500 series Level-Loc project

[Bobby Baird]

I had this noise when I had module outside lunchbox. Inside lunchbox not.
You can solve when you are testing them connecting with an alligator clip cable your audio gnd and a toggle or panel or pot body.
Have you measured total noise ? It's similar to mine?
I posted an audio file with total noise. Please can you listen it and compare to your build? I changed and tested all part of the circuit in order to understand if it depend on the project or in any part I used in mine...

Mix 100% Input on Threshold more -64dbu

 

[Aleguitarpro]

Mix 100% Input on Threshold more -64db

It's like mine.. but I haven't this 60Hz peak I see in your graph.
So it's the project that is very noisy.
Thanks

 

[Bobby Baird]

It's like mine.. but I haven't this 60Hz peak I see in your graph.
So it's the project that is very noisy.
Thanks

So from what I've read this circuit is noisy, but I have found a solution from other threads where you can filter the 9 volt power.
" install the power filter mod - just a lowpass filter on the power
line using 3 100-ohm resistors and 2 470 or 680uF capacitors!

dont need a schematic - on the other side of the 9v connection before it feeds the unit power add a 100ohm resistor then a 470uf cap to ground then another 100ohm resistor then a 470uf cap to ground then another 100ohm resistor then feed the circuit power.

Not sure where to implement this mod on the board though? After the regulator I suppose?

 

[Aleguitarpro]

So from what I've read this circuit is noisy, but I have found a solution from other threads where you can filter the 9 volt power.
" install the power filter mod - just a lowpass filter on the power
line using 3 100-ohm resistors and 2 470 or 680uF capacitors!

dont need a schematic - on the other side of the 9v connection before it feeds the unit power add a 100ohm resistor then a 470uf cap to ground then another 100ohm resistor then a 470uf cap to ground then another 100ohm resistor then feed the circuit power.

Not sure where to implement this mod on the board though? After the regulator I suppose?

I read about this mod.
Believe me that this project make me crazy because it's first time I have all this problems with a build (especially from a sort of "verified" project).
I tested 9v rail with a 9v battery to understand if noise change.
Yes, you'll solve 50/60hz hum but rest of noise remain.
I don't know if it's something related to PCB layout or not.
IC3 boosted noise that comes from the "wet" circuit after Q5 and C8.
You can check this easily in your build.

 

[bowanderror]

I read about this mod.
Believe me that this project make me crazy because it's first time I have all this problems with a build (especially from a sort of "verified" project).
I tested 9v rail with a 9v battery to understand if noise change.
Yes, you'll solve 50/60hz hum but rest of noise remain.
I don't know if it's something related to PCB layout or not.
IC3 boosted noise that comes from the "wet" circuit after Q5 and C8.
You can check this easily in your build.

I think this build is just a bit more sensitive to power supply & PCB layout.

The gain stages are super high gain and you have a high impedance trace coming from the sidechain into the JFET, which will pick up interference quite easily. That high impedance trace also carries a fully rectified, 0-9V DC version of your signal, and lacks any injection of dry signal like other, more famous FET compressors.

The original also had transformer isolated IO, which may have helped prevent any unwanted noise & interference.

 

[Rogy]

*Copy of Black Market post, in reply of Aleguitarpro's questions*

Hi,
I just came across this.

I have two of those, built by Eliani.
I noticed that in bypass the output level was 6dB hotter than the input level, so I swapped the INA134 at the input for a Thatcorp 1246 (pin-compatible but does -6dB).
After this, the gain structure is correct.

My two units do not have excessive hi freq hiss. Obviously the THD+N goes way up in normal operation since those do apply 60dB of voltage gain; remember it's been designed as a microphone-to-line mixer.

Looking at a full spectrum FFT, the noise floor is evenly spread out (no excessive noise at higher frequencies.

Hope this helps!

Best regards,

Rogy

 

[Bobby Baird]

Changing out the Burr-Brown's to ThatCorp amps you should swap the decoupling capacitors with .1uf instead of 1uf according to datasheets.
Parts: C17,C18,C19,C20
Also, curious if installing the -6db opamp on the input would increase noise floor with a +6db on output? On the other hand would a -3, +3 or a 0db, 0db input output opamp configuration have better noise floor spec in this circuit? I would like to understand this better.

 

[Eliani]

It is, and always will be an old circuit.
I picked the project up because there was nobody working on it anymore, and it did interest me, so as far as my contribution to this project goes, I've made a PCB for it... I cannot take credit for the design, nor the problems with it. But I've been wondering if we would not just scrap the 9V conversion and drive it at the regular 16V?

 

[Eliani]

I still do have 6 PCB's available...

 

[bowanderror]

But I've been wondering if we would not just scrap the 9V conversion and drive it at the regular 16V?

I haven't personally tried it, but that would probably cause more problems that it would solve. A larger Vcc would mess with the biasing of the transistor stages & change the scale of the voltage control signal from the sidechain to JFET gate.

 

[Bobby Baird]

New Panels!

 

[Bobby Baird]

RFI mod according to THAT 1246 datasheet.

 

[Bobby Baird]

Here's the FPD file rename accordingly.

 

[MonoCutter]

Thank you very much for the fpd! Yours look really great.

Dennis

 

[JMan]

Hey all, I'm a little confused about the switches needed for this project. I have Eliani's board and the BOM he sent me (v1.13 from irfrench with a few edits in blue to reflect changes to the transistors/ICs). The switch portion of the BOM doesn't seem to match up quite right to what I'm expecting, and scanning this thread hasn't helped me much since I can't quite tell what information pertains to the old versions of the project vs Eliani's boards (or if there even is a difference, although I definitely don't see the newer builds using an F2UEE).

Can anyone clarify what the four switch configurations needed are?

 

[Bobby Baird]

Hey all, I'm a little confused about the switches needed for this project. I have Eliani's board and the BOM he sent me (v1.13 from irfrench with a few edits in blue to reflect changes to the transistors/ICs). The switch portion of the BOM doesn't seem to match up quite right to what I'm expecting, and scanning this thread hasn't helped me much since I can't quite tell what information pertains to the old versions of the project vs Eliani's boards (or if there even is a difference, although I definitely don't see the newer builds using an F2UEE).

Can anyone clarify what the four switch configurations needed are?

Yeah let me look it up I used C&K switches
DPDT Bypass on-none-on 7201sdav2qe
DPDT Threshold on-off-on 7203sdav2qe
SPDT Release on-off-on 7103sdav2be
SPDT Pad on-off-on 7103sdav2be

Most all were in stock but one I had to buy from JLM audio

 

[Bobby Baird]

Eliani's board doesn't use the F2UEE switch.

 

[Bobby Baird]

use a 1246 on input istead of 1250 to have unity in bypass also if using that amps change
the decoupling capacitors with .1uf instead of 1uf according to datasheets.
Parts: C17,C18,C19,C20

 

[Bobby Baird]

These things give the drum bus with bass guitar impact and good low mid girth. Really fattens up hi-hats. Cymbals decay for days.

 

[JMan]

@Bobby Baird wow, thank you for all that info! Very helpful indeed!

Okay, so to sum up, it would look like this, right?

Bypass - DPDT on-none-on
Threshold - DPDT on-off-on
Release - SPDT on-off-on
Pad - ""

IC1 - THAT1246

C17-C20 - 0.1uF electrolytic caps

Out of curiosity, you show a picture above that appears to show several ceramic caps connected between Input and Chassis ground. Would you mind elaborating on what that is? Is that also to do with the 1246, or unrelated?