Fender Twin amp strange oscillation/ motorboating.

[jpfamps]

I've got 2 modern Fender Twin amps in the workshop both showing the same symptoms: if you turn up the drive channel and have the fx loop engaged and turned up the amp starts to oscillate at around 16Hz.

This, in my view, is being caused by coupling in the pre-amp via the power rails, ie classic motor boating.

A quick perusal at the schematic shows that 8 (eight!) triode gain stages are supplied from the same node (X), which is obviously not a good design.

If changed the filter caps, and still the problem persists.

Assuming that this isn't an inherent problem with these amps, does anybody have any further suggestions as to how to cure the problem?

I have thought about rewiring the power supply with extra decoupling, however this would not be trival on a PCB-based design such as this, and it could be troublesome to eliminate ground loops.

Thanks in advance for any help.

[Tom Phillips]

I've got 2 modern Fender Twin amps ... the schematic shows that 8 (eight!) triode gain stages are supplied from the same node (X)...

I'm not sure exactly which model you are talking about. Can you post a link to the service information?

[Jazz P Bass]

Is that the "Red Knob" Twin?
Did you have the reverb engaged?
If so, where was it set at?
Try turning it off (or down)

[g1]

More questions for you. Both amps were brought in for this fault? Disengaging the FX loop makes the problem go away? The problem occurs with signal to the input?
If they do both have the same defect and it is happening under real playing conditions, I would think the FX loop affecting it should help pinpoint the problem.

[pdf64]

The pro tube twin amp seems to fit the bill http://support.fender.com/schematics..._schematic.pdf
That cathode bypass cap C1 on the first stage seems overly large at 47uF - is that significant?
The reverb recovery uses 2 cascaded stages with a shared cathode/bypass C34; that seems bad practice also, though it's mitigated by significant LF roll off between the stages. Even so, if that cap develops sufficient ESR, the reverb recovery may be able to oscillate.
Pete.

[jpfamps]

The pro tube twin amp seems to fit the bill http://support.fender.com/schematics..._schematic.pdf
That cathode bypass cap C1 on the first stage seems overly large at 47uF - is that significant?
The reverb recovery uses 2 cascaded stages with a shared cathode/bypass C34; that seems bad practice also, though it's mitigated by significant LF roll off between the stages. Even so, if that cap develops sufficient ESR, the reverb recovery may be able to oscillate.
Pete.

Yes that is the correct schematic.

I've changed C34.

Oscillator occurs even if V1 is removed.

[pdf64]

V2B & V3A are on the 'Y+' node, the phase splitter supply; did you try a known good cap across C206? Post #1 reads as if you did, but just checking!
There's not much LF roll off in that overdrive; C10, C13 and C15 seem too high a value to me. It may go towards explaining why I thought the overdrive on that amp seemed bad when I tried one.
Other thing is tubes - are there some unusually high gain ones in these amps?
Doh, the reverb recovery is the same as regular BF, the way it was drawn confused me. The issue still holds though.
Pete.

[Enzo]

Let's not redesign the amp, since they ALL do not do this, I have to think the amp was once working. We need to find the failure.

DO not let the fact of two of them cloud the troubleshooting. Pick one, and fix it. THEN we can look at the other one. Once you know what killed the first, you can certainly look for that in the other one before continuing, but just find the problem in one first.


Having a lot of stages on one node may look weak, but they usually get away with it due to low current demand and that some of the tubes may not be in service when one channel is selected.

0K, if the loop is turned OFF the amp works OK, right? ONLY has the problem with the loop on? SO plug a patch cord from send to return. Does that stop the boating?

Look at the loop circuit. FIrst, K3A look relay can't possibly be drawn right. But I doubt it is the issue.

Make sure little R56 10 ohms is not open. The loop jacks appear to have chassis spikes, do they? And if so are they making contact with the chassis? The whole loop circuit sems to otherwise receive its ground through the shield of the output cable shield, WJ13. Or posibly WJ9 or 11. Make sure those ground conections have integrity.

[jpfamps]

Let's not redesign the amp, since they ALL do not do this, I have to think the amp was once working. We need to find the failure.

DO not let the fact of two of them cloud the troubleshooting. Pick one, and fix it. THEN we can look at the other one. Once you know what killed the first, you can certainly look for that in the other one before continuing, but just find the problem in one first.


Having a lot of stages on one node may look weak, but they usually get away with it due to low current demand and that some of the tubes may not be in service when one channel is selected.

0K, if the loop is turned OFF the amp works OK, right? ONLY has the problem with the loop on? SO plug a patch cord from send to return. Does that stop the boating?

Look at the loop circuit. FIrst, K3A look relay can't possibly be drawn right. But I doubt it is the issue.

Make sure little R56 10 ohms is not open. The loop jacks appear to have chassis spikes, do they? And if so are they making contact with the chassis? The whole loop circuit sems to otherwise receive its ground through the shield of the output cable shield, WJ13. Or posibly WJ9 or 11. Make sure those ground conections have integrity.

Thanks for the reply.

I've had several of these amps in an not seen the fault, so I assumed it wasn't an inherent feature of the amp, although as I have two in (from the same customer....) showing the same problem I was wodering it is is a known faul

I have tried a patch cord and the problem persists.

I again assumed that this was due to the extra triodes running from the same node on the power supply.

It requires the gain to be turned up to occur, and I'm fairly certain is a problem with coupling through the HT rail, as the oscillation is at 16Hz, not as mains frequency.

[pdf64]

'It requires the gain to be turned up to occur'
Even without V1 in place?

[jpfamps]

'It requires the gain to be turned up to occur'
Even without V1 in place?

Yes, even without V1.

[g1]

I'm thinking along the same lines as Enzo with the loop grounding being suspect. The fact that they are both from the same customer and the loop is being used seems a bit fishy. Maybe a blown ground trace or small value grounding resistor?