Author |
Message |
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Fri May 25, 2018 2:37 pm Post subject:
Euro 281 -- Double Dual Function Generator Subject description: Buchla 281 QFG euroclone |
|
|
I have, on the way, some PCBs for a Buchla Quad Function Generator clone. Hopefully I've done this right. Eurorack format, rather cramped, though.
I've just straight swapped the 15V supply for a 12V one -- seemed to work with the Source of Uncertainty, albeit with some component value changes.
For the 10V supply to some of the ICs, I've used a LDO voltage regulator -- the TL750L10 or something similar. I'm also going to test whether I can just run 'em on 12V or not. The panelpcb has a trace added for either 10V/12V for some switches.
The pads on the dual opamps are smaller than I like, but I wanted routing flexibility for this circuit.
For the Engineboard A trimpots, I put them close to the edge so that right-angle trimpots (Bourns type) can be used, or kludge some other one into place.
For the FETs, I'm going to try a few substitutes.
The 100nF caps in the core of the circuit, I've worked up so that either a 5mm box cap can be used, or a 10mm fancycap. (Polystyrene probably won't fit, but you can try if you want.) There are more traces around these than I like, but I'm working within size constraints.
Yes, this is four function generators (hence DoubleDual) like the original, not two function generators like most Euro adaptations! |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Mon May 28, 2018 11:31 pm Post subject:
|
|
|
Right, PCBs turned up, so I populated just the B engineboard. The two engineboards are independant, and the B one has the rear of the board exposed (along with power header) when the module is assembled, so I can probe it under power.
I tried my luck with 12V going to all sections (including the 10V section.) Nothing blew up, but nothing worked correctly, either. (I was pretty sure this would be the case, but nothing ventured...)
TL750M10 regulators are on order, they just won't be in my grubby little paws for a week or two. I'll perf up a TL071 10V source at some point. I'm also going to have to order some more LM13700's, so that my Jerkster build can have it's OTAs back. I have a horrible feeling that even with 10V added, I'm going to have to troubleshoot.
As you can see, the panel is going to be pretty damn cramped. I tried to give the Attack and Decay pots some vertical finger room (there is no horizontal room), but even so... the price we pay for a 3U synth size, I guess.
Description: |
|
Filesize: |
361.29 KB |
Viewed: |
408 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
LFLab
Joined: Dec 17, 2009 Posts: 497 Location: Rosmalen, Netherlands
|
Posted: Tue May 29, 2018 5:52 am Post subject:
|
|
|
Looks nice Alan! But yeah, troubleshooting is part of making your own PCB's
Density doesn't look too bad, right? You could switch to SMT IC's and the smaller submini switches (APEM TL36 or similar).
Thinking about PCB sandwiches as well, but I am wondering about exactly which parts in Eagle to use. |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Tue May 29, 2018 9:56 pm Post subject:
|
|
|
The stock pinhead.lbr is what I use -- be VERY careful about location and orientation between PCBs you are connecting. |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Wed May 30, 2018 2:04 am Post subject:
|
|
|
Perfed up a TL071 with 100k, 20k, and 1k resistors for a 10V source, and that is now flying-wired to the back of the module. (See, Dave, I can do wires. I just don't want to, when possible )
The module is now working, but one channel has a 5V offset (why, I have no idea...) and the attenuator for the Peak is affecting one of the channels (my suspicion is a solder bridge somewhere.) The quadrature switch is also strongly, strongly affecting the circuit, more than it should, and I think I'll have to cut some normalization traces I've done.
I'll look at this more tomorrow. Troubleshooting during week nights is... not optimal. |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Thu May 31, 2018 8:21 pm Post subject:
|
|
|
Found the offset problem -- two resistors were tied to ground, not the negative rail.
Also, found there were no 1k output resistors. I'm an idiot. Also, wrong cap value on the input switch.
Think I might get a new PCB set done. Annoyed with myself at this. |
|
Back to top
|
|
|
LFLab
Joined: Dec 17, 2009 Posts: 497 Location: Rosmalen, Netherlands
|
Posted: Sat Jun 02, 2018 4:54 am Post subject:
|
|
|
With modern opamps, you do not really need those 1k output resistors anyway.
An alternative would be to simply break the output tracks and tack on 0805 SMT resistors.
The other two resistors one could probably scratch through a track or two and hook them up to V-?
Meh, not enough to be annoyed with yourself about, it is quite a bit of circuitry, and one ecannot make an omelet without breaking some eggs.
Let me know if you'd like to get rid of a surplus set of those boards, totally do not mind crushing those particular bugs (we could do a trade again, I have a few boards you might be interested in). |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Sat Jun 02, 2018 12:29 pm Post subject:
|
|
|
I need to get panels made anyway |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Sat Jun 02, 2018 10:52 pm Post subject:
|
|
|
Worked up a front panel.
I also found that the Attack and Decay controls are reversed on my v0.2 boards. Aiyaiyai.
For the v0.3 boards, I added a set of pads for a TO220 package TL750M10 regulator, as well as the TL750L10. |
|
Back to top
|
|
|
MapacheRaper
Joined: Feb 15, 2018 Posts: 166 Location: Spain
|
Posted: Sun Jun 17, 2018 10:13 pm Post subject:
|
|
|
You are doing a fantastic work!
If you plan to sell PCBs when they are ready please put me in the list. This module is like a 2x version of Maths, more or less, isn´t?
|
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Sun Jun 17, 2018 10:47 pm Post subject:
|
|
|
Maths still confuses the hell outta me. The front panel design is willfully obtuse and hard as hell to understand. |
|
Back to top
|
|
|
LFLab
Joined: Dec 17, 2009 Posts: 497 Location: Rosmalen, Netherlands
|
Posted: Mon Jun 18, 2018 6:49 am Post subject:
|
|
|
MapacheRaper wrote: | You are doing a fantastic work!
If you plan to sell PCBs when they are ready please put me in the list. This module is like a 2x version of Maths, more or less, isn´t?
|
Maths also has 4 attenuaverters, voltage control over cycling, analog logic, and summing signals. But yeah, it is a start |
|
Back to top
|
|
|
gabbagabi
Joined: Nov 29, 2008 Posts: 651 Location: Berlin by n8
Audio files: 23
|
Posted: Mon Jun 18, 2018 8:53 am Post subject:
|
|
|
As you are talking about,
would it be to much to ask for a description of what it is doing?
ahoi, gabi |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Mon Jun 18, 2018 2:08 pm Post subject:
|
|
|
It's a copy of the Buchla Quad Function Generator. It's set up in two pairs of two.
Each Function Generator requires at least a trigger to get going (even in Cycle mode.) You can control the attack or decay slopes (like a VC Slew), and there is an Output, and an End Of Cycle output trigger. There are three modes -- Sustained (ASR, basically), Short (AR, no decay, no sustain), and Cycle (the EOC trigger goes to the input gate and sets the whole thing off again at the end of each cycle.)
Each pair of function generators has a Quad switch, and a Peak Out (with the B output going through an attenuator before the Peak Out.) Quad Mode means that when A finishes rising, B is triggered to start, and when B finishes it's rise time, A is triggered to decay. |
|
Back to top
|
|
|
MapacheRaper
Joined: Feb 15, 2018 Posts: 166 Location: Spain
|
Posted: Wed Jun 20, 2018 3:47 am Post subject:
|
|
|
Yeah, that´s quite similar to and Maths module. If you cange the attenuaters for attenuverters and do some OR fusion with the two ones you basically get an Maths. Both are great.
Will you sell PCBs once you get it finished? |
|
Back to top
|
|
|
gabbagabi
Joined: Nov 29, 2008 Posts: 651 Location: Berlin by n8
Audio files: 23
|
Posted: Wed Jun 20, 2018 11:10 am Post subject:
|
|
|
thx man!
some questions:
is the" cycle" on the frontplate an inputut/ output / switch or pot?
i would guess a switch.
when in cycle mode ASR is also working or is it then automaticly AR?
when in cycle mode it only needs one trigger to let it go around and let it cycle endlessly? later trigger is then recognized?
"Quad Mode means that when A finishes rising, B is triggered to start, and when B finishes it's rise time, A is triggered to decay" --> that means it is kind of ASR then? A can only decay when B is finished its attack?
Is it the VC over attack AND (/OR?) decay that makes the thing so interesting or is it more the whole concept?
thx again,
gabi |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Wed Jun 20, 2018 1:31 pm Post subject:
|
|
|
g.gabba wrote: | is the" cycle" on the frontplate an inputut/ output / switch or pot?
i would guess a switch. |
Yep, a switch (on-off-on). The middle position (off) is sustain mode.
g.gabba wrote: |
when in cycle mode ASR is also working or is it then automaticly AR? |
In cycle mode, the channel will with no gate/trig input keep cycling. If a gate comes in, then the output will stop cycling and stay high until the gate stops (Sustain mode out, basically), at which point with no more gate input it will go back to cycling (a very odd triangle/saw.)
g.gabba wrote: | when in cycle mode it only needs one trigger to let it go around and let it cycle endlessly? later trigger is then recognized? |
After the initial trigger, it self-triggers from the end of cycle, in cycle mode, if that's what you mean.
g.gabba wrote: | "Quad Mode means that when A finishes rising, B is triggered to start, and when B finishes it's rise time, A is triggered to decay" --> that means it is kind of ASR then? A can only decay when B is finished its attack? |
I think so (I need to stop answering these when I wake up.)
g.gabba wrote: | Is it the VC over attack AND (/OR?) decay that makes the thing so interesting or is it more the whole concept? |
The whole concept. The VC attack and decay are separate control voltages in. If you patch the output back into either of them, you get an exponential response
Dave Brown has some scopeshots here -- https://modularsynthesis.com/roman/buchla281/281_qfg.htm |
|
Back to top
|
|
|
gabbagabi
Joined: Nov 29, 2008 Posts: 651 Location: Berlin by n8
Audio files: 23
|
Posted: Fri Jun 22, 2018 3:07 am Post subject:
|
|
|
Thanks for taking the time,
i guess i really got a picture now how it works,
before it was always a mistery!
good luck with the pcbs,
cheers
gabi |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Mon Jun 25, 2018 2:26 am Post subject:
|
|
|
Got the bugger.
The only issues I had with this version:
Engineboard A rev 0.3 has some vias insanely close to solder pads, as in be VERY careful in some places to not get solder bridges.
The Atk and Decay voltage inputs need attenuators. Too sensitive right now, I might change some resistors.
Silly me forgot to include the LEDs in the panel, so I had to nip 'round to Dad's shed to drill four 3mm holes for them. ("If that's the only issue I have, I'll be glad," I said while drilling.)
Otherwise, she's a good 'un.
Oh, and the B Attenuator on the Peak Out is more a B Level (ie, it's in reverse.) |
|
Back to top
|
|
|
MapacheRaper
Joined: Feb 15, 2018 Posts: 166 Location: Spain
|
Posted: Mon Jun 25, 2018 7:25 am Post subject:
|
|
|
So sexy!. It only need some minor tweaks and is ready. Cool. |
|
Back to top
|
|
|
gabbagabi
Joined: Nov 29, 2008 Posts: 651 Location: Berlin by n8
Audio files: 23
|
Posted: Mon Jun 25, 2018 11:38 am Post subject:
|
|
|
like we say at home: Glückwunsch Alter! --> Gongrats dude!
(is there a meaning behind left and right thumb? better put both!) |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Mon Jun 25, 2018 12:32 pm Post subject:
|
|
|
The original Buchla used 180nF timing caps. Roman's schematic used 100nF timing caps.
With 100nF timing caps, with maximum decay and minimum attack (a saw waveform), the trimmer meant to adjust for 10sec decay time had a minimum decay time of about 18sec.
I had a bag of 47nF WIMA caps lying around (is this a sign I have a problem?), and so bunged them in, in place of the 100nF caps. Now, I can adjust the maximum decay for 10sec
The Atk and Decay cv inputs are still pretty responsive, but now are not insanely touchy. That seems to have fixed that problem |
|
Back to top
|
|
|
LektroiD
Joined: Aug 23, 2008 Posts: 1018 Location: Scottish Borders
Audio files: 2
G2 patch files: 2
|
Posted: Sat Aug 11, 2018 5:41 am Post subject:
|
|
|
This looks like a really nice project. I'll definitely be interested in a PCB & black panel when it's all ready...
Excellent work! _________________ LektroiD |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Sun Aug 12, 2018 2:35 am Post subject:
|
|
|
How many people would be interested in a pcb/panel set for this?
I need to do a new revision of the Engine A PCB (there are some vias that are insanely close to solder pads), and the panels I have are missing the LED holes (I can drill them out myself, but if I'm ordering more Engine A PCBs anyway...) |
|
Back to top
|
|
|
LektroiD
Joined: Aug 23, 2008 Posts: 1018 Location: Scottish Borders
Audio files: 2
G2 patch files: 2
|
Posted: Sun Aug 12, 2018 3:18 pm Post subject:
|
|
|
AlanP wrote: | How many people would be interested in a pcb/panel set for this?
I need to do a new revision of the Engine A PCB (there are some vias that are insanely close to solder pads), and the panels I have are missing the LED holes (I can drill them out myself, but if I'm ordering more Engine A PCBs anyway...) |
As stated above, I'd absolutely love a set, maybe two, depending on cost (my wallet has been hit pretty hard of late with vintage purchases)... I really like that black panel though...! _________________ LektroiD |
|
Back to top
|
|
|
|