What's making this noise? Please help me figure it out!

[Thebrainless]

Ok, thanks for the infos!
The only thing I don't get is why should I shield the whole bass when it's perfectly silent using 9v batteries, my logic would tell me to shield just the circuits generating noise, and maybe the battery, even though it should already be shielded.
That's something I can't really understand, but I guess I don't have the necessary knowledge to understand it, so I'll do my homework 
Thanks again!

[Thebrainless]

So, shielding wasn't the solution, the noise it's not from a lack of insulation, the noise is from...well, I don't know, but with a simple RC lpf filter I managed to reduce it to an acceptable level.
But there's some "strange" effect happening:

As you can see/hear I've lowered the noise to a reasonable level, but a slightly compression (?) of the signal appeared, why so?
The RC lpf filter at 60hz is made of a 265 ohms resistor and a 10 uf 50V low ESR ceramic capacitor mounted on the pcb of the step up converter, right on the output, and so were all the other filters I've tried, same capacitor, different resistors.
Can someone at least explain why the compression (or whatever it is) of the signal? I've searched online, nothing, there's no info about such thing.
Any ideas?

[GibsonGM]

Not enough info...what "LPF"? What are the values, and we need to see the schematic of how you assembled it.

Resistors limit current, perhaps they are limiting the power supply current here, and if the circuit is starved, distortion may result.  It is basically losing headroom.   You still get your 18V after the LPF?  Just a guess cuz I can't see what you built. 

I assume you mean KILOHERTZ when you write 'Hertz" in your vid?    

I suggested totally shielding your bass just cuz you are already in there doing it.   And the switching supply is RIGHT THERE, literally on top of it.  Noise will get into everything, couple into wiring etc.     Anywhere wires are passing unshielded thru the body, they'd need to have an internal shield grounded at 1 end only, things like that.   

Ambitious project, I can see it's kind of a PITA, LOL.  I think I'd prefer a 9 or 18v battery, charge it while on-board, remove charger cable and play.  But you are actually being exposed to some good stuff to learn about!  Sometimes we just CAN'T get it quiet, no matter what we try.

[anotherjim]

For easy shielding, copper slug tape comes to mind. Gardening stores will have it (and Amazon). It's exactly like self-adhesive shielding tape, even has conductive glue and can be soldered but most have a wave pattern on one edge to confuse the slugs.
You should treat the output jack shield as the centre of a star ground connection. Shielding should only connect to the star from one end - don't allow currents to travel from one place to another via shielding.

Anyway, I would consider the possibility that all is not well with that bass guitars electronics. The noise could be worsened by it taking too much current due to a fault.
Did I see a picture showing the electronics in a potted block somewhere? If so, I'd want to chuck it away - you can't fix it.
There are circuits & kits & PCBs for better alternatives out there based on classics such as the Musicman/Ernie Ball models. I know a player with a Bongo bass & he hardly ever has to change its x2 9v batteries.

[Thebrainless]

Not enough info...what "LPF"? What are the values, and we need to see the schematic of how you assembled it.

Well, all the infos are there, what values do you need? I wrote both the resistor and the capacitor values, where I put 'em, the frequency cutoff, what else am I missing?
I don't have a schematic of the step up converter, so...

Resistors limit current, perhaps they are limiting the power supply current here, and if the circuit is starved, distortion may result.  It is basically losing headroom.   You still get your 18V after the LPF?  Just a guess cuz I can't see what you built.

Yep, the output is still 18V, the LPF only takes 0.02 V away, so I had like 18.05 V now it's 18.03 V

I assume you mean KILOHERTZ when you write 'Hertz" in your vid?  

No, I mean Hertz where I wrote Hz and Kilohertz where I wrote Khz, why?

I suggested totally shielding your bass just cuz you are already in there doing it.   And the switching supply is RIGHT THERE, literally on top of it.  Noise will get into everything, couple into wiring etc.     Anywhere wires are passing unshielded thru the body, they'd need to have an internal shield grounded at 1 end only, things like that.

I get it, the fact is there's no need to do it, the bass is silent and doesn't pick up any noise with normal batteries, so I should achieve the same result with a lithium battery since that's the only thing that's changing (that, and the circuitry to charge and step up of course, but I've already tried shielding those circuits and it's useless as well).

Ambitious project, I can see it's kind of a PITA, LOL.  I think I'd prefer a 9 or 18v battery, charge it while on-board, remove charger cable and play.

Well, that's what I'm trying to achieve, to have the battery (and the circuits) permanently inside the bass, charge it when needed, then remove charger cable and play, I thought it was clear, maybe I didn't explain it clearly...

For easy shielding, copper slug tape comes to mind. Gardening stores will have it (and Amazon). It's exactly like self-adhesive shielding tape, even has conductive glue and can be soldered but most have a wave pattern on one edge to confuse the slugs.
You should treat the output jack shield as the centre of a star ground connection. Shielding should only connect to the star from one end - don't allow currents to travel from one place to another via shielding.

Again, thanks for the infos, shielding's not the solution here

Anyway, I would consider the possibility that all is not well with that bass guitars electronics. The noise could be worsened by it taking too much current due to a fault.
Did I see a picture showing the electronics in a potted block somewhere? If so, I'd want to chuck it away - you can't fix it.
There are circuits & kits & PCBs for better alternatives out there based on classics such as the Musicman/Ernie Ball models. I know a player with a Bongo bass & he hardly ever has to change its x2 9v batteries.

There's no fault in the bass circuitry, with normal batteries it works perfectly, it's just a very consuming preamp, it is been made like that, nothing wrong, nothing to fix, in fact it sounds great and I don't wanna change it, I'd just like to have a recheargable battery to power it so I don't have to buy thousands of 9V batteries and pollute

Anyway...
I've tried another step up converter, this one's bigger and apparently more precise and reliable, and in fact the output doesn't need the LPF filter like the other one, it's still less noisy by itself, but again the minimun noise is still there with almost the same curve as the filtered one, but the compression is gone (as one should've expected of course since there's no LPF involved this time)

Here's the new step up converter based on the XL6009 mosfet



Here's its noise eq:




The absurd thing that happened is this:

Let's sum it all up, so now I have a lithium 3.8V 3000 mah battery connected to the input of a charge/protection circuit based on TP4056, the circuit output going to a step up converter circuit based on the XL6009 with the output set to 18V going to the bass preamp, everything exposed outside of the bass with no shielding whatsoever.
It gives enough power for the bass to work but there's still that minimum noise, and like on the filtered circuit, the noise is low enough to be ignored, but I want a clean signal, so (here's the mindblowing solution) I randomly noticed that by touching the metal plate of the XL6009 mosfet with a screw the noise gets lowered, so I connected an alligator clip test lead to it, and with its other end I touched bass ground, output disappeared, touched circuits ground, same thing, I went randomly touching spots until I touched battery positive on the charging circuit, and magic happened! The noise was GONE!!!!!!!
The eq curve became the same as the one with the 9V disposable batteries, the noise was finally gone.

But why???

How is it possible?

I'm too noob to understand what's going on here, I'm glad the noise is gone, but at this point I wanna know why!

Thanks to anyone who will explain to me what's going on here (hopefully without telling me I gotta be shielding the circuits, and the bass, and myself, and my house... )

[anotherjim]

Hmmm...
That tab is the switch output. Connecting it to battery + ought to bypass the boost coil and the switch shorts the battery, yes? The battery wire on the charger is always connected to battery +?

Anyway, it was probably the capacitance of whatever touches the switcher tab that reduced the noise, so you might find a small value ceramic capacitor across the battery terminals - may be 10nF or smaller is needed.
Regards screening. The simplest thing to do is have the wires tightly twisted, although if it's extra-flex wire (probably the battery) it won't want to stay twisted.

[Thebrainless]

Hmmm...
That tab is the switch output. Connecting it to battery + ought to bypass the boost coil and the switch shorts the battery, yes? The battery wire on the charger is always connected to battery +?

The battery wires on the charger are both always connected to the battery, yes.
I don't know if it shorts the battery, the charger circuitry has protection against it, plus measuring with a multimeter it seems like the battery positive it's already connected to that switch cause there's continuity when checking...

Anyway, it was probably the capacitance of whatever touches the switcher tab that reduced the noise, so you might find a small value ceramic capacitor across the battery terminals - may be 10nF or smaller is needed.

You mean a capacitor between positive and negative should do the same? Should it be ceramic, electrolytic or what? You think the alligator clip wire capacitance is doing it? If that's the case then I think the capacitor should be connected between the battery and the charger circuit just on the positive and not between positive and negative, as I wrote before there's continuity between the battery positive and that mosfet switch, so I guess the battery negative isn't involved here, right?

[anotherjim]

When high frequencies are around (the converter does 400kHz and there will be higher harmonics of that) audible noise seems to happen even though the source is supersonic. The HF can act as a carrier for noise that is audible, when it meets semiconductors the noise can be detected from the HF like a good old radio receiver.

The type of filter capacitor used becomes critical at higher frequencies. A 10nF ceramic disc can outperform a 10uF electrolytic. The reason is the inductance of the structure of the capacitor. At HF inductance raises the ESR. An electro cap is a wound component - it has something of an inductor coil about it while the ceramic disc is just what it looks like but even that has some self-inductance. A belt & braces approach to filtering employs an Electro (>=10uf) with a large ceramic (100nF) and a smaller ceramic (10nF).

There could be plenty of places to fit an extra capacitor. The converter already has some ceramic caps on in and out power connections but the application example (many cheap modules are the datasheet example circuit) shows 105 value or 1uF which will be an MLCC type. Adding 10nF disc caps across those should be better. Adding a cap to the switcher tab has implications for its switching performance - could cause trouble rather than help, but won't hurt to try.

Note that as all this noise is AC, it often does not matter if the + supply or ground are used to terminate a filter cap. There should be low impedance for AC between supply and ground.

Capacitance isn't the only way to filter. Ferrite rings over the wires raise the RF impedance in the wire blocking it. These are those lumps moulded in the ends of some USB cables for instance. If you were putting this together as a commercial product, you will have to do this stuff anyway to limit radiation to the outside world.

Wire has self-inductance and at very high frequency, there is the "skin resistance" so your croc clip lead may not act as the short circuit it appears to be, but more of a load impedance. This has implications when messing around trying stuff and if you touch anything your body capacitance and resistance will change the effects.

[Thebrainless]

So, I was happy and thought I found the solution....
Oh I was wrong!
Connecting the SW to the battery positive just lasts 5 seconds, then the noise builds back up.
At this point the only thing I got rid of is the compression, and that's just by changing the step up converter with the other one...
Now while trying random things, I've put a resistor of around 300 ohms between the positive input and the positive output of the step up converter, noise gone, of course the resistor was getting hot so I took it off (no soldering, just croc leads), but what does it mean?
I'm kinda getting tired of this "solutions" without meaning....

[GibsonGM]

Oh, there's 'meaning', but you have to be a rocket scientist to really deduce what it is!   All these circuits can generally be modeled as resistances, inductances and capacitances, in series or parallel.  All the poking around you're doing is changing some ratio of those here and there, affecting the outcome.  Not that it's easy to know exactly HOW you're changing them...but sometimes that stuff gives clues.   

I used to read old ARRL Ham Radio handbooks, which dealt with this a lot because radio involves much higher freq's than we deal with.  It was pretty eye-opening (but complex).  A resistor can exhibit properties of having a capacitor in parallel and/or an inductor in series with it, depending on what frequency signal you're passing thru it, for example.  Parts are NON IDEAL in the real world because of how they're constructed and the laws of physics, meaning they come with other baggage...run 2 wires too close to each other, they can become an unwanted (or 'parasitic') capacitance.    Drop an R in the right place (even by accident), you might shunt the parasitically coupled junk away by making an accidental filter...that's why finding your noise is so difficult, esp. at a distance...it's frustrating.    Read about "non ideal properties of basic components", and that may give some meaning...

[Thebrainless]

 
Damn I opened a Pandora's box here!!!

[GibsonGM]

Yup, you SHORE did, ha ha.   In the end, some light reading about those parasitic effects - even if you don't fully understand it - might be enough to get the gist of what goes on. It's why LM386 amplifiers feedback and squeal, and so on.   Applies to audio signals, power supplies like yours...it's also why correct wiring runs and shielding and the like are so important!  Design and build 'good practices'...

[anotherjim]

So, I was happy and thought I found the solution....
Oh I was wrong!
Connecting the SW to the battery positive just lasts 5 seconds, then the noise builds back up.
At this point the only thing I got rid of is the compression, and that's just by changing the step up converter with the other one...
Now while trying random things, I've put a resistor of around 300 ohms between the positive input and the positive output of the step up converter, noise gone, of course the resistor was getting hot so I took it off (no soldering, just croc leads), but what does it mean?
I'm kinda getting tired of this "solutions" without meaning....

Some switching converters need a minimum load anyway but also any fast voltage spikes on the output will be weak and  a load resistor will pull those spike voltages lower but not drop the 18v (just make it work harder) unless the current rating of the converter is exceeded. 300R dissipates 1W at 18v so hot and wasting 60mA from the battery.

Anyway, I have a feeling the problem lies with the bass preamp. Knowing it's going to be used with battery power, the designer probably didn't waste space on power supply filters, so any extra filtering should be as close to the preamp as possible.
First, find out how much current the preamp takes. Put a 10R resistor in series with the + or - power lead to the preamp and switch on. Measure the DC voltage across the 10R. Do Ohms law: current = voltage/resistance and report back.

[Thebrainless]

First, find out how much current the preamp takes. Put a 10R resistor in series with the + or - power lead to the preamp and switch on. Measure the DC voltage across the 10R. Do Ohms law: current = voltage/resistance and report back.

The preamp takes 16.5 mah, measured with the 2 disposable 9V batteries.
It's more than certain that the preamp was made without thinking about any other power source than 9V batteries I'm afraid

[anotherjim]

Another Q
Does the PSU negative supply connect to the output jack ground? Measure resistance of this, please. Normally, you'd expect a preamp negative to go directly to ground but they may have done something clever that could catch us out.

[Thebrainless]

Another Q
Does the PSU negative supply connect to the output jack ground? Measure resistance of this, please. Normally, you'd expect a preamp negative to go directly to ground but they may have done something clever that could catch us out.

Yes, the negative from the 9V battery plug goes straight to the sleeve of the bass jack, the ring is conneted to ground, the tip I don't remember, so the circuit is closed only when a jack is in.

[anotherjim]

Ok, do you think you can find something like this in your part of the world...
https://www.amazon.com/YoCoo-Supply-Filter-Transmitter-Quadcopter/dp/B01KZHGWRW
An
LC power filter for DC-Converters.
It will work best wired as close as possible to the preamp.

[Thebrainless]

Ok, do you think you can find something like this in your part of the world...
https://www.amazon.com/YoCoo-Supply-Filter-Transmitter-Quadcopter/dp/B01KZHGWRW
An
LC power filter for DC-Converters.
It will work best wired as close as possible to the preamp.

Found that same part, 10€ shipped, it'll take one full month to arrive... Maybe it's better/faster/cheaper to diy it?
What is it made of, an inductor and a large capacitor, and what's that smd?
I've also found that there's a circuit used to eliminate ripple which is called "capacitance multiplier", made by a low value resistor and a capacitor plus an npn transistor or a mosfet, but I can't find info on what transistor choose and which values should I go for...
I've got a couple of transistors here, but who knows if they fit or not?

[anotherjim]

I'd have suggested DIY straight away, but inductors like that aren't all that well known to most of us here *. I just think an LC filter will be more effective than anything we might make with an RC filter and not waste battery power. A inductor in series is perfect for reducing high frequency noise as its impedance rises with increasing frequency while it's DC resistance is low so doesn't drop much voltage or get hot like a resistor would.
* Edit - the inductor is probably 100uH. Something like this...
https://www.bitsbox.co.uk/index.php?main_page=product_info&cPath=246_250&products_id=1877

The SMD part you can see is a ceramic capacitor. As I mentioned, an Electrolytic type will not filter out RF frequency.
Almost certainly these filters are copied from a suggested application circuit in some manufacturers product data, but need to find some. The exact inductor value used ought to match a target noise frequency but a generic filter will probably come near enough.
If you want to try making an RC filter...
68R in series with the + preamp power wire. this will drop about 1v at 16mA
A 220uF electrolytic capacitor after the resistor connecting to - preamp power wire.
A 100nF ceramic capacitor added in parallel to the 220uF.
If necessary, a 10nF ceramic in addition to the 100nF to increase RF filtering.

Now, the above may not solve your problem in one hit, but it should be fitted anyway. The reason is that not all interference problems will be apparent at home. There could be things on stage that might radiate noise at you, and you might radiate noise at them. So you really should screen everything.

[Thebrainless]

I'd have suggested DIY straight away, but inductors like that aren't all that well known to most of us here *. I just think an LC filter will be more effective than anything we might make with an RC filter and not waste battery power. A inductor in series is perfect for reducing high frequency noise as its impedance rises with increasing frequency while it's DC resistance is low so doesn't drop much voltage or get hot like a resistor would.
* Edit - the inductor is probably 100uH. Something like this...
https://www.bitsbox.co.uk/index.php?main_page=product_info&cPath=246_250&products_id=1877

The SMD part you can see is a ceramic capacitor. As I mentioned, an Electrolytic type will not filter out RF frequency.
Almost certainly these filters are copied from a suggested application circuit in some manufacturers product data, but need to find some. The exact inductor value used ought to match a target noise frequency but a generic filter will probably come near enough.
If you want to try making an RC filter...
68R in series with the + preamp power wire. this will drop about 1v at 16mA
A 220uF electrolytic capacitor after the resistor connecting to - preamp power wire.
A 100nF ceramic capacitor added in parallel to the 220uF.
If necessary, a 10nF ceramic in addition to the 100nF to increase RF filtering.

Now, the above may not solve your problem in one hit, but it should be fitted anyway. The reason is that not all interference problems will be apparent at home. There could be things on stage that might radiate noise at you, and you might radiate noise at them. So you really should screen everything.

I've already tried rc filtering, a bunch of different frequencies, the only one really effective that doesn't square the bass signal is the one at 60hz, only downside is the noise is still there, low enough to be playable on the clean channel, but to high when going overdrive, since I'm in overdrive 99% of the times...
I guess I'll try the LC filter then, that 101 inductor seems to be 100uh, I can get a bunch of them no problem, the question here's the values of other components besides that big electrolytic... Plus I've read an LC filter should be set on the frequency of the switching supply, but how do I find such frequency? Then how do I decide components values? I'll keep looking for infos then at some point I'll try random values like I did with the RC filter, and let's hope for the best!