MESA Express 5:25 - high b+, high current draw on power tubes - normal?

[wagdog]

Greetings all,

MESA Express 5:25 http://music-electronics-forum.com/a...-5-25-5-50.pdf

After redplating a new (within 4 hours of play time) MESA EL84 in my express, and now waiting for a replacement set, I poked around a bit w/a bias probe hoping to get a better idea of what was going on in the amp.

The B+ was 400v, cathode current with the new set of tubes (the one that had a redplate) was 32mA (both tubes). I didn't leave the amp on long as I didn't want the tube to redplate again. This is dissipating 12.8w at idle(!).

I put in a set of JJ's, they were running just under 32mA. I tried a few different sets of various el84's, most of them were in the 30-35mA range.

This strikes me as pretty darn high b+ for an el84, especially considering the MESA tubes are putting 32mA of current through them. I have a peavey classic 50 as well, and it too is putting 400v on the plates of the el84's.

Anyway, wanting to check the MESA out further, this morning I opened the amp up and took some measurements. Seeing 410v on the center tap of the OT (point A), and 402v on each of the plates. The screens are at 330v. I didn't measure screen current, probably should have.

Couple questions: does the B+ seem excessive for an el84? The schematic shows 401v on node A of the B+ (and then 382v on the plates, which doesn't seem likely). Is the fact that the screens are so much lower the reason this is OK? Or is it just not OK?

Adding a bias pot would allow me to run the tubes cooler, but I'm thinking that much difference could raise the B+ as much as 10 to 20 volts, and what would that then mean for the screens on the next node?

Thanks for any thoughts on this.

[soundguruman]

It's not unusual for a Mesa amp to have incorrect bias, set at the factory.
This explains the high current.
I mean, in "any" given Mesa amp, there are pretty much a "variety" of what I consider to be "design errors." OR assembly errors...
I mean 330V on screen and 400 on plate.......does indicate a problem, I would say. I mean, where's it all going?

[wagdog]

The schematic indicates 382 on the plates and 317 on the screens - my amp isn't far off that. And my grid bias was something like -12 - all pretty much in line w/the schematic. I'm just curious why MESA designed the amp to put so much voltage on the plates (as does Peavey in the Classic 50), and supplies tubes that are running so hot. There are a lot of 5:25's out there, it would seem that we'd hear of more of them going through tubes, rapidly.

[Mick Bailey]

First thing to check is your line voltage and that the transformer is strapped correctly. You also need to check the grid voltage is -11v (or close) and holds at that value.

[Mick Bailey]

Also, you need to subtract the screen current to arrive at the anode dissipation. Did just one tube red plate?

[wagdog]

Also, you need to subtract the screen current to arrive at the anode dissipation. Did just one tube red plate?

Yeah, I should have checked the screen current when I was in the amp. I'll probably have a chance to do it tomorrow. Still, was under the assumption that you could safely deduct 5% for screen current and be pretty close. Even with that, it's still high. I don't think it's doing anything that it wasn't designed for - as in, I don't think there is anything wrong with this amp specifically. The schematic clearly shows these high voltages.

I'm just wondering why it's designed with such a high b+, and why mesa is providing tubes that are running so hot when their amps are supposed to be plug and play. I'm also wondering if I do lower the idle current by modifying the - bias circuit, if the resulting increase in b+ will have an even more detrimental effect further down! such as on the screens.

And yes, one tube redplated.

[g1]

It's not uncommon to see mfgr's running 6BQ5/EL84's at 400V and idling at 12 to 14W. Whether it is technically correct or not, they consider it "class A".

[NZaudio]

Sadly I have to agree with you, have seen so many Mesa Amps and always some design mistakes.

[Mick Bailey]

Lowering the idle current will push the B+ up some. One I worked on a few weeks ago went from 400v to 417v as a result, so it can be counter-productive in that one value drops and the other increases.

In my opinion, the key to longer tube life for EL84s is controlling the screen voltage and current. I like to see a much larger screen resistor than 470 Ohm and sometimes go up to 4.7k. If you monitor screen current with a signal applied, some amps get to a critical point and then screen current rises dramatically. Using a larger resistor improves the situation quite a bit.

You may want to try the Sovtek EL84M, which can stand higher plate and screen voltages than a stock tube. My practical and listening experience with these is really positive.

[wagdog]

Lowering the idle current will push the B+ up some. One I worked on a few weeks ago went from 400v to 417v as a result, so it can be counter-productive in that one value drops and the other increases.

Yeah, that's why I'm a bit leery of "fixing" the bias problem. MESA designed the amp this way, and is furnishing tubes that are running at over 12w dissipation at idle. MESA is real big on the "use our tubes" thing for correct bias - like do they know something the rest of us don't? I tend to respect their amps, so I'm wondering what's up with that.

In my opinion, the key to longer tube life for EL84s is controlling the screen voltage and current. I like to see a much larger screen resistor than 470 Ohm and sometimes go up to 4.7k. If you monitor screen current with a signal applied, some amps get to a critical point and then screen current rises dramatically. Using a larger resistor improves the situation quite a bit.

Did think about that. I'm wondering if going that high on screen resistors will alter the tone of the amp (which sounds, really, really good, btw). Have you noticed an audible difference when upping the screen resistors like that?

You may want to try the Sovtek EL84M, which can stand higher plate and screen voltages than a stock tube. My practical and listening experience with these is really positive.

Definite possibility. Have read on the 'net that they are a beefier el84 - can't seem to find an actual spec sheet for 'em though! I'm curious to see what the replacement tubes MESA furnishes (via sweetwater) will be running at too.

[Mick Bailey]

Whether increasing the screen resistors affects the tone depends on your particular amp and your own hearing. I can't tell any tonal difference up to about 2.2k in most amps. Upwards of that you get a more pronounced effect where the drop across the screen resistor limits the rise in current through the tube, but that impacts on dynamics more than tone.

But, many of these amps run for years without problems. One red plating tube doesn't necessarily mean there's something wrong with the amp. From your measurements you need to deduct 5.5mA from your cathode current, giving a plate dissipation of 10.6w. That doesn't appear too bad to me. My approach would be to establish if there is actually an amp fault first off.

Mesa doesn't install trimpots in their amps - biasing is set through fixed resistors. The reason is to simplify tube installation by players, so they don't need the bias setting. The tubes are supplied within a narrow tolerance to go straight in - hence the 'use our tubes' approach. This makes it easy to change out power tubes at a gig, or on the road. That's the theory. The tubes are Mesa branded tubes that you can get elsewhere under their respective factory branding. I don't find them to be any better or worse for being Mesa branded.