Chasing down rattle in 835

[jjmt]

Hi. My 835 is back on the bench (see long thread in 50's section). I was prepared to call it done. After installing the Weber speaker and breaking it in for a few hours, the amp played beautifully. Very clean, of course, but it has a bell-like clarity (especially once you get the volume up a bit) that I really like. Particularly with my Parker, with which I had no trouble finding some very nice sounds. Not quite so much with my strat. In short, this amp seems to match up with my guitars very much like my R-12-R. So, I like it.

I was going to have a bass player friend over this weekend to really give it a go. But alas, it isn't ready.

Here's the thing: I've noticed that it only sounds good it's good and hot. For the first 15-30 minutes it has a very annoying rattle. Not a cabinet rattle; this comes out of the speaker. I assumes that it was a tube, so I ran through a rather large sample of 6SL7s and 6SN7s (many of which actually were microphonic ) but the problem persisted. I even swapped out the rectifier, just because. Finally, with great reluctance, I matched up and installed another pair of EL37s, but the problem remained. (Whew! I really didn't want to give up on those super-expensive power tubes.)

So, now I'm looking for suggestions. I have *no idea* where the problem is. I've pulled the lower chassis out of the cabinet and went banging around with a chopstick, but there was nothing that made any real noise and everything seemed OK. Tomorrow I'll pull the upper chassis and see if anything there is suspect.

My plan goes like this:

1. Pull both chassis and try to identify a microphonic resistor or capacitor with a chopstick.

2. Reflow all the solder joints (sigh).

3. See if noise is identifiable using a tone generator. If it is, scope it. (It would be a lot harder to scope a signal that I have to generate with a guitar.

4. Um, I don't have a 4th idea. Looking for suggestions.

OK, actually I do have a 4th idea, and that is to buy a couple OC3 tubes. It's the only tube I haven't replaced. I *think* I have one *somewhere*, but so far haven't found it. My Hickok 600A doesn't have a setting for OC3, so I can' test it. At least not that way. Is there a way to test it in amp? I only vaguely know what it does.

Anyway, I know that there are people here with way more experience that me, so toss some ideas my way? Thanks!

[jjmt]

OK folks, I really need some help. I've deox'd and cleaned the the tube sockets, and reflowed the vast majority of the solder joints.

I actually thought that I had nailed it when I discovered that the metal plate on the preamp chassis seen in this picture was loose and rattling. I had actually installed it incorrectly. I had also foolishly removed the brown sticky stuff from the edges. So I installed it correctly with a bead of hot glue for good measure. It's on solidly now.

But the noise remains. I need your ears. What does this sound like to you? the noise is good and evident starting around second 5 (and continuing after that).

[jjmt]

After swapping around about a dozen tubes (various 6SL7s and 6SN7s), I've isolated the issue to the first triode of channel 1. Doesn't seem to matter what tube I put there. With the tubes in it currently, channel 2 is clear. Soooo, back to the bench it goes.

[jjmt]

I tore the amp apart (getting really tired of doing that) with an eye on that first gain stage. The components all tested fine. The voltages on the plate and cathode were more or less identical to the other triode of that tube. There was no DC on the pot. Basically, it came down to four components: plate and cathode resistors, and cathode bypass and plate coupling caps. I just swapped them all out, as there was nothing more that I could easily do to identify the issue. I've put the amp back together (again, getting really tired of doing that) and the noise is gone.

Slider - thanks for the input. I already was thinking about the coupling cap (but it *seemed* OK). If I put everthing back together and the noise was still there, it left only the jack and the tube socket. When I cleaned the sockets I did not retention them. I know that I should, but that is an exercise that always makes me nervous. I've destroyed at least on socket by pushing a socket connector a bit too far. I need to collect some old sockets to practice on.

Anyway, I'm done with it for today. Tomorrow I'll give it a good run. I've also invited a bass player friend over, but we'll see if he has time this weekend.

[jjmt]

Glad you fixed it! The first thing I do, to a vintage amp, is replace the electrolytics; if they're original to the amp. They can cause; a voltage drop, popping, hissing, motor boating, etc. I was restoring a Gemini I, a few months back, and the first section of the cap can was 100v lower than the schematic. I disconnected that section and clipped in a temporary 40uf@500v and the voltage came up to spec.

In this case, all the components I replaced I had just replaced a few weeks ago. All tested as they should, and all voltages were correct. But I thought that I had isolated the issue to that part of the circuit, but it turns out this issue is more devious than I had originally thought.

---

Well, as is so often the case, I spoke too soon. The amp sounded great for about 10 minutes this morning, and then the exact same noise was back. On both channels. It seems like the issue materialized either after time (it was 10-15 minutes) or volume (I was continually going just a little bit louder about every minute).

It seems to be worse with the bridge pickup over the neck pickup. That could be the result of output or voicing. Right now, I am thinking parasitic oscillation. I've mentioned a few times that the first tube does not have grid-stoppers. So that is definitely something I will add the next time I crack the amp open.

But I'm shooting in the dark right now, so I guess I'll just think about it and see if I can't come up with some way to confirm a diagnosis.

[jjmt]

I spent some more time with the amp yesterday. I've also spent some time thinking about it. I really should do both of those more before deciding to fix something.

Listening to the clip that I posted, I'd have to say that the noise sounds like a resistor (or capacitor) sounds when it passes current unexpectedly. Think failing plate resistor. Except that the noise only happens when playing. The amp is really quire quiet when there is no input signal.

From my session yesterday, I couldn't get the noise to happen with channel 2. But I'm pretty sure that I'm not crazy, so I guess channel 2 is just more resistant.

But my time with channel 1 provided some interesting data. First, there is no question that the noise originates prior to the volume control. The noise can be turned down when it happens, and the signal to noise ratio remains the same. However, the noise is very much dependent on the volume pot on the guitar. Turn up that pot and at some point the noise happens. This explains why I hadn't noticed it earlier, as I tend to just set the volume on the amp around 5 and set the actual room volume using the guitar. In those sessions where the noise did not happen, I just never rolled the guitar up high enough. Working the two volume knobs (guitar and amp) has shown that it is the input signal, rather than the volume, that causes the noise. So, whereas I orginally thought of this as a "rattle", it's really a distortion.

Right now I have two theories:

1. The load resistor is failing. My understanding of load resistors is that they manage the current in the input circuit. Rolling up the volume on the guitar is going to increase the input current, so a failing resistor could cause current leakage wight at the grid, which would certainly be noisy. Replacing the load resistor at the input jack would resolve this issue.

2. I still haven't discounted the possibility of oscillation at the first amplification stage. I know that there is radio signal entering the amp from the pickups/lead (I can hear it). The HF signal could be driving the first tube into oscillation, which would significantly increase it's current draw and would cause it to distort. A grid stopper would resolve this issue.

But in recognition of the fact that there are people that are much smarter than I am, I also am reminded that slider asked about retensioning the tube sockets. A loose socket would cause arcing at the tube pin as current increases, which retensioning would resolve.

So, those three things give me something to work with. I'll report back.

[jjmt]

OK. I think this is resolved. At least, I hope so.

Thursday evening I cracked the amp open and replaced the load resistors on the input jacks (pain in the butt - all wires and resistors wrapped around the lugs in opposite directions - took ages to de-solder and pull apart). I also installed stoppers on the grids of the first 6SL7. I used 22k, which was one of the options i had, based on spare parts. The end result was absolutely no change. Noise reappeared almost immediately. I can say that the grid stoppers did reduce the amount of radio I was hearing, so that is a good thing, I guess.

Running out of options, I again rummaged through my spare parts and went back into the amp and replaced the volume pots and the input jack for channel 1. One oddity of the original input jack was that it did not ground the input when not in use (very noisy when nothing was plugged in). In fact, that part of the jack had been removed. Very odd. The new jack grounds the input. I did that through a 5.6M resistor, the same as the jack on channel 2. The ground issue is immaterial to my problem, I think, but it is possible that the jack itself was not making a firm connection to the guitar cable plug. There is no evidence of this, though. Jiggling the cable did not make any noise.

The pots also tested fine. But they were ancient, and there is the possibility that there was some residual deoxit or something left over from cleaning.

So, the amp appears to work now as expected. The problem could have been with the jack or the volume pot. It could also have been with one of the solder joints on the board, as I simplified my life by wiring the pots prior to installation, so all the connections to the board were redone. I guess it could also have been a wire.

It's been a very Dr. House kind of process. Try this. Try this other thing. Maybe try this next thing. Eventually the patient is saved. Not super satisfying, but effective in the end.