New Interfaces for Analog Synthesizers.

Paranormal Patroler
Super Deluxe Wiggler
Posts: 10676
Joined: Tue Aug 30, 2011 3:40 pm
Location: the Terminal beach
Not sure comparing a switch to a matrix switch is fair, but knowing all three modules your assessment is 100% right. The important thing is to have whatever does the job exactly right. And for me the ADDAC 214 is what I wanted to flip my drumming sequences around.

To tell you the truth, I have yet to meet anyone who fully takes advantage of their SSM!

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
Is there a manual for this anywhere? I completed my build but am a bit confused and wondering if I made a mistake. The soldering looks good... but this is happening:

one input into "1" and then rotating either by clock, or by using the "rotate" knob, outputs 1 and 3 or 2 and 4 seem to be firing simultaneously. I'm struggling to contact the most logical experiments when expected behavior is slightly murky.

Thanks!

Edit: a bit of experimentation shows:

feeding only an LFO square as "input 1" and feeding that as a trigger to 4 envelopes shows the above (1 and 3 and 2 and 4 seem tied together).

However, If I put a "0v" signal from Blinds into the other 3 inputs, it then behaves as follows (turning the rotate knob sends the destination pulse to the envelopes in turn as expected).

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
The plot thickens, and this does seem wrong. I will go double-check all my soldering, but if there are obvious candidates here, please let me know.

The setup:

4 channels on Blinds set as follows: 1 = 1v, 2 = 2v, 3 = 3v, 4 = 4v

Convenient, eh? Also convenient to have appropriate cable colors for each channel:

These were run into corresponding channels on the ADDAC214, and then run back into DATA's voltage monitor.

When sweeping the knob, results from stages indicated by LED numbers:

Stage 1: 1.92, 3.83, 5.72, 7.64
Stage 2: 3.82, 5.71, 7.63, 1.90
Stage 3: 5.7, 7.66, 1.92, 3.84
Stage 4: 7.65, 1.92, 3.83, 5.73

This is almost what I would expect in terms of the rotating progressing, but why are the outputs of each channel almost doubled? 1v in = 1.92out. 2v in = 3.83 out. 3v in = 5.72 out. 4v in = 7.64 out. Can't be right, right?

Then:

ONLY channel 1 used as input, with same 1v source. Other 3 jacks empty.

Stage 1: 1.55, 0.1, 1.56, 0.1
Stage 2: 0.0, 1.54, 0.01, 1.55
Stage 3: 1.55, 0.01, 1.56, 0.01
Stage 4: 0.0, 1.54, 0.01, 1.56

So in this case it is no longer doubled, but the 1.5v shows up on every other channel where I would expect it to only be on one. Is the additional .5 value expected? In stage one, should it be showing up on output 3, as well?

Final test:

Input 1 still receiving 1.0 volts. Input 2-4 getting explicit 0v from Blinds.

Stage 1: 1.9, 0, 0, 0
Stage 2: 0, 0, 0, 1.9
Stage 3: 0, 0, 1.9, 0
Stage 4: 0, 1.9, 0, 0

This is behaving almost as expected, but the values after stage 1 seem to be in the wrong spot. In Stage 2, shouldn't output 2 be getting the high voltage? and again, it is almost doubled.

My guess, and I am totally naive here, is that in the 2nd example above, the empty jacks should be "normaling" to zero and providing the same results as experiment 3. So something is amiss with my grounding perhaps?

Also, why are the voltages coming out higher in all cases than the inputs?

And why are they showing up in unexpected outputs?

Summa
Super Deluxe Wiggler
Posts: 2737
Joined: Fri Feb 01, 2013 6:52 am
Location: Sweden
since you DIY'ed this its almost impossible to tell what's wrong, but whenever I make a module I double and sometimes triple check resistor values and make sure capacitors are mounted in the right direction, it's easy to make mistakes.

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
Summa wrote:since you DIY'ed this its almost impossible to tell what's wrong, but whenever I make a module I double and sometimes triple check resistor values and make sure capacitors are mounted in the right direction, it's easy to make mistakes.
I wouldn't put that past myself, but in this case the boards are very much pre-populated and the soldering was just the jacks, pots, and headers.

In this case, my main frustration is that there is no documentation that I can find about truly expected behavior.

The feature list says " this module always swaps 4 inputs to 4 outputs, and can also be patched for 2 to 2 and 3 to 3 step actions." but there is no other info. Am I inadvertently "patching" like this to get the 1v (now 1.9v) coming out in 2 channels?

JohnLRice
Thanks for all the info mantid and sorry you are having problems but like Summa said, it's near impossible to know what is wrong, especially without having it on our own workbenches, or at least seeing pictures. Can you post some pictures of the PCBs?

I'm assuming you got the kit where all of the surface mount devices were pre-populated so you only had to solder the relatively few through-hole parts, meaning the jacks, LEDs, multi-pin connectors, pots and and the switch? In that case there is very little that could go wrong, in which case the thing(s) wrong might be the following, in order of most likely to least:
* solder bridges
* one or more things left unsoldered and/or cold solder joints
* through-hole parts not orientated correctly
* defective PCB / SMD parts

EDIT: Here's the build guide for reference:

JohnLRice
mantid wrote:
Summa wrote:since you DIY'ed this its almost impossible to tell what's wrong, but whenever I make a module I double and sometimes triple check resistor values and make sure capacitors are mounted in the right direction, it's easy to make mistakes.
I wouldn't put that past myself, but in this case the boards are very much pre-populated and the soldering was just the jacks, pots, and headers.

In this case, my main frustration is that there is no documentation that I can find about truly expected behavior.

The feature list says " this module always swaps 4 inputs to 4 outputs, and can also be patched for 2 to 2 and 3 to 3 step actions." but there is no other info. Am I inadvertently "patching" like this to get the 1v (now 1.9v) coming out in 2 channels?
Cross posted . . .I type and research slowly!

In general sequential switches should NOT be altering the values they are switching, they should just pass through the input value as-is or not at all.

JohnLRice
And, I'd do a quick test for you with my factory built ADDAC214 but I'm in the middle of the 50th "final" rearrangement of the case it is in and I've run into a road block so I was just taking a frustration break to calm down.

Summa
Super Deluxe Wiggler
Posts: 2737
Joined: Fri Feb 01, 2013 6:52 am
Location: Sweden
yeah post some really detailed photos of the pcb's and we can do a better job trying to find out what's wrong with your module. I do know that the pcb is mainly populated with all the smd parts, they'er also kind of delicate and if you misplace a drop of solder or heat it up too much you might fry something close by. It's always a risk you take DIYing modules but in general it all works out in the end. If nothing else helps just contact ADDAC and you'll probably end up having to send the module to them for a replacement if you can't find out by yourself what's wrong with it.

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
ADDAC has let me know that jack 1 is normalled to 3, and 2 to 4, so my "duplicate" voltages are expected (and also why forcing 0v is preventing this). Ok, that should probably be listed in the description.

The mystery remains why I am getting additional voltage out. Working with their email support.

Update: Manufacturer put the wrong resistors in a few spots on the pre-populated board. Awaiting resolution.

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
ADDAC is having me add resistors in parallel to the incorrect ones. Seems not too bad of a job but certainly an irritating hassle (and at my personal expense) to correct their mistake. If anyone else purchases the DIY kit, you may want to verify the pre-placed components and be ready for this possibility.

Looking forward to having it fully functional as intended.

JohnLRice
mantid wrote:ADDAC is having me add resistors in parallel to the incorrect ones. Seems not too bad of a job but certainly an irritating hassle (and at my personal expense) to correct their mistake. If anyone else purchases the DIY kit, you may want to verify the pre-placed components and be ready for this possibility.

Looking forward to having it fully functional as intended.
Thanks for the info and sorry you have a hassle to deal with!

I guess I'd better carefully check my factory built module to make sure it is functioning as it should since I would assume that the PCBs for both the kits and factory assembled units come from the same place?

Super Deluxe Wiggler
Posts: 1926
Joined: Thu Mar 26, 2009 11:05 am
Location: Odyssey Island
Contact:
JohnLRice wrote:
mantid wrote:ADDAC is having me add resistors in parallel to the incorrect ones. Seems not too bad of a job but certainly an irritating hassle (and at my personal expense) to correct their mistake. If anyone else purchases the DIY kit, you may want to verify the pre-placed components and be ready for this possibility.

Looking forward to having it fully functional as intended.
Thanks for the info and sorry you have a hassle to deal with!

I guess I'd better carefully check my factory built module to make sure it is functioning as it should since I would assume that the PCBs for both the kits and factory assembled units come from the same place?
Yes, and let us know! I just ordered one!

Paranormal Patroler
Super Deluxe Wiggler
Posts: 10676
Joined: Tue Aug 30, 2011 3:40 pm
Location: the Terminal beach
I'll report back when mine arrives, but I doubt there's a problem with the ones shipped from the shop.

Super Deluxe Wiggler
Posts: 1926
Joined: Thu Mar 26, 2009 11:05 am
Location: Odyssey Island
Contact:
Probably right, I ordered one with a custom black panel so it won't show up for about a month from now.

JohnLRice
ADDAC reached out to me earlier today to let me know about the resistor problem with the ADDAC214 I have. I checked my (factory built) 214 just now and it indeed has the problem, adding approximately 1.8 volts to the input. I thought something didn't seem quite right when I was testing mine initially but I was too excited to use it and had other things going on so I just shrugged it off at the time.

ADDAC said I could send it back for repairs and also provided instructions in case I want to fix it myself, which I think I'll do.

Paranormal Patroler
Super Deluxe Wiggler
Posts: 10676
Joined: Tue Aug 30, 2011 3:40 pm
Location: the Terminal beach
Received mine yesterday along with some awesomely good stuff and haven't plugged it in yet. I'll have to check, but if it can be fixed then it's no biggie. Worst case I'll send it for a repair (been having a few issues lately with several stuff, so I'm used to it by now )

Nice that they reached out to you John.

Super Deluxe Wiggler
Posts: 1926
Joined: Thu Mar 26, 2009 11:05 am
Location: Odyssey Island
Contact:
Confirmed new PCBs will have the correction so shouldn't be any issues going forward!!

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
I've attempted the repair by adding new resistors in parallel and am still getting weird results:

input 1: 1v in ---> 0.8v out
input 2: 2v in ---> 1.6v out
input 3: 3v in ---> 3.18v out
input 4: 4v in ---> 3.2v out

So... something changed! But still not there.

I've retouched each junction again with fresh solder but am not used to surface mount (which is explicitly why I bought this kit). Not sure what else to try or how to diagnose. I'm quite happy to learn but hopefully needless to say this is fairly frustrating and I am not super-impressed with ADDAC's support or response. I'm assuming I'll have to send this in for repair at my expense, or buy a new "correct" kit/module.

JohnLRice
mantid wrote:I've attempted the repair by adding new resistors in parallel and am still getting weird results:

input 1: 1v in ---> 0.8v out
input 2: 2v in ---> 1.6v out
input 3: 3v in ---> 3.18v out
input 4: 4v in ---> 3.2v out

I'm sorry to hear this! I haven't received my resistors from Mouser yet but now I don't know if it will be worth the effort to try the mod?

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
JohnLRice wrote:
mantid wrote:I've attempted the repair by adding new resistors in parallel and am still getting weird results:

input 1: 1v in ---> 0.8v out
input 2: 2v in ---> 1.6v out
input 3: 3v in ---> 3.18v out
input 4: 4v in ---> 3.2v out

I'm sorry to hear this! I haven't received my resistors from Mouser yet but now I don't know if it will be worth the effort to try the mod?
You should do it! I most-likely messed it up. I am using one of those illuminated magnifying lamps - not a microscope - so it is still a bit hard to see but the junctions do look good. I used a link to mouser part provided by ADDAC so I assume the value is correct. Nonetheless, I'm willing to take full responsibility for the soldering not going quite right since I have not touched a surface-mount part in 20 years (and even then only a couple times).

Please let me know how yours goes!

JohnLRice
mantid wrote:
JohnLRice wrote:
mantid wrote:I've attempted the repair by adding new resistors in parallel and am still getting weird results:

input 1: 1v in ---> 0.8v out
input 2: 2v in ---> 1.6v out
input 3: 3v in ---> 3.18v out
input 4: 4v in ---> 3.2v out

I'm sorry to hear this! I haven't received my resistors from Mouser yet but now I don't know if it will be worth the effort to try the mod?
You should do it! I most-likely messed it up. I am using one of those illuminated magnifying lamps - not a microscope - so it is still a bit hard to see but the junctions do look good. I used a link to mouser part provided by ADDAC so I assume the value is correct. Nonetheless, I'm willing to take full responsibility for the soldering not going quite right since I have not touched a surface-mount part in 20 years (and even then only a couple times).

Please let me know how yours goes!
I might try it, although I'm not sure how you could of done anything wrong?

For inspecting solder joints all you really need is an inexpensive Jeweler's Loupe. They are about \$5 to \$10, do search on Amazon etc. I've used one for years that looks like this:

But even though i have a couple different loupes and a good illuminated magnifying lamp, because of this and another project I ordered a inexpensive stereo microscope yesterday.

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
Somehow that photo of the stereoscope reminded my that my daughter has a little USB microscope

I took a look, and they did all look "fine" but I touched them all up again with a needle-tip iron, and now they look better (at least more squarely aligned).

I'm not sure if this is valid, but measuring across each resistor stack now, with module powered down, shows 35.97k:

The other 3 are quite close in value (+_.1k). The spec is 37k4. The original value was 100k, I added 56k. According to ye olde formula, this should yield 35897.44

So... I am pretty close (assuming my method of measuring - probing either end of each stack of 2 resistors while in-place is valid).

Nonetheless, the current results are, again feeding in 1-4v to each respective input:

outputting better than ever, but still weird:

1v --> .8v (loss of .2v)
2v --> 1.6v (loss of .4v)
3v --> 2.4v (loss of .6v)
4v --> 3.2v (loss of .8v)

So.. seems clear a consistent ratio of voltage is being lost somewhere. Can anyone smarter than I help chase this down?

JohnLRice
mantid wrote:1v --> .8v (loss of .2v)
2v --> 1.6v (loss of .4v)
3v --> 2.4v (loss of .6v)
4v --> 3.2v (loss of .8v)
Thanks much for your continuing research and data!

mantid
Common Wiggler
Posts: 55
Joined: Thu Aug 24, 2017 8:34 pm
I haven't heard back from ADDAC yet but given I would like to use this to control frequency voltage, those values are not going to work so well. Maybe I can make peace with it as a "rotating pitch shifter"?

I'm also surprised that Thonk hasn't followed up on my request to make sure their other kits are right - just trying to do my part to prevent this from happening to others.

JohnLRice please do attempt the repair, and anyone else who has purchased this, please let me know how your voltages look!