MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Cyndustries Quad LPG - Possible mod
MUFF WIGGLER Forum Index -> 5U Format Modules  
Author Cyndustries Quad LPG - Possible mod
GryphonP3
Hey guys. I wanted to ask some opinions about how the Cyndustries LPG works for you.. I have some issues with it myself. Namely, the fact that it requires the pots to be turned up applying an offset for any CV signals to work. So in order to modulate it at all, it needs to have its own constant gain turned up. There's no way of having the level controled exclusively by CV signals. Thats pretty lowsy for an LPG in my opinion.

I was thinking about replacing the pots with push/pull pots, which when pulled, bypass the pot's CW voltage source, and replace it with the signal at the CV input jack, so that the pot acts purely as an attenuator for the CV signal, and not to offset the gain. When the pot is not pulled, it would be routed as normal, mixing the offset with the external CV source.

My idea was to use DPDT switch pots, the first switch would route the CV In jacks between the CV in terminal on the board, then to lug1 on the second switch. The second switch's common lug would go to the pot's CW lug. Lug3 of the second switch would be connected to the module's +V. That way, when the switch is not pulled, it acts as normal, and when the switch is pulled, V+ is disconnected from the pot CW terminal, and it is replaced with the CV input.

Does anyone have a better idea of how to go about soing this? Going to do it to my own module next weekend. Thought I would share here in case anyone else was interested in doing this simple mod, and also to see if I might be overlooking a better solution.
groovar
Let me know how this works out I would gladly send you mine to do this
robotmakers
On my (dotcom version) Cyndustries LPG, I found a similar problem, which I interpreted to be related more to a scaling issue rather than an offset issue. My solution was to use a quad op-amp to create a CV amplifier on each of the four inputs which multiplies the incoming CV's by a factor of 2. It was pretty simple to wire up and works a charm. CV's hit the vactrols hard, which creates a response I like and opens up the filters fully. And no need to mess with the "gain" knobs, which are really "offset" knobs.

Good luck,
Roger
GryphonP3
robotmakers wrote:
On my (dotcom version) Cyndustries LPG, I found a similar problem, which I interpreted to be related more to a scaling issue rather than an offset issue. My solution was to use a quad op-amp to create a CV amplifier on each of the four inputs which multiplies the incoming CV's by a factor of 2. It was pretty simple to wire up and works a charm. CV's hit the vactrols hard, which creates a response I like and opens up the filters fully. And no need to mess with the "gain" knobs, which are really "offset" knobs.

Good luck,
Roger


Could you elaborate on how you incoporated the op amp please? Would gladly do this to mine and would be happy to help others out too.
parasitk
robotmakers wrote:
On my (dotcom version) Cyndustries LPG, I found a similar problem, which I interpreted to be related more to a scaling issue rather than an offset issue. My solution was to use a quad op-amp to create a CV amplifier on each of the four inputs which multiplies the incoming CV's by a factor of 2. It was pretty simple to wire up and works a charm. CV's hit the vactrols hard, which creates a response I like and opens up the filters fully. And no need to mess with the "gain" knobs, which are really "offset" knobs.


Very interested in this. I tweaked the CV calibration on my Modcan A version, but I think a scaling multiplier would help quite a bit.
megaohm
parasitk wrote:
robotmakers wrote:
On my (dotcom version) Cyndustries LPG, I found a similar problem, which I interpreted to be related more to a scaling issue rather than an offset issue. My solution was to use a quad op-amp to create a CV amplifier on each of the four inputs which multiplies the incoming CV's by a factor of 2. It was pretty simple to wire up and works a charm. CV's hit the vactrols hard, which creates a response I like and opens up the filters fully. And no need to mess with the "gain" knobs, which are really "offset" knobs.


Very interested in this. I tweaked the CV calibration on my Modcan A version, but I think a scaling multiplier would help quite a bit.


The CV inputs probably expect 10V levels instead of the typical 5V.
Pretty sure the original Buchla used 10V CV levels.

robotmakers solution will work
BUT
there may be an easier way.
Find the summing resistor for the CV input (probably sums with the offset/gain pot voltage at the -input of an opamp). Cut that resistor to 1/2 it's value and you should get 2X gain.
IE: if it's a 100K, remove and replace with a 50K.
if 62K, replace with 30K

Also, instead of removing the CV resistor.
Just parallel it with a resistor of the same value.
(same as replacing but quicker/easier than desoldering a resistor)
Cut the leads short and tack solder to the original resistor.
robotmakers
GryphonP3 wrote:
robotmakers wrote:
On my (dotcom version) Cyndustries LPG, I found a similar problem, which I interpreted to be related more to a scaling issue rather than an offset issue. My solution was to use a quad op-amp to create a CV amplifier on each of the four inputs which multiplies the incoming CV's by a factor of 2.


Could you elaborate on how you incoporated the op amp please? Would gladly do this to mine and would be happy to help others out too.


All I did was create four copies of a non-inverting amplifier, as described in the second circuit in this PDF doc:

http://www.ti.com/ww/en/bobpease/assets/AN-31.pdf

The CV from each of the input jacks was sent first to one of the four amplifier circuits, and then continued to whatever destination it was originally connected to. Simply four individual CV amplifiers stuck at the very front end of the module. On a tiny little perf board inelegantly stuck in there somewhere.

That said, megaohm's solution is much more clever, but I have no knowledge of exactly where to find the specified resistors on the board, etc.

Cheers,
Roger
kindredlost
megaohm wrote:
there may be an easier way.
Find the summing resistor for the CV input (probably sums with the offset/gain pot voltage at the -input of an opamp). Cut that resistor to 1/2 it's value and you should get 2X gain.
IE: if it's a 100K, remove and replace with a 50K.
if 62K, replace with 30K

Also, instead of removing the CV resistor.
Just parallel it with a resistor of the same value.
(same as replacing but quicker/easier than desoldering a resistor)
Cut the leads short and tack solder to the original resistor.

I knew it! Mr MegaOhm to the rescue. Bananallama!
I'm going to look over my QLPG and see if I can spot the summing resistor you describe. If not, I might shoot some pics and see if you can spot it. I could do the modification but I'd hate to do it to the wrong component. d'oh!

EDIT:
I've always had a love/hate relationship with the module. Good to know it's "fixable". I have been using the Q125 Signal Processor to drive my Cyndustries QLPG. I just tweak knobs until I get it to do what I am after but didn't understand what was going on. It will be nice to have this working properly.

I may have to take another look at the Mega Ohm Audio CdS VCF/A module now. I bet it is quite a bit more use-able in the 5U realm than the Cyndustries one is unmodified. What's your inventory look like Phil?
megaohm
kindredlost wrote:
megaohm wrote:
there may be an easier way.
Find the summing resistor for the CV input (probably sums with the offset/gain pot voltage at the -input of an opamp). Cut that resistor to 1/2 it's value and you should get 2X gain.
IE: if it's a 100K, remove and replace with a 50K.
if 62K, replace with 30K

Also, instead of removing the CV resistor.
Just parallel it with a resistor of the same value.
(same as replacing but quicker/easier than desoldering a resistor)
Cut the leads short and tack solder to the original resistor.

I knew it! Mr MegaOhm to the rescue. Bananallama!
I'm going to look over my QLPG and see if I can spot the summing resistor you describe. If not, I might shoot some pics and see if you can spot it. I could do the modification but I'd hate to do it to the wrong component. d'oh!



I am happy to help with this.
If you have a DMM, set it to check for continuity.
Put one probe at the jack tip and use the other probe to check the resistor leads.
May be able to do it visually. See where the wire from the jack tip leads to on the pcb and you may be able to see the trace that leads to a resistor.
It can be difficult to do this from pics, but might be easy.
Post some pics of the pcb and I'll try. Top and bottom of the pcb would be ideal if simple to do but don't go taking it all apart to get a bottom shot.
Please make it a hi-res pic!

Ideal scenario - someone in Chicago who owns one of these come to my shop (Lake and Ashland) and I'll mod it for you. Then I can take pics and show exactly how/what to do.
megaohm
kindredlost wrote:
What's your inventory look like Phil?


Not much to look at, really!
lol ...
cry

I don't have the money anymore to keep modules in stock.
Entirely built to order now.
Worse, the prices are significantly higher than my website shows.
It is because panels are much more expensive in quantity of 2-3 compared to ordering 20 at a time.
Same goes for the pcbs and some parts
BUT
whenever I'm able, I order extras so I can at least have something in stock
At the moment, I have three assembled modules.
1 x MA20 ($240) (re:synthesis panel)
2 x CdS ($300 each) (1 re:synthesis panel) (1 FSFX panel)

Sorry for the spam!
I plan on making a MOA facebook page for this kind of thing over the weekend.
kindredlost
SOLD! - the CdS VCF/A that is.

PM sent on the QLPG thingy.
megaohm
Got a QLPG on the way. I'll update the thread ASAP.
SynthBaron
How in the hell do you release a product that's so dysfunctional in a basic way that could have been fixed so easily? Sigh...
kindredlost
It could be due to the scaling issue which Phil hinted at. If the original Buchla environment used a wider voltage range for modulation then the narrower scale expected by the more recent fare of 5U and 3U would present a weaker response.

I could get some action from the circuit but it would always "bleed" when driven high enough to allow any signal through. Only by boosting the modulation signal to a higher voltage range would create a more dynamic response, thereby damping the amount of bleed through.

I do agree that selling a legacy Buchla circuit in another format without taking this level or scaling difference into consideration was a fairly ignorant or careless gesture. At the time, I was new to the contemporary modular world. This was my first non-synthesizers.com module in my system. I had taken a two decade hiatus from the field and was unfamiliar with the current array of vendors. Cyndustries was as new to me as MOTM at the time. Hindsight...
robotmakers
This sloppy sort of "take a circuit and stick it in a modular" approach is not unusual. Take the MOTM-485 filter, for example. The original GX-1 circuit was designed to run at +/-1.5 v signal levels and was simply released unchanged as a MOTM module. Of course, it clips horribly when used in a MOTM system with +/- 10v signals. As a result, the MOTM-485 rightly gained a reputation as a poorly behaved, distorted filter, but is in fact a very mellow filter - when run at the levels for which the underlying circuit was intended.

Caveat modulator.

Cheers,
Roger
megaohm
Finally finished.
Hope this helps to make these modules more enjoyable.
kindredlost
Fantastic work Phil.

Nice documentation too. The details about the rebuild are very nice to know.

This looks like something people can do to remedy this poorly designed module. The concept of four LPG's in a single module is a good one - if they actually worked. hihi

A whole bag of roasted pistachios are on the way to you. This is fun!
megaohm
kindredlost wrote:

A whole bag of roasted pistachios are on the way to you. This is fun!


Sweet!
w00t

It's a cool module after the mod.
In most cases the partial attenuation will be enough and you won't have to use any external modules to condition the CV.
bwhittington
Great work, Phil! This module has always been mounted close to a CV amplifier or four in my cabinet because of this issue. I tried to dream up messy solutions from my non-technical perspective (2x amplication after the cv input behind the panel, etc), but I wouldn't know where to begin to come up with a fix like this on my own. Very excited about this. w00t

Such a cool module, but it has always been a bit of a drag to have work around this. Thank you for for efforts. Quite a service to those that use these.
megaohm
Thanks! Glad I can help.
Tell us how the mod goes once you've done it.
diophantine
Excellent mod document, Phil! Just wish I had one of these to implement it. Instead I have to get by with just two of your excellent CdS modules. thumbs up
Zeno
Thanks! I was close to getting custom built Buchla envelopes to remedy this issue but this looks like the perfect solution!
MUFF WIGGLER Forum Index -> 5U Format Modules  
Page 1 of 1
Powered by phpBB © phpBB Group