Tips for minimizing noise

[Heavenamps]

I'm getting alot of buzzing/electrical noise in the output signal of an amp I'm building. I have an O-scope and a signal generator and from using the two, I'm fairly certain that the noise is being generated by the 6.3VAC filament supply. I've twisted the wires together to try and minimize the noise, and have tried to keep them as far away from the signal wire as possible but I still get noise. I am able to get a guitar signal through to the output but it's drowned out by noise. Does anyone have any advice on how to minimize noise and interference? I was thinking of trying to use shielded cable. Thanks!

[92Volts]

How is your power supply designed? Does the amp have multiple gain/volume/MV knobs you can adjust to determine how early in the circuit noise is being picked up?

Especially in a high-gain amp, it can be surprisingly hard to get the power supply right. Remember, in single-ended gain stages (i.e. everything in a normal preamp) power supply ripple is directly fed to the next stage as an input signal. That's the reason for multiple stages of capacitor-resistor-capacitor filters in the PSU as you head towards the input of the amp. Increasing the resistors between the stages of the power supply is probably one of the cheapest ways to improve this and to at least test whether that seems to be an issue.

[darkbluemurder]

How did you reference the heaters to ground (center tap on the transformer, 100R resistors or hum balance pot)? If the ground reference is missing, you will get a lot of hum from the heaters.


Cheers
Stephan

[Chipper]

 That much noise sounds like a grounding issue,  check for ground loops and that all grounds are “solid”.

[darkbluemurder]

And I forgot to ask: can you turn the noise down with the volume control(s)?

[Heavenamps]

My power supply is pretty standard: 25uF caps between various resistors. The amp has controls for volume, bass, mid, treble, and a master. If I turn the master pot the noise changes frequency a bit. None of the other pots affect it. I am starting to think it is in fact a grounding issue. Forgive my transformer ignorance here but I'm not sure I wired the PT correctly. I'm using a classictone 40-18050 PT:

http://www.classictone.net/40-18050.pdf

I tied the black and white leads on the primary together as well as the black/white and black/red and then wired each to the AC power input. I tied the green 6.3VAC lines to each lug of the pilot light and from there to each of the filaments. I noticed when I measure AC voltage across the filament taps (for instance across each lug of the pilot light) it measures 6.6VAC but if I measure from one tap to ground I get something like 140VAC! That's definitely seems like a problem to me.

[shooter]

sounds like you have a wiring problem, maybe a shock hazard, got a schematic n pics?

[sluckey]

but if I measure from one tap to ground I get something like 140VAC! That's definitely seems like a problem to me.

Your heater winding is not referenced to ground so that 140vac reading is bogus. Get two 100Ω 1/2 watt resistors. Connect one resistor from a lug of the pilot light to chassis. Connect the other resistor to the other lug of the pilot light. This will take care of some if not all of your hum problem.

[Heavenamps]

Thanks Sluckey, I'm going to try that! Just so I'm understanding correctly both resistors will be connected to ground from both lugs?

[shooter]

like this

[sluckey]

Thanks Sluckey, I'm going to try that! Just so I'm understanding correctly both resistors will be connected to ground from both lugs?

correct

[Meteorman]

And you are creating, mr heavenamp sir, what is known as an artificial centre tap.  I learned about it’s value on my 5e3 build. Look at the Hoffman 5e3 schematic/layout on this website.
Hope that solves your buzz!!

[tubeswell]

Recommended reading


http://www.valvewizard.co.uk/heater.html

[Heavenamps]

The grounding resistors did the trick! Thanks everyone! I've still got some minor noise from parasitics (I think) but I should be able to track it down with my O-scope. Thanks again Sluckey! The amps sounds fantastic!

[Heavenamps]

In my pursuit of tracking down noise I found another interesting quirk. I ran a 500Hz 1Vpp signal through the amp. I noticed that when I turn the volume up high I could audibly hear that 500 Hz signal somewhere in the actual circuitry. I believe it's coming from the output transformer. I had this happen before and the problem was a result of switching the primary leads on the OT to the output tube plates. I triple checked the leads this time and found them to be in the right place. Have any of you heard of this before?

[Ed_Chambley]

In my pursuit of tracking down noise I found another interesting quirk. I ran a 500Hz 1Vpp signal through the amp. I noticed that when I turn the volume up high I could audibly hear that 500 Hz signal somewhere in the actual circuitry. I believe it's coming from the output transformer. I had this happen before and the problem was a result of switching the primary leads on the OT to the output tube plates. I triple checked the leads this time and found them to be in the right place. Have any of you heard of this before?

Does the sound change with volume and tone stack?


Sure, I have had ringing transformers, but you probably have RF noise or microwave.  Maybe even a portable phone or even a lamp.  Try covering the chassis with shielding and see if it changes or disappears.  You will shield the chassis when complete anyway.  Personally, I use sheet copper, but aluminum works fine.


Use a humbucker guitar!

[shooter]

. I ran a 500Hz 1Vpp

Do you have a sine wave or square wave at the speaker?
run your scope out to high kHz, maybe MHz while you monitor the speaker, (output of the OT) if you're using a DL.  look for a coherent signal.

[PRR]

Caps and transformers "sing". More in upper midrange; try it.

If there's a speaker on the same signal, you'll never hear it.

[sluckey]

I ran a 500Hz 1Vpp signal through the amp.

That's pretty extreme. Set the amplitude for about 0.2V_pp

It's not unusual for an OT to vibrate at the frequency of the signal when driven hard.