[breez]

Hi

I have built the basic cmoy amp to a simple stripboard and there's a problem with noise. Constant hiss even with no source connected. With source connected the noise gets a notch louder (source = Chaintech AV-710 soundcard, with headphones straight to soundcard it's dead quiet). As a power supply I use single 9V battery or 12V wallwart (no difference in noise level). Also without source and with wall power there was awful noise (sounds like 50Hz crap from wall).
With headphone load and music playing the power rails were 4.48/-4.64 (battery power).
5mV DC at outputs with NE5532 (lower with other opamps).

Some info:

R2 = 100K
R3 = 130K
R4 = 470K
R5 = 0-200
Pot = 10K log (pot position makes no difference in noise)

Opamps: TL082, NE5532, NJM/JRC 4556

Input caps are 0.47uF film caps. Power supply caps are 330uF 16V electrolytics.

I'm clueless

edit: Headphones are philips hp800 (32ohm)

 

[meat01]

I am thinking it has to do with R5. Tangent's guide suggests using a resistor between 47-100 ohms with low impedance cans if you are getting hiss.

From Tangent's guide:

Quote:

here are four jumpers in the amp section, two of which are in place of the R5 resistor which I almost never use. R5 is there to quiet low-level hiss that you can hear sometimes with some low-impedance headphones. If you disconnect your source and you hear a low-level hiss with the volume control set for normal listening volumes, you can try adding a 47 to 100 ohm resistor here. It's not for quieting loud hissing noises — that's probably oscillation or some other instability problem, and adding R5 probably won't cure it. Don't add R5 if you don't have a good reason for it, though, since a higher output impedance also means the amp doesn't have as much control over the headphone drivers.

Tangent's page

 

[bpribadi]

From your description, looks like you have problem with AC hum. The solution is shielding.

I ever got experience with:
1. Hissing sound. The solution was pot, to reduce the source hiss. The source hiss probably not audible when connected directly without amp, but became audible when amplified.
2. AC Hum (50 or 60 Hz). Shielding is the solution. Use metal casing, and connect the casing to the audio ground. Also use shielded coaxial cable from source to amp interconnection.

 

[breez]

meat01:
Yes, I tried 51, 100 and 200 ohm resistors, but no effect :/ I noticed that the R5 is in the feedback loop and someone said it has more of an effect when it is placed outside. I've also eliminated the possibility for oscillation, because the amp takes only 9.5mA with NJM4556 or only 4mA with TL082.

bpribadi:

1. I now noticed that with the source connected, the source's noise was indeed attenuated away with the pot (previous testing with pot position was without source, doh). Still the "base" noise level is there unchanged (same level with no source). There must be something fundamentally wrong in my amp because of the noise without source

2. I realize that solution, but in my case it isn't really necessary, because whenever I use wallpower, the amp will be connected to a source and grounded well at the same time (signal ground = my computer's ground). The AC hum without source was just an observation.

I read about adding power supply bypassing caps in tangent's "Working with Cranky Op-Amps". I have no clue whether it has any relation to this problem, but I guess I should try. Later that is. 4:40 AM here...

 

[Earwax]

One more thing to double check is the virtual ground connections to the input and output jacks and to the pot.

Do you get the same amount of noise on both left and right channels?

 

[breez]

Amp noise is gone now! Just used smaller resistors (27K, 39K, 100K). Also powersupply bypass caps were added and input caps changed to 1uF. Biggest effect however was with the resistors.

Still some little, just audible hiss comes from the source. Too bad, but I think I can live with that slight noise and I noticed that it didn't distract me at all in quiet passages of music. For some reason the noise doesn't change at all if I change the source's volume which is a little weird I think. I'd gladly turn the source volume up to have better SNR, but then it rolls off the bass and even adds some distortion at very high settings (main reasons for building this amp).

 

[breez]

Doh! New problem... The bass rolls off when the pot is attenuating the signal! Not just a fuzzy feeling because I also measured with RMAA and the effect is very clear

Same soundcard master volume position, just played with the -0.1dB to -4.0dB calibration RMAA allows.

I read at some headwize thread that R2 should be 2-10x the pot resistance. OK I replaced 27K R2 to 47K (not quite 10x but should be better?) and all I got was more DC offset at outputs (<14mV this time) because of inbalanced inputs and the bass still rolls off (it actually worsened it a bit).

Damn troubleshooting hell

 

[breez]

I had miswired the pot!! Now it works very well, no noise with the bigger resistor values too

I can now also turn the source volume to max to get the best snr (gain = 2)

 

[bpribadi]

Good to hear that