Noise from feeding 20 distortions and fuzzes off the same 9V

[Esppse]

Hello,

I was wondering what steps I can take to reduce the noise from this massive project I took on. There are about 20 different fuzzes, distortions, and some modulation FX running off the same 9V source. All star powered and grounded.

Does noise leak through the 9V line from one effect to another?

The main 9V line also feeds a 12V voltage booster for one of the effects. I was thinking this:

For the more noisier fuzzes (fuzz factory etc...), from the 12v booster module output, I run that into a 78L09 regulator and feed that isolated 9v to the noisy effects. Maybe use multiple 78L09s to isolate it from the main 9V chain.

This does isolate the 9V line right?

When they are all chained and off the same 9V, I get a high pitch whistling when certain FX are on together, but its strangely inconsistent.

For example:

Distortion by itself = No whistling
Distortion + Compressor = Whistling
Distortion + Compressor + EQ = No whistling

This may have something to do with the input and output impedances between the circuits, but that is a bit above my technical knowledge.

Anyways, does making an array of 78L09s make sense in isolating the main 9V line from certain noisier high gain FX? Im also very tight on room in this build.

[amptramp]

If your noise is feeding from one stage to another through the ground lines, the addition of 3-terminal regulators will do nothing because it doesn't change anything on the ground line.  Noise conducted on a power or ground line is usually caused by varying current going through a common impedance.  For example, if you have a one-ohm line and you have a change of current of 2 mA, you will get 2 mV on the other end of the line.  Some effects may have enough gain that this amount of feedback can sustain an oscillation.  Without knowing any details of your setup, I can't really diagnose anything from here.  Regulators are so cheap now that you can always try them just to see what happens.

[Ben N]

Regulated and isolated are not exactly the same thing. You get some isolation like that, but nowhere near what you'd get from actual isolated PSs, which use separate transformers, or separate secondaries on a single transformer, to achieve galvanic isolation. (Have a look at the Geofex Spyder or Weber's pedalboard transformer to get an idea of what I'm talking about.) Nonetheless, ample filtering & a regulator on each 9v out may get you there. (Have a gander at this for a DIY version of this, similar to lots of commercial pedal PSs.)

A few more notes:
First, are you sure your problem has to do with power? Have you tried your setup, or the problematic parts of it, with fully isolated, guaranteed clean power, such as separate batteries, or, failing that, separate wall warts? Sometimes chained high gain pedals just don't play well together.

You say nothing about your 9v mothership source. How much current can it provide, and how much are you asking of it? Is it filtered? Does the output stay at/over what you need under load?

Also, are you trying to regulate 9v to 9v? That won't work. A regulator "wastes" some voltage, so you would need (if memory serves) at least 11 or 12v V-in under load to get your regulated 9vdc out. Thankfully, 12v DC power supplies are plentiful, just make sure it is big enough to provide your required amperage. BTW, that wasted voltage gets turned into heat, so if you are drawing any significant current, your 78L09 must be heatsinked.

"Maybe use multiple 78L09s to isolate it from the main 9V chain."
Do you mean in series? That won't work for the same reason: If you're already regulating down to 9v, your second regulator doesn't get enough to eat. If you mean in parallel, as in one per output, that could work as long as each pedal you're feeding doesn't need more than 100mA. That's how the AMZ power supply board linked above works, and lots of commercial supplies. Distortions (at least analog) probably use a lot less current, so that you could probably even daisy chain a few on a single 100mA output. If you mean paralleled on a single output, I dunno--I suppose that would double your available current to 200mA, but other than that I don't see the point, and I've never seen it done. A pedal that draws that much current is probably digital, and may not work well in this setup anyway, but, again, I don't know.

"The main 9V line also feeds a 12V voltage booster for one of the effects."
If you have enough voltage (12VDC) to feed your regulators, then you don't need a booster for 12v--just grab it (unregulated). Of course if you need regulated 12v, that's a different story (because you'd need ~15v in to get regulated 12v out), and boost/regulate or a separate PS may be your only solution.

[Esppse]

My 9V source is just an adapter here:
https://www.digikey.com/en/products/detail/triad-magnetics/WSU090-2500/3094953?utm_adgroup=AC%20DC%20Desktop%2C%20Wall%20Adapters&utm_source=google&utm_medium=cpc&utm_campaign=Shopping_Product_Power%20Supplies%20-%20External%2FInternal%20%28Off-Board%29_NEW&utm_term=&utm_content=AC%20DC%20Desktop%2C%20Wall%20Adapters&gclid=EAIaIQobChMItLj2wciP-wIVJf3jBx0DngEmEAQYASABEgJMAfD_BwE

I just put in an order for a higher current adapter here, hasnt arrived yet. I will need to flip the polarity on the plug on this:
https://www.digikey.com/en/products/detail/kaga-fei-america-inc/KTPS45-0950DT-3P-VI-C-P1/5247770?utm_medium=email&utm_source=oce&utm_campaign=4251_OCE22RT&utm_content=productdetail_US&utm_cid=3297766&so=78505246&mkt_tok=MDI4LVNYSy01MDcAAAGHzz7pCT9veEvMKrE8ZyIXSq40njI4URzIPQGsd5JUm44BVTY8wsPfvmux-0k_erELKawyhjzQx_AS4W6Nx2v8GG8H3vgjHLgmxf_VYR_f

The starting adapter must be 9V in my situation. With everything on, it stays at 9V. the 12V is coming from this:
https://www.ebay.com/itm/224996442243
I am putting in 9V, and it outputs 12v.

I meant using the 78L09s in parallel from the 12V booster output. Not for regulation, but to isolate the 9V output of the 78L09 from whats coming off the adapter, because I am unsure about where the noise is coming from at this point.

I was under the impression that the 78L09s require higher than 12V to produce 9V. Thats why I didnt want to go from 9V off the adapter into the 78L09s.

Am I mistaken here?

[antonis]

Without knowing any details of your setup, I can't really diagnose anything from here.

What Ron said..

If you post a schematic diagram of the way the supply is shared among effects, what kind of supply it is and what are the chained effects, we might be able to tell you more..

edit: Didn't see your post above..

P.S.
A first brute guess is for supply inadequate current capability together with poor filtering..
 

[Ben N]

My 9V source is just an adapter here:
...

I just put in an order for a higher current adapter here, hasnt arrived yet. I will need to flip the polarity on the plug on this:
...

2.5A would seem to me to be plenty for a bunch of distortions, even 20 of them, but what do I know?

The starting adapter must be 9V in my situation.

Why?

With everything on, it stays at 9V. the 12V is coming from this:
https://www.ebay.com/itm/224996442243
I am putting in 9V, and it outputs 12v.

But it may also be contributing noise. It seems to me that it makes more sense to start with the voltage you actually need (12v) and drop it down as necessary by regulating than it does to start low and then have to boost and regulate back down.

I meant using the 78L09s in parallel from the 12V booster output. Not for regulation, but to isolate the 9V output of the 78L09 from whats coming off the adapter, because I am unsure about where the noise is coming from at this point.

In parallel to a single output or to two outputs? See my comments above. I doubt doubling the 78L09 to a single output will help with noise, unless 100mA is not enough current for that output. Liberal use of 100uf or so caps to ground might be more effective.

I was under the impression that the 78L09s require higher than 12V to produce 9V. Thats why I didnt want to go from 9V off the adapter into the 78L09s.

According to one datasheet I looked at, 11.5v to produce 9v, so a robust 12v output under load should be fine, but you are quite right.

[Esppse]

I have a weird 13 pin setup, that only spits out 9v, why is why the 9V setup is required.

This was my plan with the 78L09.

9V off adapter ---> Boost to 12v --->
1st 78L09 to first noisy effect
2nd 78L09 to second noisy effect

etc..

So the 78L09s all draw from a single 12v point, but they all go to separate individual effects.

Does this sound right?

[ElectricDruid]

Does this sound right?

No, it sounds a bit crazy to be honest.

If you want lots of 9V outputs, why are you starting with a 9V input and then trying to boost to 12V before going back down to 9V for the outputs? I'm not surprised you're having problems. You're doing it "the hard way", at best.

Start with a proper transformer for all your 9V supplies and work from there. Trying to force something to do something it *really* doesn't want to do is never going to end well!

[antonis]

If you want lots of 9V outputs, why are you starting with a 9V input and then trying to boost to 12V before going back down to 9V for the outputs?

Probably because he wants the benefits of a regulated 9V supply, via 78L09 regulators which need a higher than 9V unregulated input..

P.S.
I think we deal with a noisy booster and not noisy effects..

[amptramp]

We haven't asked one pertinent question - I doubt all of the distortions and fuzzes are running at once, so for the ones that are off, are their inputs grounded?  You can get a lot of noise by leaving the inputs open, often enough to oscillate or at least modulate the power supply current and voltage into the distortions and fuzzes that are in use.

[Esppse]

We haven't asked one pertinent question - I doubt all of the distortions and fuzzes are running at once, so for the ones that are off, are their inputs grounded?  You can get a lot of noise by leaving the inputs open, often enough to oscillate or at least modulate the power supply current and voltage into the distortions and fuzzes that are in use.

The inputs and outputs are not grounded when off. Does it make a difference whether the input or output is grounded when not active? I tried grounding the output of a noisy ring modulator when off, the oscillation of that ring mod is still bleeding through somehow. Would grounding the input make more sense?

The thing is that many are on a 2 pole rotary switch and it's basically selecting between the circuits. But most are on their separate switch. I see a way to do that with a 3pdt switch, but is there a way to ground the input with a 2pdt?

[FiveseveN]

is there a way to ground the input with a 2pdt?

Code Select Expand



IN  -•                        from effect -•
       \___ to effect                        \___ OUT

GND -•                                 IN -•

      SW A                                  SW B



[Esppse]

Ok I have just done a revision run with the following mods:

Changed non-grounding True Bypass to input-grounding True Bypass
3 Effects that were feeding off the main 9V line now receive power from the 12V Voltage booster into their own 78L09s for isolated 9V.
Hex Fuzz, Fuzz Factory, Thing Modulator

Used these caps for the 78L09s, they are not electrolytic, not sure if that matters.
https://protosupplies.com/wp-content/uploads/2017/11/7809-Basic-Circuit.jpg

Result:
Somehow my bypass tone has lost some high end. I am not sure why, my best guess is that there might be some part of the main true bypass line leaking to ground on certain switches. I will change some suspicious ones back to non-grounding True bypass to see if it helps.

The noise on the effects remains unchanged. Good news is that using the 78L09s after the voltage booster did not cause anymore noise, but it did not take any away either. The Ring mod (thing modulator modified with 10nf cap for higher frequency ringing, and a B2M pot for more range) is still bleeding through somehow. This effect received the dedicated 9V from one of the 78L09s. Im going to try to pull the LM567 chip from the Thing Mod, and 4070 from the digital octaver fuzz to see if the noise bleed stops. Even when the input is grounded, there is still noise leaking from the Thing mod. The noise must be from the ground path.

This is the spec for the adapter I am using:
https://secureservercdn.net/72.167.243.77/hnx.63b.myftpupload.com/wp-content/uploads/2019/10/volgen_KTPS50-VIRev2.pdf

The 9V is 45 watts, should be wayyyyyy enough current for everything.

All of my noise is higher pitched whine and whistling. I assume this is much easier to deal with than low ground loop type noise. I star grounded everything so I do not have the low noise.

Some effects I believe are suspicious are the following:
Thing Modulator (used LM567 instead of LMC567) (just ordered the LMC version today on ebay, hope its original and will take care of noise)
Digital Octave Fuzz (http://guitar-fx-layouts.238.s1.nabble.com/Escobedo-Digital-Octaver-Fuzz-td31124.html)
Parasit Multiwave (pcb version) (https://www.parasitstudio.se/multiwave.html)

This circuits use CMOS chips in them. I was curious to know if these are more susceptible to noise. And also, using CMOS chips, at what point does an effect be considered digital, rather than analog? Because they are generating octaves not present in the original signal.

Anyways, I have been reading up on power filtering. I am not sure if any of the ones on this thread will be useful:
https://www.diystompboxes.com/smfforum/index.php?topic=116491.0

Would they be of any use or was that already built into the adapter? I was thinking maybe power filter that utilizes an inductor, because all the noise is in the upper frequencies. It actually sounds like digital noise, super high whine and ringing. Would putting the filter before everything help, or do filters have to be implemented before each circuit individually?

[FiveseveN]

Im going to try to pull the LM567 chip from the Thing Mod, and 4070 from the digital octaver fuzz to see if the noise bleed stops.

Just unplug them from power. And please try it with a proper PSU designed for audio. You don't need 45 W, a tenth will do but you can't trust a random switchmode "Desktop Adapter" to be quiet.

at what point does an effect be considered digital

When it goes through an ADC.

they are generating octaves not present in the original signal

You don't need digits to do math.

[Esppse]

I forgot to mention that I am also using a Tube circuit in this build, which is why the high current one was used:

http://tagboardeffects.blogspot.com/2012/04/boob-tube-v15.html

I was unsure of its total current draw.

I had a lower current Triad adapter, the noise was still the same, that is for audio is it not?

[amptramp]

A three-terminal regulator is always going to pass current fluctuations back onto the power line even if the voltage to the effect circuitry remains the same, so maybe you should be looking at shunt regulation.  A simple resistor and zener diode at the input to the effect will maintain close to constant current so that current fluctuations in the circuitry are matched by almost equal and opposite current fluctuations in the zener, keeping a relatively constant current on the power lines.  If you are willing to go to a larger power supply than you would need to run the effects circuits, you can spend some of that extra power on shunt regulation so that the fluctuations in ground current and changes in ground voltage are reduced.

[antonis]

A TL431 should be convenient for what Ron said above..