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

Random*Source Serge 4x4 Random Source locking up
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Author Random*Source Serge 4x4 Random Source locking up
diophantine
Hopefully others here have successfully built their R*S 4x4 SSG/RS module... Having never used either before, I can't entirely judge their behavior.

SSG-only PCB seems fine. I calibrated it as described and it generally behaves as I'd expect.

I calibrated the SSG/Noise PCB by putting it behind the SSG pot/jack PCB. It went similar to the SSG calibration, and appears to operate similarly. Max smooth rate is approx. 4.1 kHz. Max stepped rate (when clocked from the smooth) is ~715 Hz.

When I put the SSG/Noise PCB behind the RS pot/jack PCB, the noise and S&H work fine. However, the random voltage source side is problematic: at some settings it works fine, but at other times it "locks up" and produces a steady high voltage. Scenarios like:

- Rate knob CCW, turn on, high steady signal. Turn knob CW, around 12:00 it starts outputting expected signal, and subsequently works as expected regardless of knob position.
- Rate knob 12:00, turn on, brief normal signal, then locks up on steady high signal after a few seconds.
- Rate knob CW, turn on, high steady signal. Turning knob has no effect.

Here's a video showing the middle scenario:


I've tried changing the trimpot settings a bunch, which usually results in some change of behavior, but it is usually minor or just gets rid of the first of those scenarios I mentioned.

Given that the SSG seems to work properly, I'm guessing that this has to do with the configuration of these into the random voltage source. (I have a schematic of this, but don't know if it is 100% accurate.) If anyone can help out here, or let me know how your configuration went, that would be much appreciated!
diophantine
Addendum:

I wired up (via bananas) the SSG (SSG-only board) to be a Random Voltage Source:
Top Coupler Out -> Smooth In
Top Coupler Out -> Stepped Sample
RS S/H Src -> Stepped In

I was able to get it to lock up twice, in the same way as the RVS in the RS, though it took a lot of work... and I'm honestly not sure what I did to cause it.

. . . hours later after being distracted by dinner, etc. . . .

I thought the S/H Unipolar trimpot was just for the S/H output, but it turns out that it affects this issue. When I turn the trimmer fully CCW it shuts off the S/H Src output. When I turn it barely back on it seems to mostly solve my problem! There's still some pot positions (CCW) where the RVS output is high when the system turns on, but turning it pushes it into normal operation mode.

I might play with it more tomorrow to see if I can get it perfect, but now it is at least totally usable!
bananeurysm
I've been having the exact same issue calibrating mine.

I was able to mitigate it somewhat by a LOT of trial and error with the unipolar noise trimpot - and also I needed to tweak the smooth and stepped trim pots as well... they all seem to affect each other, and the problem. I'm sure I could still get it better (safer from lockups) and also with a more ideal range as I ended up having to tweak the smooth and stepped rate trimmers a good deal as well.

I was mostly getting lockups with the RVG rate pot fully clockwise.. the unipolar noise trimmer seems to set the range of the noise which can limit lockups but this also affects the range of the smooth and stepped outputs. FWIW I was getting lockups (not always at the same time) on the smooth/stepped outputs - and also on the S+H outputs. (By the way... are these meant to be the same or what?)

Ken Stone does talk about calibrating around these lockups a little bit about towards the bottom of this page:
http://www.cgs.synth.net/modules/cgs97_noise.html

but unfortunately doesn't offer a very specific calibration routine. I'm sure some old serge heads know how to get it right tho!
nordlead
I've just checked my RandomSource module here, in particular the 3 scenarios diophantine describes above and all three of them work fine. Some thoughts:

- some Serge modules like the DSG or the SSG can occasionally hang - happened in my Creature from time to time. It's not clear to me when / why that happens. On the DSG it's fairly easy to patch something complex that will kill the cycle, but temporarily removing a cord into trig in usually fixes that. However, I didn't run into this when playing with the R*S Random Source so far.

- The Noise source is a quirky ("analog") design and linking it to the SSG kind of turns it into a ... Random Source! So it's not so easy to figure out what's going on or what might be going on. Since I can't reproduce those lockups here, it's hard to remote-diagnose.

- When the RATE is turned down, the signal moves quite slowly, so depending on where it's at, it might appear to be locked.

- It's been a while that I built mine and wrote the docs. Still as far as I remember I took a very straightforward calibration approach: calibrated the SSG as described to get the max range. Then I believe I simply calibrated the noise side so that the lower S/H Out is (fairly) bipolar - my scope now shows an average of appr. -500mV. The (white) S/H out shows an average of appr. 2.3V, i.e. seems to sit above the 0V line, i.e. most of the lowest points are around 1V, some occasional spikes go lower (this is random!):



My S&H SRC looks like this:



If your S/H output looks different, try adjusting it using the S/H trimmer till it looks similar.

- I don't think I changed any SSG calibration afterwards - I'd rather try to avoid that.

- Just in case: might be worth checking that the power supply is actually delivering +/-12V. Any power issues could lead to unforeseen consequences in this setup.

Hope this helps a bit!
xonetacular
I was having the same issue the other week- messing with the trimpots seemed to mostly fix it.


I have another issue/question- how is the trigger button and trigger i/o supposed to work?

If I'm understanding it should send a trigger from the trigger i/o jack when the button is pressed, but hitting the button doesn't seem to do anything when I hook it to the trigger in on the dsg or anything else.

not really understanding how to use the two s/h out jacks and what they are supposed to do
bananeurysm
xonetacular wrote:
I was having the same issue the other week- messing with the trimpots seemed to mostly fix it.


I have another issue/question- how is the trigger button and trigger i/o supposed to work?

If I'm understanding it should send a trigger from the trigger i/o jack when the button is pressed, but hitting the button doesn't seem to do anything when I hook it to the trigger in on the dsg or anything else.

not really understanding how to use the two s/h out jacks and what they are supposed to do


My guess is you're getting a high DC value out of the SH jacks that doesn't change when you press the trigger? If you fiddle with the trimpot on the noise pcb and power cycle it - you should be able to find a value where you get a different voltage out of the SH jacks as you press the trigger.

I had to do this a number of times until it (mostly) works without locking up. This is what Ken is referring to on his page linked above.
xonetacular
bananeurysm wrote:


My guess is you're getting a high DC value out of the SH jacks that doesn't change when you press the trigger? If you fiddle with the trimpot on the noise pcb and power cycle it - you should be able to find a value where you get a different voltage out of the SH jacks as you press the trigger.

I had to do this a number of times until it (mostly) works without locking up. This is what Ken is referring to on his page linked above.


yes but the rest isn't locked up.

Hmm. The S/H src outs seem to work properly and change when I send a gate into the trig i/o jack, but pressing the button still doesn't do anything. hmmm.....

been fiddling with that trimpot and no luck yet. I wonder if I did something dumb. would sending a trig in work if the button isn't?

is it possible to put that switch in backwards?
xonetacular
before I rip apart the component pcb to flip that switch can anyone confirm if it needs to be installed in a certain direction? looking at the datasheet it looks like it matters but I'm not sure.

it was a while since I built it but I don't remember if the pcb had any indication of direction
the bad producer
Looking at the datasheet, it seems that you can put the switch in backwards - as it were - having said that I've not seen the PCB or anything:

http://www.mouser.com/ds/2/60/-6481.pdf
xonetacular
All fixed now, pin 1 was marked on the pcb but I missed it. d'oh!
EarlJemmings
I'm having some issue with my DUSGs as well. Both sides are simply outputting a steady 11.5V signal, nothing changing that, and about 2.5V coming out of the input. Two of the ICs on the board get very hot- too hot to touch even.
Similar on my Carnivore, and same two ICs, the two eight legged ones sitting next to each other, can't make out the lettering.
The left side of the carnivore was working perfectly, but then started behaving like that. The right side of the carnivore's dusg responds to input, but that's all. Stepped random I can't get to cycle, but can't test past that just yet.

It would seem like a power issue to me, or a short somewhere by the power on the board, but I've checked everything over, and can't find a single power issue.
Any help would be much, much appreciated!


Leturbo
EarlJemmings wrote:
I'm having some issue with my DUSGs as well. Both sides are simply outputting a steady 11.5V signal, nothing changing that, and about 2.5V coming out of the input. Two of the ICs on the board get very hot- too hot to touch even.
Similar on my Carnivore, and same two ICs, the two eight legged ones sitting next to each other, can't make out the lettering.
The left side of the carnivore was working perfectly, but then started behaving like that. The right side of the carnivore's dusg responds to input, but that's all. Stepped random I can't get to cycle, but can't test past that just yet.

It would seem like a power issue to me, or a short somewhere by the power on the board, but I've checked everything over, and can't find a single power issue.



I don't think this will resolve all of the issues here (truth be told I'm having some myself with the carnivore), but for the DSG on the carnivore you need a jumper on the lower COM pads:



For the regular DSG this jumper isn't needed, according to the build notes.

Denny
tojpeters
Your ICs should not be getting hot.
Usually that means your power lines are backwards.
Even if the power was reversed momentarily and you quickly unplugged it and fixed it the damage is done.
EarlJemmings
Leturbo wrote:
I don't think this will resolve all of the issues here (truth be told I'm having some myself with the carnivore), but for the DSG on the carnivore you need a jumper on the lower COM pads... For the regular DSG this jumper isn't needed, according to the build notes.

Denny


Thanks for the heads up! I'll add that on
EarlJemmings
tojpeters wrote:
Your ICs should not be getting hot.
Usually that means your power lines are backwards.
Even if the power was reversed momentarily and you quickly unplugged it and fixed it the damage is done.


That might be the issue, thanks. At one point the cgs power distro board touched the boat and shorted. I don't remember what was plugged in at the time, but it's possible. Do I need to replace the ones that are getting hot then, or is it something earlier in the flow?
If it is those ones, does anybody know what they are? the heat took the lettering off, I can't make out the lettering anymore
EarlJemmings

(these are the ICs btw)
cygmu
EarlJemmings wrote:

If it is those ones, does anybody know what they are? the heat took the lettering off, I can't make out the lettering anymore


My understanding is that R*S remove the markings on some of the ICs they use to protect information about their design upgrades on these circuits. If that is the case for these ICs then perhaps the only people who know what they are are Ralf, Matthias and maybe Serge.
EarlJemmings

So CGS has what I think are the corresponding ics listed as 4558 and LF353, alternatively TL072 for each, which seems more likely to me.
I thought those were pretty hard to break chips, but I'll try swapping them out.
Does anybody happen to know the size of SMT used?
cygmu
OK, so those are op-amps, and TL072 etc will work ok in the circuit. Random Source have done quite a bit of work to identify the particular op amps that they think work best in this circuit and I am willing to bet they're not TL072s so you will perhaps sacrifice a bit of performance.

They look like SOIC-8 packages from the photos. Someone who has an actual unit may be able to correct me on that; and someone with better knowledge than I have might be able to correct everything I've said...
EarlJemmings
cygmu wrote:
OK, so those are op-amps, and TL072 etc will work ok in the circuit. Random Source have done quite a bit of work to identify the particular op amps that they think work best in this circuit and I am willing to bet they're not TL072s so you will perhaps sacrifice a bit of performance.

They look like SOIC-8 packages from the photos. Someone who has an actual unit may be able to correct me on that; and someone with better knowledge than I have might be able to correct everything I've said...


Thanks! If anyone could confirm that the names have been removed from their DUSG, I'd be much appreciative. I'll also send an email to random source to see if they'll help me out with identifying the chips
Leturbo
EarlJemmings wrote:
cygmu wrote:
OK, so those are op-amps, and TL072 etc will work ok in the circuit. Random Source have done quite a bit of work to identify the particular op amps that they think work best in this circuit and I am willing to bet they're not TL072s so you will perhaps sacrifice a bit of performance.

They look like SOIC-8 packages from the photos. Someone who has an actual unit may be able to correct me on that; and someone with better knowledge than I have might be able to correct everything I've said...


Thanks! If anyone could confirm that the names have been removed from their DUSG, I'd be much appreciative. I'll also send an email to random source to see if they'll help me out with identifying the chips



Confirmed: the IDs on those chips are wiped off.
EarlJemmings
Leturbo wrote:
Confirmed: the IDs on those chips are wiped off.


Cheers, I've sent an email to R*S, hope this can be solved easily
Leturbo
I just noticed as well... you appear to be using the wrong SSG PCB for the carnivore. Again, not related to the DSG at all, but I'm guessing either you got the wrong PCB with the carnivore, or mixed it up with the regular SSG module. The components are all missing in the bottom half.

Otherwise, any luck hearing from R*S?
EarlJemmings
Leturbo wrote:
I just noticed as well... you appear to be using the wrong SSG PCB for the carnivore. Again, not related to the DSG at all, but I'm guessing either you got the wrong PCB with the carnivore, or mixed it up with the regular SSG module. The components are all missing in the bottom half.

Otherwise, any luck hearing from R*S?


Hm, it's one I got as a package, I think from R*S, but it may have been from Modularaddict, but I don't remember. It is working perfectly though, aside from the noise section, but I believe that's due to the transistor. I checked the pinout, but I may have gotten it backwards.
What do the extra components do?

And no, no response from R*S, I think I'll just order the 4558 and LF353, as well as the other ICs on the board and just swap out the lot to be safe. Nothing appears scorched, the caps are fine, so I assume it's just one of the chips.
Leturbo
Quote:
Hm, it's one I got as a package, I think from R*S, but it may have been from Modularaddict, but I don't remember. It is working perfectly though, aside from the noise section, but I believe that's due to the transistor. I checked the pinout, but I may have gotten it backwards.
What do the extra components do?

And no, no response from R*S, I think I'll just order the 4558 and LF353, as well as the other ICs on the board and just swap out the lot to be safe. Nothing appears scorched, the caps are fine, so I assume it's just one of the chips.


According to the build notes for the carnivore (http://randomsource.net/docs/RandomSource_Serge_Carnivore.pdf)

"Both the SSG and DSG pcbs
exist in different versions with different
funcionality. For Carnivore you must
use an SSG pcb that does not only
contain parts in the upper area. "

The picture of the finished module has the parts showing as well:



One of the ICs is a CD4007UBM... as to how it works... I'm not at that level yet. lol. I'm literally following build docs to a T.... which sometimes gets me into trouble since there are errors in them...

Denny
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group