GSSL HELP THREAD!!!

[MarioXLR]

Hello all.

I have made my own and is working perfectly, but...
when i put the Release in the AUTO position it starts to distort... strange, because works fine in any other position...
After some time analysing the schematic and the PCB Layout i found an error...

https://dl.dropboxusercontent.com/u/9196627/ssl-1.jpg
https://dl.dropboxusercontent.com/u/9196627/ssl-2.jpg

if you see on the previous links, When Release is ON, we have the 91K resist in parallel with 6u8 cap (on the schematic, but on PCB is with u47 cap).
Should i change the resistors on the PCB to keep schematic???

i´m not using any kit for now... just made the PCB and brought the components....

THX all

regards

 

[main repairman]

Hello all.

I have made my own and is working perfectly, but...
when i put the Release in the AUTO position it starts to distort... strange, because works fine in any other position...
After some time analysing the schematic and the PCB Layout i found an error...

https://dl.dropboxusercontent.com/u/9196627/ssl-1.jpg
https://dl.dropboxusercontent.com/u/9196627/ssl-2.jpg

if you see on the previous links, When Release is ON, we have the 91K resist in parallel with 6u8 cap (on the schematic, but on PCB is with u47 cap).
Should i change the resistors on the PCB to keep schematic???

i´m not using any kit for now... just made the PCB and brought the components....

THX all

regards

Correct values ​​are 91K+u47 and 750K+6u8

 

[gyraf]

PCB Layout is verified and known working.

Make sure your "auto" position actually connects to the time constant circuit (and that your rotary switch is programmed correctly).

Jakob E.

 

[MarioXLR]

PCB Layout is verified and known working.

Make sure your "auto" position actually connects to the time constant circuit (and that your rotary switch is programmed correctly).

Jakob E.

Hello,

yes, the rotary is working fine... in all other release positions is working fine... gonna change for new caps... maybe that...


THX

 

[main repairman]

Hi guys.
I simulate in the program LTspice GSSL compressor for measurement.
In the first graph, is the input signal and the signal at the junction R20k,10k, katode 1N4148. Is this the half-wave correct?
In the second graph, is the output signal with Tl074/pin14 2:1 ratio setting is this the opposite half-wave and +0.2 V offset correct?
Thank you for your help.

 

[main repairman]

Hi guys.
I simulate in the program LTspice GSSL compressor for measurement.
In the first graph, is the input signal and the signal at the junction R20k,10k, katode 1N4148. Is this the half-wave correct?
In the second graph, is the output signal with Tl074/pin14 2:1 ratio setting is this the opposite half-wave and +0.2 V offset correct?
Thank you for your help.

Guys suffice me answer correct -  incorrect.
I need to solve this problem.
Thanks

 

[main repairman]

No help?

 

[Harpo]

Need an echo answer for your 1st.pic?

with 0.14VAC 400Hz sine signal in (0.2Vpk, 0.4Vppk) it will swing to +0.2Vpk for the negative halfcycle at input and leave out (0V) the positive halfcycle

Waveshape looks as already described. Level will read the same with given values. (Your input signal level is 0.07VAC, 0.1Vpk, 0.2Vppk).
2nd.pic maybe yes, assuming you might have the complete schematic behind, except the bypassed sidechain-VCA, and not only up to the rectifiers. This 2:1 ratio setting is non-linear, so result will differ with different input level and a different circuit-simulator might have different parts models behind.
I have no urge to install your simulator, draw the complete schematic from scratch and run some numbers. Maybe some other members have more time to burn to get your request in a meaningful context.

 

[main repairman]

Need an echo answer for your 1st.pic?

with 0.14VAC 400Hz sine signal in (0.2Vpk, 0.4Vppk) it will swing to +0.2Vpk for the negative halfcycle at input and leave out (0V) the positive halfcycle

Waveshape looks as already described. Level will read the same with given values. (Your input signal level is 0.07VAC, 0.1Vpk, 0.2Vppk).
2nd.pic maybe yes, assuming you might have the complete schematic behind, except the bypassed sidechain-VCA, and not only up to the rectifiers. This 2:1 ratio setting is non-linear, so result will differ with different input level and a different circuit-simulator might have different parts models behind.
I have no urge to install your simulator, draw the complete schematic from scratch and run some numbers. Maybe some other members have more time to burn to get your request in a meaningful context.

Harpo YES I need to verify that the compressor is working properly for further building.
I'm sorry for the many questions, but my measurements are other in parts of SC
and this is my problem - I need to repair.
Input and output part is the same as you with your measurements - SC part no. Why?
All is checked. Values ​​are correct.

 

[gyraf]

Sorry, I don't see the need for discussing simulator software and it's possible errors in this thread? Why not just build the unit and measure it irl?

Jakob E.

 

[main repairman]

Sorry, I don't see the need for discussing simulator software and it's possible errors in this thread? Why not just build the unit and measure it irl?

Jakob E.

Simulation is the reason because Harpo's measured values are different than my (problem SC part - TL074/pin1/14). I need to find out the reason of different measurement results. I want to know all the details, all measured values ​​etc. for other building GSSL.

 

[MeToo2]

Hello all.

I have made my own and is working perfectly, but...
when i put the Release in the AUTO position it starts to distort... strange, because works fine in any other position...
After some time analysing the schematic and the PCB Layout i found an error...

https://dl.dropboxusercontent.com/u/9196627/ssl-1.jpg
https://dl.dropboxusercontent.com/u/9196627/ssl-2.jpg

if you see on the previous links, When Release is ON, we have the 91K resist in parallel with 6u8 cap (on the schematic, but on PCB is with u47 cap).
Should i change the resistors on the PCB to keep schematic???

i´m not using any kit for now... just made the PCB and brought the components....

THX all

regards

I can confirm the published PCB (and the PCB's from Gustav) are both correct.
Also confirm the redrawn schematic here is incorrect.

If you look at the original *SSL* schematic here you can also confirm which cap and which resistor go in parallel for yourself.

 

[frazzman]

Howdy folks - having built about 5 or so of these in the past without any major dramas... I was convinced I could build the GSSL with my hands tied behind my back until now.

I decided to build a unit for a friend and its turning out to be the biggest mistake I made Anyway....

I've got a unit that won't pass any audio at all on the right channel... all the obvious things have been checked:
- XLR wiring is ok
- Replaced all opamps and VCAs
- The fault doesnt follow the VCAs - i.e. if I swap left and right - the fault doesnt follow
- Checked all resistor values - everything checks out.

I was convinced that there must be a short that I can't see in the 5534 input opamp section... despite hours of scouring this area I couldnt find anything. In some late night desperation I then went to the drastic measure of gutting the entire input and output opamp sections, I removed every cap, socket and resistor in these areas and solder suckered every hole. Then applied a flux remover to clean things up....

Just finished reassembly in this area only to find the problem is still present.

If I disconnect the left channel signal the meter stops moving so I know that the signal isn't making it through to the output stage.

Having checked all of the above about what feels like a million times .... What are the usual suspects that result in no audio at all in a particular channel - particularly when all VCAs and opamps are OK?

I wanted to give this to my buddy as a gift but at this rate its going to end up in the trash! Its been many a late night troubleshooting... does anyone else have trouble going to bed when they know theres a build laying on the desk not working!!?

Many thanks

 

[Harpo]

If I disconnect the left channel signal the meter stops moving so I know that the signal isn't making it through to the output stage.

So signal doesn't make it to the output of the right side 5534 line receiver (input stage).
Double check it reaches the right side 22uF||100nF input caps by ohming out between XLR-pins 2/3 and input side of prementioned caps to exclude a bad molex connection.

 

[frazzman]

If I disconnect the left channel signal the meter stops moving so I know that the signal isn't making it through to the output stage.

So signal doesn't make it to the output of the right side 5534 line receiver (input stage).
Double check it reaches the right side 22uF||100nF input caps by ohming out between XLR-pins 2/3 and input side of prementioned caps to exclude a bad molex connection.

 

Thanks  Harpo - sorry I just realised I made a typo - this:  "if I disconnect the left channel signal the meter stops moving so I know that the signal isn't making it through to the output stage."
Should have read "if disconnect the right channel...." - I.e. if I disconnect this working channel the meter does not move so I figured its not making it across to output.

Just to clarify what you want me check - are we looking to measure resistance between right side xlr and the 22uf caps??

 

[Harpo]

If I disconnect the left channel signal the meter stops moving so I know that the signal isn't making it through to the output stage.

So signal doesn't make it to the output of the right side 5534 line receiver (input stage).
Double check it reaches the right side 22uF||100nF input caps by ohming out between XLR-pins 2/3 and input side of prementioned caps to exclude a bad molex connection.

 

Thanks  Harpo - sorry I just realised I made a typo - this:  "if I disconnect the left channel signal the meter stops moving so I know that the signal isn't making it through to the output stage."
Should have read "if disconnect the right channel...." - I.e. if I disconnect this working channel the meter does not move so I figured its not making it across to output.

Just to clarify what you want me check - are we looking to measure resistance between right side xlr and the 22uf caps??

Nothing to do with the output stage. Fault is at the now obviously left -not working- input side. With unit powered off and input disconnected check for close to zero ohms between input-xlr pins 2/3 and both 22uF or 100nF caps (doesn't matter which cap of each pair, they are in parallel). If you don't measure continuity/0 ohm between xlr-pin2 and cap and xlr-pin3 and the other cap, something is interrupting signal flow, most likeley at or inside the molex connection. A broken or lifted trace is another possibility. If you measure continuity/0 ohm between both cap pairs, there is a short on the path to the caps, most likley at the molex pcb solder connection.
Your sidechain is working, else the meter wouldn't move. Sidechain receives signal from summed line receivers and when meter stops moving when you disconnect the working right side, fault is at the left side line receiver. You already checked parts values and opamp working, so fault most likeley will be in front of this opamp.

 

[Olegarich]

frazzman, do you have any oscilloscope? Few weeks ago i had problem and I've been searching for a short or wrong value resistor or cap orientation for a 3 days and everything looked well. After I printed schematics on the paper put it in front of the board, 1khz wave to compressor and started to check all the way the sound travels, that lead me (and Harpo help) to the sidechain problem and in the end i found short in the corner, but what i want to say is, in my case best way to find eror was to look on schematics and check components all the way sound travels, if you getting no sound where it should be probably there is gonna be a problem (because you sidechain looks good). Defenitely check out does your parts receiving needed volatges.

 

[Rockitdog]

Hi Guys.

I wonder if someone is willing to help me. I have a power transformer for my GSSL thats 50VA (230V primary / 2X15V secondary) says on the box that its rated at 1.66Amps.

I am new to power transformers, I have successfully fitted and tested another transformer for a 1176 rev a I built. But that was from Hairball audio and obviously selected by the guys there.

I know that 50VA is quite a bit bigger than the recommended 2X15v / 10VA from Gyrafs BOM. It says "10VA or more" but how much more before armageddon?? I am quite nervous about messing around with high voltage and obviously want to be 100% certain before committing.

Question is, If I fit this transformer, will I cause any damage to the board or components? I will obviously ensure that no IC's are socketed during testing of the power rails. 

Please help and thanks in advance.

Cheers
Richard

 

[Harpo]

Question is, If I fit this transformer, ...

Question is, how to make it fit in your maybe 1HE box because of your boxes inner dimension limits. 50VA is oversized, but won't cause damage for its rating.

 

[Rockitdog]

Question is, how to make it fit in your maybe 1HE box because of your boxes inner dimension limits. 50VA is oversized, but won't cause damage for its rating.

Thanks for the help. Thats what I thought, but wasn't sure.

It measures +/- 37mm in thickness, I've ordered but havent received the chassis yet. It looks like its going to fit (Just measured it with one of my other 1 HE units). Knowing my luck it wont fit at all, if thats the case I'll order a smaller one and put it down to school fees.

Thanks again for the help, very much appreciated.