Sta Level Distortion, Grrrrr....

[orangechili]

I've noticed a few others in various threads have had a distortion issue with their Sta Level builds, not the pleasant distortion we all like. Crappy, blotchy spitting distortion to be exact. I built a version point to point Sta Level and have the distortion issue with the build. The unit sounds a little thin as well. Others with pcb builds have had a similar issue so i'm starting to wonder if this is a component or something that needs to be tweaked.... Here's some notes on the build:

-I have traced the unit connection by connection 3 times against the schematic. I have reflowed solder joints, etc to no avail
-I used a NOS RCA 6386 tube, I tried connecting 2 6BA6 tubes instead of he 6386 and the issue is still present so it is not the tube
- I have tried two different input transformers, no dice
-I am using the Edcor output transformer, the extra screen tap leads that aren't needed are taped off
-I am using a custom output pad attenuator and no input pad

If anyone is down to help me crack this da vinci code I would greatly appreciate it! I'm hoping someone on here has fought with this thing before and has some insight

Attached is the schematic I used

 

[orangechili]

If someone can double check the input pot wiring on the schematic against this turret diagram and tell me if it's correct I would appreciate it. My head is getting turned around from this thing. The input pot appears to function well but I am trying to beat down every path...

I also should mention I used a 500v power transformer instead of 600v so my voltages are lower than what's posted on the schematic. I wonder if the lower voltages may be starving the circuit maybe?

 

[emrr]

No input pad:  what input transformers, and what signal level. 

 

[orangechili]

Correct, no input pad. Really just any decent signal level it sounds like blotchy pops of slight distortion when I talk into a microphone. With a faster release the distortion is worse as you can imagine. I'm currently using a 1:4 UTC S-2 transformer. I used a Cinemag 1:4 transformer before with the same result.

 

[mjrippe]

If the distortion occurs when you cross the threshold, it is due to unbalance in the 6386 or 6BA6 tubes.  Run a signal with low enough level that it does not cause compression, see if it distorts.

 

[bernbrue]

Pull out the 6AL5 tube and measure only the amp with RMAA, not with a microphone. When the results are good, you at least know, that the mistake must be somewhere in the sidechain.
In my build thread of the sta level there are measurements for comparison.  Under normal circumstances neither you nor builders using my pcb should experience any kind of audible distortion. The amp in this compressor is one of the cleanest tube I've ever heard.
Bernd

 

[orangechili]

In this case i don't think it's in the side chain or threshold related. I took the 6aL5 tube out and still hear the distortion pops when I talk. I attach a mp3 to this post to show an example of the distortion. Just remove the .txt from the file extension. 

The OB2 measurements are right in line with the schematic. I know the PT called for is 600v and I'm using a Triad 500v one. I'm wondering if the lower voltages in the rest of the circuit may cause this? They are about 30v lower coming off the B+ line.  I have tried different rectifier tubes, different 12at7 tubes, different 6al5 tubes and the 6BA6 pair instead of the 6386 tube.

It's gotta be a connection or something off i'm thinking...

Bernbrue, I don't have a PC(only mac)  so can't download the RMAA program you mentioned.

 

[emrr]

PT voltage wouldn't make that sort of difference. 

Can't make a meaningful comment on the sound in that sample.  Too many unknown variables.  Doesn't sound anything like the pops that occur with unbalanced gain reduction tube though, that's like a momentary mute. 

 

[orangechili]

That's what i figured about the GR tube. I'm gonna try a few things here on my end and try to trace things yet again. As far as eliminating variables it's the exact same circuit as the schematic i posted minus the input pad and the output section has a custom T pad attenuator. I'll try and remove the the T pad on the output and see if that helps the issue at all. It's frustrating when things work just enough to give you a hard time  :-\

 

[bernbrue]

Bias voltage for the 6AL5 tube is derived from B+. When your B+ is low, your bias voltage will be too low as well. It should be around 13-15V at the junction of R32 and R39. The voltage of 25V shown in the schematic is RMS.
Bernd

 

[orangechili]

Yeah,  all my other voltages are lower but that voltage at the junction of R32 and R39 is 25v exactly.

 

[emrr]

That bias being lower does nothing other than lower threshold and ratio.

 

[orangechili]

That distorted clip I posted was with the 6aL5 tube pulled as well. No compression present at all.

 

[emrr]

It really just sounds like too much gain.  Please describe chain. 

 

[orangechili]

Just my normal chain I test things out with for initial sound tests, a cheap condenser mic through a Redd47 preamp with about 48db of gain.

My gut tells me it's something on the input of the circuit. That's why I wanted to double check the input pot wiring. I may pull the custom T pad output attenuator and wire like the original just to eliminate more things

 

[emrr]

If the output of the Redd47 is stout enough to print directly, then it's probably too hot for the input of an unpadded Stalevel, and that could be your splat.  It's what it sounds like.  No pad on a Stalevel wants to see something like -40dBFS.  But maybe I'm down a rabbit hole, and still missing something.  Do you have another Stalevel that does work in the same circumstance?

 

[CJ]

what's shakin with this thing? any success yet?

what we need is first off, the schematic labeled with DC voltages at plate and cathode for each tube plus 6V6 GT plate screen and cathode voltage.

label the DC with blu ink, then we will inject signal and write in red all the ac voltages

if that don't work then we break out the o scope, if one is available.
what would be a good test voltage for the input at 1 K Hz, maybe 1 volt RMS?

 

[orangechili]

Negative amounts of success yet, thanks for asking CJ!

Attached is the schematic I used, it's the only one i've found online really. I will try to go through and measure again and label the dc in blue like you suggested

Don't have a solid scope right now but can send a 1K sine wave out of my daw for testing as well

I will try to do this in the day or so, thx!

 

[emrr]

You still haven't answered the fundamental question about the level you're putting into it. 

CJ is pulling your leg, 1V RMS is +13dBm, which will blow the shit out of an unpadded Stalevel.  Like I said, test with -30 to -40dBm input.

 

[orangechili]

I just read over the post real quick and missed the 1V RMS comment at the end, I was just stating I can use a sine wave coming out of my daw at any level.

It makes sense that overloading the input would cause the splat but it seems like removing the input pad on the sta level is something most folks regularly do. The Redd47 only puts out a max of 48db i believe so not too hot of a signal going into it when testing with a mic.

I'll try to get measurements and do more testing when i get off work later this evening, thx!