[mminutel]

I recently built a Pimeta and didn't think there was a problem until I was listening to a soft song. The right channel has a weird noise. I cannot describe it. It is not a hiss or a crackle, it just sounds almost like the buzz of electricity (if that makes sense). It is only in the right channel.

I have single buffers. AD8610/20 as OPAmps. Class A biasing.

The amps do not seem like they are oscillating. The 8620 seems a little hot but I could hold my finger on it for as long as I want so it isn't really hot like my other amp got when oscillating. There are no solder bridges. I have also checked with a multimeter. Each channel has around 20mV of offset. The only thing that has a significant voltage difference is the Left Buffer which is a difference of around .2V but it is in the right channel.

I have checked about everything that I can think of. No pics right now but I cannot see any solder bridges even under magnifying glass.

Has anyone had this problem before and fixed it? I will keep tinkering with it but I have checked everything I can possibly think of. I have checked resistor values to make sure they are the same and everything. Also, what is the main cause of offset? I get around 21mV which is higher than I would like but I don't know how to minimize it. The last thing is that when I turn off the amp, there is a "pop". It isn't like it hurts my ears or anything but I measure around 1V when I turn it off but it slowly depletes.

Thanks all. Like said, I have tried everything I can think of just looking for further advice.

 

[achina]

Check your output jack wiring. I once had L-L R-G G-R. It played but sounded weird.

 

[mono]

Are you using an unregulated power supply? If so, try regulated or batteries and see if the noise persists. If using the computer sound, maybe it's the source or connection, try another.

 

[mminutel]

Ha. I had actually never thought about testing my sound card. I had not heard the noise until I built the Pimeta so I didn't even think that would be the problem. Anyways, shortly after the post I was afraid of blowing my amp so I just plugged my phones into an extension cable to watch a movie and I got the same thing. So, that would be it. Thanks anyway guys. Sorry for the bad post.

Anyways, I am still having the problem of the offset jumping pretty high at power off. Any suggestions?

 

[achina]

Time to build Alien DAC!

 

[mono]

Quote:

Anyways, I am still having the problem of the offset jumping pretty high at power off. Any suggestions?

Easiest suggestion(s) might be using a dual pole, dual throw switch for power. Wire amp V+ rail to center lug, PSU V+ rail to top lug, and a low ohm resistor to ground on the bottom lug. Thus, when you flip the power switch off the resistor drains the caps quite quickly (depending on resistor value chosen). Since the total power and especially duration is short, you wouldn't even need a very large wattage resistor.

Another option is to use a multimeter and carefully monitor which rail is draining slower with a goal of switching which rail the LED is on or slightly changing current drain of the LED or adding a high KOhm resistor to the slower draining rail till they come close to even drain rate.

You could use a relay on the output instead, but it's a bit more involved to tack onto an existing amp rather than planning ahead for it in a design.

 

[mminutel]

Thanks. I am not exactly sure why would would need a dual pole switch though. I don't think I am fully understanding it. Couldn't you just use a SPST switch?

 

[mono]

Dual pole because the common pole is the amp capacitor bank, one flip of switch connects that to your power supply and flipping the switch the other way connets the capacitor bank to the resistor-ground to quickly drain it. However, on second thought this may have problems when using a railsplitter for virtual ground, with it's limited current capability. No you can't use a SPST.