Noise from compressor - overdrive

[ghiekorg]

Hi Guys,
I have a problem with my compressor and/or overdrive.
Here is a clip: https://youtu.be/QY98tgPWZVQ

In the clip there are other pedals but the results is same even if I just have compressor and overdrive alone. I tried to use the 9v battery on the mxr and also place it after the b3k but the noise was still there (maybe slightly less). Both pedal have a separate input from the fuel tank, so isolated and filtered.
On the b3k I used 2x 2N5457.
Is it normal you think? 
Thanks

[11-90-an]

Does this happen before or only now?

forgive my eyes, but is that compressor -> overdrive or overdrive -> compressor?

if it's overdrive -> compressor, compressors brings up low signals and lower down the loud, and since noise is generally considered 'low' by the compressor, it brings the signal up... (or something like that, I'm bad at explaining...  )

if it's compressor -> overdrive, noise is brought up by compressor, then amplified greatly by the overdrive...

noise gate as next build? 

[ghiekorg]

in the video the chain is: comp-od. But as i tried also to reverse and didn't change much.
I understand the things you said, but amplified SO MUCH? I mean, it's not even an extreme distortion pedal, it's just a bass OD :-/

I don't have any more space for a noise gate unless is big like a coin

[11-90-an]

I understand the things you said, but amplified SO MUCH? I mean, it's not even an extreme distortion pedal, it's just a bass OD :-/

Amplification is still amplification...

I don't think this is as small as a coin, but since it's 1590A, maybe mount it under your board? 
I've yet to try making this...

[ghiekorg]

I checked that too. mmm maybe i could fit into a plastic case and use a toogle switch without instead of a 3pdt. and a mini knob. When i finish all the pedals and i see how much space i have left i will give it a try. But it will be at the end of the chain, i don't know if i want it there.

[Mark Hammer]

It is the duty and obligation of EVERY compressor to make everything it receives equally audible.  It does so by pulling back on the gain/amplification of louder portions of the input signal, and boosting the bejeezus out of the quiet portions of the input signal.  Unfortunately, it cannot tell, without a great deal of outside help, whether the "quieter" parts are lightly-strummed strings or simply residual hiss.  It would be wonderful if the gain was automatically adjusted on the basis of the spectral content of the "input" (i.e., IF mostly broadband noise, THEN do not increase gain), but you'd probably have to move into digital compression for that miracle.

This is also why, even though there may be some musical utility to doing so, I counsel against placing a compressor after a drive pedal.  Drive pedals rely on high gain, which results in amplified input noise, even after you stop playing, which the compressor is only too happy to boost.....so, you know, it can be heard.

Compressors generally work best when the input signal they receive is as dynamic, clean and noise-free as possible.  They will still work, if those principles are not observed, just not as well or as quietly.

One term you will often hear with respect to compressors is "breathing".  As noted, when you pick a string, the gain of the compressor will be automatically reduced.  If there was any residual hiss at the input, accompanying the guitar signal, it will also be amplified less.  When you stop picking, the gain will be gradually increased (referred to as "gain recovery").  And since the only thing to be gradually amplified is the residual input noise, that progressively louder hiss sounds like a person inhaling - "breathing".

[ghiekorg]

Makes sense... That's also why i place the compressor at the begining of the chain, just after the octaver (because i want to have its 2 ports available, as i use it with both guitar and bass)

I just tried with the tentacle an even on the highest gain level everything is quite like an evening in a swiss village. The problem is the b3k.
Is there anything i can test? some ground problem, some noisy transistor, i have no clue.

[Mark Hammer]

It's not exactly a panacea, but it would be good if the resistors or the input stage on the compressor were 1% metal-film low-noise types.

refresh my memory: what compressor circuit are you using?

[ghiekorg]

It's a bought one, i haven't built it. I built the b3k. The comp is an MXR studio compressor, i don't know its components

[ghiekorg]

Reading the comments on effeects layout i read this "Also make a boot's trap before the 9V input of the board, it means, put a 100R in series, like the rat, that gonna improve the noise performance"
shall i just put a 100R between the input DC jack and the "IN" of the board? Do you think it would work?
What about also changing the 5457 for something else?

[11-90-an]

Yes, that will help, but make sure to put it in between +9v and the +9v board input, not in the signal "IN"...!

[duck_arse]

.... make a boot's trap before the 9V input ......

you're gonna cop it when Antonis reads this!


after your in-line resistor, you need to add an electro cap to ground to form the filter. 22uF - 100uF, bigger = more carp caught.

[ghiekorg]

oh. I just closed it :-/ would it make a big difference (the cap, i mean)?

The noise seems slight better now. i will post a clip later. Now i have another problem: when i set the attack switch on one of the two ON positions and i have the drive maxed out i get a "thump thump thump" noise. I will post later, i have to go out now. :-/  maybe it didn't like to be opened again...

[ghiekorg]

here is the weird noise problem, I don't think i had it before today's "mod" but not sure 100%. I think i went often max drive and attack to le right..
https://www.youtube.com/watch?v=a_Of-Ew-XCs&ab_channel=ghiekorg
Do you ever experienced that?

[Mark Hammer]

It sounds like LFO ticking, but overdriven and filtered.  Is there an LFO anywhere in there?

[ghiekorg]

i thought so too, but i don't think there is any. It's also true i understand nothing of components
here are the components

and here the schematic

It's not a big deal, i can play with the pot till 80% gain and no noise, it's just weird and it shouldn't happen i guess

[Mark Hammer]

Verify that you don't have bias voltages leaking where they shouldn't.  Sometimes that can result in artifacts that sound like oscillation.

[ghiekorg]

Thank you Mark. I am not sure i know how to verify that in practice :/

[Mark Hammer]

R21 connects to the Bias voltage (Vdd) and so does R2. 

There is nothing to block the DC voltage coming out of IC1a before it reaches the Blend pot.

The bias voltage coming in via R21 has a more or less straight path to the Level pot, via R22-thru-R27, without any cap to block that DC.  The Level pot, meanwhile, is also tied to the bias voltage.  C21 blocks any DC from leaving the circuit at the output jack, but I suspect that there might be a need for a cap between the output of IC4a and the Level pot.  If you can, stick a 1uf between IC4a and the input lug of the Level pot, and see if that makes the throbbing go away.  A similar cap between the output of IC1a and lug #1 of the Blend pot might also be called for.

[Eb7+9]

It is the duty and obligation of EVERY compressor to make everything it receives equally audible. 

sorry, but that's just a bunch of non-sense

there are NO real floating-gain compressors on the market, unless you count my "Envelope Lock" unit ... which requires careful training to use

http://www.ovnilab.com/reviews/envelope.shtml

I did warn the reviewer beforehand ...

---

just because a pedal says compressor on it (eg., MXR Dynacomp) doesn't mean it's not a Limiter, or Limiting Amplifier (re., UREI 1176) ... in fact, they're all limiters

the bottom line is, many limiters are poorly designed as far as noise goes ... if you stick a noisy circuit in front of a gainy one, such as a fuzz unit, guess what ??!

you need real knowledge in electronics to play with this stuff and not get bit in the arse ... in this case I would almost suggest using modified LA-2A divider network without a make-up amp (where the main source of noise originates) - in essence, a passive Limiter stage - and then have your fuzz or whatever makeup for the roughly 2/3 factor loss associated with an LA-2A style optical divider network

but, that's not how most players do it of course ...

instead, they like to use the makeup gain to drive their fuzz harder while having the Limiting effect present as well ... the key in that case is to design a Limiter circuit with ultra-low output noise // which is what I aim for in my creations, incl. the Envelope Lock reviewed above ... in fact, I won't consider a design finalized unless you can't tell it's engaged (with the side-chain turned off) by total absence of noise ... of course, you need good low-noise op-amps if you go that route