[andrzejpw]

Just a quick question on my just finished meta42. Whenever I turn it on or off, I can here it pop through my headphones. Is this DC? It's a bit louder when I turn it ON, but its also there when I turn it off. I don't want to keep on unplugging my headphones: I'm afraid I'll wear out my neutrik jack.

 

[tangent]

Which op-amp are you using? Many chips just have that kind of character. The AD843, for example, is rather noisy when powering up and down. The AD8620 and OPA2132, by contrast, are almost silent.

One tip with noisy op-amps: you'll get less noise if you turn the music on after the amp. If the chip is coming up while there's a signal, noisy chips tend to munch on the music rather nastily.

 

[andrzejpw]

I'm using the 8610. And this doesn't matter if there's music playing or there isn't. Whenever I turn it on, there's a crack, a snap, a pop, whatever you want to call it.

 

[tangent]

Do some measurements. Put a DC meter on it on a millivolt scale and turn the amp on. This will work best with an analog meter, or if you can swing it a digital scope. A DMM won't be fast enough to see a turn-on pop.

 

[tangent]

I decided to follow my own advice: I powered up an AD8620 in a META42 (v1) with a DSO attached and I see a -1V (approx.) pop on startup. This won't hurt anything -- that's about the normal excursion for most headphones.

What I hear is a plainly-audible pop, but it's not loud enough that I'm worried for my hearing or my headphones. If the pop you're hearing is loud enough that you're worried about your hearing, you should still measure your amp to see if it's behaving differently from mine. I tested 4 different chips, so what I'm seeing is the normal behavior, but that doesn't mean that your amp is normal.

For what it's worth, if your pop is the same as mine, even an analog meter won't be able to see it. It's a sharp 1V spike that tapers exponentially towards 0 -- it's nearly at 0 at 200 microseconds from the initial edge. If it were a wave, it would be roughly 2.5 kHz, which is obviously well within the audio range, but still it's pretty darn quick. By contrast, the AD843 turn-off crunch is much longer and goes up and down rapidly as it dies -- it isn't just a single spike that tapers rapidly to 0.

 

[andrzejpw]

Well, I'm not sure, but it doesn't seem as bad now. I don't think its a big problem. I only have a digital multimeter. . . is it safe to say if I hooked up the meter, and there was nothing, it was ok, and if there was a problem, something would appear?

 

[tangent]

The really bad problem would be if you had a constant DC level on the headphones, which will sound pretty much the same, because there's still that sharp vertical, but this time no decay back to 0. Whether your headphones can stand a constant DC level depends on the level and the headphones themselves, but the worst part is that you end up distorting the music coming through, which can be bad for the headphones in their own way.

So yes, do a DC measurement on the amp. If it's over about 20 mV, something's wrong. In fact, with a low-offset chip like the 8610, if it's over about 5 mV, be worried. All of my 8610/20 amps have been in the 1 mV area for DC offset, IIRC. It's a wonderful little chip that way.

It's not that 5 mV is going to hurt anything, but rather that it's abnormal for this chip, so it's worth seeing if there is a real problem or if you've just got an out-of-the-ordinary situation somehow.

 

[Gariver]

Some PS electrolytic caps give off turn-on and turn-off pops. For some reason (unknown to me), Elna Cerafine electrolytic caps do NOT pop at all. They are completely silent! That's the reason I use them! I mean, I don't like hearing those pops through my headphones.

Erix has had the same experience with Elna Cerafines. He also likes the silence of the Elnas Cerafines.