Red plating due to speaker cable touching tube?

SunDevil

Why is it always long weekends where 'stuff' happens?

I have a Cathode biased amp running a factory supplied quad of EL84s (2 pairs of push /pull with a single shared cathode resister as I understand it.) 

A pair of the EL84s red plated yesterday and I noticed my speaker cable was touching one of them (melting a little of the cable sheath) - could the cable contact have caused a bias issue for the pair?

After moving the cable and letting everything cool down for an hour or so, the amp played fine with no red plating (for the 5 min or so I was prepared to risk the tubes, as clearly they are now on their last legs. )New tubes are on order

The only other info I can offer, is that the reason the back of the amp was off, was that I'd bought a set of NOS Ei tubes for it. I'd taken the stock tubes out and had installed one of the NOS tubes, when I noticed that the 2nd NOS tube had bent pins, so I stopped there and then, removed the installed NOS tubes and re-installed the stock tubes in their original positions. I played the amp for 30min or so after this point with no issues

Any help much appreciated!

Baz

ecc83

Several points come to mind...

How "red" did the valves get? If just a dull glow for a few minutes and no HT fuse blew (if it has one!) then they might be quite ok, certainly if they work you can keep them as back ups.

The melted cable probably caused the amplifier to go into supersonic instability ( a car stop lamp makes a good RF detector!) . Is the cable melted such that both conductors are exposed? If so, split it with a craft knife and insulate the wires.

I found that some brands of EL84 would re plate slightly (in a darkened room) whereas others would not but if this is the case it would indicate to me that the cathode bias resistor is too small in value. ICBM has said in the past that many CB amps run way too hot and nothing is lost by upping the cathode resistor by a few tens of Ohms. I found 47R (4 valves) too low and this was upped to 68R. Blind A/B tests showed no sonic differences and red plating was cured for all valve brands.

Dave.

ICBM

What Dave said.

Having the speaker cable too close to the valves can sometimes cause instability even if the cable isn't damaged, due to electromagnetic coupling between the cable and the valve - rare, but I've come across it at least once. (Can't remember what amp it was.)

Which pair of valves red-plated - both on the same side? If so that will be caused by one valve overloading and raising the grid voltage for the other one in parallel. Only one of the valves will be the culprit.

SunDevil

Hi Gents,

Firstly, thanks as ever for the help!

The 2 tubes (next to each other, so I asume in a pair) were practically glowing - very similar actually to a pic I found on here posted by ICBM

The maker has clarified that the amp has no HT fuse - could that lead to damage and should I consider a retrofit?

The cable itself is still sound - a little melted superficially, but not melted through, to the point of a short.

Cathode resistor value is 56k - apparently, they upped this from 47k on the original run, but felt the amp (Komet Connie) lost dynamics if they went higher.

Sounds like my next step is to get some replacement tubes and see if I was unlucky this time - the maker has recommended a nice quad of NOS Mullards as being up to the job (but not offered to gift them to me

SunDevil

The end of my post was cut off - any recommendations on tubes that may be tougher than the (low IP) Sovteks?

Cheers

Baz

ICBM

JJs. The best modern EL84s.

Karltone sells them.

jpfamps

SunDevil said:

Hi Gents,

Firstly, thanks as ever for the help!

The 2 tubes (next to each other, so I asume in a pair) were practically glowing - very similar actually to a pic I found on here posted by ICBM

The maker has clarified that the amp has no HT fuse - could that lead to damage and should I consider a retrofit?

The cable itself is still sound - a little melted superficially, but not melted through, to the point of a short.

Cathode resistor value is 56k - apparently, they upped this from 47k on the original run, but felt the amp (Komet Connie) lost dynamics if they went higher.

Sounds like my next step is to get some replacement tubes and see if I was unlucky this time - the maker has recommended a nice quad of NOS Mullards as being up to the job (but not offered to gift them to me

I suspect you mean 56 ohms rather than 56k!

No HT fuse is daft, and may contravene the regs.

If the valves on one side the of the push pair are glowing I would suspect loss of grid reference on that side, although oscillation is also a possibility.

I've actually seen oscillation due to the speaker cable being too near the valves / input quite frequently when testing of the bench, but is less common in amps as the speaker cable is usually routed away from the danger zones as it were.

SunDevil

Indeed - 56 ohm rather than 56k, thanks Frank

..no wonder my pedals used to be a bit hit and miss!

ICBM

American amps still very often don't have HT or any other secondary fuses, which is just stupid in my book - and should in fact contravene current EU regs as jpfamps said. An HT fuse is extremely important to prevent secondary damage in the event of a valve failure. Interestingly this is one of the few changes Marshall made to the Bassman circuit when they copied it to make the JTM45 -rightly so.

Worse, some have fuses for things like the filament supplies, which are not really necessary at all, and still no HT fuse. This sounds ridiculous, but I recently repaired a very expensive Fender Custom Shop amp exactly like this, and where a valve failure had caused a cascade of damage which resulted in quite an expensive repair, and the unnecessary fuse didn't even blow! I fixed it by removing it and converting the holder to an HT fuse, so it won't do the same again…

I'm very surprised Komet don't fit them since the chap who builds them is very knowledgable and a huge fan of old Marshalls.

SunDevil

Thanks - I know the Conestellation is the only amp they do without an HT fuse (Holger's first 'stand-alone' Komet)

I'll have a chat with him and see if I can get his sign-off on an appropriate mod without tearing up my warranty


Baz

ecc83

"Cathode resistor value is 56k - apparently, they upped this from 47k on the original run, but felt the amp (Komet Connie) lost dynamics if they went higher."

Hmm? Well my test consisted of me switching in and out the extra 21 Ohms whilst CEO (top "voicing" man) and another director auditioned the amp. They did not know the state of the bias moment to moment and could not tell which state was which.

The bottom line is to measure Vk, Va and Vg2 and see what the dissipation actually is.. If it is much over 12 watts per bottle only YOU can decide if the risk is worth any audible benfit, real or imagined!

And yes you do need an HT fuse. If you don't want to drill the chassis you can fit an insulated inline holder. Not so convenient but it will save your butt!

Dave.