[Project] Turing Machine Mk2

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

Moderators: Kent, Joe., luketeaford, lisa

Post Reply
User avatar
synthetek
Wiggling with Experience
Posts: 499
Joined: Fri Jun 05, 2015 3:32 am
Location: Orange County, California
Contact:

Post by synthetek » Sun Jul 08, 2018 5:17 am

I would check the 3904 and the links between boards

User avatar
Bartelby
Common Wiggler
Posts: 195
Joined: Mon May 07, 2018 12:40 pm
Location: Penperlleni, Sunny Wales
Contact:

Post by Bartelby » Sun Jul 08, 2018 5:29 am

synthetek wrote:I would check the 3904 and the links between boards
Links all have continuity and base to the emitter on the 3904 it's reading around 0.7v

typicalsystem
Learning to Wiggle
Posts: 8
Joined: Wed Apr 19, 2017 3:10 pm

Post by typicalsystem » Wed Aug 01, 2018 1:56 pm

Looking for some advice on how to troubleshoot my build. This was my first attempt at euro DIY, so I'm guessing it's almost certainly going to be inexperience/soldering issues.

Module is powering up fine and I'm getting all the top LED's lighting up, as well as a working noise output. I seem to be getting a voltage output from the CV out, I can increase the voltage by turning the scale knob. Don't have a multimeter but I'm seeing this reflected by input led's when I send to my Maths/Disting.

However I'm getting no response at all when sending a square wave into the clock input and I have tried different sources and audio rates as suggested in this thread. Nothing happens when I turn the main dial, change the pattern length, and the clock LED does not light up.

I've tried to reflow solder around the components that I can see are in the clock input signal flow but that hasn't changed anything. Any tips for where I should be checking given that the module may be partially working?

Is the wisest move to invest in a flux pen and redo any vaguely iffy connections on both PCB's? I am just cautious that I may accidentally damage the board.

Appreciate any help suggested.

User avatar
emmaker
Veteran Wiggler
Posts: 711
Joined: Sat Mar 10, 2012 5:07 pm
Location: PDX

Post by emmaker » Wed Aug 01, 2018 3:02 pm

typicalsystem wrote:Looking for some advice on how to troubleshoot my build. This was my first attempt at euro DIY, so I'm guessing it's almost certainly going to be inexperience/soldering issues.

Module is powering up fine and I'm getting all the top LED's lighting up, as well as a working noise output. I seem to be getting a voltage output from the CV out, I can increase the voltage by turning the scale knob. Don't have a multimeter but I'm seeing this reflected by input led's when I send to my Maths/Disting.

However I'm getting no response at all when sending a square wave into the clock input and I have tried different sources and audio rates as suggested in this thread. Nothing happens when I turn the main dial, change the pattern length, and the clock LED does not light up.

I've tried to reflow solder around the components that I can see are in the clock input signal flow but that hasn't changed anything. Any tips for where I should be checking given that the module may be partially working?

Is the wisest move to invest in a flux pen and redo any vaguely iffy connections on both PCB's? I am just cautious that I may accidentally damage the board.

Appreciate any help suggested.
Check the components values and the diode polarity in the Clock input circuit on the 'front' schematic.

Do yourself a favor a buy a cheap DVM (can get for less than $10) and test the circuit. U1B pin 5 should change by putting the clock input high and low. It should be a little less voltage than the clock input voltage. U1B pin 7 should go high/low (maybe about +/- 10-9 volts) and the junction of D2/R12/R7 should go between high and gound.

User avatar
arthurdent
Veteran Wiggler
Posts: 663
Joined: Fri Nov 25, 2016 3:21 pm
Location: Ohio, along the North Coast

Post by arthurdent » Sun Aug 12, 2018 9:35 pm

I just spent the weekend building the TM MkII, it was the full kit from Thonk. This was the first time that I've ever built anything electronic. I had bought a practice board and some resistors from a local electronics place and spent several days last week practicing my soldering technique. The kit went together very well and so far, after the last 4 hours having it powered up, everything appears to be working OK. Yeah, tuning in the lock is tricky but I think I have it right - at least for now :nana: :nana:

metabaron24
Learning to Wiggle
Posts: 21
Joined: Sat Oct 28, 2017 11:14 am

Post by metabaron24 » Thu Aug 23, 2018 9:09 am

This should be a quick one to answer, but for the life of me I haven't been able to find an answer in this thread or anywhere else in the web.

So I've just built a Volts expansion rev3 PCB and I'm trying to figure out which of the two headers I connect to the Turing machine's Gates output and which one do I use to daisychain to more Volts expanders?? It says on the PCB: Either header will work, but does that really mean they both work as in or out.

And if I can't tell from what it says on the PCB what would be the way to figure it out without trying? (I'm afraid of frying something if I connect the Turing Machine's output to the output of the Volts.)

Am I missing something super obvious? Why has no one ever commented on this or asked the question? :despair:

User avatar
synthetek
Wiggling with Experience
Posts: 499
Joined: Fri Jun 05, 2015 3:32 am
Location: Orange County, California
Contact:

Post by synthetek » Thu Aug 23, 2018 2:52 pm

metabaron24 wrote:Am I missing something super obvious? Why has no one ever commented on this or asked the question? :despair:


Yes, either header will work. It doesn't matter just connect the gates header from the main module to either one on volts. iIf you want to chain another one use the available header on volts. just make sure the red stripe is orientated correctly on all of them. If you check with a multimeter the pins on both volts headers are connected together.

metabaron24
Learning to Wiggle
Posts: 21
Joined: Sat Oct 28, 2017 11:14 am

Post by metabaron24 » Thu Aug 23, 2018 3:02 pm

Fantastic! Thank you so much. Time to connect and play! :banana:

User avatar
arthurdent
Veteran Wiggler
Posts: 663
Joined: Fri Nov 25, 2016 3:21 pm
Location: Ohio, along the North Coast

Post by arthurdent » Fri Sep 07, 2018 5:14 pm

I just finished building the Turing Machine Expander VOLTAGES kit. I have power to it, for the most part it appears to be working, but I have a question about the two rotary pots.

As I understand it, the left-hand pot is for Scale adjustment, to adjust the scale/range of the sequence playing. This pot is supposed to be functional for both outputs - the regular and the inverted - CORRECT??

The right-hand pot is a shift/offset pot to be used if the inverted output is being used so that the output can be offset to drive positive-only-input modules. Therefore this pot is ONLY functional with the inverted output and has no effect on the regular output - IS THIS CORRECT?? Because right now, that's the way it's working on my module. So if this is not correct then I have a problem with my build.

User avatar
efm-7
Common Wiggler
Posts: 170
Joined: Thu Oct 15, 2015 11:55 am
Location: Canada

Post by efm-7 » Thu Nov 29, 2018 9:04 pm

Hi Everyone,
I'm having an issue where it does not loop when I have it "locked"

The step selector switch appears to do nothing.

When I have it locked in either direction, the LEDs are either all lit up or all off.

That being said - when I try to do a calibration, I never get anything resembling a constant tone - it's either random/noise or silence

I've done all the obvious checks such as the solder joints etc..

Please advise!

[video][/video]

Edit: It was a defective rotary switch

User avatar
Umeshu
Learning to Wiggle
Posts: 26
Joined: Thu Nov 12, 2015 3:43 pm
Location: France

Post by Umeshu » Sun Feb 10, 2019 4:49 am

Hi guys

This morning i was not completly wake up .
I tried to reinstall my Turing machine with pulse and volts in my case, but i made a mistake with the ribbon cables.
I plugged the out of the Pulses module into the Turing Machine power entry and so, the pulses out of the Turing Machine in a bus board. :doh:

The result was that nothing happend, no lights, and a little burning smell :bang:

I corrected my mistake and plug it again but now, all the Pulses out are at the same clock (like the main clock) and the Pulse out of my turing Machine does not work anymore.

So before i checked all the resistors and maybe other components, do you have an idea of what component could have burned?
Do i have to verify both my Pulse module and my Turing Machine or just the Turing Machine ?

Thanks in advance for your answers

c0rkindustry
1-Post Wiggler
Posts: 1
Joined: Mon Apr 01, 2019 12:15 pm
Location: Kentucky

Post by c0rkindustry » Sun Apr 07, 2019 11:28 pm

Hi
Just finished my mkii build and everything seems to be working with one exception. The third LED (at the top) is not lighting up in the sequence, or at power on. I've checked everything that i can think of with my multimeter and re-soldered a few joints, but no luck.
LED works through the multimeter, but can't get it to light via the link3 pin header (as the others do). Successfully checked resistance on all of the 2.2k resistors attached to the LEDs.
I'm stumped. Could it be a faulty PCB? Anyone have any other ideas?

This is only my second DIY build so it's entirely possible that i did something wrong :)

Thanks for your help!

EDIT: ...and if i had read through this entire thread i would have found that others have experienced similar issues with the 2nd LED.

moogah
Wiggling with Experience
Posts: 444
Joined: Mon Nov 22, 2010 10:27 am

Post by moogah » Mon Apr 08, 2019 1:39 pm

Does anybody have a set of PCBs and magpie panel for this? I really want to get my grubby mitts on one :)

Moudular
Learning to Wiggle
Posts: 8
Joined: Sat Mar 23, 2019 3:25 am
Location: Versailles

Post by Moudular » Thu Apr 25, 2019 4:16 pm

Hello

I've got an issue the Turing machine only works after I press Write
and couldnt make the tone calibration

the out works
the clock light and make the TM move (led and cv out)
Noise out (does it suppose to change like cv out ?)
pulse out when led1 turns black (or something like that)

from where should I try to cheek the build ?

[video] [/video]

mgams
1-Post Wiggler
Posts: 1
Joined: Sat May 03, 2014 1:42 pm
Location: Salt Spring Island, Canada

Post by mgams » Sat Jun 08, 2019 11:13 am

Calibrating Turing Machine MK2

I ran into the same problems as other wigglers: the margin between losing noise and getting a lock is very narrow. I had to turn the pot almost all the way CCW. I measured the value with a multimeter (unplugged the module!): app. 24kOhm, which is almost 1/40 of the pot value of 1MOhms. No wonder there is almost no way to adjust it properly.

My first thought was to change the pot to a 50kOhm (which you can do, it is a Bourns 3362P series pot) but I wanted a higher noise output. So I decided against it.

The pot adjusts the noise level to the noise output (via R19 1kOhm) and IC7B pin5 located on the TuringBack PCB. There it creates a clock signal depending on the position of the change pot via IC7A. If the noise level is to high it creates a clock signal even if the Change pot is fully CW. But if you reduce the noise level to IC7B pin5 you also reduce the noise
output level.

So all you have to do is reduce the noise level at IC7B pin5 without reducing the noise output level too much. There is a
convenient way to do it: insert a voltage divider (two resitors in series) instead of R11 100kOhm and connect the connection between them to pin1 of the Link1 connector on the TuringFront PCB (cut the trace leading to pin1 right next to it).

Values of the voltage divider: At first I used 2 47kOhm resistors which worked fine. But I wanted to set the "sweet spot" right at midnight of the adjusting pot. So I temporarily soldered in a 100kOhm pot after removing R11 and adjusted it to locking (the adjusting VR1 1MOhm pot being at midnight). Then I measured the resistor values of the temporary pot. In my case it was 33kOhm at the lower side (towards GND) and 67kOhm towards the higher side (at C11 0.47µF).

So my values for the voltage divider are 33kOhm and 68kOhm (standard resistor values) and the noise level going to IC7B pin5 is about 1/3 compared to before. You might get different results, though (depending on the 2N3904 transistor and other component variations).

So the Turing Machine locks, noise level is higher than before and it works great!
You do not have the required permissions to view the files attached to this post.

Wartimetapestry
Common Wiggler
Posts: 75
Joined: Sun Dec 23, 2018 5:36 am

Post by Wartimetapestry » Sun Jun 30, 2019 12:50 pm

Hey everyone. I hope you are all well.

I'm looking for a little advice. I've built the Turing Machine Mk II and I'm having trouble with the calibration. I only get a steady tone if the trimmer is fully CCW, leaving very little noise for random values. I've seen throughout this entire thread that a few people have had this issue. I cant, however, seem to see any fix, despite searching high and low and tracing every single pad with my multimeter.

From what I can tell, everything is hooked up right, correct values, no shorts etc. I'm following along with the calibration video but my CV output seems to have a lot more noise in the signal... by that I mean, I actually hear a lot of white noise coming through with the oscillation tone, with the change dial set at fully CW. Is this normal?

I've followed fully through the schematic but unfortunately I'm limited to a multimeter, so continuity and resistor checks are about as far as I've gone.

Two potential pitfalls - I've used generic ferrite beads and my electros are touching the top PCB when I join them together. Are either of these things likely to cause any issues?

Any advice on how any of you have solved this issue would be greatly appreciated. I've been pulling my hair out for the last few days, neglecting my relationships and not making music. It's a stress!

The solution right above this one seems like something I COULD do but since the circuit seems to work for most people as is, I'd rather really figure it out within those parameters. I'm up for learning, I'm just at a bit of a loss.

Thanks for your time.

User avatar
Umeshu
Learning to Wiggle
Posts: 26
Joined: Thu Nov 12, 2015 3:43 pm
Location: France

Post by Umeshu » Sun Dec 15, 2019 9:43 am

Hi guys, its been a long time that i had my problem with the Turing Machine when i pluged the pulses out of the Turing Machine in the bus board. (see my post above)
The result was a little burning smell and the pulse of the turing machine that does not work anymore.

I decided today to test all resistors and diodes of the expander board, but all seems fine.
I don't know how to test other components.

Could the problem be with theCD4081BE chips ?

Thanks in advance for your answer

User avatar
emmaker
Veteran Wiggler
Posts: 711
Joined: Sat Mar 10, 2012 5:07 pm
Location: PDX

Post by emmaker » Sat Dec 21, 2019 12:13 am

Umeshu wrote:Hi guys, its been a long time that i had my problem with the Turing Machine when i pluged the pulses out of the Turing Machine in the bus board. (see my post above)
The result was a little burning smell and the pulse of the turing machine that does not work anymore.

I decided today to test all resistors and diodes of the expander board, but all seems fine.
I don't know how to test other components.

Could the problem be with theCD4081BE chips ?

Thanks in advance for your answer
Hard to say what burnt up.

Do a visual inspection under good light and with a magnifier looking for what burnt. If that doesn't work actually smell the board and see if you can find an area that has that burnt smell. That may show you the area the component is in and maybe get you to the actual component.

User avatar
Umeshu
Learning to Wiggle
Posts: 26
Joined: Thu Nov 12, 2015 3:43 pm
Location: France

Re: [Project] Turing Machine Mk2

Post by Umeshu » Tue Dec 24, 2019 4:23 am

Here are pictures of the two fuse on the expander outputs board. I don't know if it was like that before the incident, i have no experience of the burn of these type of components and don't know how to test them. Could it be the cause of my problem ? If yes i will order new ones and change them

Image
Image

Synthiq
Veteran Wiggler
Posts: 595
Joined: Mon Feb 06, 2017 3:10 pm
Location: California

Re: [Project] Turing Machine Mk2

Post by Synthiq » Tue Dec 24, 2019 2:22 pm

I don't have any personal experience with these specific fuses, but I'm sure they shouldn't look like that. If you have a multimeter, measure the resistance between the pins on both of them. According to the specs, the small one should have around 5ohms and the larger one 10ohms. If you see values far from these, I would definitely replace them. But if there was enough current flowing to cause this damage to the fuses, there is a good probability that something else has been exposed to the same currents and has met the same fate even if it isn't directly visible.

User avatar
Umeshu
Learning to Wiggle
Posts: 26
Joined: Thu Nov 12, 2015 3:43 pm
Location: France

Re: [Project] Turing Machine Mk2

Post by Umeshu » Sun Feb 02, 2020 4:06 am

Thanks for your answer Synthiq.
I measured the value of the 2 fuses and while the bigger is 10ohm, the small one is 5.9 but it's maybe a normal value.
I will order new fuses to replace them and hope it will correct the issue.
I will let you know if it correct the issue.

User avatar
rduif
Learning to Wiggle
Posts: 22
Joined: Thu Sep 06, 2018 8:45 am
Location: Rotterdam, Netherlands

Re: [Project] Turing Machine Mk2

Post by rduif » Sun Mar 08, 2020 4:14 pm

Hi everyone,

I've made my first turing machine mkII and I'm having some issues that I don't understand, so I'm hoping someone can point me in the right direction. I've been through this complete thread but it's still not clear to me where the problem lies exactly.

My module started up fine, all leds are working and it loops when clocked both by VCO and normal clock. I was able to get it calibrated without losing the noise output completely. At 12h the sequence would be random, but at both 5h and 7h it would always lock in the same sequence. At 5h (8steps) it would be 11000000 and at 7h 11000000 00000011, which follows what I understand from the module (apart from it being the same sequence all the time.

Since then ive:
1) checked all resistors
2) reflowed all joints
3) swapped CD4050 and CD4081 to see if behaviour changed, which it did not
4) swapped 1M trimmer with 50K trimmer to help with finetuning calibration

Upon further inspection, I seem to hear some repeated sequences namely a ramp up from 0V and some periods of 0V. It seems that it loses bits sometimes and 'resets' other times. This happens when calibrated but also when introducing more noise. Im quite sure that there are no faulty connections as Ive check everything multiple times. I can only imagine one of the IC's is bad (Ive sources myself, some from china), but as I dont really understand how sequences are generated, I dont know which one it would be. Does anyone have a clue based on the information above?

Thanks in advance!

User avatar
rduif
Learning to Wiggle
Posts: 22
Joined: Thu Sep 06, 2018 8:45 am
Location: Rotterdam, Netherlands

Re: [Project] Turing Machine Mk2

Post by rduif » Mon Mar 16, 2020 9:11 am

Hi everyone!

Im still having problems and Im running out of ideas. I hope someone can help me out. I feel a bit like everytime I take it apart and retest it, the behaviour changes but remains clearly wrong.

The current status when clocked by a pulse wave at audio frequency from Zlob Dual VCO:

With change knob fully clockwise, it seems like it only outputs cv intermittently. the first led lights very faintly irregardless of the trimmer setting (which is currently a 50K btw). Every second or so, a stream of cv does seem to come out and all leds light up, but then it stops again. Putting the trimmer fully clockwise seems to make this occur more often, but for the most part it just stays at 0.

When I put the change knob fully counter-clockwise however, all leds light up and I seem to be able to calibrate as normal (remove glitches until steady tone is achieved). However, when I reduce the clock rate, I find that it just reverts to the same sequence every time. I have plenty of noise on the noise output at this stage, so it is not trimmed too far. And putting the change knob at noon gives me a seemingly random datastream every time.

Does anyone have any recommendations? Ive tried to look at the schematics to see if this behaviour can be explained in some way but Im a bit out of my depth there. Im a step away from just changing every IC on there but Id rather learn to find the root cause (also I dont have spares for everything).

Cheers

User avatar
arthurdent
Veteran Wiggler
Posts: 663
Joined: Fri Nov 25, 2016 3:21 pm
Location: Ohio, along the North Coast

Re: [Project] Turing Machine Mk2

Post by arthurdent » Mon Mar 16, 2020 6:47 pm

This might be a long shot but you can give it a try. When I built my TM I had trouble with it when using only certain VCO's as clock sources but no trouble with others. Here's the link where I discussed it and got help, maybe it will work for you :despair: :despair:

viewtopic.php?t=207175&highlight=

User avatar
rduif
Learning to Wiggle
Posts: 22
Joined: Thu Sep 06, 2018 8:45 am
Location: Rotterdam, Netherlands

Re: [Project] Turing Machine Mk2

Post by rduif » Wed Mar 18, 2020 1:25 pm

arthurdent wrote:
Mon Mar 16, 2020 6:47 pm
This might be a long shot but you can give it a try. When I built my TM I had trouble with it when using only certain VCO's as clock sources but no trouble with others. Here's the link where I discussed it and got help, maybe it will work for you :despair: :despair:

viewtopic.php?t=207175&highlight=
Thanks for the tip! I've tried to run the oscillator through an lpf (AI synthesis OTA VCF) as suggested in your topic. Unfortunately now the TM simply doesnt clock at all, even when I boost the output of the lpf (which is by default quite low) to appropriate levels. Soooo now I just have more questions haha. I also tried using the sine and triangle waves btw, which made no difference in the behaviour.

I did clock the TM from a few other sources, namely the clock out of a drumbrute at 1ppq (which thonk told me should work properly) and just a looping envelope. It definitely changes the behaviour somewhat (though its difficult to compare because the clock rates are not audio range) but in the end it reverts to the same problem.

Post Reply

Return to “Music Tech DIY”