Fender Vibrolux Reverb motorboating noise

[goldtop5]

I have a Fender Vibrolux Reverb that's making motorboating noises. I ran a scope trace. When I got to the phase inverter section of the circuit, I didn't get a full sine wave. I replaced any resistor that was reading out of tolerance, but still it seems there is distorted movement at the top slopes of the sine wave. The signal seems to go back to normal when I pull the phase inverter tube and/or the power tubes out. Any ideas of what I'm overlooking? Thanks!

[xtian]

Motorboating is a common symptom of failing filter caps. You can clip in a new one, in parallel with existing caps, to see if that solves the problem. This is a common approach to troubleshooting Hot Rod Deluxe amps.

[stokes]

Filter caps.I've had an unusual number of newer Fenders with this issue over the years.

[Justin Thomas]

Good question - is this a vintage, or one of the "reissues?"

Justin

[goldtop5]

This is a vintage one. Cosmetically, this amp is in really good shape despite its age. I've replaced all the filter caps, the screen and control grid resistors, and tried out new tubes. I even tried swapping the NP caps. I can't seem to make the noise go away. However, I haven't looked at the disk caps yet.

[pdf64]

A scope trace taken inside a NFB loop can be confusing; the loop is trying to correct errors, so it's likely to look weird.
And I've seen reports where it's postulated that some of those original wimpy OTs cause so much phase shift that the loop can be positive at high frequencies, causing parasitic nasties to the wave form.
I suggest that the loop is opened, eg link out the 100 ohm resistor, and then see what you get.
A voltage survey at idle is a good idea too, ie Vdc / Vac (as appropriate) at all tube electrodes and power supply nodes.

[goldtop5]

I got the problem solved. It took a while to realize it was something simple. There was a disconnected chassis ground going to one of the filter caps (25uf, 25v) on the preamp. I resoldered it and the problem went away. Things like that are the last thing I'd suspect because I never see broken chassis soldered connections.

[Justin Thomas]

I reckon I the ground is disconnected, the filter cap may as well not be there... I got a 1966 Princeton Reverb one time, if you so much as turn the Reverb a hair over zero, it screamed like mad. The ground from a cap in the Reverb had popped right out of its solder blob and dangled in midair. This amp had had absolutely nothing done to it except maybe tube changes (it came with a couple Mullards). It happens.

Justin

[Leo_Gnardo]

There was a disconnected chassis ground going to one of the filter caps (25uf, 25v) on the preamp. I resoldered it and the problem went away.

That's a cathode bypass cap, not a supply filter. If two triodes are sharing a common cathode connection, removal or malfunction of that cap will allow signal to pass from one tube to the other via the cathodes, resulting in noises you don't want to hear.

Now you know to check all the connections on the board side of the chassis too. Everything counts!

[Justin Thomas]

Dagnabbit, I made the same mistake! Read it and didn't check the numbers, only the words! But, it IS a filter of some sort, right?

:P

Justin

[Leo_Gnardo]

Dagnabbit, I made the same mistake! Read it and didn't check the numbers, only the words! But, it IS a filter of some sort, right?

As it's a resistor and capacitor in parallel, you can look at it as a filter. Applied to a single triode, all you need is a resistor. By paralleling a cap, you're defeating the tube's propensity to limit its own gain - in the AC realm. The fancy word is antidegeneration. So you can give the gain a wide band boost with a large value cap, or from some mid or hi frequency point upwards, with a smaller cap. The typical 1K5 paralleled with 25 uF gives us a wideband boost with -3dB point at about 4 Hz, hi fi enough? All I can think of, when Fender and other brands parallel cathode R/C it's a bid to save a little cash. You'll never do any harm by separating these combination setups and providing each cathode with its own R and C. The only exception I can think of is if you're paralleling ALL triode elements to try and minimize tube noise, as Rivera and a couple others do, then there's no point in separating the cathodes is there?

[g1]

Sounds like it was one of the wires that goes to chassis from the eyelet board?
You may not have seen any broken before but it's not that uncommon in this style of vintage Fender. I try to always check them and the solder from the brass plate to chassis (the plate under the pots).

[goldtop5]

Yes it was. I guess I've been lucky enough to not find a broken connection like that in almost 10 years!

[52 Bill]

It's not all that common, but you do see it every now and then.

I find it most often when the fiberboard has been subjected to a lot of moisture and has swelled and warped a lot. The board can rise enough to pull the ground wires right out of the solder.