What you read above. It's a very simple amp - A gain block that goes into a diode clipper followed by another gain stage, some low frequency filtering and then power output using a class B output within a unity follower feedback loop. The amplifier itself works very well without any discernible noise (this is for headphones btw, so the noise would definitely be noticeable). But when I use the only >10V supply I have (It's a pretty crappy SMPS), it induces a lot of noise into the circuit. I've tried filtering it through a CLC filter, which didn't work well. I also tried a capacitance multiplier - didn't work well either. Strangely however, if I ground myself and then touch any metallic parts on my electric guitar, the noise drops noticeably. It doesn't go away though. I'd like to know why that is, it's very interesting to me.
I'd appreciate advice on how to fix the noise as well, but the above is what I'm really here for.
I'd appreciate advice on how to fix the noise as well, but the above is what I'm really here for.
Electrical leakage with nowhere to go but through you when you ground yourself ---I would bin the SMPS .
I absolutely plan on throwing it in the trash once I can afford to build a linear supply.
Would connecting the SMPS outputs to ground be enough to get rid of it?
And are there any resources I can reference to understand this more? The output of the SMPS should be galvanically isolated from the mains, so then why does it have leakage?
Would connecting the SMPS outputs to ground be enough to get rid of it?
And are there any resources I can reference to understand this more? The output of the SMPS should be galvanically isolated from the mains, so then why does it have leakage?
Its a design characteristic of a lot of SMPS,s , capacitor filters are used to smooth out the power supply but they must have somewhere to remove the electrical noise and that is to an earth point .
The problem is how much voltage/current is running through the capacitor filter as a cheap SMPS can have quite a bit and earthing it when its not supposed to be earthed could cause a fault condition but that is down to the design .
When you earth it through your body ,your body acts like a resistor that's why there is still noise , it depends on whether you have thick or thin skin and whether your hands are moist or not .
Those with thin moist skin will get higher voltage electrical shocks , I have thick dry skin otherwise I would not be here posting this .
The problem is how much voltage/current is running through the capacitor filter as a cheap SMPS can have quite a bit and earthing it when its not supposed to be earthed could cause a fault condition but that is down to the design .
When you earth it through your body ,your body acts like a resistor that's why there is still noise , it depends on whether you have thick or thin skin and whether your hands are moist or not .
Those with thin moist skin will get higher voltage electrical shocks , I have thick dry skin otherwise I would not be here posting this .
This problem is inherent to all 2-prong smps ac-adapters (wall warts). The elephant in the room are their y-caps injecting hi-distorted 100Hz ripple common mode noise currents into the secondary circuit. As you noticed any filtering and capacitance multipliers do not help. This is to be expected as they are fine only for suppression of differential mode noise.
Your test indicates that the only way to remove this noise is grounding your secondary circuitry to protective earth (PE) - or use an old fashioned 50Hz iron PSU not exhibiting these problems.
For this reason all pro guitar amps in the world have a 3-prong mains plug.
Your test indicates that the only way to remove this noise is grounding your secondary circuitry to protective earth (PE) - or use an old fashioned 50Hz iron PSU not exhibiting these problems.
For this reason all pro guitar amps in the world have a 3-prong mains plug.
Last edited:
Find a scrap amp in working condition that uses a linear supply.
Use that, and you will find voltages inside that will be useful for the guitar.
Also, run a ground right up to the guitar.
If a switching transistor fails in short mode, you will get mains voltage, which is dangerous.
And look around, laptop SMPS (brick) with a 3 prong supply may be available.
Use that, and you will find voltages inside that will be useful for the guitar.
Also, run a ground right up to the guitar.
If a switching transistor fails in short mode, you will get mains voltage, which is dangerous.
And look around, laptop SMPS (brick) with a 3 prong supply may be available.
I've found that cheap SMPS from *insert mass market website name here* sometimes come with a 3-pin connection but when you break the unit open, it is not connected to anything. These have given me grief with grounding issues before and now I just pony up for the more expensive SMPS units that have a datasheet.
How much power / voltage is the amp running on? Sometimes it's possible to cobble together a low power / voltage linear supply for not too many $$$
How much power / voltage is the amp running on? Sometimes it's possible to cobble together a low power / voltage linear supply for not too many $$$
Then how exactly would I go about removing common mode nois? Common mode chokes and ferrite beads on the power cable?
I wish I could find a scrap amp somewhere.
As for the transistor failure, correct me if I’m wrong but the mains voltage should never reach the output regardless of failure because the output is galvanically isolated. If it does fail, a fuse or PTC should stop the transformer from burning out.
I also tried using a DELL power brick - still has the same issue.
Right now I’m just using my headphones, but I do want to make it into a getter amp - 20 to 30 Watts of output power should be enough. Of course that would require a fairly beefy power transformer considering the losses I’ll incur on voltage regulation and a class AB output stage
FWIW, so far in this thread I have just read anecdote and some incorrect statements about guitar amps. Suggest you have a more open mind about whether the SMPS is really the problem. Passive guitar circuits are high impedance and are non trivial to shield and avoid hum.
Maybe so, but if I use batteries the noise disappears from the output, as stated in my first post. Well designed switchers can be good and low noise enough for guitar amp use, but I doubt mine is
Well, it did cost me $3
But then what do you think is the cause of the noise? It’s definitely not the 50Hz hum off of mains, much higher in frequency. I’ve changed from an smps to a battery source, and from what I can tell, the amplifier and guitar should be fine. How would the guitar noise not be coupled in if I’m using batteries.
But then what do you think is the cause of the noise? It’s definitely not the 50Hz hum off of mains, much higher in frequency. I’ve changed from an smps to a battery source, and from what I can tell, the amplifier and guitar should be fine. How would the guitar noise not be coupled in if I’m using batteries.
The point is using battery gets rid of both AF and RF interference at the same time. You haven't figured which is your problem yet. And if your body helps shield, it could well be RF. If you use any other AC supply of any design or quality, you will have to deal with RF.
I buy scrap amps at the flea market.
Old combination sets - double cassette, FM, aux and / or CD in.
Mostly Philips, sometimes Sony, Aiwa etc.
$2 each without speakers.
Philips 6 " speakers, with tweeters, about 3$ a pair, in cabinet, sometimes the MDF board is moisture damaged.
Philips India are mostly Philips / Japanese capacitors, NEC or Matsushita chip amps, the FM tuner and pre amp are mostly top Japanese or Philips chips. Toshiba selector chips mostly.
5 band equalizer too.
A good wash with soapy water, then plain water, then sun dry.
Next day blow away the dust inside, and spray all the pots with contact cleaner.
Ready to go.
And if it does break, parts are available...or find another.
Old combination sets - double cassette, FM, aux and / or CD in.
Mostly Philips, sometimes Sony, Aiwa etc.
$2 each without speakers.
Philips 6 " speakers, with tweeters, about 3$ a pair, in cabinet, sometimes the MDF board is moisture damaged.
Philips India are mostly Philips / Japanese capacitors, NEC or Matsushita chip amps, the FM tuner and pre amp are mostly top Japanese or Philips chips. Toshiba selector chips mostly.
5 band equalizer too.
A good wash with soapy water, then plain water, then sun dry.
Next day blow away the dust inside, and spray all the pots with contact cleaner.
Ready to go.
And if it does break, parts are available...or find another.
Last edited:
The chopper transistor induces a lower voltage in the secondary coil which is rectified to output.
So in failure mode, it should not pass on a high voltage to secondary.
But if you look at the transformers, the windings can burn or flash over, the wires are thin, with a thin layer of insulation.
So in failure mode, it should not pass on a high voltage to secondary.
But if you look at the transformers, the windings can burn or flash over, the wires are thin, with a thin layer of insulation.
My body doesn’t help me shield it I think. Correct me if I’m wrong, but the noise decreases **only** when I touch both the ground as well as any metallic part of the guitar. The noise exists even without the guitar plugged in (Although that’s normal behavior I hear) but again the noise disappears when I touch the ground pole and the floor.
Btw how would I go about testing if the issue is AF or RF?
Try using a combination of 100 / 470 / 10 uF in parallel with a 100 pF as a noise filter, the circuit will look like a ladder... capacitors across rails.
Pretty cool. I’d love to try it out right now, but alas, COVID will be the bane of all that constitutes my happiness