Vintage Modern 2266C - blown bias resistor

[gswearengin]

Howdy folks,

This is my first post in this section. I’m just hoping to get some insight from people who know more about this stuff then I do (not difficult!).

I recently purchased a VM 2266C that had been pretty lightly used and sat for most of the last 7-10 years. It’s in fantastic condition.

When I tried to set the bias, I was getting significantly different readings for each side (30/50). One of the power tubes felt like it had come separated from the base, so I assumed that was the issue and ordered some new KT66’s.

When they arrived, I randomly stumbled upon this thread when searching to verify what I should set the bias to. After opening up my amp, I realized that I have the same problem:



R83 is toast. I’m assuming this is what’s causing the problem in my case, as well.

I quickly ordered a couple replacement resistors, but while I wait for them to arrive, I have a few questions:

1.) Can I do this myself? I’ve built pedals and an amp (and know basic safety precautions), but I don’t think I’ve ever worked on an existing circuit board before…

2.) I’m assuming it makes sense to remove the daughter board to work on it? The resistor is tucked away pretty well under the chassis. Lots of cables to remove/mark/remember to do so!

3.) In the meantime, is it safe to use the amp? Any ideas of what I can expect after I replace the resistor and hopefully get the bias set correctly? And any ideas what May have caused this in the first place?

Any input appreciated.

Thanks!

Gary

 

[TheKman76]

Almost certainly a design flaw and definitely the cause of your bias issue. This position in a combo is exposed to quite a bit of heat to begin with and under high load this resistor will see power close to it's rating.

The bias measurement resistors should be rated for at least 1W just to be reliable, even 2-3W to provide reasonable temperature stability. It should also be mounted off the board to give it better convection and radiation. I realise you've ordered parts already, but if you want to fix it right I'd suggest a couple of 3W 1% tolerance resistors here and replace R83 & R84.

In answer to your questions:
1) Yes, if you have some soldering experience. Re-work is a slightly different set of skills and you'll need either a solder sucker (hate them) or some de-soldering braid to clean up the through holes after removing the existing resistors.
2) I would, yes. Take lots of photos before you start and take your time.
3) Probably. If you have a multimeter you should be able to measure the resistor in circuit and I expect it will read higher that 1 Ohm. If this is the case, set your bias according to the lower of the two measurements you take for bias current. V = I * R, so the higher resistance will show a higher voltage at the bias test points, which will be false in this case. This resistor will effect operation very little, until it goes open circuit, then it will be obvious as one valve isn't working at all.

I hope someone with specific knowledge of this model can chime in, I've not seen one in the flesh.

 

[Jon Snell]

This is undoubtably caused by a KT66/EL34/6L6 failure, (whichever is/was fitted). KT66 is the recomended valve but beware of Chinesium poor quality types.
There is good reason to use the 1W fire proof resistor and if you fit other types, will quickly learn, that it can wrote off the board next time it fails!
Replace the output valves, replace the 1R 1W fire proof resistors, check the bias voltage to rebias the new output valves, (KT66s) and it will give you years of sterling service.

Edit; More info added.
Set the bias to 46mA with a new pair of matched KT66s.

 

[TheKman76]

This is undoubtably caused by a KT66/EL34/6L6 failure

What failure mode would cause this?

I ask out of ignorance, I'm having trouble imagining it.

 

[gswearengin]

Thanks y’all.

The amp has the stock Marshall-branded KT66’s. They supposedly have low usage, but I just got the amp so I can’t say for sure. I initially panicked because one of them had come loose from the base, but it really doesn’t seem as bad as I initially thought. I did order some replacement KT66’s, so I’m happy to use those instead.

Yes, I’ve already ordered replacement resistors. I did some quick searching, general recommendation seemed to be that using 1/2 watt was a good idea, as they act almost as a fuse for the rest of the amp. I ended up ordering what looked to most similar to what was in there (in the right value, of course), without a whole lot of extra information to go on.

I can definitely see @TheKman76 ’s point that it’s a potential design flaw as it’s located in a pretty precarious area and, based on my small sample of 2, seems to occur with R83 on the combos. I haven’t done an exhaustive search, but I haven’t seen any mention of R84 failing, which is in a different location, further away from the chassis.

If this gets it up and running and biasing properly but it then happens again, I’m happy to repeat the process with beefier resistors.

Thank you,

Gary

 

[Spanngitter]

When repalcing that R I would cut it off directly at the component, bend the remaining lead up and tag solder a replacement to these lead stumps.
This saves you having to rip out the PCB what you have to do anyway as Marshall bends the leads on the solder side and therefore you cannot remove the component from above without a high change of damaging PCB.

 

[gswearengin]

Interesting idea (thanks for the suggestion!). That would make it quick and easy and also get the component floating over the PCB which might help with heat dissipation.

I may need to remove the daughter board anyway, since it’s buried in a pretty hard to access area, but I will definitely think about this and maybe start with this idea.

 

[Ken Underwood]

When repalcing that R I would cut it off directly at the component, bend the remaining lead up and tag solder a replacement to these lead stumps.
This saves you having to rip out the PCB what you have to do anyway as Marshall bends the leads on the solder side and therefore you cannot remove the component from above without a high change of damaging PCB.

Thats a bodge, do it properly

 

[TheKman76]

Thats a bodge, do it properly

It can work just fine, but, I agree. Do it right and there's no reason to revisit this.

I've done stuff like this in situations where there's no real power or heat or vibration... All those things are present here.

 

[Ken Underwood]

When repairing any electronic equipment, then no matter what type it is, then do it once and you will not visit it again, always keep a high standard in all what you do, trust me, been doing it for a very long time.

 

[FleshOnGear]

What failure mode would cause this?

I ask out of ignorance, I'm having trouble imagining it.

Probably something that causes the tube to lose its bias, like the grid shorting to the cathode.

 

[TheKman76]

@gswearengin mentioned physical damage to a tube. Sounds plausible.

 

[gswearengin]

I *think* the tube is okay. Apparently this is somewhat common with KT66s. I’ve seen several mentions of people basically gluing them back together.

 

[TheKman76]

Can you tell us if it's the same part of the circuit?

 

[Jon Snell]

What failure mode would cause this?

I ask out of ignorance, I'm having trouble imagining it.

Internal flashover within a valve, very common with worn valves.
The screen grid bends when over heated and can touch the suppressor grid sometimes the control grid causing catastrophic damage to the valve, it then flashes over within the envelope, especially when the valve is old or has had a hard life. Usually they still perform after repairs are made to the amplifier but is unreliable and likely to repeat the failure. That in turn takes out the weakest link; the cathode resistor.
The suppressor grid sags when stressed and can touch the anode. The suppressor grid is connected to the cathode either internally, in a KT66 or externally in an EL34. No more explanation is required.

 

[gswearengin]

Can you tell us if it's the same part of the circuit?

Good point/question. I’m going from memory here and the tube has since been removed, but I’m fairly confident that the loose tube is on the opposite side from the blown resistor.

This amp has supposedly led a pretty tame life. That’s based on statements of the previous/original owner, but it certainly has the looks to prove it (with one obvious exception, see above…)

 

[PelliX]

I *think* the tube is okay. Apparently this is somewhat common with KT66s. I’ve seen several mentions of people basically gluing them back together.

Possibly, but it's hard to say without vigorous testing. The plastic 'base' can come loose, that's not an issue. If the pins move in relationship to the actual glass bottle - then you should look no further and chuck it (or use it as an ornament). Check the getter flash for any signs of leakage.

 

[gswearengin]

Yup, the plastic base has come loose. Pretty sure that’s it. I freaked when I first noticed (and ordered new tubes), but it doesn’t seem as bad as I initially thought, and I have found another culprit for the bias mismatch.

 

[TheKman76]

@Jon Snell Thank you for the detailed explanation!

So, it sounds like the consensus is new output valves and new cathode resistors and bed rest. The doctors will check in tomorrow morning.

 

[gswearengin]

OP checking in here. Good news and bad news...

I finally had a chance to change out the blown resistor last night. Man, that was a PITA...so much stuff to unplug and move around. I would say it went pretty well. This is what I ended up with (elevated, to hopefully deal with the heat better):



Look okay to y'all? Anyway, I was super excited to put the power tubes back in and set the bias.

Aaaaaand...same(ish) problem! When I set one side to 48mV, the other is around 36mV.

But, if I switch the tubes the problem moves to the other side. That sounds like mismatched tubes to me. Agree?

I just wanted to check with y'all before ordering yet another set of power tubes.

(These were sold to me as new and unused, and the specs certainly matched, but who knows. Sounds like I may have been duped, hopefully not intentionally. I also tried the original tubes again, and it was closer but still not perfect. But one tube is so loose that I really don't trust it and don't want to use it anymore.)

So, new power tubes? Anything else to look into?

Thanks again to everyone for your help!

Gary