noise debugging

[poiureza]

Hello,

Anyone familiar with the noise in attachment ?
http://sndup.net/sd2s

What the hell is it ?

I have this every time I check one of my breadboard builds or even complete PCB's.
It's no ground hum, no white/pink/whatever noise
I disconnect everything I can in the room (PC, screen, lights, ...) but nope still there

It's definitely lower than any audio signal coming out of the circuit, maybe 10-20dB lower.
But still it makes "regular" noise optimization (IC, PCB trace layout, resistor choice etc ...) impossible

[GibsonGM]

What are you using to power your FX with?  It sounds like switching noise, maybe some kind of aliasing.  If you power your stuff with a battery, does it still do this??

[Dormammu]

Hi.
This may be interference from an external powerful source. Try check out the circuit in another location 500-1000 feet away. If you have a factory combo - how does it behave?

[poiureza]

It is battery powered ...
No offending external powerful source I know of (I'm in the basement)

Checking it out 500ft away ?
It would be worth a test (need to build me one battery powered headphone amplifier though)

FWIW it's ticking at exactly 25Hz, half the AC power over here

[FSFX]

That sound is like you get with 'motorboating'.

Check for proper decoupling in the circuit. Also see if changing decoupling capacitor values alters the frequency or rate of the sound.

It is often seen in high gain circuits with poor power supply decoupling allowing low frequency feedback to get into the input stages via the bias networks or poor grounding.
The circuit starts to act as a 'relaxation oscillator'. This can be caused by inappropriate choice of high value coupling capacitors in a high impedance part of a circuit.
Coupling capacitor values should be chosen to be appropriate for the impedance and frequency response of a circuit design.
 

https://en.wikipedia.org/wiki/Motorboating_(electronics)

[GibsonGM]

Problem with the breadboard or something? He says it does this all the time if I'm not mistaken, not just with 1 build.  Totally sounds like motorboating, yup!

I wonder if it's the house wiring?   Does this happen to just an amp that's plugged in?  How about a pedal that's built (in an enclosure)?

[FSFX]

Problem with the breadboard or something? He says it does this all the time if I'm not mistaken, not just with 1 build.  Totally sounds like motorboating, yup!

I wonder if it's the house wiring?   Does this happen to just an amp that's plugged in?  How about a pedal that's built (in an enclosure)?

It is difficult to pinpoint the source without suitable test equipment such as an oscilloscope, radio scanner or spectrum analyser. I said it sounds like motorboating, that's all, just one thing to eliminate. But if it occurs with every circuit and is continuous then it may be some nearby pulsed RF source like some home automation system and so adding some RF filtering may help. It sounds quite regular and so not likely to be a discharge source like an internal combustion engine or arcing.

[poiureza]

Thx for the links to motorboating and relaxation oscillators, I will dig into that.

The soundfile into an FFT software shows rectangular pulses (only positive swing - no negative) every 40 msec.
I wonder how this 25Hz can actually be heard ...
Anyway, the circuit here is low gain (~10dB and no clipping involved) but yes, it happens will almost everything I put on breadboard ...

I'l try different decoupling caps and see if anything changes.
All HP and LP filters created by caps and resistors in the circuit are supposed to have corner frequencies above 60Hz and below 20kHz but I will triple check that.  I even decoupled the powersupply with 100uF and 100nF although it's a battery lol.
and I tested with a 9V DC adapter and the noise is exactly the same, not louder, not quieter, same frequency.

And it's been a VERY long time since I actually boxed a circuit, sorry I don't remember if this noise was still present.

Okay, now about the suggested RF decoupling in case it's a nearby pulsating RF source.
How would I proceed ?

[FSFX]

Okay, now about the suggested RF decoupling in case it's a nearby pulsating RF source.
How would I proceed ?

Reducing the effect of RFI depends a lot on how it is getting into your circuit. If it is being picked up on the input lead or signal source then having a small RC filter from the input jack to the circuit is often the easiest solution. A 1k resistor and 470 pF capacitor should block most RF frequencies whilst not really affecting the audio by any noticeable amount. Small ferrite beads on the input wire can help but the inductance they create is very small and so they are really only effective for helping block very high frequency RF such as TV, FM radio, cellphone and WiFi type of signals. They are not effective for long wave or medium wave AM signals.

If you are building on a breadboard then the long patch wires can act as radio antennae so keep the wires as short as possible. Also having a grounded metal plate under the breadboard may help screen it from picking up RF signals. You can cover it with an insulated layer to prevent accidental shorting of your circuit or wires to ground. Keep your breadboard as far away as you can from cellphones, cordless phones, Bluetooth devices, computer screens and switched mode power supplies and other devices that may emit radio signals.

If you can, try running your breadboard circuits at another location like a friend's house to see if the problem moves with location.

And you don't by any chance live close to any military facilities or other places which may operate high power, low PRF pulsed radar systems, do you?

[stallik]

And you don't by any chance live close to any military facilities or other places which may operate high power, low PRF pulsed radar systems, do you?

Now that jogs a really interesting memory. Story for another time perhaps - don't want to derail this thread

[Dormammu]

It is battery powered ...
No offending external powerful source I know of (I'm in the basement)

Checking it out 500ft away ?
It would be worth a test (need to build me one battery powered headphone amplifier though)

FWIW it's ticking at exactly 25Hz, half the AC power over here

Sometimes, someone has to deal with this :

[poiureza]

Sometimes, someone has to deal with this :

I definitely feel like the guy in the video but I don't have the problem with guitar and amp plugged in.

[poiureza]

Reducing the effect of RFI depends a lot on how it is getting into your circuit. If it is being picked up on the input lead or signal source then having a small RC filter from the input jack to the circuit is often the easiest solution. A 1k resistor and 470 pF capacitor should block most RF frequencies whilst not really affecting the audio by any noticeable amount. Small ferrite beads on the input wire can help but the inductance they create is very small and so they are really only effective for helping block very high frequency RF such as TV, FM radio, cellphone and WiFi type of signals. They are not effective for long wave or medium wave AM signals.

If you are building on a breadboard then the long patch wires can act as radio antennae so keep the wires as short as possible. Also having a grounded metal plate under the breadboard may help screen it from picking up RF signals. You can cover it with an insulated layer to prevent accidental shorting of your circuit or wires to ground. Keep your breadboard as far away as you can from cellphones, cordless phones, Bluetooth devices, computer screens and switched mode power supplies and other devices that may emit radio signals.

If you can, try running your breadboard circuits at another location like a friend's house to see if the problem moves with location.

And you don't by any chance live close to any military facilities or other places which may operate high power, low PRF pulsed radar systems, do you?

Lots of great info thanks !

I'll try the MHz filter at the inputs though I think that the patch cable antenna thing is a very good candidate.
Actually I do have a metal plate under the breadboard but come to think it isn't connected to anything duuuh ...
Good idea also to cover the circuit with an aluminum foil, I'll try that (I suppose I need to connect that one to ground as well ?)
And yes I'm going to test my stuff at another location later this week

There's a military facility 2 miles away, not sure what they're doing but they do have a big tower with radio antennas, and I have an electricity/phone company 2 miles away in the opposite direction.  Mmmmh ...


Thanks a lot guys, I do some tests and let you know
In case it's the military facility I'll have to blow it up I'm afraid

[poiureza]

Reducing the effect of RFI depends a lot on how it is getting into your circuit. If it is being picked up on the input lead or signal source then having a small RC filter from the input jack to the circuit is often the easiest solution. A 1k resistor and 470 pF capacitor should block most RF frequencies whilst not really affecting the audio by any noticeable amount. Small ferrite beads on the input wire can help but the inductance they create is very small and so they are really only effective for helping block very high frequency RF such as TV, FM radio, cellphone and WiFi type of signals. They are not effective for long wave or medium wave AM signals.

If you are building on a breadboard then the long patch wires can act as radio antennae so keep the wires as short as possible. Also having a grounded metal plate under the breadboard may help screen it from picking up RF signals. You can cover it with an insulated layer to prevent accidental shorting of your circuit or wires to ground. Keep your breadboard as far away as you can from cellphones, cordless phones, Bluetooth devices, computer screens and switched mode power supplies and other devices that may emit radio signals.

If you can, try running your breadboard circuits at another location like a friend's house to see if the problem moves with location.

And you don't by any chance live close to any military facilities or other places which may operate high power, low PRF pulsed radar systems, do you?

Lots of great info thanks !

I'll try the MHz filter at the inputs, that the patch cable antenna thing is a very good candidate.
Actually I do have a metal plate under the breadboard but come to think it isn't connected to anything duuuh ...
Good idea also to cover the circuit with an aluminum foil, I'll try that (I suppose I need to connect that one to ground as well ?)
And yes I'm going to test my stuff at another location later this week

To be honest I'm currently having this problem with a populated PCB, not a breadboard but I do have the same problem with breadboards too.
The PCB is just laying around without any shielding ...


There's a military facility 2 miles away, not sure what they're doing but they do have a big tower with radio antennas, and I have an electricity/phone company 2 miles away in the opposite direction.  Mmmmh ...


Thanks a lot guys, I do some tests and let you know
In case it's the military facility I'll have to blow it up I'm afraid

[anotherjim]

Submarines use low-frequency communication but I read that it's at least 30Hz. Maybe they are experimenting.
More humdrum is an SCR power control that only needs to trigger on alternate cycles. Perhaps a fluorescent or LED light fitting?

[Dormammu]

Submarines use low-frequency communication but I read that it's at least 30Hz.

I heard that they communicate at frequencies up to 300 kHz.

[FSFX]

Submarines use low-frequency communication but I read that it's at least 30Hz.

I heard that they communicate at frequencies up to 300 kHz.

But submarine communications does not use modulation like that in the posted sound clip which is more like over-the-horizon radar.
The pulse repetition rate of radars and the frequency at which they operate depends a lot on the range of the radar.
If it is a short range mortar tracking radar then it uses extremely high frequencies with a high repetition rate to be able to detect small object and determine the distance to a fine accuracy. Large, very long range targets require lower frequencies to allow reflection off of the ionosphere or bending of the wave around the Earth's surface and a low repetition rate to for accurate ranging and prevent aliasing of the reflected signal generating false distance measurements.   

[GibsonGM]

Our sub comms here on the US east coast use VLF, and probably ELF too.  That said, 3-30kHz seems to be very common.  And in practice, it's doubtful that us 'not cleared civvies really know exactly what's up with that, LOL.  I used to live next to a station on the Maine coast...it was interesting. No interference from them tho.

If I had that noise and couldn't figure it out, I'd ask a neighbor to let me bring my amp and breadboard over to see if it occurs at THEIR house as well.  The worst I ever get here is the local sheriff's dept or someone speaking French thru my amp when I turn down my guitar volume and am playing thru an unshielded fuzz.

[FSFX]

And in practice, it's doubtful that us 'not cleared civvies really know exactly what's up with that, LOL.

I am sure that the principles haven't changed since I worked on ELINT systems, just the technologies implementing it.

[GibsonGM]

They probably have not, unless the laws of physics have changed    Only that regular civilian folks don't really have access to the 'little tricks' that are involved in its deployment.  It's kind of...secret.  We did hear rumors (I'm sure you know which area I used to live in, if you worked in this).   Off topic, anyway.  I support the idea that there is some kind of control signal nearby that's bugging the OP.    "Smart meter"?