Hairball Audio: 1176 Stereo Linking PCBs Support Thread

[mathflan]

I finished to wire the stereo link but It doesn't work 
I check all voltages and wiring, All seems OK.

Do I miss something?


thanks

 

[Echo North]

Did you add it to a previously working build or is it a new build?

What is the issue?

Also make sure you are familiar with how the circuit works.  It's different then the UA 1176 link box.


Mike

 

[ethervalve]

Hi Mike,
Do you have a finished unit with the Stereo Linking board installed?
If so, and if it's not too much trouble, would you mind uploading a photo of the innards at some point?
I'm a bit stuck troubleshooting an oscillation issue with a rev D with the stereo board (I think it's a somewhat common issue: the oscillation occurs when the output pot is near maximum).
I'd just like to see a photo of a successful build to get a better idea of where i should run the wires.

 

[dmp]

I built two revD units with the stereo link boards. I ran into the same problem with oscillations as the output pot was turned up (actually before I even added the link boards) and it was a shielding / grounding issue. Run shielded wire to the link board and the output pot and connect the shield at one end to the chassis ground, not the audio ground. At first I soldered the shield to the pcb ground, which is audio ground, and got oscillation like you are describing.

 

[ethervalve]

Thanks for sharing that dmp.
I'm at work so I'm just going by memory, but I believe I also connected the shields to the pcb ground. I'll try switching up the shield connections ASAP.

 

[ethervalve]

Hi all,
I think you were right dmp. Re-wired it and the oscillation went away. Thanks!

Now, I'm still unable to get the stereo matching to work perfectly though. At slowest attack/fastest release i'm getting about a 3 dB discrepancy in gain reduction between units whenever I got beyond about 5dB or 6dB in GR.
A couple things I'm wondering about:
1) there seems to be a great deal of Gain reduction interaction when I adjust the release pot on either unit--would switching to tighter tolerance release pots be worth attempting? I'm currently using 20% alphas. (I noticed this older post from Gswan that suggested checking the 5meg pot when attempting the "connect pad 7" style stereo match: http://www.groupdiy.com/index.php?topic=646.msg439565#msg439565 )
2) I had to drop in a 4.7k trimmer to complete the tracking adjust calibration for one of the units (the other had enough range with the standard 2k trimmer). Does this suggest anything fishy? Both units seem to behave identically outside of the stereo link mode when fed the same programme.

 

[TheJames]

I've been pondering stereo-linking FET compressors lately.  This circuit makes alot of sense, tap off both inputs, sum them together, then feed the output to both side-chains.  However, as mentioned, you've got to get both side-chains' controls matched for it to behave the same.

I'm not too familiar with the intricacies of the 1176, however strictly audio speaking....Assuming you have matched GR FETs, summing the inputs, feeding one side-chain, and then buffering that output and feeding both GR FET gates should provide the best stereo tracking and only one set of attack/release controls to deal with.  I'm probably missing something, but that seems more elegant to me.  There might be some issue with GR meter tracking on the "slave" unit...But it would seem this would at least get the audio signals tracking right.

 

[Soeren_DK]

Is this correct? ???

Cheers
Søren

EDIT: Picture not correct

 

[Echo North]

Is this correct? ???

Cheers
Søren

Yes.

Edit: No.

See posts below.

 

[dmp]

Ethervalve,
I haven't gotten my units to match very well either, so I can't be too much help... IF your units are matched well at fast  attack / slow release, you are doing better than I am.
But, at slow attack / fast release, your 5Meg pot is not dominating the circuit because the wiper is turned such that R57 is in parallel with the pot resistance. If R57 is matched very well, it is in parallel with 5meg+/-20%, which reduces to 256k +/- ~1% (if my math is right). So I would not think the 5meg pot is the likeliest suspect. Similarly for the attack pot, be sure that R54/C27 are matched very well.  Is discrepancy in GR measured at the input/output signal, or is it a apparent difference from the meters?
About the tracking adj trimmer - I *think* that a large resistance indicates Q12/Q13 are not matched very well for hfe and the larger trimmer is necessary to compensate for this. I don't think it would affect operation if you are able to get the GR meter calibrated correctly. 

 

[ricothetroll]

Hi !

I tried to make my Stereo Link work, but for the moment it doesn't... I wired it the way shown on the picture posted by Soeren_DK on the current thread.

I drew a principle schematic in order to have a better understanding of how the circuit works. You'll find it here :

https://drive.google.com/open?id=1H3SaXUZYbqDIisaIh44ibAI24Vrb8IPk (sorry for the ugly hand-drawing)

My interpretation was based on the schematic on page 4 of that pdf :

http://mnats.net/files/STLNK1.1DOC.pdf

I might be wrong but it appears to me that Soeren_DK's wiring doesn't make sense : pad 15 of the main board mustn't be connected directly to pad 15 of the ratio board as it used to without the stereo link. Pad 15 of the ratio board is where the signal enters the sidechain, connecting it to pad 15 of the main board means that when the link is activated, both the output of the preamp AND the sum (L+R) goes to the sidechain.

I suggest this wiring :
https://drive.google.com/open?id=1Izp6w0IfLxoo1KfgV5yeDNNGCqPe5XuJ

Is it right ?

Best regards.

Eric

[edit] : moved the image files from the long gone Wuala server to Google Drive

 

[craigg]

HI there.  I have two mnats 1176s.  One is a Rev A, and one Rev D (with hairball kits).  work great.  Sound great.  Got the stereo linking kit and put it together.  Not sure if the stereo linking works yet because I'm having a problem powering the stereo link pcb in the Rev A.  The Rev D seems fine.  When I try to power the SL pcb on the 30v test point in the Rev A, my audio level drops, the compression goes away (turning the compression on or off makes no difference, and my 30v test point reads about 26v and my -10v test point (which read about -8.5v) goes to -12v.  Unhook the 30v power from the SL pcb and the unit works fine - compression, all switches, etc.  Anybody have any idea?  I hooked up this pcb to the rev d that I have and it worked fine (didn't have this problem), so I feel confident that the SL pcb isn't the culprit.  Thanks everyone.

 

[Echo North]

HI there.  I have two mnats 1176s.  One is a Rev A, and one Rev D (with hairball kits).  work great.  Sound great.  Got the stereo linking kit and put it together.  Not sure if the stereo linking works yet because I'm having a problem powering the stereo link pcb in the Rev A.  The Rev D seems fine.  When I try to power the SL pcb on the 30v test point in the Rev A, my audio level drops, the compression goes away (turning the compression on or off makes no difference, and my 30v test point reads about 26v and my -10v test point (which read about -8.5v) goes to -12v.  Unhook the 30v power from the SL pcb and the unit works fine - compression, all switches, etc.  Anybody have any idea?  I hooked up this pcb to the rev d that I have and it worked fine (didn't have this problem), so I feel confident that the SL pcb isn't the culprit.  Thanks everyone.

Did you triple check all the components on the PCB (resistor values specifically)?

Seems odd.  I would start there.  Also check the TL071 to see if you are getting +15V at the output.

Mike

 

[Echo North]

I might be wrong but it appears to me that Soeren_DK's wiring doesn't make sense : pad 15 of the main board mustn't be connected directly to pad 15 of the ratio board as it used to without the stereo link. Pad 15 of the ratio board is where the signal enters the sidechain, connecting it to pad 15 of the main board means that when the link is activated, both the output of the preamp AND the sum (L+R) goes to the sidechain.

I must admit I know little about the ROTARY ratio board wiring. 

If you look at the PCB traces on the ROTARY ratio board Pad 15 is directly connected to the CW lug (left lug from the back) of the output pot.  So if you follow the Purple schematic, pad 15 of the main board (Pre Amp Output) connects to the output pot lug and is then wired to the "loop" terminal.

I guess what I'm trying to say is even though the ROTARY ratio board shows pad 15 connected to the main board pad 15, it's only traced to the output pot and not the actual ratio circuitry (by looking at).

Mike

 

[Echo North]

Hi all,
I think you were right dmp. Re-wired it and the oscillation went away. Thanks!

Now, I'm still unable to get the stereo matching to work perfectly though. At slowest attack/fastest release i'm getting about a 3 dB discrepancy in gain reduction between units whenever I got beyond about 5dB or 6dB in GR.
A couple things I'm wondering about:
1) there seems to be a great deal of Gain reduction interaction when I adjust the release pot on either unit--would switching to tighter tolerance release pots be worth attempting? I'm currently using 20% alphas. (I noticed this older post from Gswan that suggested checking the 5meg pot when attempting the "connect pad 7" style stereo match: http://www.groupdiy.com/index.php?topic=646.msg439565#msg439565 )
2) I had to drop in a 4.7k trimmer to complete the tracking adjust calibration for one of the units (the other had enough range with the standard 2k trimmer). Does this suggest anything fishy? Both units seem to behave identically outside of the stereo link mode when fed the same programme.

Definitely what DMP said.  The parallel resistor on each pot is dominating the over all resistance.

Maybe your fets are out?  Did you get them from me? In the past you could order 2 sets of FETs from me and they would give you stereo linking withing 1db up to about 10db GR.  UNLESS you were unlucky enough to get the two that were at either extreme of my matching window.  I've since addressed and fixed this issue.  If you got FETs from me and think they may be the issue let me know and I'll send you new sets.

Mike

 

[ricothetroll]

Hi Mike,

Thanx for your answer !

I'm using the Gyraf Audio version, maybe that's the reason why we have those differences.
http://gyraf.dk/gy_pd/1176/1176.pdf
If you take a look at the ratio board on the first page (superimposed view), you see that pin 15 feeds on the one side the pot, and on the other side the sidechain (rotary switch) through the top left 56k resistor.
I'll try my wiring and let you know the result !
Best regards.
Eric

 

[Echo North]

Hi Mike,

Thanx for your answer !

I'm using the Gyraf Audio version, maybe that's the reason why we have those differences.
http://gyraf.dk/gy_pd/1176/1176.pdf
If you take a look at the ratio board on the first page (superimposed view), you see that pin 15 feeds on the one side the pot, and on the other side the sidechain (rotary switch) through the top left 56k resistor.
I'll try my wiring and let you know the result !
Best regards.
Eric

I see what you are saying and see the issue.

Yes you want wire the switch between the pot and the sidechain, so some creative wiring needs to happen on the rotary boards.  As you stated earlier you'll probably just want to wire pad 15 from the main board directly to the pot and keep it away from the ratio board.

You have it figured out, I'll stop confusing things

Mike

 

[mathflan]

Hi,


thank ricothetroll,

your wiring works greats.
I know now why it did not work before.


Thanks echo north for the PCBs


MAth

 

[Echo North]

Thanks echo north for the PCBs

Don't forget confusion.  I provided this thread with a lot of confusion.  ;D

Mike

 

[ethervalve]

Definitely what DMP said.  The parallel resistor on each pot is dominating the over all resistance.

Maybe your fets are out?  Did you get them from me? In the past you could order 2 sets of FETs from me and they would give you stereo linking withing 1db up to about 10db GR.  UNLESS you were unlucky enough to get the two that were at either extreme of my matching window.  I've since addressed and fixed this issue.  If you got FETs from me and think they may be the issue let me know and I'll send you new sets.

Mike

Yes I did use a couple sets of fets i bought from you.
I was actually planning on ordering another two sets of fets plus another stereo matching board set to add stereo capabilities to the pair of rev As I built last year.
If it's not too much trouble, I would very much appreciate it if you could throw in another couple sets of fets when i place the order. Thanks so much for the offer!