Calibrating the G-SSL with THAT2181 - share thread !!!

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Purusha

Well-known member
Joined
Jan 14, 2006
Messages
1,868
Location
Somewhere on the Planet
I invite you all to give your own piece of advice and share what ever you did to calibrate your own G-SSL with THAT2181 inside. This Thread may also become a part of the SSL meta thread one day :wink:
 
OK, So far my G-SSL's voltage is correct, sound is passing through but the unit doesn't sound calibrated yet.

I am using the 2181LB and have done these changes so far :

- took out the 68R next to VCAs

- replaced the 10K with the jumper next to the VCAs

- replaced the 1M resistor with the 220K

- replaced the 100K to 120K on the right of the TL074

- replaced the 100R with 51R right of the TL072

- replaced the two 47K summing to 120K

- soldered the 47K to the +leg of the threshold pot

RESULTS:

Less sensitive threshold but still getting distortion on 6dB GR
strange attack and release behavior, faster release shows more GR but less distortion than slower release, usually is the way around.

Anyway, threshold 12 o'clock, still reading 10dB GR.... any suggestions what should I try next?
 
Purusha I did some of the modification on My SSL using 2181LB VCA but I didn't modifie this :

- replaced the 100R with 51R right of the TL072

- replaced the two 47K summing to 120K

What this modifications done?
 
I don't know about the 100R, but the 120K will lower the sensitivty on the treshold. I got the info about the changes from Reggie.

What about your G-SSL mathflan? I saw your post in the SSL help thread. Does your threshold work properly? Any distortion when you compress 6dB or more?
 
Here's what I did:

took out the 68R next to VCAs
YES
- replaced the 10K with the jumper next to the VCAs
YES
- replaced the 1M resistor with the 220K
YES
- replaced the 100K to 120K on the right of the TL074
USED 122k
- replaced the 100R with 51R right of the TL072
NO
- replaced the two 47K summing to 120K
YES
- soldered the 47K to the +leg of the threshold pot
USED 27k

ALSO: changed the two 15k resistors by the power tranny to 27k and switched the meter sensitivity resistor to 1k. Also used the "LB" VCA

Steve
 
Guitarmaker, how does your G-SSL work now? Have you checked if your ratio is accurate?

I am not so much concerned with my threshold being so sensitive, I am sure this can be adjusted, what concerns me the most is the aggressiveness which I don't like and the distortion I hear too soon.

I hope this is not part of the G-SSLs charm :green:
 
Purusha

My SSL worked nice one weeks ago, But now I have some trouble that I try to fix, I don't know why It's doesn't works properly...

but when My SSL worked good, I can say that yes the Threshold was a little too sensitive, My ratio became good with the resistor 120 k.
But I don't understand why my SSL compressed when the threshold is at 0; at 12 O'clock, with 0db signal at the inputs, is it normal... :?:
 
I have the same problem, my compressor is compressing although the threshold is at 12 o'clock. I haven't checked the ratio yet.

Maybe some of the gentleman here can help us in this regard.
 
this is not a problem, it is just normal.
the threshold is NOT labeled like "compression starts at this level",
and the threshold changes with the ratios, too.
at the end it shouldn't be labeled in dB ...
 
Thanks Volker. What about the distortion I am hearing. It's not like an overdriven electrical guitar distortion but it's there in a subtle way. I don't like it, it reminds me of RNC on the low frequencies.

My ratios are:

2:1 - Input increased 10dB, Output increased 5.5dB.
4:1 - Input increased 10dB, Output increased 2dB.
10:1 - Input increased 10dB, Output increased 0,8dB.

I don't know how to translate this to the proper ratio figure but I know it's not right yet.

Can someone suggest what should I change next to get closer to the desired ratios. A 1K in parallel with other resistor was mentioned in the
"SSL Clone compressor ratio measurements" thread. Is this the *1K near the VCA?
 
[quote author="Purusha"]My ratios are:
2:1 - Input increased 10dB, Output increased 5.5dB.
4:1 - Input increased 10dB, Output increased 2dB.
10:1 - Input increased 10dB, Output increased 0,8dB[/quote]

your results will change depending on how much gain reduction is going on, because the GSSL has different knees with each ratio.
your results are not the exactly math, but are looking near enough.
 
I initially didn't like the sound of this compressor at all until I "tamed the threshold" I felt it was waaay too distorted with the LB VCA anyway. With all the mods I posted earlier the ratios are spot on and it sounds very nice. Much more subtle and useable IMO. Initially it was harsh, grainy, and compressed too early. Hope this helps.

Steve
 
Edit 3-27-06:
After re-reading this post, I think that I am confusing the issue here a little. My point was to comment that using the pre-trimmed VCA's is easier. However, this is a thread about using 2181's. Sorry if I was confusing. If you don't have VCA's yet, you can buy 2180's and get away with a shorter set of mods.

Original Post:
My previous build used the pre-trimmed 2180's, and here is what worked for me. This was originally suggested by Steffen and seems to be simpler than the mods suggested here. This mod works perfectly in my opinion, with no distortion or pumping. Note that the 2180 and 2181 are the same chips, anly the 2180 is pre trimmed. The trimmer circuitry needs to be left in for the 2181. I am almost finished with another GSSL and will be using 2181's this time. I'll report back with when I am done, and I'll also run some ratio curves too.

1) 100K resistor driving neg input of TL072 to VCA CV's replaced with 122K to get proper ratios for THAT 2180LA's.

2) 2K resistor feeding meter replaced with 1K +/-5% for 0-20dB range.
Note: I tried to get a 0-10 dB range, but there wasn't enough current with the resistor completely bypassed to drive the meter to full scale at 10 dB reduction (this is what the trimmer is for that you see by the ammeter).

3) Threshold potentimeter changed from +12/-12V range to 0/-12V range to get better precision (+12 leg replaced with ground). This has the threshold kick in at about -50 to -40 dBFS with the threshold at it's lowest setting (ratio dependant).

4) Pin 4 on the VCA's are bent up and away from the IC sockets to disconect from the trimmer circuit (I'm not sure if this is nesc, data sheets have pin 4 as N.C.) NOTE: THIS DOESN'T APPLY TO THAT2181 VCA's.
 
Thanks Emperor-TK. Please report back how you calibrated your new G-SSL with 2181 when finished. :thumb:

About your 3. point, can you explain a little more how to make this mod? ... connect the +leg instead of to the middle leg to the PCB and leave the place where previous +leg was unconnected?

BTW, you mentioned you are at the end with your SSL with 2181. This means a day or two or more?
 
[quote author="Emperor-TK"]My previous build used the pre-trimmed 2180's, and here is what worked for me. This was originally suggested by Steffen and seems to be simpler than the mods suggested here. This mod works perfectly in my opinion, with no distortion or pumping. Note that the 2180 and 2181 are the same chips, anly the 2180 is pre trimmed. The trimmer circuitry needs to be left in for the 2181. I am almost finished with another GSSL and will be using 2181's this time. I'll report back with when I am done, and I'll also run some ratio curves too.

1) 100K resistor driving neg input of TL072 to VCA CV's replaced with 122K to get proper ratios for THAT 2180LA's.

2) 2K resistor feeding meter replaced with 1K +/-5% for 0-20dB range.
Note: I tried to get a 0-10 dB range, but there wasn't enough current with the resistor completely bypassed to drive the meter to full scale at 10 dB reduction (this is what the trimmer is for that you see by the ammeter).

3) Threshold potentimeter changed from +12/-12V range to 0/-12V range to get better precision (+12 leg replaced with ground). This has the threshold kick in at about -50 to -40 dBFS with the threshold at it's lowest setting (ratio dependant).

4) Pin 4 on the VCA's are bent up and away from the IC sockets to disconect from the trimmer circuit (I'm not sure if this is nesc, data sheets have pin 4 as N.C.) NOTE: THIS DOESN'T APPLY TO THAT2181 VCA's.[/quote]

so i guess you didnt use hipassfilter board ?
 
[quote author="Purusha"]

BTW, you mentioned you are at the end with your SSL with 2181. This means a day or two or more?[/quote]

Yes, I'm working on it today and hope to have it finished up.

[quote author="digichild"]
so i guess you didnt use hipassfilter board ?[/quote]

No, I must have missed that one. I haven't been checking in on the latest GSSL info. Is that a hi-pass filter for the sidechain?

-Chris
 
exactly...i was wondering cos you mention the adjustment of threshold...that can be done now with a trimpot on the sidechain board

http://www.groupdiy.com/index.php?topic=12024&highlight=sidechain+highpass
 
You can also add the overkill version :green:

http://home.hetnet.nl/~chickennerdpig/FILES/API/API-Thrust-HPF-circuit_20060310.pdf

This is what I choose to add to my GSSLs. Mine are finished, but not tested with the GSSL yet.

More about the overkill HP filter in this thread : http://www.groupdiy.com/index.php?topic=8897&postdays=0&postorder=asc&start=0

3) Threshold potentimeter changed from +12/-12V range to 0/-12V range to get better precision (+12 leg replaced with ground). This has the threshold kick in at about -50 to -40 dBFS with the threshold at it's lowest setting (ratio dependant).

Can you explain to me how to wire this one please.
 

Latest posts

Back
Top