GSSL HELP THREAD!!!

[ArnoldLayne]

Whatever magnitude 'so much' might be ... heatsinks have been invented for this reason.

It's too hot to keep a finger on it. Obviously I can put a heatsink on it.

The differential between vregs input and output in V times connected load in A is the generated heat in W. A single incandescent lamp (FI illumination for GR-meter, illuminated switches, ...) might draw more current than the complete GSSL and up to now only you know what and how you have connected any additional parts. For usual the vregs in the GSSL, when built as shown on schematic, do not require additional heatsinking with a 2*15VAC or 30VAC center tapped mains transformer in front.

For now, i have only assembled gustav's PCB, no mods. Power transformer is a 2x15VAC, 1A (but I measure 18V on the PCB..).
I've been looking for shorts around that area, but haven't found any yet.

 

[thepraqtice]

Hi Harpo, thanks for your response.

In GSSL mode or turbo mode and both SSCF boards switches to bypass ?
Sure the pole of the turbo/gssl toggle connects to the top of the 15K (double value for balanced out operation) input resistor through the -from your wiring overview- light blue wire ? (connecting it to the top of the right side removed 47K resistor, just next to the top of the left side removed 47K resistor that feeds the upper SSCF board through the violet wire would be less confusing with a pic showing a previous revision board from year 2002).

Yes the problem happens in both GSSL/Turbo modes and If the Sidechains are in or bypassed and if compression is in or bypassed. The distortion seems to less the more compression is dialled in, which made me think it might be in input/output stages?

I soldered the pole from the GSSL/Turbo to the top of the 15k because the trace on my board became damaged by my violent solder pump. This one of the reasons why I can't strip the board back to just GSSL easily.
I just traced it to the nearest spot, this is ok isn't it? Do you mean the 15k resistor should be changed for this configuration?

Ratio connections are hooked up to the correct points on the GSSL/Lorlin pcb. Does wiring diagram seem correct other than this?
Where do you think the problem could be on SSC or Turbo boards?

I thought it could maybe be something to do with how the SSC boards were summing L + R and maybe sending an offset CV signal to the right VCA modulated by a signal from the left input (?) but I can't work out where this could be happening. Both SSC boards seem to be ok...

You'd need a Y cable with two T/S plugs for your external SCF XLR socket, just saying.

I wondered about how i'd done the ext sidechain wiring. What is the best way to wire the xlr if I just want to send it either a balanced or unbalanced signal from a balanced mono aux send on my desk?

Am confused about why a signal coming into the left input distorts through the right output.

I tried to swap the NE5532's from one of the sidechain boards with the opamps in the output stage and the problem went from signal at left input comes out left output fine but distorts on right output + signal at right input coming out of right output fine (no distortion on left).

To ---> Signal at left input with no signal at left output but distorted signal in right output and signal at right input does not come out right or left output at all.

I swapped the opamps back and the problem went back to how originally was. If I swap the opamps of left and right the problem stays on the sides. If I swap the VCA's the problem stays on same sides also.

The VCAs pin5 is a current source that connects to the -15V rail through a 5K1 resistor for the audio VCAs and to the -12V rail through 3K9 resistor(s) for the sidechain VCA(s).

What do you think the best course of action is? To replace opamps and VCA's with THAT 2180Bs without Pin 4 snipped?

Many thanks,

Sam

 

[Harpo]

I soldered the pole from the GSSL/Turbo to the top of the 15k because the trace on my board became damaged by my violent solder pump. This one of the reasons why I can't strip the board back to just GSSL easily.

: there might be a 3cm piece of wire or a resistor snippoff somewhere ...

I just traced it to the nearest spot, this is ok isn't it?

Yes, same spot, only 2cm apart.

Do you mean the 15k resistor should be changed for this configuration?

: you never answered what 'this' is. This is related to GSSLs balanced-out or unbalanced-out connection to your next piece of gear (a/d-converter, mixing desk, monitor amp, ...) and to be read in context with the (15K) feedback resistor value of the VCA following opamp.

Ratio connections are hooked up to the correct points on the GSSL/Lorlin pcb. Does wiring diagram seem correct other than this?

This ratio connection at GSSL-control pcb is missing in your diagram and you'd not be the 1st.to connect it wrong.

Where do you think the problem could be on SSC or Turbo boards?

Oh these blurry cristal balls, but I'd suspect your GSSL mainboard or control pcb 1st.

I thought it could maybe be something to do with how the SSC boards were summing L + R and maybe sending an offset CV signal to the right VCA modulated by a signal from the left input (?) but I can't work out where this could be happening. Both SSC boards seem to be ok...

After rectification and ratio circuit, the higher of both L/R CV signals is timed and controls all 4 VCAs. Again the connection of the turbo boards ratio wires wrongly connecting to the same pole and throw positions on control pcb might be the culprit.

You'd need a Y cable with two T/S plugs for your external SCF XLR socket, just saying.

Ignore this nonsense. I mixed up your white wire 0V reference voltage with send. Keep it as shown.

I tried to swap the NE5532's from one of the sidechain boards with the opamps in the output stage and the problem went from signal at left input comes out left output fine but distorts on right output + signal at right input coming out of right output fine (no distortion on left).
To ---> Signal at left input with no signal at left output but distorted signal in right output and signal at right input does not come out right or left output at all.
I swapped the opamps back and the problem went back to how originally was. If I swap the opamps of left and right the problem stays on the sides. If I swap the VCA's the problem stays on same sides also.

Broken opamp(s), a pin might have bent inwards when pushing opamp into its socket, (once was an) opamp fitted in wrong orientation when powered on, shorting solder connection at output stage, ...
Check if your audio path is working correctly with audio-VCAs pulled out of their socket and both audio-VCA sockets pins 1/8 linked with a bypassing piece of wire, so you can exclude broken opamps in this area. Powering on again, this should pass clean audio (obviously without compression). As the generated control voltage should be the same for both audio VCAs, double check for same DC millivolt measurement at both VCA sockets pin3 with audio testtone feeding both L/R inputs. All correct, power down, pull the wire links and reseat the VCAs back in their socket.

What do you think the best course of action is? To replace opamps and VCA's with THAT 2180Bs without Pin 4 snipped?

1st.check is for all 4 supply voltages are working correctly. Check wiering, parts values, parts orientation and check for shorts. No need to blindly replace all maybe still healthy opamps and/or VCAs. Isolate and identify the fault.
You can bend the 2180 VCAs pin4 sideways, so it doesn't connect to its socket (or just don't fit the 47R/68R/10K resistors that connect to this pin).
Good luck

 

[vp]

Hi everyone,

I would really appreciate some help with the following GSSL issue. I've searched the thread for any solutions and tried a few but still can't solve it...

I built the GSSL from Gustav's kit with the SSC board and it has been working happily for a couple of months, but recently it has developed an unpleasant distortion-like noise.

The distortion is only present when the unit is compressing and gets more pronounced with stronger compression (higher ratio, lower threshold). I examined, scoped, cleaned, tested everything but could not fix it. Apart from the distortion, the unit works as expected.

I think that the distortion is caused by an unstable CV signal. If I scope the point labeled 'C' in the schematics (with input e.g. 1kHz sinewave or any kind of steady signal), it resembles a low frequency sawtooth wave - I imagine that it should be a smooth line, not rippled, right?

I can measure the rippled signal throughout the control board and on the TL072 outputs.

Another suspicious signal is on pin 14 of TL074 - clipped upwards, but the negative peaks extend 5 or more times lower than signal on e.g.  pin 1 of TL074.

The supply rails are clean and stable (+-15V, +-12V), I swapped the TL074, TL072 and some diodes around TL074 to no effect. I also ruled out the relay (issues persist if I bypass it with wire).

Any ideas?
Thanks in advance!

 

[gyraf]

VP,

Strange. The only thing I can think of would be if the tantalum release-timing capacitors on the control board are oriented the wrong way around, destroying them over time..

Jakob E.

 

[vp]

Hi Jakob,

thank you for your answer - and of course many thanks for sharing the compressor (and other great circuits) in the first place!

Unfortunately it seems that it's not the release capacitors - I tried swaping one and the crackle persists... As the matter of the fact, I think I pretty much ruled out the part between 'C' and 'D' (methodically tested - took me a couple of hours).

Do you think it would be possible that the source of the problem is somewhere in the precision rectifier part? I cannot completely understand this part of the circuit with its feedback loops including the Ratio circuit etc. Could it be that something makes it produce some too big peaks at pin 14 of TL074 for the Attack/Relase smoothing circuit to handle?

It's a nightmare, I must have spent more than 16 full hours on this already...

edit:
I forgot to mention - the issue seems to be on the 'attack side' - the CV resembles reverse sawtooth.

 

[green-fuzz]

Hello everyone, this is my first post ever on a diy forum and it is about a problem on my very first diy project... So before I start : thank you all for the incredible amount of support this forum gives to a newbie like myself.

This is my porblem :

1/ When i fire up my gssl (no mod no nothing added for now) nothing visibliy happens. So i checked, and the ts7815 transistor seems to latch up ( as indicated by Jakob in this thread http://groupdiy.com/index.php?topic=440.0 ) : 23 volts go in, but +0.7v goes out instead of +15v. I guessed replacing the transistor should be the way to go BUT there came point two...

2/ If I do a little "power cycling" (i'm not sure that's the correct expression : i mean turn on, wait, turn off -the diode curiously lights up a little bit after i turned off- and turn on again), then miracle, the unit lights up and compresses stuff - to my great amazement-... but things don't get better that easy : first, the ts7815 heats up to the point where I can't touch it. Second, when i measure the voltage at the output of the transistor, i can see it decreasing slowly (5 or 10 minutes) from +14.96v to +14.87v. After that it doesn't move anymore. I don't know, maybe it's nothing, but the -15 doesn't behave like that and reads -14.97 all the way.... besides don't forget this is my first project and i know absolutely nothing....


So this is my question : should replacing the ts7815 transistor solve point 1 and point 2 ? Or are the overheating and little voltage loss a completely different problem ?

Thanks a lot everybody, and sorry for the long post...

Denis

 

[dave293]

Hi guys,

I have have just finished up my first diy project, a GSSL compressor and am having some issues.

On the power transformers secondaries I am getting a reading of 3vac.

Is there anything in particular I should be looking at as a cause of this problem?

The transformer is a Amveco 62063. Hooked up for 240v operation as shown here. (http://www.amveco.com/Miniature_Low_Profile_Transformers_2.htm

Any help is appreciated, thanks in advance.

 

[Majestic12]

On the power transformers secondaries I am getting a reading of 3vac.

Is there anything in particular I should be looking at as a cause of this problem?

The transformer is a Amveco 62063. Hooked up for 240v operation as shown here. (http://www.amveco.com/Miniature_Low_Profile_Transformers_2.htm

Any help is appreciated, thanks in advance.

Maybe you've connected your secondaries with inverted phase to each other.
Disconnect the two secondaries from everything else and measure the voltage on each of them seperately. If you still read something around 3Vac than you've got most likely a defective transformer.

 

[green-fuzz]

A little update on my problem, so that the next newbie like me can benefit from my experience : the transistor was faulty and did latch up. I changed it and everything is working just fine by now...

I still have a question, though as i took measure from +15 and -15 rails : +15 reads 14.88v (fluctuates to 14.87) and -15 reads -14.95. Is this difference a problem?

 

[dave293]

On the power transformers secondaries I am getting a reading of 3vac.

Is there anything in particular I should be looking at as a cause of this problem?

The transformer is a Amveco 62063. Hooked up for 240v operation as shown here. (http://www.amveco.com/Miniature_Low_Profile_Transformers_2.htm

Any help is appreciated, thanks in advance.

Maybe you've connected your secondaries with inverted phase to each other.
Disconnect the two secondaries from everything else and measure the voltage on each of them seperately. If you still read something around 3Vac than you've got most likely a defective transformer.

I disconnecting the secondaries and still got strange results (1.2v and 63v). I have ordered a replacement transformer, thanks!

 

[Gustav]

I still have a question, though as i took measure from +15 and -15 rails : +15 reads 14.88v (fluctuates to 14.87) and -15 reads -14.95. Is this difference a problem?

No, thats not a problem.

Gustav

 

[Harpo]

On the power transformers secondaries I am getting a reading of 3vac.
..
I disconnecting the secondaries and still got strange results (1.2v and 63v). I have ordered a replacement transformer, thanks!

Maybe order a replacement battery for your multimeter, giving such inconsistent readout values.
(you measured AC voltage across each winding ?)

 

[vp]

Hi,

I am still failing each day after another to fix this unfortunate CV issue on my GSSL...

For now, I would kindly ask anyone to please explain to me the following (as my issue might be attack-related = too fast attack):

Which components do affect the attack time?

Is it only the:
- selected attack series resistor
- selected release capacitor+resistor  to ground
- fixed 3M3 resistor to ground

Or is it also influenced by:
- circuit before point 'C'
- opamp stage after point 'D'
- the whole sidechain loop

I would deeply appreciate any explanations (or even some pointers to relevant literature, I wouldn't mind acquiring some more general electronics knowledge...).

Thanks in advance!
Vasja

 

[gyraf]

Attack time is determined by the (switched) attack resistor, and the (by release control) selected tantalum capacitor.

 

[vp]

Thanks Jakob!

Hmmm.... I can't figure this one out then...

And on the TL074 pin 1, after the diode, should be a half-wave rectified positive signal and on pin 14 a full wave rectified negative signal, right? (both a bit non-linear maybe)

 

[vp]

Hi again,

I'm really sorry to bother you all with my questions, but I am really desperate about my 'crackling' GSSL build... I hope I am not spaming, but rather contributing.

I've done some oscilloscope measurements of a 2kHz sine wave, amplitude modulated (100%) with a 100Hz square wave.

I would be really grateful if anyone could confirm if the measurements seem correct, and if not, what could be the cause...

Much obliged!

EDIT: I forgot to mention the compression settings:  threshold="fully CCW", ratio="1:10", attack="fully CCW", release="one before fully CW, i.e. through 1M2 resistor", makeup="fully CCW"

 

[green-fuzz]

Thank you so much Gustav,

it's been only two weeks since i recieved my kit from pcb grinder and thanks to your help i already have it up and running... besides i'm so happy with the sound that i've been compressing random stuff all evening, from my own work to youtube drum loops and Richard Dawkins lectures.

great fun.

Anyway, next step : super sidechain...

Thanks again to everybody on this forum (especially Jakob, regarding the latching up 7815). See you all... soon, I guess.  ;D

 

[thepraqtice]

Broken opamp(s), a pin might have bent inwards when pushing opamp into its socket, (once was an) opamp fitted in wrong orientation when powered on, shorting solder connection at output stage, ...
Check if your audio path is working correctly with audio-VCAs pulled out of their socket and both audio-VCA sockets pins 1/8 linked with a bypassing piece of wire, so you can exclude broken opamps in this area. Powering on again, this should pass clean audio (obviously without compression). As the generated control voltage should be the same for both audio VCAs, double check for same DC millivolt measurement at both VCA sockets pin3 with audio testtone feeding both L/R inputs. All correct, power down, pull the wire links and reseat the VCAs back in their socket.

Thanks again for all your help Harpo, think I'm starting to narrow this down.

I took out the VCA's and swapped with a small piece of wire as you suggested and the compressor passes very nice clean signal through both sides and good L/R balance with a mono signal. No distortion at all. Pop the VCA's back in and it returns on the right side as before.
Does this mean we can exclude the opamps as source of mysterious distortion? Do you think a faulty VCA might be the problem?

With VCA's out I checked mV's at Pin 3 with a test tone. Can't remember exact readings but they were the same on both sides at varying amounts of compression, so it's presumably not a CV offset issue.

I also checked the voltages at Pins 5 and 7 and found this a bit confusing.
Pin 7 is +14.95V on both L+R. But the reading on Pin 5 was a steady -10.72V on both sides. Should this not be closer to -15V? I seem to remember Pin 5 being -14.95V with opamps out before. Would this effect the VCA's performance?

Many thanks,

Sam

 

[Harpo]

Pop the VCA's back in and it returns on the right side as before.
Does this mean we can exclude the opamps as source of mysterious distortion? Do you think a faulty VCA might be the problem?

Either faulty right side audio-VCA or a solder bridge at the VCA on pcb. You could confirm this by swapping left side with right side audio-VCA. With faulty VCA, distortion will follow the faulty part. Distortion still at right side, use a magnifying glass to spot and remove the solder bridge.

I also checked the voltages at Pins 5 and 7 and found this a bit confusing.
Pin 7 is +14.95V on both L+R. But the reading on Pin 5 was a steady -10.72V on both sides. Should this not be closer to -15V? I seem to remember Pin 5 being -14.95V with opamps out before. Would this effect the VCA's performance?

Pin 5 is a current source that connects to the -15V rail thru a 5K1 resistor for the audio-VCAs or to the -12V rail thru a 3K9 resistor for the sidechain-VCA. This is not a direct connection to the supply rail.