SSL! :D

[tmbg]

It's all built! I have no way to test it though, because I don't have anything that makes noise! I'm in the middle of moving, and all my studio stuff is at the old house.

I have a mic and a crappy stereo preamp here though, and running the preamp out, into the SSL left channel, then the left out back into the other channel of the preamp, and watching the meters on the preamp, it is definitely passing audio, and looks like it's pprobably even compressing!


BUT, the meter on the SSL no worky

When I first put it together, I let the magic smoke out of one of the VCAs I did some tracing and figured out one of the 5534s was bad, replaced that and the voltages on the VCA pins matched, so I popped in a new pair, and no magic smoke this time.

Next I'm thinking I may throw the function generator into it, and scope the output, see if I can actually SEE the compression, and check for distortion and frequency response if I can get either of my stupid scopes to behave.

PICS!
http://gromit.usedforcomparison.com/~tmbg/ssl1.jpg
http://gromit.usedforcomparison.com/~tmbg/ssl2.jpg
http://gromit.usedforcomparison.com/~tmbg/ssl3.jpg

 

[radiance]

This is what I call DIY: a home made chassis right?
Nice work!!

 

[tony dB]

cool! Well done mate!
you'll get this debugged soon,
as the Lab is THE source!

Whished i could be of more help straightaway, but didn't build one yet myself

:thumb:

Tony

 

[tmbg]

The meter was actually working, I just had it hooked up backward! But now the meter seems WAY sensitive, pretty much any input completely pegs the meter.

Of course, I still havent been able to LISTEN to it, so maybe something is awry in the audio path

 

[tubejay]

Nice work! I like the box!

We've got the same soldering station/iron. I love mine, it's cheap, simple, and works fantastically!!

What kind of meter is that you're using? Maybe you need a bigger resistor in there?

 

[daArry]

try using a trimmer for the meter resistor so u can adjust it, when u find a suitable value u could then poke in a resistor on brd...

 

[tmbg]

I need to actually hear what's going on before I start messing with the meter. It's a DC 1mA meter (at least it's supposed to be), and I used the 2k that was indicated on the scheme.

I ran my func gen thru it and scoped the output, and it appears to be doing its job, and with low distortion (if you can believe the straight lines of a triangle wave Of course, the triangle wave doesn't bow even if I turn the distortion trimmer all the way in either direction, so maybe that's not an accurate enough test.

I'm gonna go pick up my board and my events today, and I'll put the SSL between them, and have some fun

The best part is, I have enough parts to build another one! Oh wait, no I don't, I fried a VCA :/ Debating over either trying to get some 202s for the second one, or some 2180s and compare verbos's VCAs with the 2180s or 202s.

 

[verbos]

[quote author="tmbg"]The best part is, I have enough parts to build another one! Oh wait, no I don't, I fried a VCA :/ Debating over either trying to get some 202s for the second one, or some 2180s and compare verbos's VCAs with the 2180s or 202s.[/quote]

nooooooo!!!!!!

Just kidding.

It will be nice to see how they really stack up. Although I put a 2180 on one channel and a c1252 on the other on stereo material and I couldn't notice a difference. Maybe I didn't look hard enough. Maybe my home studio conditions weren't critical enough. Anyway, congrats on your build!

Mark

 

[tmbg]

Wow, this thing sounds really good! Seems to add a fair amount of hiss... I guess thats a given with any compressor. But that's ok, I'm just thrilled that it passes clean, equal gain audio on its first try!

It's a pretty nice, flexible compressor. Overall I'm very happy, but does anyone have any tips on cleaning up the noise? Better resistors in the audio path or something? I used crappy 5% carbon film resistors, but I sprung for the polyester film 0.1uf box caps like everyone uses, so I'm guessing those are ok. I bought generic ceramic caps for all the other non-lytics.

Is there a better brand of 'lytic that might help that I could try?

 

[tubejay]

I used 1% metal film resistors, and mine has no hiss problems at all.

 

[tmbg]

I may make the next one with metal films and 202's. I didnt go with metal film this time because there's eight MILLION resistor values necessary!

Maybe I can find a nice assortment

Also I used an EI transformer, because my toroid is in the mail. Wonder if the toroid will clean it up any.

 

[BladeSG]

Well done tmbg. I've just got my first GSSL working and I know what it feels like to have a successful build. :grin: (mine still needs to be racked)

 

[Greg]

Nice work!!!

but I don't think this unit should be adding hiss... hmmm?!?

 

[StephenGiles]

Looks really good, hiss could be worrying though.
Stephen

 

[tmbg]

the hiss is proportional to the makeup gain... and it's in both channels.

 

[gyraf]

You will always have noise in gaining circuits - but how much noise do you have - referred to 775mV- at what gain?

And how about with no input, inputs shorted to ground? If you feed it a noisy signal, then there will be noise at it's output.

Jakob E.

 

[tmbg]

I can get 2% metal films locally, think it would be worth trying them? I think they're all 1/8W, there's nothing in the audio path that carries more than that is there?

 

[gyraf]

I don't think you get any important part of the noise from your resistors.

 

[tmbg]

With the inputs open (not grounded, no input), here's what I get scoping the left channel, with the threshold fully counterclockwise:

makeup gain at 0 (fully CCW):
http://gromit.usedforcomparison.com/~tmbg/ssl-noise-L1.jpg

makeup gain at 100% (fully CW):
http://gromit.usedforcomparison.com/~tmbg/ssl-noise-L2.jpg

There's noise and hum. They're both there when the makeup gain is at 0, but most appartent when it's at 100%.


I'll try it with the input grounded next, does that need to be +/- shorted? or both shorted to ground?

 

[gyraf]

Noise measurements are ALWAYS with inputs shorted to ground.

+, -, and 0 (XLRF pins 2,3,1) all shorted together.

Jakob E.