TTSH Ver.3 (rev.8) General build thread

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

Moderators: luketeaford, Joe., lisa, Kent

Post Reply
KSS
Super Deluxe Wiggler
Posts: 2698
Joined: Mon Jan 25, 2016 7:28 am

Re: TTSH Ver.3 (rev.8) General build thread

Post by KSS » Wed Jul 08, 2020 11:24 pm

It's too bad Jon removed the ability to easily take different parts of the later TTSH's offline. That makes finding an issue like this *much* easier.

Not as a first step, but there *are* ways to isolate different 'modules' of a 4 layer PCB'd TTSH.

Go fact finding first. Stop trying to fix "the" problem and instead concentrate on *fully* understanding *your* project. A notebook would be a good tool. Especially if you could look back through it right now to see clearly every thing you've already done. And the results of each thing you did. Both the at first results, AND the later issues *after* those initial results.

User avatar
LED-man
Super Deluxe Wiggler
Posts: 1470
Joined: Wed Aug 08, 2012 1:48 pm
Location: Wolfsburg Germany

Re: TTSH Ver.3 (rev.8) General build thread

Post by LED-man » Thu Jul 09, 2020 1:44 am

Amir,
I think we gave you the info to separate the power of each section by cutting the traces and installing jumpers at the 2x3 pin pads. This helps you to find the faulty section/area.
Furthermore, are you sure that your LM301 are fine ?
I got few months ago 100 fake LM301 from a trusted eBay seller, they worked limited - I got signals in and out but the current was 40/50mA per IC and the circuit starts to swing.
################################
TTSH Support Page and other DIY Projects:
https://www.DSL-man.de

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Thu Jul 09, 2020 5:22 pm

LED-man wrote:
Thu Jul 09, 2020 1:44 am
Amir,
I think we gave you the info to separate the power of each section by cutting the traces and installing jumpers at the 2x3 pin pads. This helps you to find the faulty section/area.
Furthermore, are you sure that your LM301 are fine ?
I got few months ago 100 fake LM301 from a trusted eBay seller, they worked limited - I got signals in and out but the current was 40/50mA per IC and the circuit starts to swing.
Thanks guys! The lm301 were originally from mouser so i am certain those are good. I bought some more recently from jameco and they seem to be working fine but ill swap some around (not randomly) to see if it makes a difference. I want to avoid cutting traces if possible.

KSS: I get what you are saying and thanks. I am not doing this at random. I am continuously measuring everything with the synth on. Everything seems fine except the final outputs for the VCOs and noise seem to be slightly high.

Today I rescanned the entire board with it on and working (gate switch down). things I have found:

1: ground pin all over the board shows about 1.49000mv, it does not fluctuate

2: On vco 2 I followed the saw out of the sub board and at the 1k resistor it is measuring exactly a perfect 10vpp. After that point as it starts entering the opamps and surrounding circuits it gradually increases to 10.7vpp.

3: I started to trace the noise section going to the pwm of vco2 to see if there was anything fishy going on around it at VCO2. I found that it is 20vpp at the socket as well. However, while I was in this section I decided to check the pins on the sliders, I suddenly noticed that the E pin of all the sliders in VCO2 are at negative voltages while in other section they are at positive voltages. The E pin based on the data sheet is supposed to be the negative side of the led. Any idea why this would be reversed only in one section (VCO2)? the pins don't look shorted, I have checked the other sections as well. I have in the past disconnected the led power section completely and it hasnt changed anything in VCO vpp outputs or the noise output but haven't done so with the new sub boards that I built. The led section of the board really confuses me and I honestly wish it wasn't even there. Do you think I have bad leds in there or one of the bc337 bad? Dont the oscillators share led load?


I hope I am finally on to something and I really appreciate all the help from all of you! thanks :)

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Sat Jul 11, 2020 8:48 pm

Made some progress today. I decided to check the voltages at top of the electrolytic caps and I know this isnt a sure way to figure out if they are working or not but comparing all of them together I found two sections that were way off. In the filter section the negative cap was at around -13.6V at the top and the positive was at .8v, most places around the board these were -14v and 128mv. In the noise section they were -10.8v and .65v. I took them out and replaced them with new 1uf caps (I only had a few left). Afterwards I was getting -14v and same 128ish mv! The frequency range immediately changed on the oscillators down and I had to use the trimmer to get them back within range which wasn't an issue. There are still a few that I need to change but will have to reorder new caps. I noticed both in ring mod as well as all oscillator sections that these readings were .3v and -14v. Which makes me think the positive sides are leaking a bit also.

Any idea what would make them leak like this all of the sudden? My noise vpp and osc vpp are still high but I am guessing/hoping this will change once I swap the other caps as well. They seem a lot more stable now tuning wise. These were the original caps that were in the bom. I ordered some ELNA caps, they were the only ones with correct lead spacing and voltage rating I wanted. Will test some more once I have them.

KSS
Super Deluxe Wiggler
Posts: 2698
Joined: Mon Jan 25, 2016 7:28 am

Re: TTSH Ver.3 (rev.8) General build thread

Post by KSS » Sun Jul 12, 2020 12:12 am

Nice work! Now you're getting it. There's a reason long time tech's stress the use of GOOD quality caps without getting too much involved in the recap wars.

Caps on the power rails in synths need to be high quality. It's not the place to try and save a little money. It's also not the place to buy into audiophoolery and try to get 'magic' caps. Nichicon's designed for power rails are fine.

Caps die from any number of reasons -usually heat, voltage stress or early mortality- and it would be tough to diagnose from afar.

But I *do* think you're finally getting to -at least one of- the root issues.

I hope it doesn't sound condescending to say I'm proud of you!

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Sun Jul 12, 2020 10:18 am

Thanks, I've been messing with this thing for so long that even a small step forward is amazing! I've had less trouble building MI modules and I've built most of those. The size of this board makes it impossible to find problems like this. :)

These were the caps that were in the original mouser bom: https://www.mouser.com/ProductDetail/647-UPM1H010MDD1TD
I used them and it says they have 2000 hour life span. Not sure if that's a long time but I have definitely not used the synth that many hours. I have it in the original metal case but I am using the altitude linear power supply. Maybe its getting too hot in there.

The only 2.5mm lead spacing 50v caps that they had in stock were the ELNA (mouser: 555-RFS50V010MF3#5). I've read a lot about caps in the past and think these are good for the application. I don't know much about ELNA and what makes these audio caps though. Will they be ok on power rails? Are they magical caps?

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Sun Jul 26, 2020 9:33 am

Some updates/question:

I got the ELNA caps. They are massive but fit. So far I've replaced all the caps that weren't the same as the others. VCO 1 & 2 are outputting around 10.5vpp now on their square outs. VCO3 is still acting funny though. I noticed that the PWM slider is a little off. The slider at exactly half way is showing up as a 40%-ish square. VCO2 is very close to 50% at same position. When I have the slider all the way to the right VPP goes down to 10.4vpp and when I have it all the way to the left it goes up to 11.5vpp. Something is still off somewhere. :bang: I don't think it is actually within VCO3 because I have changed the opamp, diodes... The only thing that I've noticed different is that the balancing outputs of the opamp seem to be fluctuating a lot while they dont on vco 1. VCO2 doesn't have this cap. I've never dealt with a bad slider, but could this be the result of a bad pwm slider?

KSS
Super Deluxe Wiggler
Posts: 2698
Joined: Mon Jan 25, 2016 7:28 am

Re: TTSH Ver.3 (rev.8) General build thread

Post by KSS » Sun Jul 26, 2020 9:58 am

Bad slider is possible, but unlikely.

Measure the voltage at the wiper in middle and at both ends. Compare that with the reference voltage going into the PWM comparator AKA square wave 'symmetry' trim on VCO1

In each case look at the resistors going from the slider ends to rails and GND. These determine the 'center' point and range of slider operation and are not necessarily optimal for all individual cases. They're statistically correct enough for a production synth.

Here's an important question. When you use your TTSH, how does it sound? What is it like to play with it?

Sometimes chasing perfection in specs makes us miss what really matters. Baseball was still fun before Sabermetrics.

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Sun Jul 26, 2020 12:11 pm

KSS wrote:
Sun Jul 26, 2020 9:58 am
Bad slider is possible, but unlikely.

Measure the voltage at the wiper in middle and at both ends. Compare that with the reference voltage going into the PWM comparator AKA square wave 'symmetry' trim on VCO1

In each case look at the resistors going from the slider ends to rails and GND. These determine the 'center' point and range of slider operation and are not necessarily optimal for all individual cases. They're statistically correct enough for a production synth.

Here's an important question. When you use your TTSH, how does it sound? What is it like to play with it?

Sometimes chasing perfection in specs makes us miss what really matters. Baseball was still fun before Sabermetrics.
Thank you, I will take a look at those values later today and see if I can find anything. The synth sounds amazing and seems to be working and tracking fine. I am more worried about something still causing a short and later causing damage down the line somewhere else that will take forever to find when the synth stops making sounds all of the sudden again :)

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Sun Jul 26, 2020 7:06 pm

KSS,

I tried checking the trimmers against each other and honestly couldn’t really make a comparison. The circuits between each oscillator for the pulse wave is very different. I am assuming lowering one of the two resistors a few hundred ohms would fix it to be center. The range is a few percent wider than oscillator 2.

That aside I think I finally made a discovery of what actually went wrong and need your help because I am not familiar with the ring mod circuit at all. I never even use it unless I am using it as an extra vca. I started calibrating it and noticed that the positive null trimmer isn’t making any difference on the output voltage. More importantly it’s stuck close to 2V!!! The original service manual says that it shouldn’t output more than 10mv.

I don’t even know where to begin with this one. I started tracing the output of the trimmer to the resistors and it was around -6v up to that point. After the resistors it drops down to millivolts. I am assuming one of the transistors is dead and causing the issue but not sure. I changed the opamps and it didn’t make a difference. I tested as a vca and it was working as it should and on the audio side I see manipulated waves coming out but not sure where that extra 2V is or where it’s going. Any of you guys really familiar with that part of the board? Any ideas where I could start?

Edit: never mind I realized the trimming instruction on Patrick’s website are wrong for the positive null. You are supposed to raise vco2 slider not vco1 slider like his instructions say. With vco2 up, nothing happens when I turn the trimmer still but it is already at -1.4mv. Something is either right or wrong here :)

KSS
Super Deluxe Wiggler
Posts: 2698
Joined: Mon Jan 25, 2016 7:28 am

Re: TTSH Ver.3 (rev.8) General build thread

Post by KSS » Mon Jul 27, 2020 6:23 am

Where did you get your trimmers? If part of a kit, ask the provider. I'd be far more skeptical of trims than OPAs. There are some really crappy trimpots floating around the usual sellers.

Feed the sine into both inputs of the RM. It should double freq and look symmetrical.

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Mon Jul 27, 2020 9:30 am

The trimmer checked out fine. It’s from mouser/bom. It has positive rail on one side and negative on the other. The output is adjusting and sitting around -6v right now but nothing changes in the RM output Jack as I turn it.

I am having a really difficult time with this because checking the rails, I am not seeing any shorts across the board. All the rails show around 500-600ohm of resistance against each other. But some component is in a failed state and incorrectly routing voltage somewhere.

edit: v/oct tracking is working. I was following bad instructions again. Using the original arp instructions it is working and locking. :)

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2081
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Re: TTSH Ver.3 (rev.8) General build thread

Post by fuzzbass » Mon Jul 27, 2020 2:25 pm

amir wrote:
Sun Jul 26, 2020 9:33 am
I noticed that the PWM slider is a little off. The slider at exactly half way is showing up as a 40%-ish square. VCO2 is very close to 50% at same position.
Every TTSH is like this, usually a bit more off-center for VCO3 than VCO2, but both off-center. Every time I kvetch about this, I put the phones on and listen, decide I can't really hear it very much. But as KSS says, you can sub a resistor somewhere here to put 50% duty cycle right dead center on the slider if you want to. Also not at all unusual for the saw and pulse outputs of the VCOs to be up to +/- 10% amplitude. Literally every single place you can connect these is a gain stage or mixer, so little differences don't matter. If you think your VCOs are not hot enough, adjust up the gain trimmer on the VCF, VCA, ect. Same thing with outputs of the EGs...
Wired for weird

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Mon Jul 27, 2020 3:45 pm

fuzzbass wrote:
Mon Jul 27, 2020 2:25 pm
amir wrote:
Sun Jul 26, 2020 9:33 am
I noticed that the PWM slider is a little off. The slider at exactly half way is showing up as a 40%-ish square. VCO2 is very close to 50% at same position.
Every TTSH is like this, usually a bit more off-center for VCO3 than VCO2, but both off-center. Every time I kvetch about this, I put the phones on and listen, decide I can't really hear it very much. But as KSS says, you can sub a resistor somewhere here to put 50% duty cycle right dead center on the slider if you want to. Also not at all unusual for the saw and pulse outputs of the VCOs to be up to +/- 10% amplitude. Literally every single place you can connect these is a gain stage or mixer, so little differences don't matter. If you think your VCOs are not hot enough, adjust up the gain trimmer on the VCF, VCA, ect. Same thing with outputs of the EGs...

Yea, my vco2 is dead center but vco3 is almost 3/4. The only output that is too high is the square off vco3. its outputting 11.5v with pwm slider to the left and 10.4 with it all the way right. I don't understand why they made the circuit for the square out of vco3 so different from vco1. With the cores removed, vco1 output at the square is 1.5mv but coming out of vco3 is ~-13v because of the 1M resistor going to the negative rail. It doesn't make sense if they are both outputting the same thing out of the core to have that there just because vco3 is using a slider for the pwm amount vs vco using a trimmer. Both are 100k. I've read things in the past about designers doing this to confuse people back then who wanted to copy the circuit. I wonder what would happen if I removed that resistor. Probably get something much closer to vco1's square out.

side note, I've finished 2 peaks, a shades, and two more ripples while fixing this on the side with no issues. (hopefully i don't jinx them) but jesus the sound out of the TTSH doesn't even compare to the eurorack stuff. It is a monster. :D

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Mon Jul 27, 2020 5:45 pm

KSS wrote:
Mon Jul 27, 2020 6:23 am
Where did you get your trimmers? If part of a kit, ask the provider. I'd be far more skeptical of trims than OPAs. There are some really crappy trimpots floating around the usual sellers.

Feed the sine into both inputs of the RM. It should double freq and look symmetrical.

Sine freq doubled but positive null trimmer still has no impact on output. Seems the output is grounded because it is reading the same 1.5ish mv that I get at ground across the board.
You do not have the required permissions to view the files attached to this post.

KSS
Super Deluxe Wiggler
Posts: 2698
Joined: Mon Jan 25, 2016 7:28 am

Re: TTSH Ver.3 (rev.8) General build thread

Post by KSS » Mon Jul 27, 2020 6:47 pm

Does the Ring Mod Audio/DC switch position make any difference?

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Mon Jul 27, 2020 7:02 pm

KSS wrote:
Mon Jul 27, 2020 6:47 pm
Does the Ring Mod Audio/DC switch position make any difference?

Not for the positive null. Negative null it does. It calibrates only in the correct setting. But the positive voltage at the does not move. It’s frozen at about 1.5mv.

KSS
Super Deluxe Wiggler
Posts: 2698
Joined: Mon Jan 25, 2016 7:28 am

Re: TTSH Ver.3 (rev.8) General build thread

Post by KSS » Mon Jul 27, 2020 7:18 pm

I meant does the double fre sine output image on your scope of the dual sine input jump vertically when the Audio-DC slide switch is moved?

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Tue Jul 28, 2020 1:22 pm

KSS wrote:
Mon Jul 27, 2020 7:18 pm
I meant does the double fre sine output image on your scope of the dual sine input jump vertically when the Audio-DC slide switch is moved?
Nothing happens on the scope when I flip the switch but the audio coming out does change. When I flip it to dc it sounds like I’ve added a sub oscillator. Also when used as a vca with the cv going into the second jack, it kind of modulates the sound in the audio position but works as it should once I flip it into dc. Same with the cv patched into the first jack but it seems to bleed a little bit of audio that way. The envelope doesn’t fully close.

User avatar
hofmann25
Common Wiggler
Posts: 54
Joined: Thu May 05, 2016 6:12 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by hofmann25 » Tue Jul 28, 2020 8:44 pm

Hey All, haven't been here in a while.

My TTSH recently started behaving strangely.

CV into KBD CV no longer plays in tune. I can tune the oscillators but going up and down the keyboard the keys are no longer in tune.

BUT, multing the CV and then patching into the Oscillators individual KBD CV everything is fine...

I have the MIDI2CV hard wired to the KBD CV also but this is the first time I've experienced this oddity.

Anyone experienced this?

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Tue Jul 28, 2020 10:57 pm

hofmann25 wrote:
Tue Jul 28, 2020 8:44 pm
Hey All, haven't been here in a while.

My TTSH recently started behaving strangely.

CV into KBD CV no longer plays in tune. I can tune the oscillators but going up and down the keyboard the keys are no longer in tune.

BUT, multing the CV and then patching into the Oscillators individual KBD CV everything is fine...

I have the MIDI2CV hard wired to the KBD CV also but this is the first time I've experienced this oddity.

Anyone experienced this?
Measure your cv jack without anything plugged in. If the voltage is fluctuating then likely your midi cv device has a failed component feeding into the line. To be sure disconnect and measure again. If you still see fluctuations then you likely have an issue on the board. Then things get trickier.

User avatar
hofmann25
Common Wiggler
Posts: 54
Joined: Thu May 05, 2016 6:12 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by hofmann25 » Thu Jul 30, 2020 4:29 am

amir wrote:
Tue Jul 28, 2020 10:57 pm
hofmann25 wrote:
Tue Jul 28, 2020 8:44 pm
Hey All, haven't been here in a while.

My TTSH recently started behaving strangely.

CV into KBD CV no longer plays in tune. I can tune the oscillators but going up and down the keyboard the keys are no longer in tune.

BUT, multing the CV and then patching into the Oscillators individual KBD CV everything is fine...

I have the MIDI2CV hard wired to the KBD CV also but this is the first time I've experienced this oddity.

Anyone experienced this?
Measure your cv jack without anything plugged in. If the voltage is fluctuating then likely your midi cv device has a failed component feeding into the line. To be sure disconnect and measure again. If you still see fluctuations then you likely have an issue on the board. Then things get trickier.
Oh I should have stated. CV from my Rene MKII into the KBD CV yield the same errors as the Midi2CV.

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Thu Jul 30, 2020 12:50 pm

If you have anything internally connected to the cv input, remove it and measure the jack and hope that is the issue because just like my issues above with the oscillators, if it’s a failed component anywhere on the main board... it’s next to impossible to find without butchering the power lines on these new boards before you can isolate the problem.

If it comes down to it, I would trace the cv line where it is normalled and start looking for problems in those areas first. For example, try plugging dummy cables in all cv inputs (filter, osc 1, osc 2) then input your cv at the normal keyboard jack. Turn up only osc 3 and see if you still have same issue. Keep swapping around the dummy plugs and test each part on its own. If the issue goes away when you place a dummy in a particular combo then remove them one by one until problem returns and you will at least have an idea of where the problem coming from.

User avatar
hofmann25
Common Wiggler
Posts: 54
Joined: Thu May 05, 2016 6:12 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by hofmann25 » Thu Jul 30, 2020 7:39 pm

Thanks Amir. I will try that.

amir
Common Wiggler
Posts: 199
Joined: Thu Dec 15, 2016 12:59 pm

Re: TTSH Ver.3 (rev.8) General build thread

Post by amir » Tue Aug 04, 2020 10:13 am

I wanted to post an update about the ELNA caps. Caps definitely do make a difference. I am not done replacing all ~30 electrolytic caps. Still have the envelope and vca sections left and will do measurements after. If you think soldering 30 caps takes time, desoldering and then soldering is 10x worse. I found that dragging a bit of solder across both leads allows them to easily drop under the board on their own after a few seconds. Then its a matter of using the whick to empty the holes and cleaning after before soldering the new ones in.

On to the sound. Its definitely gotten deeper. At first I thought it was in my head but its not. To me most arp 2600s including videos of other ttsh that I have seen sound very tin like. Almost a violent metallic sound, like something is off. Thats what this changed. Still sounds violent but much more low end heavy until I increase the resonance slider of course. You can feel the dynamics even on those little tiny built in speakers. They improved greatly in sound as well. :)

Post Reply

Return to “Music Tech DIY”