[BUILD] Bytes Turing Machine expander

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

Moderators: luketeaford, Joe., lisa, Kent

david_r
Wiggling with Experience
Posts: 476
Joined: Fri Dec 12, 2014 12:23 am
Location: SF Bay Area

Post by david_r » Wed Jun 03, 2015 10:20 pm

Great expander! And great service from windspirit/circuitshaman for building and making the appropriate modifications to my TM.

Can't wait to really sit down with it and play with the loop length modulation.

User avatar
windspirit
Super Deluxe Wiggler
Posts: 3050
Joined: Thu Sep 20, 2012 11:43 pm
Location: krellifornia

Post by windspirit » Thu Jun 04, 2015 2:56 am

Thanks for the kind words, I promise an updated build guide is in the pipeline! I have successfully upgraded a few Turing Machines (including David's) by cutting out R14 instead of doing the trace cut and so it is officially going in the manual.

User avatar
drip.feed
Roll it off at 30 Hz
Posts: 3179
Joined: Thu Sep 15, 2011 10:05 am
Location: Belfast, Northern Ireland

Post by drip.feed » Wed Jun 10, 2015 5:25 am

Hi windspirit, I have just had an enquiry about building a customer a Bytes expander for his Turing RLS.

Does the Bytes need a Backpack?

Can the Bytes work with just an RLS? Or do I need the other Turing modules too?

Thanks. :guinness:
Dripfeed

User avatar
windspirit
Super Deluxe Wiggler
Posts: 3050
Joined: Thu Sep 20, 2012 11:43 pm
Location: krellifornia

Post by windspirit » Wed Jun 10, 2015 5:30 am

Hi Drip, you need a backpack and a Turing Machine and that is it. No other expanders necessary.

User avatar
drip.feed
Roll it off at 30 Hz
Posts: 3179
Joined: Thu Sep 15, 2011 10:05 am
Location: Belfast, Northern Ireland

Post by drip.feed » Wed Jun 10, 2015 5:39 am

windspirit wrote:Hi Drip, you need a backpack and a Turing Machine and that is it. No other expanders necessary.
Cheers m'dear. :tu:
Dripfeed

User avatar
NrkBm
Learning to Wiggle
Posts: 43
Joined: Sat Apr 12, 2014 11:27 pm
Location: Canada

Post by NrkBm » Sun Jun 14, 2015 10:24 pm

Ok so i finished Turing machine with both expander and back pack 2 months ago. They were working as expected but now that i have also finished Bytes i think there's something wrong. I'm not sure about the behavior... It seems like there is a lot of repeating sequences going from high notes to lower notes since i built Bytes.. Anyone have this behavior ? Is it normal ? I think it's less random than it was before building Bytes.. Thanks for you help :)

Heres a video to explain it better than words.



Thanks!

User avatar
windspirit
Super Deluxe Wiggler
Posts: 3050
Joined: Thu Sep 20, 2012 11:43 pm
Location: krellifornia

Post by windspirit » Mon Jun 15, 2015 3:50 pm

Hey, I see what you are talking about, I would say that it is not normal but also not unusual. Do you happen to have an oscilloscope? If you could give some scope shots of pin 14 on the tl074 of the main turing machine while the prob knob is at 12 oclock and all the way clockwise and counter clockwise that would be handy.

Also does locking work correctly?

rsponholtz
Learning to Wiggle
Posts: 2
Joined: Tue May 27, 2014 1:44 pm
Location: St. Louis, MO

Sourcing pots FAQ

Post by rsponholtz » Mon Jun 15, 2015 10:30 pm

I'm sure this is a FAQ, but where are you all getting your potentiometers? Mouser has a 1k min order (which is about 997 more than I need), and eBay parts seem pretty dodgy. Any suggestions?

User avatar
woodster
Super Deluxe Wiggler
Posts: 1340
Joined: Mon Mar 05, 2012 9:36 am
Location: UK

Post by woodster » Tue Jun 16, 2015 5:51 am

Thonk(UK), Smallbear(US) or Banzai(EU) are all good for those 9mm Alpha pots.

User avatar
windspirit
Super Deluxe Wiggler
Posts: 3050
Joined: Thu Sep 20, 2012 11:43 pm
Location: krellifornia

Post by windspirit » Tue Jun 16, 2015 6:50 am

Since writing the build docs and BOMs alpha has raised their minimum order for buying through mouser for this particular kind of pot. Here is the link listed on my site for the same part through smallbear http://smallbear-electronics.mybigcomme ... -pc-mount/ you want x3 of the 100k

User avatar
NrkBm
Learning to Wiggle
Posts: 43
Joined: Sat Apr 12, 2014 11:27 pm
Location: Canada

Post by NrkBm » Tue Jun 16, 2015 6:56 pm

I got a reply from Circuit Shaman by email for my issue and he asked to reply here on that thread to help others and/or maybe be helped by someone that had the same issue. So here is it :

From Circuit Shaman :
I certainly can see what you are talking about and I believe that I have noticed it before but I never figured out what the cause was since technically it is still working correctly. Unfortunately the Turing Machine has a great number of variables which influence the flavor of randomness generated and it makes it really difficult to get them to behave consistently. This is why I ended up not making them anymore.

Question #1
"Hey, I see what you are talking about, I would say that it is not normal but also not unusual. Do you happen to have an oscilloscope? If you could give some scope shots of pin 14 on the tl074 of the main turing machine while the prob knob is at 12 oclock and all the way clockwise and counter clockwise that would be handy.

Question #2
Also does locking work correctly?"
So my answer to question #1 is : No i don't have an oscilloscope :S maybe in the future, but i'm not enough experienced for that at this moment.

And my answer to question #2 is : Yes the lock knob is working correctly, also maybe i should specify that the lenght knobs on bytes is changing the sequence when i turn the sequence lengh knobs from bytes. I'm not sure about that but i'm almost sure that the lock knob is working properly.

Here's the video i made so demonstrate this behavior :
And here's another better video in HD : (20 minutes)

The only modification i made other than what is in the manual or in the build documentation is the higher value for the resistor network for the fader leds and the faders leds (for green ones). I also changed the "604-DC-10EWA" (LED Bars and Arrays) for a green one. Same brand, same specs but green. And last thing : the leds on pulses expander. Nothing else than that.

I just did the "CHECKING THAT YOUR BYTES EXPANDER WORKS" from the manual page 6-7 for bytes and everything looks okay... - https://docs.google.com/file/d/0B1jw1Qp ... VlibTZoTkU

Please tell me if it can help having HD pictures of many angles. If yes, i will take some.

Thanks for your help :)

edoway
Learning to Wiggle
Posts: 32
Joined: Thu Jan 03, 2013 6:09 pm
Location: Berlin, Germany

Post by edoway » Sat Oct 31, 2015 7:31 am

Hi there! I have a Turing Machine with Pulses, Voltages & Bytes and have a question (and maybe a problem):
When I have the TM locked CW I can empty (eventually all) bits holding down the WRITE switch and I can add bits holding the WRITE switch up. I think this is how it should work, so no problems here.
When I have the TM locked CCW I cannot empty any bits: as soon as I release the WRITE switch (that I am holding down to delete the bits) I have 8 bits on and 8 off.
When I have the TM locked CCW empty works only if I move down for a short moment the WRITE switch: in this case the some of the 8 empty bits become full and some of the 8 full become empty. Same (or maybe inverted) happens if I move up for a short moment the WRITE switch.
Is this normal behavior?
Thank you so much!

AndrewHP
Learning to Wiggle
Posts: 31
Joined: Thu Jul 26, 2012 5:32 am
Location: Preston

Post by AndrewHP » Sun Jan 31, 2016 7:54 am

I've finally had the time to get my bytes together and had an issue with it that I thought I'd share. It was similar to the locking problem but seemed fine on the longer sequences. going through the testing on the short sequences I got the following with the pot:

1 = 00000000
2 = 0-0-0-0-
3 = 00000000
4 = 00--00--
5 = 00000000
6 = 000---00

This seemed strange so I pulled it out and had a look at it. I checked the trace cut under a magnifier and decided to give it another go. I've just tried it again and it all seems to be functioning correctly:

1 = 0-0-0-0-
2 = 00--00--
3 = 000---00

Basically - If you've had the same issue as I've described then have another look at your trace cut or remove R14. I only read about this method after I'd done the cut - avoiding my issue is probably another good reason to remove the resistor.

User avatar
windspirit
Super Deluxe Wiggler
Posts: 3050
Joined: Thu Sep 20, 2012 11:43 pm
Location: krellifornia

Post by windspirit » Sun Jan 31, 2016 8:35 pm

edoway wrote:Hi there! I have a Turing Machine with Pulses, Voltages & Bytes and have a question (and maybe a problem):
When I have the TM locked CW I can empty (eventually all) bits holding down the WRITE switch and I can add bits holding the WRITE switch up. I think this is how it should work, so no problems here.
When I have the TM locked CCW I cannot empty any bits: as soon as I release the WRITE switch (that I am holding down to delete the bits) I have 8 bits on and 8 off.
When I have the TM locked CCW empty works only if I move down for a short moment the WRITE switch: in this case the some of the 8 empty bits become full and some of the 8 full become empty. Same (or maybe inverted) happens if I move up for a short moment the WRITE switch.
Is this normal behavior?
Thank you so much!
Hi edoway, for some reason I was not alerted to this post. Yes this is the correct behavior. If you have not already I would highly suggest reading Tom Whitwell's manual over on the music thing website. It is very well written and if you take a few minutes to really digest it you will understand why it behaves this way. You can find it here: http://musicthing.co.uk/modular/wp-cont ... n-v2-1.pdf
I've finally had the time to get my bytes together and had an issue with it that I thought I'd share. It was similar to the locking problem but seemed fine on the longer sequences. going through the testing on the short sequences I got the following with the pot:

1 = 00000000
2 = 0-0-0-0-
3 = 00000000
4 = 00--00--
5 = 00000000
6 = 000---00

This seemed strange so I pulled it out and had a look at it. I checked the trace cut under a magnifier and decided to give it another go. I've just tried it again and it all seems to be functioning correctly:

1 = 0-0-0-0-
2 = 00--00--
3 = 000---00

Basically - If you've had the same issue as I've described then have another look at your trace cut or remove R14. I only read about this method after I'd done the cut - avoiding my issue is probably another good reason to remove the resistor.
Huh this is a weird one. Maybe you were getting a combination of the sequence coming out of bit 8 and also whatever sequence was coming from the Bytes expander. Let us know if you end up having this issue resurface.

User avatar
maltemark
Ultra Wiggler
Posts: 758
Joined: Tue Dec 16, 2014 11:39 am
Location: Gothenburg

Post by maltemark » Sat Mar 05, 2016 9:15 am

Image

Built mine last week after getting the ADC's and right-angle stuff. It was an easy build, even though build instructions felt spread out across several places at times - maybe just because I tend to have lots of browser tabs open all the time?

Anyway, it seems to work as it should. Only trouble I encountered was the standard "should have gone to bed 2 hours ago but can't stop soldering, so I make a routine mistake" thing. And I accidentally swapped the 4015 and 4051, just as a previous builder had done - as noticed in the original post. Put everything in its right place and the bytes seems to have passed the calibration routine.

Going to have to build a second turing machine and use it to control the pattern length of the expanded one I think. It would be a great kind of Meshuggah machine.

sammy123
Super Deluxe Wiggler
Posts: 1216
Joined: Sun Mar 24, 2013 12:36 pm

Post by sammy123 » Sat Mar 26, 2016 1:39 pm

I've been going through the testing steps of the bytes on page 6 of the manual and am not able to get 2 bytes or 7. Any ideas? There is nothing between 1 and 3 and nothing between 6 and 8. My 1-8 length knob is kind of off in general.

User avatar
windspirit
Super Deluxe Wiggler
Posts: 3050
Joined: Thu Sep 20, 2012 11:43 pm
Location: krellifornia

Post by windspirit » Sat Mar 26, 2016 3:35 pm

That is def not normal, can you post some pics of your build? Top and bottom?

sammy123
Super Deluxe Wiggler
Posts: 1216
Joined: Sun Mar 24, 2013 12:36 pm

Post by sammy123 » Sat Mar 26, 2016 5:01 pm

Ok...missed a pin on one of the maxims. All set now I think. Time to read the manual again. Thanks

User avatar
squarewavesurfer
Wiggling with Experience
Posts: 252
Joined: Tue Jan 27, 2015 2:43 pm
Location: Canada

Post by squarewavesurfer » Fri Apr 01, 2016 12:06 pm

I have read through the build instructions and this thread but have not seen anyone mention that the current mouser BOM seems to have the wrong trimmer.

The build doc says the 3 legs should all be inline as is apparent on the PCB. The problem I have is that my trimmer (3296Y-1-104LF as in the BOM) has 2 outer legs inline with the middle leg offset to the front of the trimmer (closer to the trimmer screw).

I see now that the pictured trimmer has "W 104 321C0" marked on top, wheras mine is marked "Y 104 602C0".

Has anyone encountered this and if so, can you simply bend the offset leg to be inline or do I need to order the correct one?

sammy123
Super Deluxe Wiggler
Posts: 1216
Joined: Sun Mar 24, 2013 12:36 pm

Post by sammy123 » Fri Apr 01, 2016 12:28 pm

If the pinout is the same then you can just bend the leads. I have done it a million times. The pinouts can be confirmed in the data sheet or by simply looking at the numbers on the side.

User avatar
squarewavesurfer
Wiggling with Experience
Posts: 252
Joined: Tue Jan 27, 2015 2:43 pm
Location: Canada

Post by squarewavesurfer » Fri Apr 01, 2016 1:04 pm

Thanks, I will see how bending the lead goes.

I think the correct part number on mouser is: 652-3296W-1-104LF, someone may want to confirm this.

In a mouser comparison chart between the one I ordered and the above part you can not tell the difference aside from the W vs Y in the part number.


Picture file

User avatar
windspirit
Super Deluxe Wiggler
Posts: 3050
Joined: Thu Sep 20, 2012 11:43 pm
Location: krellifornia

Post by windspirit » Fri Apr 01, 2016 1:56 pm

Hey, I havent built any bytes for awhile but I am always using the mouser bom that I made amd the trimmers are usually in a straight line instead of staggered. I have had all kinds of funny things happen with old mouser Boms and I wouldnt be surprised if they simply replaced one trimmer with another, either by accident or on purpose.

User avatar
elcoco
Common Wiggler
Posts: 71
Joined: Thu Sep 10, 2015 10:56 am
Location: Wherever my hat is
Contact:

Post by elcoco » Sun Apr 10, 2016 4:07 pm

Hi I just recently built a Turing machine with all its expanders and everything was working well until I installed the bytes expander. When I plug into the seq out it seems to not be holding the pitch, instead it plays the pitch corresponding to the 8 bits on the screen and then when the clock changes instead of holding the voltaje output it falls to zero. I'm sure that this didn't happen before. Any one have any idea what could be going on?

User avatar
windspirit
Super Deluxe Wiggler
Posts: 3050
Joined: Thu Sep 20, 2012 11:43 pm
Location: krellifornia

Post by windspirit » Sun Apr 10, 2016 4:15 pm

Sorry, do you mean that the bits all clear or that the voltage coming from the output of the module drops even though there are still bits in the register? Can you upload a video to show us?

User avatar
elcoco
Common Wiggler
Posts: 71
Joined: Thu Sep 10, 2015 10:56 am
Location: Wherever my hat is
Contact:

Post by elcoco » Sun Apr 10, 2016 4:58 pm

i think its just that the output drops because the leds on the screen are still the same. basically whenever the sequence moves forward the output voltage goes to 0

Post Reply

Return to “Music Tech DIY”