[SOLVED] Guitar Amplifier noise when a DC adaptor used.

KlausST said:

Hi,

To reduce noise:
There are at least 5 GND connections on/to the PCB. Please use only one PCB GND connection and connect here all periferal GNDs with extra wires..in star point manner.
Additionally connect the cases of the pots to the star GND.

Klaus

Click to expand...

Would you be so kind to describe the "star point manner" please? I'm looking online but there are many examples for other kind of equipments and applications and its confusing to me. Logic tells me that I should add one extra cable to every ground connection that there is in the circuit, and join them all together in one mesh and connect them to the ground terminal on the PCB? and also the wire that runs from the ground terminal of the PCB to the metal plate?

Did you try a search (forum, internet)?

Klaus

KlausST said:

Did you try a search (forum, internet)?

Klaus

Click to expand...

Yes, I did found something. It is what I was thinking about. I'll try to see if that stops the amp from humming, I'll attach and example of what I was thinking about to see if its correct.. Did you saw the video that I attached in post #33 ?

Hi,

Reread post #40.

Klaus

The red wire from the input jack to the pcb is not shielded so it is an antenna that picks up hum.
All the wires from the volume control and tone controls are also picking up interference because the pcb was designed to have the controls mounted directly with no wires.
I agree that the cases of the pots should be grounded (mounted on a grounded metal box or plate).

KlausST said:

Hi,

Reread post #40.

Klaus

Click to expand...

I've tried the star point and it didn't work out, it still makes noise even tough I put a wire from the star point with all the ground connections onto the metal control plate where the switch, jack, dc jack and LED are mounted. When I touch that metal plate, the hums reduces drastically. The only pot making noise is the Bass pot (2nd from top to bottom) when turn to "0" the hum is the loudest. I don't see how is not working if all the ground connections are put into the ground port on the PCB. Sorry about the messy job. I'm pointing to the ground wire to the metal plate in the 3rd pic.

Audioguru said:

The red wire from the input jack to the pcb is not shielded so it is an antenna that picks up hum.
All the wires from the volume control and tone controls are also picking up interference because the pcb was designed to have the controls mounted directly with no wires.
I agree that the cases of the pots should be grounded (mounted on a grounded metal box or plate).

Click to expand...

You are right, the pots should be mounted in a metal plate and I will try to use shielded wire on the input jack (like in the telephone mic that I've showed you in other post) but what I don't understand is why is the bass pot the only one humming loudly when I turn it down? all the other pots work fine. It happened to me the 3 times that I've tried to build this amp.

Hi,

bzblues said:

I try to inform myself, but the more I see the more confusing it gets.

Click to expand...

...join the club, as they say, that's completely normal .

bzblues said:

... this forum is for people who have advanced knowledge in the field.

Click to expand...

No, it's for everyone. I don't have advanced knowledge.

Maybe the unshielded wires connecting the bass pot are very sensitive to picking up hum.
Maybe the bass pot is connected backwards then turning it "down" boosts the bass (hum).

Hi,

The star point in best case is very close at the PCB. But no need to change this now.

But - very important - really all GND signals go to the star point. No other GND wires are allowed.
Check these: LED_GND, signal_input_GND, master_volume_gnd, DC_input_GND... really all.
And check that there are no other GND wires. Especially check DC input..
--> Star wiring ensures that all parts in your circuit refer to the same reference voltage.

But star wiring can't suppress power supply fluctuations coming from the power supply. It can't replace a bad power supply.
Star wiring here just can reduce the effects of heavy supply current to cause GND bounce.

Now one may say there is no heavy supply current. Yes and no. If the power supply voltage is clean, the current will be clean, too.
But when the power supply goes into burst mode it will run and push the supply voltage to a desired level...then it will stop working (maybe for a couple milliseconds) ... during this time the supply voltage will drop a litle ... then it starts to run again to push the supply voltage..
But this "push" may create relatively heavy currents (peaks) ... to charge your 1000uF capacitor.

Did you try "base load" recommended in post#7? Please give feedback how this changes th noise.

Try a 1ohm series resistance in the "+" power supply line from connector to PCB. Give feedback.

Klaus

Hi Audioguru,

Audioguru said:

Maybe the unshielded wires connecting the bass pot are very sensitive to picking up hum.
Maybe the bass pot is connected backwards then turning it "down" boosts the bass (hum).

Click to expand...

I have a question: The bass trimpot is being used as a - I think the term is - rheostat. What would be the difference, if any, between connecting the unused pin 1 of it to the wiper or soldering pin 1 to a dead-end with no connection to ground or anything else on the PCB (as the schematic shows)?

Thanks.

Audioguru said:

Maybe the unshielded wires connecting the bass pot are very sensitive to picking up hum.
Maybe the bass pot is connected backwards then turning it "down" boosts the bass (hum).

Click to expand...

The bass pot is connected the right way. It can be that the wires are causing the humming, but I don't get why its only that pot and not the others you see? ... maybe its because of its reading of 1 Meg, or its components in that part of the circuit. When I bought it, the manufacturer offers you an option of "wiring" so you can mount the pots away from the PCB. On the 2nd amp that I've built a few months ago, just like this one, I had the same problem and the pots where mounted onto a metal plate. So, I guees that I should concentrate on:

1) Replacing the power supply for a higher quality one
2) Try to make a noise filter (investigating how can it be done)

KlausST said:

Hi,

The star point in best case is very close at the PCB. But no need to change this now.

But - very important - really all GND signals go to the star point. No other GND wires are allowed.
Check these: LED_GND, signal_input_GND, master_volume_gnd, DC_input_GND... really all.
And check that there are no other GND wires. Especially check DC input..
--> Star wiring ensures that all parts in your circuit refer to the same reference voltage.

But star wiring can't suppress power supply fluctuations coming from the power supply. It can't replace a bad power supply.
Star wiring here just can reduce the effects of heavy supply current to cause GND bounce.

Now one may say there is no heavy supply current. Yes and no. If the power supply voltage is clean, the current will be clean, too.
But when the power supply goes into burst mode it will run and push the supply voltage to a desired level...then it will stop working (maybe for a couple milliseconds) ... during this time the supply voltage will drop a litle ... then it starts to run again to push the supply voltage..
But this "push" may create relatively heavy currents (peaks) ... to charge your 1000uF capacitor.

Did you try "base load" recommended in post#7? Please give feedback how this changes th noise.

Try a 1ohm series resistance in the "+" power supply line from connector to PCB. Give feedback.

Klaus

Click to expand...

I realize that I didn't connect the speaker negative wire to the star point and it killed the guitar signal completly, the interference its gone, but so the guitar, so it didn't work. I don't have a 1 ohm resistor at hand, but I'll try with the lowest value that I have in my tool box.

I'll also try a lowpass filter but I have a doubt. I'm on a website that has a RC calculator. You just put 2 out of 3 of the information and it gives you what you need. But here is the thing. The humming sound appears to be 50 or 60hz, the sound that makes when the circuit is not grounded ...wich it is...in star point! but still, I touch the plate and the noise goes down pretty much, but still there. So, I calculate what kind of Resistor would I need to work with the 0.1mF capacitor to cuttoff freq of 60hz.

But its a low pass filter. It woudl mean that it would cut my freq only allowing the 60hz ? thats what logic its telling me! or would I need a HIGH PASS FILTER? I think its the later one! the filter goes into the guitar jack? or into the speaker terminals, or in the DC jack?

Anyways... I'll keep on trying.

Audioguru said:

Maybe the unshielded wires connecting the bass pot are very sensitive to picking up hum.
Maybe the bass pot is connected backwards then turning it "down" boosts the bass (hum).

Click to expand...

Audioguru said:

The red wire from the input jack to the pcb is not shielded so it is an antenna that picks up hum.
All the wires from the volume control and tone controls are also picking up interference because the pcb was designed to have the controls mounted directly with no wires.
I agree that the cases of the pots should be grounded (mounted on a grounded metal box or plate).

Click to expand...

I put a shielded wire from the jack to the PCB and it didn't work. I still don't understan why everything is "grounded" and still there is huming. I touch the metal plate where the jack, LED, DC jack and switch are mounted and the hum goes a way a bit but its still there.

Only thing left for me to try is the RC filter, but I'm not sure where to put ir. In the DC jack? in the guitar jack? in the speaker's terminals?

The AC-DC adapter is a switching one but the line hum is passed over. The line hum you are hearing is most likely 100-120Hz, the filter capacitor is likely to be under value. In other words, the adapter is unable to provide the

Quick attempt: try to add a 2200 uF capacitor to the DC supply jack of the amplifier. Does the noise disappear? It may not completely disappear but may reduce substantially.

Another: use a 9V 2A rated adapter and see whether the noise goes away. How to test whether the adapter is the culprit? Use the adapter to drive a small speaker with a 1 uF capacitor in series . If you can hear the hum, the adapter is not good (do not do this without the series capacitor because the speaker will be cooked!).

Another: make a pi filter for 100 Hz and use the output for the amplifier. The amplifier assumes that you have a decent power supply. It does not work well with a noisy supply.

c_mitra said:

The AC-DC adapter is a switching one but the line hum is passed over. The line hum you are hearing is most likely 100-120Hz, the filter capacitor is likely to be under value. In other words, the adapter is unable to provide the

Quick attempt: try to add a 2200 uF capacitor to the DC supply jack of the amplifier. Does the noise disappear? It may not completely disappear but may reduce substantially.

Another: use a 9V 2A rated adapter and see whether the noise goes away. How to test whether the adapter is the culprit? Use the adapter to drive a small speaker with a 1 uF capacitor in series . If you can hear the hum, the adapter is not good (do not do this without the series capacitor because the speaker will be cooked!).

Another: make a pi filter for 100 Hz and use the output for the amplifier. The amplifier assumes that you have a decent power supply. It does not work well with a noisy supply.

Click to expand...

Hello! thank you for your reply.

I've tried a 9V 2A higher quality adapter that I have and the noise is still there just like with the other 9V 1A cheap adapter.

I'll try the 2200 uF (microfarad, right?) capacitor to the positive and negative terminals of the DC jack to see if it works. The thing that I'm worried about is that the guitar signal is compromised if I add something to quiet down the noise.

bzblues said:

I realize that I didn't connect the speaker negative wire to the star point

Click to expand...

As far as I see the negative speake is not GND .. instead it is one side of the LM386 bridge.

bzblues said:

So, I calculate what kind of Resistor would I need to work with the 0.1mF capacitor to cuttoff freq of 60hz

Click to expand...

Standard formula: fc = 1 / ( 2 x Pi x R x C)
But have in mind: cutoff is just 70% of non filtered. But usually you want it to be reduced to 10% or 1%.... thus fc needs to be far below 60Hz.

you need a low pass filter for the supply. DC is veeeerrrryyy low frequency. You let DC pass (you let low frequencies pass) while you want to suppress high frequency (60Hz) noise.

Mind: this is not an audio signal filter. Here indeed a high pass filter could be a benefit. But usually you want 60Hz audio but don´t want 60Hz hum.

***
in the picture of post#51 I still see non_star_point GND wiress. All the black ones (but not the speaker black one).
Remove those black connections.

****
there is an (unconnected?) aluminum plate where the PCB is mounted. Connet it to the star point.
****
Don´t connetct the effect I/O unless you really need it.

Klaus

Yes, the bass pot is used as a rheostat. Usually, a rheostat has its unused pin connected to the wiper (the middle pin) so that if the wiper gets disconnected by some dirt in the track then the total resistance of the pot is there. It has no effect on how the pot works.

You connected the negative speaker wire to ground without seeing on the schematic that it connects to the output of one of the LM386 power amplifier ICs?? Then that LM386 is probably destroyed, even if it was not playing sounds. I spoke earlier that both LM386 ICs are already overloaded
since they are connected in a "bridged" mode. Modern amplifier ICs are designed to be bridged and are shown bridged in the datasheet. Not the old LM386.

You talked about using an LC lowpass filter. It is wrong because it passes low frequencies like hum but cuts all the high frequencies that you want to hear. A highpass filter would cut the hum but it would also cut all low frequencies that you want to hear.

Do you know if the hum is 50Hz from stray pickup or is 100Hz from the rectified power supply? It sounds like a 100Hz buzz to me. Maybe the main filter capacitor C24 (1000uF) or bias filter capacitor C23 (10uF) is defective or is mounted backwards.

bzblues said:

The thing that I'm worried about is that the guitar signal is compromised if I add something to quiet down the noise.

Click to expand...

Removing noise from the power supply should not and must not affect the sound being amplified.

If the capacitor across the DC jack does not help, then the hum is being picked up from air. First of all, you need to figure out whether the hum is 50 or 100 Hz. Compare with a pure tone generator like https://www.szynalski.com/tone-generator/. If the mains frequency is 60Hz, then listen for that.

while running on battery, if you bring the adapter (powered on but output not connected to the amplifier) power connector close to the signal input, do you get any hum pickup? If you touch the signal input with your finger, how much hum it picks up?

Also a very good design practice: bypass the power supply pins with a 0.1uF ceramic cap (i.e., the pins 8 and 4 shorted with a 0.1uF ceramic cap). It will not eliminate but reduce the noise. Do that for both the op-amps.

also re-read post #16

Audioguru said:

Yes, the bass pot is used as a rheostat. Usually, a rheostat has its unused pin connected to the wiper (the middle pin) so that if the wiper gets disconnected by some dirt in the track then the total resistance of the pot is there. It has no effect on how the pot works.

You connected the negative speaker wire to ground without seeing on the schematic that it connects to the output of one of the LM386 power amplifier ICs?? Then that LM386 is probably destroyed, even if it was not playing sounds. I spoke earlier that both LM386 ICs are already overloaded
since they are connected in a "bridged" mode. Modern amplifier ICs are designed to be bridged and are shown bridged in the datasheet. Not the old LM386.

You talked about using an LC lowpass filter. It is wrong because it passes low frequencies like hum but cuts all the high frequencies that you want to hear. A highpass filter would cut the hum but it would also cut all low frequencies that you want to hear.

Do you know if the hum is 50Hz from stray pickup or is 100Hz from the rectified power supply? It sounds like a 100Hz buzz to me. Maybe the main filter capacitor C24 (1000uF) or bias filter capacitor C23 (10uF) is defective or is mounted backwards.

Click to expand...

Fortunately the amp still works fine after connecting the negative wire of the speaker to ground. About the LC filter, that was my main concern, that the guitar signal loose some frequencies. I'll make some recordings of the sound to see at what exact frequency is the hum at. The capacitors are the correct way, if they where defective then the amp wouln't work? Thank you!

Your post #33 has a video that plays the 100Hz buzz triangle waveform when the pot or switch lowpass filter cuts all the microphone high frequencies.
The lowpass filter is connected only to the microphone and has no effect on the sound of the electric guitar. The filter would affect the guitar sound if its sound was acoustic on the microphone.