Yes, this circuit has been built a couple of thousand times and was found working
Earlier in this thread we have already discussed that biasing thing - it has to do with reducing distortion as per THAT's vca usage application notes
/Jakob E.
GSSL HELP THREAD!!!
- Thread starter hairyandy
- Start date
Help Support GroupDIY Audio Forum:
Yes, this circuit has been built a couple of thousand times and was found working
Earlier in this thread we have already discussed that biasing thing - it has to do with reducing distortion as per THAT's vca usage application notes
/Jakob E.
front247
Member
I have been using my GSSL for over two years and everything has worked well from the very beginning, the system is quiet, almost inaudible, no noise or buzzing, it compresses both channels perfectly, the bypass works great... but from the very beginning there is one small problem: the right R channel is quieter by exactly 4 dB, generally it didn't bother me because I always equalized the levels on the sources, but now I have some free time and it doesn't bother me. XLR inputs and outputs checked, all op amps checked, working and replaced, vco - checked and replaced, working, the entire audio input and output path checked element by element - everything is working. I've lost ideas at this point... the rest of the system, as I read, influences the signal but only controls it. What else can cause one channel to be 4 decibels quieter? Do you have any idea where to look?
Read (precise) voldtages at control-inputs to the two main vcs's?
I am assuming it is 4db lower in bypass right? I had a similar issue and what I did was run a 1khz sine wave with both input and output connected and started at the output XLR connectors and checked voltage back towards the inputs following the signal path and see where the voltage changes. Sounds like you've got an incorrect resistor in the signal path or short somewhere if I had to guess.
front247
Member
Hello
I measured the voltages: input -15.24V +15.24V
At the outputs of the potentiometers to the VCO - right channel - 3.20V, left channel 3.19V.
On operational amplifiers Left channel - 0.62V and -0.63V. Voltages measured within the VCA circuits
front247
Member
Today I will do the same, I will pass sine waves and look for the path, thank you for the tip, although, as I wrote, I measured the voltages and measured all the elements where the analog signal goes and there is perfect symmetry, I selected the resistors so that they are identical, the capacitors too, but I will check, thank you for the tip
I have a 4 dB drop on the right channel all the time, both during compression and bypass.update: I passed a 1kHz sine signal through the signal path: at the inputs I have: 28mV and 28.4mV but... I checked how the cables are connected to each other without GSSL and they have a perfect 28mV for both channels. I don't know if .4 mV is a lot...? I went further and at the input of the ne5532 opamp I already have 6.8mV - R channel and 6.2mV - L channel... at the input of the operational amplifiers after the capacitor I already have 1.2mV left channel and 1.8mV right channel. So there is a difference, but I don't know if it is so big that it affects the 4 dB drop? I can't compare it, but there is a difference. But I don't know what causes it, because from the very beginning the left channel is getting bigger and bigger.I followed the entire signal chain and the output of one of the VCAs is 10.8mV and the output of the other is 6.2mV, so this is a significant difference, I only increased the sine input signal a bit... but at the outputs of 22uf capacitors, from which the signal through the res\ The 27K ystor goes into the VCO and is respectively: 1.8mV and the second one is 1.2mV. I replaced the VCA channels, I have the 2181 and it is identical, so the VCOs are functional, I think, there is also this difference at the JRC 5584 outputs, but... at the inputs it is much smaller, i.e. 6.8 mV and at the other 6.3 mV... but it is I also swapped places and it's exactly the same, so I don't know where the signal is lost... I have no idea how to check it to locate the element of the audio track that is responsible for it. I will also add as a side note that when changing the settings of the attack and realise controls, the difference increases slightly to about 6 dB at extreme attack and release settings. But the problem, in my opinion, definitely lies in the signal path.
Last edited:
Are you checking AC voltage for the audio signal and not DC? If your meter is on AC then sounds like you either have a really low level going through the unit (and you need to raise it) or your not checking the correct components.Today I will do the same, I will pass sine waves and look for the path, thank you for the tip, although, as I wrote, I measured the voltages and measured all the elements where the analog signal goes and there is perfect symmetry, I selected the resistors so that they are identical, the capacitors too, but I will check, thank you for the tip
update: I passed a 1kHz sine signal through the signal path: at the inputs I have: 28mV and 28.4mV but... I checked how the cables are connected to each other without GSSL and they have a perfect 28mV for both channels. I don't know if .4 mV is a lot...? I went further and at the input of the ne5532 opamp I already have 6.8mV - R channel and 6.2mV - L channel... at the input of the operational amplifiers after the capacitor I already have 1.2mV left channel and 1.8mV right channel. So there is a difference, but I don't know if it is so big that it affects the 4 dB drop? I can't compare it, but there is a difference. But I don't know what causes it, because from the very beginning the left channel is getting bigger and bigger.I followed the entire signal chain and the output of one of the VCAs is 10.8mV and the output of the other is 6.2mV, so this is a significant difference, I only increased the sine input signal a bit... but at the outputs of 22uf capacitors, from which the signal through the res\ The 27K ystor goes into the VCO and is respectively: 1.8mV and the second one is 1.2mV. I replaced the VCA channels, I have the 2181 and it is identical, so the VCOs are functional, I think, there is also this difference at the JRC 5584 outputs, but... at the inputs it is much smaller, i.e. 6.8 mV and at the other 6.3 mV... but it is I also swapped places and it's exactly the same, so I don't know where the signal is lost... I have no idea how to check it to locate the element of the audio track that is responsible for it. I will also add as a side note that when changing the settings of the attack and realise controls, the difference increases slightly to about 6 dB at extreme attack and release settings. But the problem, in my opinion, definitely lies in the signal path.
front247
Member
Of course, I check on the DC range, but in fact I did the tests at a very low signal amplitude, after raising the signal to half, instead of e.g. and the right channel
I don't have much experience in such testing with a 1kHz sine wave signal. So what amplitude of the signal transmitted through GSSL should be set?
Audio is AC. Think of it as a waveform, like in a waveform editor. It goes into plus and minus to move the speaker cone out and in. A drop of 4dB should clearly show when measuring AC as a difference in the order of V (not mV) .
If you measured DC, you now know what the DC offset is. Think of it as the zero crossing of a waveform. Some mV of DC is not unusual. If too high, your zero crossing is off.
If you measured DC, you now know what the DC offset is. Think of it as the zero crossing of a waveform. Some mV of DC is not unusual. If too high, your zero crossing is off.
The relevant voltages are the control voltages going to VCA's - in the "202-emulation" circuit, NE5534's pin6's
If these are the same, there probably is a resistor with wrong value before or after the VCA
hint: 4dB is factor 1.58 or 0.63
If these are the same, there probably is a resistor with wrong value before or after the VCA
hint: 4dB is factor 1.58 or 0.63
front247
Member
On pin 6 in the VCA system on NE5534 I have perfectly equal 15.26V on both channels... is it bad?
I measured more precisely and the entire INPUT input up to the 22uF/35V coupling capacitors had identical signal values. Then the signal enters VCA 2181 through the ne5534 chip and here, too, the voltage values on both channels are perfectly equal... The signal voltage changes only after the VCA on the 15k resistor and is 4.6V for the left channel and 8.2V for the right channel. So it turns out that the error is at the VCA output, but... I replaced several pieces interchangeably in both channels and it is still the same, only the right channel somehow produces a lower signal than the left one, even though the VCAs are functional.I don't understand one more thing - at the VCA pin 8 outputs I have 0.001mv DC and behind the 15K resistor, i.e. at the output of the first half of the NE5532 I have 4.6V and on the second channel 8.2V... and I don't understand why it is like this I have to switch the meter between AC and DC because otherwise I cannot measure anything at certain points. I thought the signal was linear point by point, but something is wrong here. But the compressor works perfectly fine except for this 4 dB drop. I can't find it because I don't understand how it is possible?
I measured more precisely and the entire INPUT input up to the 22uF/35V coupling capacitors had identical signal values. Then the signal enters VCA 2181 through the ne5534 chip and here, too, the voltage values on both channels are perfectly equal... The signal voltage changes only after the VCA on the 15k resistor and is 4.6V for the left channel and 8.2V for the right channel. So it turns out that the error is at the VCA output, but... I replaced several pieces interchangeably in both channels and it is still the same, only the right channel somehow produces a lower signal than the left one, even though the VCAs are functional.I don't understand one more thing - at the VCA pin 8 outputs I have 0.001mv DC and behind the 15K resistor, i.e. at the output of the first half of the NE5532 I have 4.6V and on the second channel 8.2V... and I don't understand why it is like this I have to switch the meter between AC and DC because otherwise I cannot measure anything at certain points. I thought the signal was linear point by point, but something is wrong here. But the compressor works perfectly fine except for this 4 dB drop. I can't find it because I don't understand how it is possible?
Attachments
Last edited:
15.26V is wrong, try again?
front247
Member
What try again?
Wrong voltage - perhaps measured wrong or at a wrong pin
Hello Everyone ! just finished my Gssl with Super Sidechain mod. and well, I get no signal at all. I know, I should have build the vanilla version first, check and then put the side chain mod in ... me stupid...now I have to pay for it.... I will do this tmrw. But before I go to sleep I was thinking putting it here. maybe someone finds an error immediately, or maybe I get some tipps on how to start troubleshooting. Will reverse to vanilla version asap to see if the error is in the mod. I only checked for 15v + and - , nothing more, this was fine. send audio in but got nothing out, with compressor engaged or bypassed. leds are on, so that's a good start
Attachments
Last edited:
the_racker_reilly
Member
do you have an audio probe, or can you make one? thinking you could follow the audio path from the input with it and hear where the signal goes missing...
Missing th hpf jumper?Hello Everyone ! just finished my Gssl with Super Sidechain mod. and well, I get no signal at all. I know, I should have build the vanilla version first, check and then put the side chain mod in ... me stupid...now I have to pay for it.... I will do this tmrw. But before I go to sleep I was thinking putting it here. maybe someone finds an error immediately, or maybe I get some tipps on how to start troubleshooting. Will reverse to vanilla version asap to see if the error is in the mod. I only checked for 15v + and - , nothing more, this was fine. send audio in but got nothing out, with compressor engaged or bypassed. leds are on, so that's a good start
I'm not sure if the ssidechain uses it still?
Last edited:
You have to put a jumper in it, or the connection will not be made through the side chain.
The connection is just a convenient spot for adding a capacitor in series with the path, but if you dont put anything in there, it leaves a gap - even if you have the SSC connected elsewhere.
Gustav
Thanx ! I will put jumper there right now. was confused, cause on some pictures of other builds there ws nom jumper either...hopefully this it !
OK, update
I put in the Jumper.
-Meter is moving now
-rtio, attack, release seem to work , judging from meter movement changes.
-threshold, and ,,of course,,
makeup doing nothing-still no Audio Out, also not in bypass mode
