Turing Machine Old vs. New - Differences

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

Moderators: Kent, Joe., luketeaford

Post Reply
User avatar
galingong
Common Wiggler
Posts: 86
Joined: Tue Feb 04, 2014 2:46 am

Turing Machine Old vs. New - Differences

Post by galingong »

I'm planning to build a Turing Machine soon, so I've been looking at the schematics. Since it was updated just recently, I thought it makes sense to build the new version. Well, turns out, not necessarily.

So the new version adds 2 "AND-Gate" ICs (4081) and two "Buffer" ICs (4050). What they do, is they produce "Pulse" and "Gate" signals from B1-B8 signals, and output them to headers on the back of the PCB - for expansion board purposes I guess. Only one of the 8 "Pulse" signals is routed to the front panel as "Pulse Out". It probably spits out a pulse, I'm not sure when exactly though.

An other modification is that now the circuit utilizes all 4 units in the 4016 Switch IC. Can anybody tell me what's the reason behind this?

Also in Mark 2 an additional 4015 Shift Register IC is added, the outputs of which seem to be not connected. This is also beyond me.

The "Length" switch is also replaced by a rotary switch, that allows more control over loop length. This is a really simple modification, and actually the only feature from the new version that I'd like to implement in my build.

Would it be a better idea to use the old schematic and route B1-B8 and "End of Loop" to a rotary switch to get the new "Length" functionality, or use the new schematic, and get rid of the unused expansion-purposed circuitry? I guess it's a long shot, and the one who would be able to answer this is probably the designer or someone who knows the circuit inside-out, but worth a try. Thanks!

edit: I design and etch my PCB, that's why my dilemma is relevant.
Grumskiz
Wiggling with Experience
Posts: 346
Joined: Thu Apr 03, 2014 12:51 pm
Location: Germany

Post by Grumskiz »

The Turing Machine really comes alive when you have any or some of the expanders. The original module required an additional PCB called the "backpack" to hook up expanders. This "backpack" has been merged into the main PCB on the 2nd version which I believe is why a couple of ICs have been added, but don't seem to do anything.

If you really want to design and etch your own PCB, I think you should add the expansion ports to it. Or you could merge the Turing Machine and expander designs on one PCB if you have enough space.
User avatar
galingong
Common Wiggler
Posts: 86
Joined: Tue Feb 04, 2014 2:46 am

Post by galingong »

Thank you, I initially planned to combine it with a Barton CV Quantizer. To be honest I wasn't aware what the expanders did, but will look into it.

On the old board the expander header only received the B1-B8 signals and 1-2 additional ones, and I have access to those even if I leave out the recently added 4 ICs, so using expanders should not be a problem. Although rack space will be :D
User avatar
Tombola
Super Deluxe Wiggler
Posts: 1879
Joined: Sun Jan 23, 2011 11:33 am
Location: London, England

Re: Turing Machine Old vs. New - Differences

Post by Tombola »

galingong wrote: So the new version adds 2 "AND-Gate" ICs (4081) and two "Buffer" ICs (4050). What they do, is they produce "Pulse" and "Gate" signals from B1-B8 signals, and output them to headers on the back of the PCB - for expansion board purposes I guess. Only one of the 8 "Pulse" signals is routed to the front panel as "Pulse Out". It probably spits out a pulse, I'm not sure when exactly though.
These power the expanders, were previously in the backpack

There are also several changes in the clock circuit and the noise circuit
The DAC circuit now has an proper voltage reference
The entire switching circuit operates at a lower voltage, further from the rails
There's a trimmer for the noise level, which makes locking more reliable
galingong wrote: An other modification is that now the circuit utilizes all 4 units in the 4016 Switch IC. Can anybody tell me what's the reason behind this?
The old design had a couple of transistors + resistors doing this job, so it saves components
galingong wrote: Also in Mark 2 an additional 4015 Shift Register IC is added, the outputs of which seem to be not connected. This is also beyond me.
Both old and new have 2 x shift registers to make 16 steps. One has connections on all the outputs, the other only has connections to the first and last gate.
galingong wrote: The "Length" switch is also replaced by a rotary switch, that allows more control over loop length. This is a really simple modification, and actually the only feature from the new version that I'd like to implement in my build.
Would it be a better idea to use the old schematic and route B1-B8 and "End of Loop" to a rotary switch to get the new "Length" functionality, or use the new schematic, and get rid of the unused expansion-purposed circuitry?
Build whichever you like, if you don't ever want to use the expanders, cut them off. If it was me, I'd definitely start with the new schematic, because there are numerous revisions all through the circuit.
galingong wrote: the one who would be able to answer this is probably the designer
I'm not that hard to find!
User avatar
galingong
Common Wiggler
Posts: 86
Joined: Tue Feb 04, 2014 2:46 am

Post by galingong »

Hi Tom,

Thank you for the detailed answer, last night I managed to put it together but did not have much time for testing. I've built the new circuit simply leaving out the 4 mentioned ICs, I don't plan to use expanders, but if for some reason I would, I have access to needed signals through headers.

It seems to be working fine, I'll have to experiment with it though. What I've figured out is that exact trimming is needed, at first I thought it didn't work until I started turning the trimmer (good thing I've used a 16-turn, I don't always do).

What's the proper way to set the trimmer? Is there a voltage I should test on the board, or just "turn it until works"? :)

Also I'm sure I'm not the first nor the last one, but thank you for creating and sharing such an amazing piece of hardware.
User avatar
Tombola
Super Deluxe Wiggler
Posts: 1879
Joined: Sun Jan 23, 2011 11:33 am
Location: London, England

Post by Tombola »

galingong wrote: What I've figured out is that exact trimming is needed, at first I thought it didn't work until I started turning the trimmer (good thing I've used a 16-turn, I don't always do).

What's the proper way to set the trimmer? Is there a voltage I should test on the board, or just "turn it until works"? :)
Trimming doesn't need to be super exact, the kits use a 1-turn trimmer.

Clock in, ouput to an oscillator in audio range, scale so it's comfortable to listen to.
Set the main pot to 12 o'clock - you should have fully random sequences
Set the pot to 5 o'clock - you should have a sequence that locks/repeats

The trimmer sets the point at which locking occurs.

You can also test it at Audio Rate

DivKid's video is the best manual of how it's supposed to work:
User avatar
galingong
Common Wiggler
Posts: 86
Joined: Tue Feb 04, 2014 2:46 am

Post by galingong »

OK, everything is perfect, amazing stuff really. This is how it turned out in 7HP (because rack space).
Not the most beautiful thing in the world but anyway, does the job and fits. Thanks again for your help, cheers!

Image
User avatar
Tombola
Super Deluxe Wiggler
Posts: 1879
Joined: Sun Jan 23, 2011 11:33 am
Location: London, England

Post by Tombola »

galingong wrote:OK, everything is perfect, amazing stuff really. This is how it turned out in 7HP (because rack space).
Not the most beautiful thing in the world but anyway, does the job and fits. Thanks again for your help, cheers!

Image
That's really beautiful
Grumskiz
Wiggling with Experience
Posts: 346
Joined: Thu Apr 03, 2014 12:51 pm
Location: Germany

Post by Grumskiz »

Wiring looks like a lot of work, but damn that panel is nice and compact!
DSC
Super Deluxe Wiggler
Posts: 2295
Joined: Thu Apr 11, 2013 11:54 pm

Post by DSC »

Going the other way with the largest panel possible would be the one I just finished. Clocking in at 39HP, this includes the 'Bytes' Expander from Mystic Circuits and a dedicated clock from Ginko.

Image
Last edited by Flatscan on Tue Jan 05, 2021 3:46 am, edited 1 time in total.
Reason: undoing the damage
User avatar
james2098
Common Wiggler
Posts: 97
Joined: Mon Oct 26, 2015 4:04 pm
Location: uk

Post by james2098 »

DSC wrote:Going the other way with the largest panel possible would be the one I just finished. Clocking in at 39HP, this includes the 'Bytes' Expander from Mystic Circuits and a dedicated clock from Ginko.

Image
that looks great!
Post Reply

Return to “Music Tech DIY”