Trapezoid VCO

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

Moderators: Joe., lisa, luketeaford, Kent

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Trapezoid VCO

Post by Haralds:Werk » Tue Apr 30, 2019 3:48 am

I am working on a trapezoid VCO on the basis from Don Tillman's article from July 2003. I have made some changes but kept the basic architecture as given in his block diagram. I replaced the exponentiator circuitry and changed the control for the address line. So far I have a have a (faulty) PCB for the core. After some debugging I got it running. All four trapezoid outputs and the thru zero FM.

The next Version PCB for the core is on its way and will include a LFO switch and the quadrature square wave outputs. The wave shaper for triangle, sine, pulse andramp up/down is still in breadboard mode bit looks promising.

Image

Image

Image

Image
www.haraldswerk.de
Home of Next Generation Formant

Jarno
Super Deluxe Wiggler
Posts: 2767
Joined: Mon Nov 15, 2010 6:34 am
Location: Rosmalen, NL

Post by Jarno » Tue Apr 30, 2019 3:03 pm

Interesting VCO, I take it you did not use the 3280 OTA? Regular LM13700?

The article:
http://www.till.com/articles/QuadTrapVC ... ssion.html
http://www.till.com/articles/QuadTrapVCO/vcocore.html

User avatar
dksynth
Common Wiggler
Posts: 231
Joined: Wed Jul 01, 2015 12:56 pm

Post by dksynth » Tue Apr 30, 2019 7:40 pm

very interested to see this!

User avatar
BugBrand
Knowledge of Bugs
Posts: 7186
Joined: Tue Jul 15, 2008 11:59 am

Post by BugBrand » Wed May 01, 2019 2:25 am

Jarno wrote:Interesting VCO, I take it you did not use the 3280 OTA? Regular LM13700?
Sure Harald will comment, but Alfa remade the 3280 recently: http://www.alfarzpp.lv/eng/sc/AS3280.php

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Thu May 02, 2019 1:56 am

Jarno wrote:Interesting VCO, I take it you did not use the 3280 OTA? Regular LM13700?

The article:
http://www.till.com/articles/QuadTrapVC ... ssion.html
http://www.till.com/articles/QuadTrapVCO/vcocore.html
Thank you for the links to the original articles. I have forgotten them in my first post.

I am using the 3280 OTA and the SSM2220 matched pair in this design. I have some NOS of both around and want to use them up. The SSM2220 is easily replaced with a hand matched transistor pair and the 3280 is available again as AS3280 as stated above. I have no AS3280 to check if it is a drop in replacement though.

Using the LM13700 might be possible but I suspect some loss in performance quality. It should be an easy change with some adaption in the Iabc (exponentiator) and a few other resistor changes. And a new PCB of course.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
dot matrix madness
Wiggling with Experience
Posts: 355
Joined: Fri Aug 23, 2013 2:48 am
Location: Germany

Post by dot matrix madness » Thu May 02, 2019 4:04 am

Does somebody own the Doepfer A-110-6 Trapezoid VCO and would check which chips and OTAs they use?
Licenced to solder since 1993

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1449
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Post by cygmu » Thu May 02, 2019 4:48 am

Good point! I don't have one, but there are some photos here:
https://www.thomannmusic.com/doepfer_a_110_6.htm

There is a V2164 in there.

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Fri May 03, 2019 3:36 am

dot matrix madness wrote:Does somebody own the Doepfer A-110-6 Trapezoid VCO and would check which chips and OTAs they use?
A quick search on the net reveals this one:
http://www.doepfer.de/a100_man/A110_6_t ... umpers.pdf
You can see Doepfer is using LM393N, DG409DJ, LF374 and SSM2164. So we can assume that they have replaced the 3280 with the SSM2164.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Mon May 06, 2019 1:28 am

dksynth wrote:very interested to see this!
It is still an ongoing project. I have a first PCB for the core (the red one in the picture) the second one is already shipped from the vendor and should arrive this week. The waveshaper is partly in breadboard mode. A first PCB for the waveshapers should be ready in two or three weeks.

Image
www.haraldswerk.de
Home of Next Generation Formant

User avatar
guest
Super Deluxe Wiggler
Posts: 3951
Joined: Mon Aug 19, 2013 11:49 am

Post by guest » Mon May 06, 2019 2:29 pm

hey Heralds:Werk, this is a bit off topic, but i just wanted to say that i appreciate your industriousness, attention to detail, and that you share your work here. its great following along with all the stuff youre making.
openmusiclabs.com

User avatar
dksynth
Common Wiggler
Posts: 231
Joined: Wed Jul 01, 2015 12:56 pm

Post by dksynth » Mon May 06, 2019 6:24 pm

Same sentiment here as from guest, haraldswerk has been bookmarked in my home computer for quite awhile. Your write ups are great!

Curious as to what you changed in the address bits, seems to me that’s where you’d adjust things to modify phase? The further info on how to do that never got written up.

I’m going to spin up a version with the 2164 cells, see what I can manage!

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Wed May 08, 2019 3:38 am

guest wrote:hey Heralds:Werk, this is a bit off topic, but i just wanted to say that i appreciate your industriousness, attention to detail, and that you share your work here. its great following along with all the stuff youre making.
Thank you. Every feedback is welcome. That keeps me publishing. Feedback is often scarce. I also appreciate critical feedback, questions and suggestions to improve my designs.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1449
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Post by cygmu » Wed May 08, 2019 7:25 am

Haralds:Werk wrote:Every feedback is welcome. That keeps me publishing. Feedback is often scarce. I also appreciate critical feedback, questions and suggestions to improve my designs.
I always feel bad not saying anything in your threads, because I enjoy them a great deal, but don't have anything much to add, given that your work is so accomplished and well executed.

I have been intrigued by this Tillman design for a long time so I am very keen to see how this one develops. If anyone can figure out how to implement the voltage controlled phase shift that is referred to in the original article but never written up I will be very excited!

I asked about it here
http://www.muffwiggler.com/forum/viewtopic.php?t=203284
but I don't think there was a conclusion. I also asked Don Tillman who said he didn't want to reveal the secrets because he was planning to commercialise the design (this was after the Doepfer appeared) so perhaps we shouldn't try to figure it out, but I am sooo curious...

User avatar
djthopa
Super Deluxe Wiggler
Posts: 1746
Joined: Thu Mar 22, 2012 5:48 pm
Location: Somewhere between a nappy and a wire

Post by djthopa » Wed May 08, 2019 8:00 am

Great work as always, and very kind from you to openly share your projects! :hail:

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Thu May 09, 2019 1:09 am

dksynth wrote: I’m going to spin up a version with the 2164 cells, see what I can manage!
The 2146 should work, Doepfer is using it. If you start building from the original Tillman schematic have a close look to the voltages at the EN pin and the address lines A0/A1 of the DG409. And check the value of the pull up resistors.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Fri May 10, 2019 1:17 am

cygmu wrote:
I always feel bad not saying anything in your threads, because I enjoy them a great deal, but don't have anything much to add, given that your work is so accomplished and well executed.
Thank you for this remark. It got me thinking. I show the same behavior. In most cases when reading interesting threads I think "fine work" but did not comment if I have nothing substantial to add. I don't want to add chatter.
cygmu wrote: I have been intrigued by this Tillman design for a long time so I am very keen to see how this one develops. If anyone can figure out how to implement the voltage controlled phase shift that is referred to in the original article but never written up I will be very excited!
Back on topic. I have an idea how to add it. But before diving into it I first want to have a stable running version of the core and the waveshapers. So far I have the usual triangle, sine, ramp up/down, square and pulse. But I stumbled over a lot of interesting wave forms while tinkering with adding / subtracting / mixing the above. Maybe I should add sort of a patch/mix section for experiments.

The second run off PCB's for the core arrived today. I'll stuff them over the weekend.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Wed May 22, 2019 8:30 am

So, the second PCB was populated and tested. I had to correct a newly introduced error and ran int some subtle timing problems with the comparators. It was strange. Turning the thru zero FM pot to one side everything worked fine. On the other side the oscillation first gets weird and then stops all together. Took me some time to fix that. Now I am waiting for the next PCB run. Meanwhile, for those who are interested what I am talking about: Here is the draft of the circuitry.


Edit: Don't try to build this one. The emphasis is on DRAFT!
Last edited by Haralds:Werk on Sun Aug 25, 2019 10:55 am, edited 1 time in total.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
diablojoy
Super Deluxe Wiggler
Posts: 1616
Joined: Fri Jan 22, 2010 9:03 pm
Location: melbourne

Post by diablojoy » Fri May 24, 2019 4:31 am

Hmm nice progress been meaning to try out this circuit for a while.
about the only thing I don't actually have on hand is any TL052's .
It probably wont happen today but if it does it definitely wont go smoothly.

User avatar
LektroiD
Super Deluxe Wiggler
Posts: 1376
Joined: Tue Oct 20, 2009 1:55 am
Location: Scotland

Post by LektroiD » Fri May 24, 2019 5:09 pm

This looks like a really interesting oscillator. If PCBs & panels become available, please count me in...

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Sat May 25, 2019 10:05 am

diablojoy wrote:Hmm nice progress been meaning to try out this circuit for a while.
about the only thing I don't actually have on hand is any TL052's .
I tried the TL052 because I want something faster then the original used LF412. This VCO is still in draft mode and I'll decide what OpAmp's to use when the final version is running. The next PCB should arrive this week. You can use the LF412 or TL062 for testing as well. If you use OpAmp's with higher power consumption you'll need to replace the LM317L/LM337L with LM317/LM337 in TO220 package. Note, this is only page one of two. The circuitry for the square output is not shown. I'll build two versions and the swap the OpAmp's around to have a direct comparison.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Wed May 29, 2019 2:55 am

LektroiD wrote:This looks like a really interesting oscillator. If PCBs & panels become available, please count me in...
For the moment I build this one for my 15V banana system. When I've got everything running I'll ask if there is some demand for a Eurorack version.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
diablojoy
Super Deluxe Wiggler
Posts: 1616
Joined: Fri Jan 22, 2010 9:03 pm
Location: melbourne

Post by diablojoy » Wed May 29, 2019 3:14 am

ahh cool good to know thanks :tu:
It probably wont happen today but if it does it definitely wont go smoothly.

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Tue Jun 18, 2019 1:33 am

Coming closer. On the left you can see the quad waveshaper, top right the latest (third) PCB for the trapezoid VCO. This one still had some subtle issues. To my surprise it worked not symmetrical with regards to the zero crossing on the lower end of the frequency range. On the "left" side the trapezoid VCO stops working below about 150Hz. On the "right" side everything worked OK. It took me some time to figure out what's going on. Turned out a timing problem when switching the direction of the address rotation. I hopefully fixed it with some additional circuitry (on breadboard). Waiting now for the next PCB.

Image
www.haraldswerk.de
Home of Next Generation Formant

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 266
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Tue Aug 20, 2019 3:25 am

Coming soon.

Image

Image
www.haraldswerk.de
Home of Next Generation Formant

B0bcat
Common Wiggler
Posts: 154
Joined: Sun Jun 10, 2018 9:07 pm

Post by B0bcat » Tue Aug 20, 2019 10:55 am

I want to express interest in a Euro version :)
WTB: Plan B or SubCon Model 15 VCO, Intellijel Dixie v1, Doepfer A-129-5

Post Reply

Return to “Music Tech DIY”