909
That's fine, there is no bottom silk screen for those boards
ehafh
Altitude909 wrote:
That's fine, there is no bottom silk screen for those boards


nice, i went ahead and ordered some.
i'll post some pics when they show up
whoop_john
To those that have recently asked, I am ordering another batch of PCBs soon. I never had any feedback from anyone on the first batch as to how they got on with the tester. Presumably everyone is happy? I know one got lost in the post - envelope arrived, contents missing. To this person, I will be sending you another one.
lcf
https://midisizer.com/other/vbe-matching/
What kind of power suplly could i use to power this please?

Edit: So if i understood correctly, we can't mesure VBE with a voltmeter on Ian's but on MFOS yes? Sorry I really have trouble to understand this one. (hides)
DomMorley
Please add me to the list for a pcb too.
Musicology of Dreams
whoop_john wrote:
To those that have recently asked, I am ordering another batch of PCBs soon.


Great thanks John
Best regards smile
sonicwarrior
whoop_john wrote:
I never had any feedback from anyone on the first batch as to how they got on with the tester. Presumably everyone is happy?

I have not build mine yet. Returning cancer and chemo therapy which attacks my hands/skin got between me and building it.
Altitude909
lcf wrote:
https://midisizer.com/other/vbe-matching/
What kind of power suplly could i use to power this please?

Edit: So if i understood correctly, we can't mesure VBE with a voltmeter on Ian's but on MFOS yes? Sorry I really have trouble to understand this one. (hides)


This requires a +/- 12V PSU

I have not looked at Ian's design but the MFOS one you measure voltage. Since Vbe is a voltage, I would assume you do the same with Ian's design. You need an exceptionally good meter (better than 0.001V resolution), I use an old 6 digit Agilent
elmegil
I power my build of Ian's with 2 9V batteries. It's fairly voltage agnostic.
lcf
whoop_john wrote:
I am ordering another batch of PCBs soon.

One board for me John please.
I'm located in France.
Thanks.
Rocket_Salad
Hey guys,

I'm gona be building this later today, and was wondering whether a 9V battery would work as a power source?

Cheers
Dead Banana
Rocket_Salad
For some reason this question didn't come up before: just saw it.

Cheers
anto32
whoop_john wrote:
To those that have recently asked, I am ordering another batch of PCBs soon.


Hi, Interested in a pcb,

thanks,
elmegil
I've lost track of which matching circuit this is; I use a 9V for Ian Fritz's version, certainly.
whoop_john
elmegil wrote:
I've lost track of which matching circuit this is; I use a 9V for Ian Fritz's version, certainly.

The circuit version I am working to is this revised version as posted earlier in this thread.

Yes you can use a 9v battery, it only needs a single supply.



More PCB boards will be available soon. I will post here when they arrive.
Musicology of Dreams
whoop_john wrote:
More PCB boards will be available soon. I will post here when they arrive.


Nice thanks for last news.
Greetings
lcf
whoop_john wrote:
More PCB boards will be available soon. I will post here when they arrive.

Thanks for the update.
duff
whoop_john wrote:
I never had any feedback from anyone on the first batch as to how they got on with the tester.


Sorry for the lack of feedback. I built mine up and matched the resistors but my first attempts gave odd readings that I put down to my rubbish DMM. Having now purchased a more accurate DMM I am having another go but still not getting the results I expected.

I can measure an mV value across the pads, but when I flip the switch I don't see the sign change. The reading seems to stabilize at the same value I had before the switch was flipped.

I can't believe I have built it incorrectly given the low part count. So I assume I must be doing something wrong, but I am not sure what. hmmm.....
whoop_john
duff wrote:
whoop_john wrote:
I never had any feedback from anyone on the first batch as to how they got on with the tester.


Sorry for the lack of feedback. I built mine up and matched the resistors but my first attempts gave odd readings that I put down to my rubbish DMM. Having now purchased a more accurate DMM I am having another go but still not getting the results I expected.

I can measure an mV value across the pads, but when I flip the switch I don't see the sign change. The reading seems to stabilize at the same value I had before the switch was flipped.

I can't believe I have built it incorrectly given the low part count. So I assume I must be doing something wrong, but I am not sure what. hmmm.....

OK. I am away from home until Friday pm. We need to get to the bottom of this before I get any more boards made.

I did double-check the circuit but will do so again, but I really think a fresh pair of eyes would be better, in case I missed anything. As you say the circuit is very simple. I will post my PCB layout when I get back home. It's probably higher up in this thread already.

I did a few tests when I built mine and was getting similar results to you - I did not get the expected change of sign, but I have not needed to do any matching myself so I assumed too I was doing something wrong.

If there is a problem that is down to me I will send out new boards to those ten I sent boards out to.
SlightlyNasty
You shouldn't see a sign change when switching S2 on that design, all the switch is doing is swapping which resistor each transistor is connected to - the meter probes are still connected to the same transistor either way.

All the switch does is let you check that the resistors are correctly balanced - ie. you shouldn't see a change in value when the switch is flicked from one position to another. When the trimmer is set up correctly the switch is basically redundant.
duff
SlightlyNasty wrote:
You shouldn't see a sign change when switching S2 on that design, all the switch is doing is swapping which resistor each transistor is connected to - the meter probes are still connected to the same transistor either way.

All the switch does is let you check that the resistors are correctly balanced - ie. you shouldn't see a change in value when the switch is flicked from one position to another. When the trimmer is set up correctly the switch is basically redundant.


I wondered if that was the case. Certainly the value I measured seemed to match with the value I calculated when re-positioning the probes to measure the Vbe of each transistor individually and then subtracting them. Given the massive error margin my old DMM introduced I wasn't certain, but the transistors I matched using this PCB do appear to work fine in a Minimoog VCF clone.
whoop_john
Then I think the circuit is good to go. I did check it several times.

You will want to make sure you are checking in a room where there is not a fluctuation of temperature due to heating systems or doors opening and closing. I am sure this goes without saying - test under different air conditions and all bets are off. You may as well simply hope that two transistors off the same bandolier are similar.

I have read two methods of stabilising the ambient temperature when testing transistors.

One is to wire up an old PC fan and blow ambient air over the tester so that it always enjoys a similar representative sample of the room's air when testing.

The other is to put the assembly in a shoe box so that it is not so affected by slight changes in air movement within the room.

The transistors do take a little time to settle down, so don't be surprised to see the figures scroll at first. Try not to handle with warm fingers when inserting test components.
qfactor
whoop_john: I was looking through this volt meter display and wondering if this could be wired up with your pcb to display the mV readings when matching transistors to become a "stand-alone" transistor matching device.

I wonder if it's possible to make this read in mV, good to 2 decimal places? seriously, i just don't get it
FiltroMan
Hi qfactor, hi all
why not use one of these?
http://es.aliexpress.com/item/5-Digit-0-36-DC-0-33-000V-high-accuracy- Digital-Voltmeter-Volt-Voltage-Panel-Meter/32354601094.html?spm=2114.1 2010408.1000023.7.EqS0k0

0,001V and 5 digit.
EATyourGUITAR
wasting your time with a 5 digit voltmeter spec for +/-3% @ 2 digits. also, even if it was real, how would you power it without batteries and maintain low noise supply for lab use? try getting a real multimeter, the best one you can afford.

https://www.amazon.com/Keysight-U3401A-Digital-Multimeter-Display/dp/B 01D0QYE0W/

https://www.amazon.com/Keysight-U3402A-Digital-Multimeter-Display/dp/B 01D0QYEGG/

https://www.amazon.com/Fluke-8808A-5-5-Digit-Digital-Multimeter/dp/B00 46C3UGO/
FiltroMan
EATyourGUITAR wrote:
wasting your time with a 5 digit voltmeter spec for +/-3% @ 2 digits. also, even if it was real, how would you power it without batteries and maintain low noise supply for lab use? try getting a real multimeter, the best one you can afford.

https://www.amazon.com/Keysight-U3401A-Digital-Multimeter-Display/dp/B 01D0QYE0W/

https://www.amazon.com/Keysight-U3402A-Digital-Multimeter-Display/dp/B 01D0QYEGG/

https://www.amazon.com/Fluke-8808A-5-5-Digit-Digital-Multimeter/dp/B00 46C3UGO/


Thanks for the help.
I ordered one of these, to match transistors only. http://www.newark.com/keysight-technologies/b2987a/electrometer-high-r esistance-meter/dp/82X2282
humax5600
In Germany we say : shoot with cannons on sparrows
EATyourGUITAR
FiltroMan wrote:
Thanks for the help.
I ordered one of these, to match transistors only. http://www.newark.com/keysight-technologies/b2987a/electrometer-high-r esistance-meter/dp/82X2282


I posted good information and you mock me by filling the thread with garbage. not only that but you made a new account to do it. you have two posts and one of them is disruptive and distracting to the real conversation for people that came here to learn.
FiltroMan
You're wrong, you have mocked me, recommending a multimeter 800 € to match transistors. I can think so, but I prefer to think more positively.
My comment was not to mock you, was to see that I can not afford it.
No disruptive and no distracting to the real conversation.
Sorry for the misunderstanding
cygmu
Question for those in the know: how accurate/fit for purpose is Ken Stone's method?
http://www.cgs.synth.net/modules/simple_transistor_matching.html

Obviously it will depend on the accuracy of the DMM being used. But what do people think of that method in general?
EATyourGUITAR
FiltroMan wrote:
You're wrong, you have mocked me, recommending a multimeter 800 € to match transistors. I can think so, but I prefer to think more positively.
My comment was not to mock you, was to see that I can not afford it.
No disruptive and no distracting to the real conversation.
Sorry for the misunderstanding


in that case I am sorry. I am poor also. I have discovered that you can buy a batch of 2N3904 from mouser and they will likely be pretty close by being in the same batch. no testing needed. its not perfect but that is the cheap way. I would not waste money on a $50 meter for transistor matching and waste another $10 on the transistor matching PCB. you can buy 8 of these SSM2210 for the same price. for my computer I purchased 5 $20 USB mouse before I got a $70 one I liked. it would be cheaper to just do it right the first time and not waste $20 $20 $20 $20 $20. the SSM2210 is matched as good as a 5 digit multi meter that costs $500. obviously SSM2210 is the cheaper option for people like you and me.
FiltroMan
EATyourGUITAR wrote:

in that case I am sorry. I am poor also. I have discovered that you can buy a batch of 2N3904 from mouser and they will likely be pretty close by being in the same batch. no testing needed. its not perfect but that is the cheap way. I would not waste money on a $50 meter for transistor matching and waste another $10 on the transistor matching PCB. you can buy 8 of these SSM2210 for the same price. for my computer I purchased 5 $20 USB mouse before I got a $70 one I liked. it would be cheaper to just do it right the first time and not waste $20 $20 $20 $20 $20. the SSM2210 is matched as good as a 5 digit multi meter that costs $500. obviously SSM2210 is the cheaper option for people like you and me.


Interesting. Thank you very much.
Ssm2210 in Aliexpress $ 2,53 O.o
Spending $ 45 in a uni t 61e or use ssm2210 ? What do you recommend me?
some completely diy module using monolithic matched transistors ?
EATyourGUITAR
You can build a daughter board for the 2210 to change the pinout. I cut the legs off 1N4001 to make legs. I would not buy from china or aliexpress. Only mouser or digikey.
cygmu
Mouser, digikey and farnell no longer have 2210s. There are a few 2220s still around if you need PNPs.

Are there other dual NPN chips one could use instead?
sonicwarrior
That's pretty OT. The question has been asked a few times, e.g. here:
https://www.muffwiggler.com/forum/viewtopic.php?t=59667&highlight=
cygmu
sonicwarrior wrote:
That's pretty OT. The question has been asked a few times, e.g. here:
https://www.muffwiggler.com/forum/viewtopic.php?t=59667&highlight=


Thanks.

(I didn't think it was off topic since the previous three posts had been a discussion of using SSM2210s instead of hand matching.)
elmegil
cygmu wrote:
Mouser, digikey and farnell no longer have 2210s. There are a few 2220s still around if you need PNPs.

Are there other dual NPN chips one could use instead?


SSM2212. if you read the data sheet it looks like they did a global search/replace of 2210, they're all highlighted in blue :-D (To be fair, the graphs do look different)
whoop_john
Anybody tried this method?

FiltroMan
elmegil wrote:
cygmu wrote:
Mouser, digikey and farnell no longer have 2210s. There are a few 2220s still around if you need PNPs.

Are there other dual NPN chips one could use instead?


SSM2212. if you read the data sheet it looks like they did a global search/replace of 2210, they're all highlighted in blue :-D (To be fair, the graphs do look different)


SSM2212 mouser: $6 to expensive

whoop_john wrote:
Anybody tried this method?



Its real? It would be amazing woah but dont understand, measure the resistance?
elmegil
FiltroMan wrote:
elmegil wrote:
cygmu wrote:
Mouser, digikey and farnell no longer have 2210s. There are a few 2220s still around if you need PNPs.

Are there other dual NPN chips one could use instead?


SSM2212. if you read the data sheet it looks like they did a global search/replace of 2210, they're all highlighted in blue :-D (To be fair, the graphs do look different)


SSM2212 mouser: $6 to expensive


Were you reading what EATmyGuitar was saying?

ALL the monolithic matched pairs are in that range. It can still be more cost effective than buying a meter just for matching, buying a PCB and parts to build a matcher.
FiltroMan
elmegil wrote:
FiltroMan wrote:
elmegil wrote:
cygmu wrote:
Mouser, digikey and farnell no longer have 2210s. There are a few 2220s still around if you need PNPs.

Are there other dual NPN chips one could use instead?


SSM2212. if you read the data sheet it looks like they did a global search/replace of 2210, they're all highlighted in blue :-D (To be fair, the graphs do look different)


SSM2212 mouser: $6 to expensive


Were you reading what EATmyGuitar was saying?

ALL the monolithic matched pairs are in that range. It can still be more cost effective than buying a meter just for matching, buying a PCB and parts to build a matcher.


Yes, I have read what he said . And he has not said that all monolithic matched transistors are in that price range . seriously, i just don't get it

I have been researching, for example Befaco use this http://www.mouser.es/Search/Refine.aspx?Keyword=bcm847 in the VCO. It is not a substitute for ssm but it is a good answer to my previous question.
elmegil
Quote:
I would not waste money on a $50 meter for transistor matching and waste another $10 on the transistor matching PCB. you can buy 8 of these SSM2210 for the same price.


$60 / 8 is more than $6 per pair.
elmegil
In all honesty, if you buy your transistors on tape, they are almost always matched plenty well enough to use without doing individual matching. So if budget is your concern, don't bother with a matching rig.
GryphonP3
elmegil wrote:
In all honesty, if you buy your transistors on tape, they are almost always matched plenty well enough to use without doing individual matching. So if budget is your concern, don't bother with a matching rig.


+1. For super critical transistors in the circuit, go with a 2210. For others, just use transistors from the same tape strip. Doubt you will ever have problems with this rule of thumb.. If you truly notice it and it is worth the investment to you, thats your call. Great multimeters for super precision jobs get expensive.
keninverse
Another option is just to buy the Erica synths 394 replacement. But like others have said just buy some on tape and throw it in a circuit and see if it works out. Most of the time the matching will be pretty close.
audiohawk
whoop_john wrote:
Then I think the circuit is good to go. I did check it several times.


Is it still on the go?
If yes, please count me in for a pcb.
Thank you for your work we're not worthy
thresholdpeople
whoop_john wrote:
I never had any feedback from anyone on the first batch as to how they got on with the tester.


Ahh yep, mine is still sitting half built. I'm moving now, but in a month or so I'll actually need it when I get to building.
ManicPenguin
Forgive me if this is a stupid question, as I'm fairly new to all this. I sourced three boards using fonik's eagle file through OSHpark, then stopped to realize that this uses a ten pin header for power.

My understanding was that this circuit was designed to run on two 9v batteries...is that still possible with the eurorack footprint PSU connector? If so, how would one accomplish that?

EDIT: After closer inspection of the board, and comparing it to Ian's original schematic, I figured out that the two outside pairs of pins are the V+ and V-, respectively. Sorry for the idiotic question.

All that being said, I have two of these boards available, if anyone is interested. thumbs up
Bamboombaps
hi ive got a few of the fonik boards if anyone needs them
EDIT ALL GONE

roger
hi,

anyone has an oppinion about cheap chinese transitor tester? like this?[/url]
AonFLuX
Bamboombaps wrote:
hi ive got a few of the fonik boards if anyone needs them

4.50GBP inc european shipping
5.00GBP elsewhere



Possible to run these with a 9V DC? (with small mod of connector)
Bamboombaps
hm. might need resistor changes?
AonFLuX
Bamboombaps wrote:
hm. might need resistor changes?


Ok, thanks. hmm. Maybe I'll just rip out some of the perfboards I bought and never used smile and follow the darwing I found here

And about the Chinese transistor tester I read that they are good for getting specs of the transistor and perhaps spotting fakes - but do not have the resolution needed to match transistors.
Sammus
Bamboombaps wrote:
hi ive got a few of the fonik boards if anyone needs them

4.50GBP inc european shipping
5.00GBP elsewhere


PM'd
whoop_john
AonFLuX wrote:
Bamboombaps wrote:
hm. might need resistor changes?


Ok, thanks. hmm. Maybe I'll just rip out some of the perfboards I bought and never used smile and follow the darwing I found here

And about the Chinese transistor tester I read that they are good for getting specs of the transistor and perhaps spotting fakes - but do not have the resolution needed to match transistors.

As I found out, if you balance the two sides with the preset pot, then you can eliminate one of the switches, as effectively both ways should read the same, if that makes any sense?
AonFLuX
whoop_john wrote:

As I found out, if you balance the two sides with the preset pot, then you can eliminate one of the switches, as effectively both ways should read the same, if that makes any sense?


Aha, ok - yeah, that got to be the case. Well even easier then.. smile Thanks

Edit: But if I match using this circuit and keeping one transistor stationary and just change the others and match the others based on the deviation to the stationary - does the both sides even be perfectly matched? The resistance will not change much and will have the same "unbalance" at all times. So maybe just keep the sides close balanced in resistance and not bother about shifting place or using a switch? Or am I missing something?
jean_schön
will a Uni-T UT61E have the required resolution and accuracy to match transistors with ian fritz's circuit?
acsww4
What is the advantage of building a tester versus just using the diode check on a 4 1/2 digit multimeter?
Altitude909
acsww4 wrote:
What is the advantage of building a tester versus just using the diode check on a 4 1/2 digit multimeter?


Highly accurate current source. The MFOS one has +/-10V Power
acsww4
Ahh, thank you! Then I shall build a tester...
rithma
So I breadboarded MFOS NPN transistor tester, got the precise +/- 10v and nulled out my op amp,
but when I power up the circuit, the last TL071 (the op amp whos output feeds the multimeter) holds steady at 0v. When I disconnect the emitter of DUT
from the op-amp's non inverting input (leaving 10v at the collector and ground at the base), the transistors Emitter ramps up to 350mv then back to 0v, over and over again,
maybe 2 times a second.
What could possibly be going on? I'd love to figure this one out...
[/img]
Midiot
I built the tester found on page 2 of this thread, (by Ian Fritz and Daverj) on pref-board, (point-to-point method).....using spare parts I already had.
I'm using a bench power supply to power it.
It works !!
I added various transistor sockets for flexibility (only 1 pair used at a time). Soon I'll install a BNC jack to go to my 6.5 digit multimeter, rather than un-shielded alligator clips.
My two 100k resistors were already better than 1%, so a 1k trim pot was adequate to balance them.

I left room on the board for other useful circuits.
Where power enters the board, I used the typical 10uf + 0.1uf decoupling caps, and it's set-up for bi-polar power, when needed. LED's show when power is "on".



daverj wrote:
Seeing the mention of 9 volt batteries above got me thinking that Ian's circuit could be simplified to run off a single battery instead of two. So I came up with a simple change and PM'd Ian about it. He came back with an even simpler modification to do it. We went back and forth a couple of times and ended up with the following schematics.
................
-- The same simplified circuit could also be connected to a regular power supply too instead of a battery. It just needs one power rail plus ground for that.


whoop_john
Midiot wrote:
I built the tester found on page 2 of this thread

If you have used the R3 preset to balance out the resistance to both legs, the reading should be the same whether switched through R1 or R2, thus S2 is not necessary and will appear to do nothing.
Grumble
Quote:
thus S2 is not necessary and will appear to do nothing.

Yes, S2 is essential to pair the transistors: If you balance the CURRENT (!) in both legs and the transistors are perfect twins, then the switch appears to do nothing, but if there is a small difference between the two dut's, switching S2 will show a different voltage between the points M+ and M-.
cygmu
Grumble wrote:
Quote:
thus S2 is not necessary and will appear to do nothing.

Yes, S2 is essential to pair the transistors: If you balance the CURRENT (!) in both legs and the transistors are perfect twins, then the switch appears to do nothing, but if there is a small difference between the two dut's, switching S2 will show a different voltage between the points M+ and M-.


No, the switch is to cancel out any imbalance in the resistors. If the resistors are accurately matched then the circuit will directly reveal any mismatch in the transistors without the need for the switch; indeed flipping the switch will just change the sign of the reading.
Grumble
If it was to balance the resistors, why not take high precision resistors?

Place two transistors, turn R3 until you measure 0 volt, flip the switch and if you still measure 0 volt you have a twin transistor.
This way you can achieve a very high accuracy in comparing the transistors because the used voltmeter can be set to measure low voltages.
cygmu
Grumble wrote:
If it was to balance the resistors, why not take high precision resistors?


That works. I used well-matched resistors in my version of the circuit and the two readings are always of the form x and -x, as far as the precision of my multimeter can tell.

EDIT: Ahh, but wait, I am using Ian's original circuit rather than the daverj version. If I had built the daverj version with the same resistors and used it with the same not-that-good multimeter that I am using, rather than changing sign, the readings would just stay the same when I flipped the switch.
frijitz
Grumble wrote:
If it was to balance the resistors, why not take high precision resistors?

This is stated in the doc.
analogdata
Apologies if this has been answered a million times.
Chinese transistor tester.
Is this not accurate enough for matching?

MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 16
Powered by phpBB © phpBB Group