[AVAILABLE] TTSH V4 Build Thread

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, Joe.

Post Reply
resynthesize
Wiggling with Experience
Posts: 305
Joined: Tue Nov 01, 2016 1:29 pm
Contact:

Re: [AVAILABLE] TTSH V4 Build Thread

Post by resynthesize » Tue Jan 26, 2021 1:14 am

synthcube wrote:
Mon Jan 25, 2021 5:32 pm
resynthesize wrote:
Thu Jan 14, 2021 12:56 am
ebfx wrote:
Wed Jan 13, 2021 7:23 pm
resynthesize wrote:
Wed Jan 13, 2021 10:51 am
fallout wrote:
Wed Jan 13, 2021 10:20 am
It looks like you have a reversed diode to the right of the LM301 at the top.
:doh: how did i not see that?! I’ll fix after work today and report back. thank you!
I'm curious to know if this solves your problem. The diode in question is part of the ADSR circuit so I wouldn't expect much effect on the AR operation.

I've been trying to track down a similar problem with my v4 build for a few nights now. After reviewing the v3 schematics and spending a few nights probing my board, I narrowed the problem down to the 2N5460 (Q4) in the ADSR section. I pulled Q4 from the board and put it in my (admittedly very cheap) transistor checker and it came back as a BJT instead of a JFET. So, I replaced both 2N5460's in the ADSR/AR section and everything started working as expected. You may want to check if you have a similar problem.

I checked the markings on the original transistors from Synthcube and they are definitely labeled 5460, so I suspect they either got a bad batch or I was a bit too heavy-handed during installation. My replacement NTE 2N5460's have a different pinout (S-D-G instead of D-S-G) so I had to do a bit of pin swapping on the board, but they read as P-channel JFET's on my tester and now my circuit is working like it is supposed to.
I just replace the diode and the ADSR behavior is different but still not right. The release slider is still just generating a fixed voltage at the ADSR output when I adjust it as well. Thank for the tip on the 2N5460, I'll pull that next and test it after I go through and verify the rest of the components with the unpopulated pcb photo that @fallout helpfully provided.

edit: I just pulled a 2N5460 and tested and sure enough it tests as a BJT-PNP as well. i sourced mine from utsource originally. I'll order some replacements and see how it goes. thanks!
please do post what you find regarding the 2N5460-- thanks!
I replaced both 2N5460s in the envelope section. The AR envelope started working right away, so that's sorted (I think the utsource 2N5460s were indeed bad). The attack slider in the ADSR section still has no effect (the rise in voltage is nearly instantaneous) so I've still got an issue there. I've been puzzling my way through the circuit description in the service manual in section 2.10.1 with limited success - it seems like likely sources of instantaneous charging of C1 and C2 could be an incorrect value at R12 or the attack slider itself (R179), but I'm admittedly a novice at this stuff. any hints appreciated!
Attachments
Screen Shot 2021-01-25 at 10.18.33 PM.png

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Tue Jan 26, 2021 7:55 am

Check to make sure you swapped the correct pins on the replacement 2N5460's.

There's some good troubleshooting information of the ADSR section in the following thread (starts about 1/2 way into the page and continues on the following):

viewtopic.php?f=17&t=98954&start=2850

ebfx
Learning to Wiggle
Posts: 15
Joined: Thu Oct 15, 2020 7:46 am
Location: Massachusetts

Re: [AVAILABLE] TTSH V4 Build Thread

Post by ebfx » Tue Jan 26, 2021 8:52 am

KSS wrote:
Mon Jan 25, 2021 5:28 pm
fallout wrote:
Sun Jan 24, 2021 2:12 pm
I've attached a pic of the part of the circuit that I have on my breadboard.
Good news! That drawing is missing R2. The 100K to ground from pin 2 of the op amp. Add that and let us know how it goes.

Edit: That resistor is missing from the schematics of the 2600 in both version of the service manual and on both pages where its drawn in them. But the 4022 PCB layout in both those manuals clearly show its existence.The PCB is correct.
Thanks for the info KSS... however there is only one 100k resistor in the Noise section (NOISE-R1) and that one connects the emitter of Q1 to +15V. We may indeed be missing a resistor (NOISE-R2 is nowhere to be found on the schematics) but I don't believe it's on the PCB.

That said, I will try adding the resistor to my breadboard setup and report back.

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Tue Jan 26, 2021 10:26 am

ebfx wrote:
Tue Jan 26, 2021 8:52 am
We may indeed be missing a resistor (NOISE-R2 is nowhere to be found on the schematics) but I don't believe it's on the PCB.

That said, I will try adding the resistor to my breadboard setup and report back.
I think he was referring to the PCB scan in the 2600 service manual. It doesnt exist on the TTSH PCB.

That said, it didn't resolve the problem for me.

@ebfx are you also seeing that noisy sine wave after the 1uf and not on the emitter?

ebfx
Learning to Wiggle
Posts: 15
Joined: Thu Oct 15, 2020 7:46 am
Location: Massachusetts

Re: [AVAILABLE] TTSH V4 Build Thread

Post by ebfx » Tue Jan 26, 2021 11:02 am

fallout wrote:
Tue Jan 26, 2021 10:26 am
ebfx wrote:
Tue Jan 26, 2021 8:52 am
We may indeed be missing a resistor (NOISE-R2 is nowhere to be found on the schematics) but I don't believe it's on the PCB.

That said, I will try adding the resistor to my breadboard setup and report back.
I think he was referring to the PCB scan in the 2600 service manual. It doesnt exist on the TTSH PCB.

That said, it didn't resolve the problem for me.

@ebfx are you also seeing that noisy sine wave after the 1uf and not on the emitter?
Ahhh, I see. I misunderstood what he meant by "the PCB".

Yes, the sine wave is after the 1uF cap. The emitter looks as expected with a ~7V offset.

I replaced the 1uF cap with a 10uF cap on my breadboard setup which pretty much eliminated the sine wave. Unfortunately, while the opa output didn't look nearly as clean, it was still basically a sine wave.

I did notice that on the schematic in the service manual, Q1 is listed as 2N5172 selected (their emphasis). I'm not familiar with the "selected" terminology. Anyone know what that's about?

User avatar
Jim the Oldbie
Super Deluxe Wiggler
Posts: 1091
Joined: Mon Jun 10, 2019 12:16 am
Location: Midwest USA

Re: [AVAILABLE] TTSH V4 Build Thread

Post by Jim the Oldbie » Tue Jan 26, 2021 8:52 pm

I think that means selected for proper noise characteristics (amplitude, spectrum etc.)?

User avatar
LED-man
Super Deluxe Wiggler
Posts: 1528
Joined: Wed Aug 08, 2012 1:48 pm
Location: Wolfsburg Germany
Contact:

Re: [AVAILABLE] TTSH V4 Build Thread

Post by LED-man » Wed Jan 27, 2021 5:42 am

Please look in the TTsH rev1 build Thread or on my website.
i use mostly an BC337 16 (different pinout).
Or BC337 25 with additional 10k resistor in series on the electrolyte cap.
viewtopic.php?t=98954&start=1175

https://www.dsl-man.de/display/TTSH/TTSH+Noise+Mod
################################
TTSH Support Page and other DIY Projects:
https://www.DSL-man.de

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Wed Jan 27, 2021 10:26 am

*deleted*
Last edited by fallout on Wed Jan 27, 2021 10:40 am, edited 1 time in total.

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Wed Jan 27, 2021 10:30 am

*deleted*
Last edited by fallout on Wed Jan 27, 2021 10:39 am, edited 1 time in total.

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Wed Jan 27, 2021 10:31 am

Finally some good news! I found a few LM301's and even better, it fixed the noise issue. I still have the LM301AP's from Synthcube being delivered, I will test those as well. My guess is that they will work too.

I'm using a BC547 and it sounds pretty good. I'll have to try the BC337 as LED-man suggested but I don't have any on hand ATM.

@ebfx pm me if you need a 301, I found a few.

For my own knowledge, does anyone know why the TL071, LM741 and a different batch of LM301AN's didn't work in the circuit?

ebfx
Learning to Wiggle
Posts: 15
Joined: Thu Oct 15, 2020 7:46 am
Location: Massachusetts

Re: [AVAILABLE] TTSH V4 Build Thread

Post by ebfx » Wed Jan 27, 2021 10:53 am

fallout wrote:
Wed Jan 27, 2021 10:31 am
Finally some good news! I found a few LM301's and even better, it fixed the noise issue. I still have the LM301AP's from Synthcube being delivered, I will test those as well. My guess is that they will work too.

I'm using a BC547 and it sounds pretty good. I'll have to try the BC337 as LED-man suggested but I don't have any on hand ATM.

@ebfx pm me if you need a 301, I found a few.

For my own knowledge, does anyone know why the TL071, LM741 and a different batch of LM301AN's didn't work in the circuit?
Are you saying that a new LM301 fixed your issue in the original circuit, or the combination of the new LM301 and the BC547 fixed it? Based on LED-man's links I got the impression the problem we're having is more related to the BJT than the op-amp.

Thanks for the offer fallout, but I've got a couple LM301's already.

Also, thanks to LED-man for the info -- Gonna order some BC337's this afternoon.

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Wed Jan 27, 2021 11:11 am

Yes, I didn't even bother to breadboard it. Popped it in the TTSH and it worked right away (I swapped out the 2n5172 when initially troubleshooting). It seems that whatever batch of LM301AN's Synthcube is currently shipping doesnt work in some circuits. Someone was having an issue with their 1602 sequencer build that was caused by Synthcube's LM301AN's, that's why I mentioned it early in the discussion and why Synthcube sent out a few LM301AP's to test. FWIW, I doubt the transistor had anything to do with the issue we were having. I believe the recommendations for different transistors is more for the characteristics/quality of the noise generated. I've seen a few posts about the 2n5172's sounding a bit too distorted? I didn't have a chance to listen to mine.

Either way, if you have an LM301 from another batch (not from the Synthcube kit), try it!

ebfx
Learning to Wiggle
Posts: 15
Joined: Thu Oct 15, 2020 7:46 am
Location: Massachusetts

Re: [AVAILABLE] TTSH V4 Build Thread

Post by ebfx » Wed Jan 27, 2021 11:38 am

fallout wrote:
Wed Jan 27, 2021 11:11 am
Yes, I didn't even bother to breadboard it. Popped it in the TTSH and it worked right away (I swapped out the 2n5172 when initially troubleshooting). It seems that whatever batch of LM301AN's Synthcube is currently shipping doesnt work in some circuits. Someone was having an issue with their 1602 sequencer build that was caused by Synthcube's LM301AN's, that's why I mentioned it early in the discussion and why Synthcube sent out a few LM301AP's to test. FWIW, I doubt the transistor had anything to do with the issue we were having. I believe the recommendations for different transistors is more for the characteristics/quality of the noise generated. I've seen a few posts about the 2n5172's sounding a bit too distorted? I didn't have a chance to listen to mine.

Either way, if you have an LM301 from another batch (not from the Synthcube kit), try it!
Got it. Glad you got it working! I think I have some LM301's from another (non-Synthcube) project laying around, but if I don't I'll PM you.

My gut said the problem was with the op-amp all along, but since swapping it with one from another section didn't change anything I moved on. Let me know what happens with the new Synthcube ones.

kristop
Common Wiggler
Posts: 62
Joined: Thu Mar 24, 2016 12:31 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by kristop » Wed Jan 27, 2021 3:13 pm



Oh boy! What a fun build. And it worked on the first power up.... we'll see how calibration goes.

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Wed Jan 27, 2021 3:56 pm

Now that the noise problem is fixed, I've encountered another problem caused by the LM301AN.

When calibrating the internal clock, I was not able to adjust the period to slower than 1 second. The rate init trimmer would reach a point and the square wave would go flat. I swapped out A42 for one of my spare LM301AN's and the problem was resolved.

@ebfx I'm curious if you encounter this as well.

User avatar
synthcube
you will be assimilated
Posts: 2214
Joined: Wed Dec 21, 2011 6:42 pm
Location: Boston
Contact:

Re: [AVAILABLE] TTSH V4 Build Thread

Post by synthcube » Wed Jan 27, 2021 3:58 pm

it may have been listed earlier but what is the date code on your LM301AN please? Trying to narrow this down as best we can. Thank you.
facebook: https://www.facebook.com/2SynthCube

synthcube webstore www.synthcube.com/cart
music from outer space: www.musicfromouterspace.com
blacet research: www.blacet.com
MOTM DIY Analog Synthesizers: www.motmsynthesizers.com
music from outer space euro smt: www.mfoseuro.com

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Wed Jan 27, 2021 4:14 pm

synthcube wrote:
Wed Jan 27, 2021 3:58 pm
it may have been listed earlier but what is the date code on your LM301AN please? Trying to narrow this down as best we can. Thank you.
JR67ACE3

The numbers on the bottom of the chips are all different, or at least the ones I've looked at.

ebfx
Learning to Wiggle
Posts: 15
Joined: Thu Oct 15, 2020 7:46 am
Location: Massachusetts

Re: [AVAILABLE] TTSH V4 Build Thread

Post by ebfx » Wed Jan 27, 2021 6:49 pm

fallout wrote:
Wed Jan 27, 2021 4:14 pm
synthcube wrote:
Wed Jan 27, 2021 3:58 pm
it may have been listed earlier but what is the date code on your LM301AN please? Trying to narrow this down as best we can. Thank you.
JR67ACE3

The numbers on the bottom of the chips are all different, or at least the ones I've looked at.
Same for me. All my LM301AN's say JR67ACE3.

ebfx
Learning to Wiggle
Posts: 15
Joined: Thu Oct 15, 2020 7:46 am
Location: Massachusetts

Re: [AVAILABLE] TTSH V4 Build Thread

Post by ebfx » Wed Jan 27, 2021 6:52 pm

fallout wrote:
Wed Jan 27, 2021 3:56 pm
Now that the noise problem is fixed, I've encountered another problem caused by the LM301AN.

When calibrating the internal clock, I was not able to adjust the period to slower than 1 second. The rate init trimmer would reach a point and the square wave would go flat. I swapped out A42 for one of my spare LM301AN's and the problem was resolved.

@ebfx I'm curious if you encounter this as well.
Hadn't gotten there yet, so I just took a look and I'm seeing the same thing. I've got a friend who's supposed to bring over a couple of LM301's from his stock later this week (all of mine are from Synthcube, his are not). I'll let you know what I find.

User avatar
KSS
Super Deluxe Wiggler
Posts: 4803
Joined: Mon Jan 25, 2016 7:28 am

Re: [AVAILABLE] TTSH V4 Build Thread

Post by KSS » Wed Jan 27, 2021 7:28 pm

fallout wrote:
Wed Jan 27, 2021 10:31 am
For my own knowledge, does anyone know why the TL071, LM741 and a different batch of LM301AN's didn't work in the circuit?
Did you try the 100K to GND?

We've seen several posts in these TTSH build threads about overly high noise levels. I never understood why the people reporting those didn't have what the 2600 did. Until now. THC copied an error in the 2600 manual and no one has fixed it yet. Besides the one person who noted in this thread the conclusion it had no effect.

As answer to your direct question, what happens to an opamp whose inputs are too high? And what is the effect of a 100K to GND just before that input?

---------------
Earlier someone asked about th meaning of 'select'. "Select" or "SEL" on ARP schematics means the part is chosen for some particular parameter result. Which parameter result -and therefore which parameter of the part- changes depending on the circuit. There are far more SELs in a 2500 than a 2600.

But this is one of the possible 'gotchas' in cloning old circuits with newer parts. Their basic parameters may not fall along the same curves as the originals, even though they have the same part number. All well-designed circuits try to place as many parts in the middle of their parameter bands and tolerances as is possible.

Which is why we end up with things like the added resistor on LED mans noise transistor substitution.
Treat utility modules as stars instead of backup singers.
Treat power supplies like Rockstars instead of roadies!
Chase magic sound, not magic parts.

User avatar
KSS
Super Deluxe Wiggler
Posts: 4803
Joined: Mon Jan 25, 2016 7:28 am

Re: [AVAILABLE] TTSH V4 Build Thread

Post by KSS » Wed Jan 27, 2021 7:37 pm

FWIW, there's a missing 100K resistor* in the ARP 2600 manual for the 4020 also. I did check that one in the past, and AIRC it had been accounted for by THC. But I'd look to see if that's another issue like the above described situation.

*R21-100K to GND off pin 1 of the 4020

Edit: I should note that there are also some other values in that circuit which are labeled incorrectly. and worse, Rob of AMSynth promoted some false statements about them and how to correct them. The end result is that the THC TTSH might have different values than the ARP original due to how this ARP mistake and its evaluation and 'fix' by Rob were handled by THC in the various TTSH revs. Which obviously mostly work fine in the TTSHs. So knowing the original ARP values may *not* be the correct solution for a TTSH. <--Which is why I'm not adding them here.

But all of this may still be a factor into your TTSH ADSR problem.
Treat utility modules as stars instead of backup singers.
Treat power supplies like Rockstars instead of roadies!
Chase magic sound, not magic parts.

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Wed Jan 27, 2021 7:56 pm

KSS wrote:
Wed Jan 27, 2021 7:28 pm
fallout wrote:
Wed Jan 27, 2021 10:31 am
For my own knowledge, does anyone know why the TL071, LM741 and a different batch of LM301AN's didn't work in the circuit?
Did you try the 100K to GND?

We've seen several posts in these TTSH build threads about overly high noise levels. I never understood why the people reporting those didn't have what the 2600 did. Until now. THC copied an error in the 2600 manual and no one has fixed it yet. Besides the one person who noted in this thread the conclusion it had no effect.
A few posts back, you'll see that I did try your suggestion of putting a 100k resistor from pin 2 to ground and it had no effect whatsoever on the problem. The problem wasn't high noise levels on the output. It was a perfect 10Hz square wave at the opamp's output. My thought is that the offending opamps are unstable running at such high gain and feeding back the instability/oscillation to its input. I could be way off, I'm guessing. Anyhow, the fix was simply putting another LM301AN from a different batch into Z1.
KSS wrote:
Wed Jan 27, 2021 7:28 pm
As answer to your direct question, what happens to an opamp whose inputs are too high? And what is the effect of a 100K to GND just before that input?
I understand that an opamp's output will hit the -/+ rails when the input is too high but I don't think that was the whole story in this case. That said, I honestly can't answer what a 100k resistor from pin 2 to GND would do? Based on my test, my guess would be it does nothing! :lol: What would be it's purpose? I appreciate all of your help!

User avatar
KSS
Super Deluxe Wiggler
Posts: 4803
Joined: Mon Jan 25, 2016 7:28 am

Re: [AVAILABLE] TTSH V4 Build Thread

Post by KSS » Wed Jan 27, 2021 8:13 pm

fallout wrote:
Wed Jan 27, 2021 7:56 pm
KSS wrote:
Wed Jan 27, 2021 7:28 pm
fallout wrote:
Wed Jan 27, 2021 10:31 am
For my own knowledge, does anyone know why the TL071, LM741 and a different batch of LM301AN's didn't work in the circuit?
Did you try the 100K to GND?

We've seen several posts in these TTSH build threads about overly high noise levels. I never understood why the people reporting those didn't have what the 2600 did. Until now. THC copied an error in the 2600 manual and no one has fixed it yet. Besides the one person who noted in this thread the conclusion it had no effect.

As answer to your direct question, what happens to an opamp whose inputs are too high? And what is the effect of a 100K to GND just before that input?
A few posts back, you'll see that I did try putting a 100k from pin 2 to ground and it had no effect whatsoever on the problem. The problem wasn't high noise levels. It was a perfect 10Hz square wave at the opamp's output. My thought is that the offending opamps are unstable running at such high gain and feeding back the instability/oscillation to its input. The fix was simply putting another LM301AN from a different batch.

I understand that an opamp's output will hit the -/+ rails when the input is too high but I don't think that was the whole story in this case. That said, I honestly can't answer what a 100k resistor from pin 2 to GND would do? Based on my test, my guess would be it does nothing! :lol: What would be it's purpose? I appreciate all of your help!
Please note the underlined above in your quote of me. ;)

The opamp has a high input impedance. A resistor to ground just before it gives the electrons an alternate place to go. the input therefore receives a lower level than if the resistor is missing. When you added it to yours, it did reduce the input level. But apparently not enough to overcome the apparently poor specs of the supplied OPA part. I'd keep the 100K in place based on other threads complaining about too high noise levels and others saying the TTSH doesn't sound like the 2600.

User avatar
fallout
Common Wiggler
Posts: 83
Joined: Fri Dec 18, 2020 2:45 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by fallout » Wed Jan 27, 2021 8:23 pm

KSS wrote:
Wed Jan 27, 2021 8:13 pm
The opamp has a high input impedance. A resistor to ground just before it gives the electrons an alternate place to go. the input therefore receives a lower level than if the resistor is missing. When you added it to yours, it did reduce the input level. But apparently not enough to overcome the apparently poor specs of the supplied OPA part. I'd keep the 100K in place based on other threads complaining about too high noise levels and others saying the TTSH doesn't sound like the 2600.
Excellent! Thank you for the explanation. Always good to learn something new.

kristop
Common Wiggler
Posts: 62
Joined: Thu Mar 24, 2016 12:31 pm

Re: [AVAILABLE] TTSH V4 Build Thread

Post by kristop » Thu Jan 28, 2021 12:04 am

fallout wrote:
Wed Jan 27, 2021 11:11 am
Yes, I didn't even bother to breadboard it. Popped it in the TTSH and it worked right away (I swapped out the 2n5172 when initially troubleshooting). It seems that whatever batch of LM301AN's Synthcube is currently shipping doesnt work in some circuits. Someone was having an issue with their 1602 sequencer build that was caused by Synthcube's LM301AN's, that's why I mentioned it early in the discussion and why Synthcube sent out a few LM301AP's to test. FWIW, I doubt the transistor had anything to do with the issue we were having. I believe the recommendations for different transistors is more for the characteristics/quality of the noise generated. I've seen a few posts about the 2n5172's sounding a bit too distorted? I didn't have a chance to listen to mine.

Either way, if you have an LM301 from another batch (not from the Synthcube kit), try it!
Glad this issue was on this page. Noticed while actually sitting down to start calibration that my noise just thumped and my sample & hold didn't work. Swapped the LM301 with one I had in my parts drawer and presto. Glad I have 9 more in case other issues arise.

Post Reply

Return to “Music Tech DIY”