Pop/crackle in one channel

[Saurav]

Hi,

My amp started doing this today, after I dusted it. Mental note - never dust amp again Anyway, it seems like it only happens after the amp's heated up for a while. Mostly it's random pops in one channel, one time it turned into this rushing/crackling sound. I've swapped external cabling and it stayed with the channel. I'll try swapping tubes too to make sure that it's in the electronics, but I suspect it is.

I took a look through the archives. Assuming it's not the tubes (which would be easy to fix/replace) - this sounds like a bad solder connection or a resistor dying, right? The amp's been working fine for several months now with no modifications to it, so could that still be a solder joint? Can a solder joint fail after several months of just sitting in one place and playing music? Is there any way to find out where the problem might be? I guess I could reflow every solder joint for that channel. Is there any way to identify a resistor that might be the culprit? Could it be something else entirely?

One more thing I noticed - if I turn the amp off and then turn it back again within a few seconds, the 5AR4 flashes white for an instant. I don't remember seeing it do that before, and I *think* I've turned it off and on again before. Is that bad for an amp? The last time I had any problem was this, which was several months ago and it disappeared on its own.

Thanks in advance,

Saurav

 

[Geek]

Anyway, it seems like it only happens after the amp's heated up for a while.

There's your key right there

Get yourself a $3 can of freeze-spray from Rat Shack and hit capacitors and resistors until it stops.

To narrow it down, is it the same volume, regardless of volume control? If so, it's after your volume control (assuming you have an integrated amp).

 

[Saurav]

Good idea. I'd heard of that before, but forgot about it. It's not an integrated amp.

 

[SHiFTY]

I would try wiggling the tubes around, i bet you just have a loose tube. 9 pins are esp bad for this, octals never seem to get loose.

Also turning your amp off and on again quickly will destroy the rectifier and output tubes, because there is a massive current surge (no bias and full b+). Wait for it to cool a little.

 

[Sch3mat1c]

Don't hit tubes with it

I had an old tired 6GH8A, probably removed from a TV, which would do that in Frankenhouse. I think I even tried to record it once. Ah, here we go:
http://webpages.charter.net/dawill/Images/AmpHiss.mp3
Kind of a swishing hissing popping sound. Not really the crackling record it sounds like. Changes a bit when the tube is bumped/moved (as if oxidized) but the pins don't look any different from any others...

Keep swapping tubes and if it follows one, replace it.

Oh, and don't overlook the obvious - the contacts might just be on a spot of crud. Wiggle and reseat the tubes.

Tim

 

[Saurav]

Well, when I posted the question I'd removed all the tubes, put the amp upside down on the floor and tapped every solder joint/resistor/cap with a pair of pliers. I also moved apart some of the 9-pin socket tabs that looked like they might touch. It still popped after I did that, and the crackling/rushing sound happened after that too.

I'll try swapping the tubes next. Maybe one of my 2A3s is dying. I'm using Sovteks at a pretty high operating point - maybe they're not quite as super-robust as everyone seemed to believe. Though they have lasted well over 1.5 years now I think. I might drop my B+ back down, go back to the higher cathode resistor, and try fancy 2A3s.

Also turning your amp off and on again quickly will destroy the rectifier and output tubes, because there is a massive current surge (no bias and full b+). Wait for it to cool a little.

Thanks. I'd heard that from non-DIY 'audiophile' friends, and I always wondered if there was some basis to that. Makes sense though. My whole reason behind using the 5AR4 was to slow down the initial powering up process.

 

[Sch3mat1c]

About the surge, that's because the capacitor is discharged by the circuits, but all the tubes (including the rectifier) are still hot and able to conduct. This puts full voltage *directly across the rectifier plates* as the capacitor charges in the first cycle or two of new supply voltage, instead of the usual gingerly climb when starting cold. Most tubes are rated for a hot-switching current in the amperes range, but maybe you've exceeded it.

Tim

 

[cogsncogs]

I've had that happen many times in the past. Turned out most of the time to be a bad, loose, or wore out SOCKET. 9 and 7 pin sockets are nortorious for this. Fix: re-tensioning/tightening the pin contacts in the socket with a very small jewelers flat blade screwdriver. If the crackling, popping follows the tube, then the tube's your suspect! If wiggling the tube causes the popping to go away or let's say causes it to go in and out, I'd suspect the socket. And please don't fill you sockets w/s*** like WD40 or some other TWEAK!

Wayne

 

[Burnedfingers]

Wayne has a good point there.

Tighten the sockets and leave the WD40 or other crap in the garage where it belongs.

Note* Make sure the B+ is discharged before sticking objects into sockets.

 

[Saurav]

No WD40 in my sockets I'd never heard of that tweak before.

The problem didn't follow the tubes. I found a grid stopper resistor on the bad channel that seemed like it was close to the lead separating from the body (you know when the lead starts to twist a little too freely right at the end-cap?). Replaced that, in the process found another resistor that looked close to breaking so I replaced those on both channels. It's playing clean now, let's hope that's all it was.

 

[cogsncogs]

Note* Make sure the B+ is discharged before sticking objects into sockets.

Burnedfingers has a good very good point also! You could Burn A Finger or two doing that with juice still in your sockets!

Wayne

 

[cogsncogs]

The grid stoppers to the 2A3's? Good to hear you've solved your problem! I've seen WD40 in places you wouldn't believe

Wayne

 

[Saurav]

To the ECC99 driver stage tubes, actually. I replaced them with some Vishay/Dales I had lying around. So of course now the amp sounds better too And I don't know if I've solved it yet, but it seems better, it ran longer without popping than it has in the last few days.

 

[Saurav]

OK, I didn't solve anything. The problem is still in the same channel, it doesn't follow the tubes (tried moving both the 2A3s and the ECC99s). It happens about 15 minutes after the amp is turned on (usually within one LP side). It'll start with a pop, followed immediately by a crackling sound. The crackle lasts for a few seconds up to a minute. Then it pops again and is silent. This usually repeats after several minutes if I continue listening to music. I've swapped tubes, replaced the resistors around the ECC99 (grid leak and grid stopper), and re-flowed all solder joints in that channel all the way up to where the power supply splits for the two channels.

The intermittent nature of the problem makes debugging a little cumbersome. I hooked up my scope while listening to music today, and found that the noise shows up on the plate of the ECC99 (i.e. at the driver end of the coupling capacitor). I think that means there's a good chance it's something around the ECC99, and the 2A3 is probably fine. That leaves the C4S board, the battery under the cathode, and the tube socket itself. None of the resistors on the C4S board look burnt, and they all measure the correct value when cold. Before I start ripping things out, does anyone have any idea about what the likely candidate could be? Does this silence-pop-crackle-pop-silence pattern sound like anything in particular?

Thanks,
Saurav

 

[Burnedfingers]

I'm not a tube expert like Frank and others here but I would suggest resoldering the tube socket as well as the parts associated with it. Is there a possibility of a flux joint? I would also pull the battery and clean the connections there also. Is there a possibility the coupling cap is letting some DC thru? I would change the coupling cap and see if there is a difference. Is it a PC board amp by chance? If it is I would solder both sides of the board instead of depending on the feed thru's for contact.

 

[Saurav]

> I would suggest resoldering the tube socket as well as the parts associated with it.

Did that.

> Is there a possibility of a flux joint?

Well, my soldering is pretty ugly, so I guess so I went through and made sure none of the pins were touching though. Even bent some pins outwards to give them good clearance from adjacent pins.

> I would also pull the battery and clean the connections there also.

The leads from the battery (it's a cordless phone battery) are soldered in, so no contacts there that could get dirty.

> Is there a possibility the coupling cap is letting some DC thru?

I'm not sure I understand. The only coupling cap is betwen the input/driver tube and the power tube, so the DC is on the side where the problem is. If that doesn't make sense, then I've totally misunderstood your question. Or do you mean the coupling cap is shorting and letting DC go through itself?

> I would change the coupling cap and see if there is a difference.

I'm going to try that.

> Is it a PC board amp by chance?

Nope. The CCS circuits are on PCBs, but that's it.

> If it is I would solder both sides of the board instead of depending on the feed thru's for contact.

I usually do, or pour enough solder in there to ensure that I see solder on both sides of the board.

Thanks for the help. I'll try replacing the coupling cap. If that doesn't fix it, I guess I'll have to replace parts one by one until I find it.

 

[Saurav]

Well, an update - I found a wire that was borderline, it broke while I was removing the capacitor. So I replaced the wire instead, let's see how that works. Hopefully that was it. No more stranded hookup wires for me.