Mesa Studio and Bias

[justonwo]

I just purchased a pair of EL84s to install in a friend's Mesa Boogie Studio 22. I haven't been able to find any schematics anywhere for the purpose of locating the bias pot, but KCA NOS Tubes has a brief statement at the top of their EL84 page that says most of the amps that use these tubes don't require bias adjustment. Can anyone comment on this? Is the bias unadjustable on this amp?http://www.kcanostubes.com/EL84.html

 

[John Phillips]

There is no bias pot.

Many EL84 amps are cathode-biased, or 'self biasing', but actually the Studio .22 is fixed bias. Whether you think it needs adjustment or not is another question... Mesa don't (read the info on their website for the reasons), so they don't fit a pot.

FWIW, I have never had to rebias any Mesa amp when fitted with good tubes.

Actually, there are cathode-biased amps which require adjusting (involving replacing the cathode resistor with a different value) to run correctly too.

 

[justonwo]

So I guess in many ways this Mesa is like my Champ - more or less plug and play. Assuming the plate dissipation is within a normal range, no modifications are necessary.

That makes it easy. Thanks for the information.

 

[fractal]

Hey justonwo,

Not to give a commercial to Groove Tubes, but the #5 apparently matches the bias that Mesa sets at the factory. I've put in #5's in mine, although you can probably put in something close. Someone correct me if I'm wrong, but I think I remember the advice correctly. I would think that you could call any particular company, and get similar info if they rate tubes by current flow.

Hope this helps,
-Garrett

 

[RussB]

I like the JJ EL84 in my Studio 22+. I have been buying them from Bob @ Eurotubes.com for the last 5 or-so years...He gives me tubes that are in the "proper" bias range for my 22.

you'll find schematics here

 

[justonwo]

I already ordered some NOS Teslas from KCA NOS tubes. Hopefully, they'll put the plate dissipation around the right spot. We'll see. Thanks for the suggestions, guys.

 

[justonwo]

This evening, after biasing my Rivera, I decided to double check the plate dissipation on my buddy's Mesa Studio 22 before replacing his tubes for him (I think this is an old one with only one channel, not the Studio 22+). Anyway, I checked the current and voltage on the Studio with my Weber Bias Rite. I measured 390 volts and 38 mA. This is WAY too high a plate dissipation for EL84s.

I think the tubes are the original tubes from several years ago so I figured it was just bad tubes. I installed a set of NOS Teslas in place of the original tubes and got the same measurement - 38 mA at 390 Volts, for a total dissipation of 14.8 Watts. At this voltage, I figure the current should be somewhere around 18 mA. Anyone have any thoughts? Maybe I'm doing something wrong? Is the Bias Rite somehow measuring the total current draw of both tubes?

For reference, I had the master and volume all the way down, with nothing in the inputs, and the speaker was plugged in. I figured this was the correct procedure.

 

[John Phillips]

That does sound very high.

You can't draw the current from both tubes through one socket, if the Bias Rite is between the tube and the socket. 390V is the correct B+ too.

Have you checked the bias voltage? The schematic shows 11.6V, if something's gone out of spec in the supply that could have fallen.

 

[justonwo]

Hi, John. I'm just now learning how to service my own amplifiers and haven't yet figured out how to measure the bias voltage. I guess I need to buy a tube amp book. Is this a simple procedure?

What is the bias voltage a measure of? The voltage drop across the cathode? Might the amp simply need a new cathode resistor (or, better yet, a bias pot)?

 

[John Phillips]

It's a fixed-bias amp. This means that there is no cathode resistor, and the bias voltage is supplied by a separate circuit. This is why it's called 'fixed', because it is not dependent on the tube current - not that it's non-adjustable, you can have adjustable-fixed-bias and non-adjustable fixed-bias amps, depending on whether they have a trimmer... Mesas don't.

(All amps I know of with a bias pot are actually fixed bias, which is a source of much confusion! You could theoretically have an adjustable cathode-bias amp too, but I'm not aware of any.)

In a fixed-bias amp the bias voltage comes from a separate part of the power supply, containing a diode, resistors and caps to set the voltage. If one of these has gone out of spec, the voltage will change, which affects the bias setting of the amp, just as if you had adjusted a trimmer.

The voltage is measured at the grid of the power tubes, to ground. The grid is pin 2 of the power tube sockets - counting clockwise from the space, when viewed from above (easy on this amp, the circuit board and tube connections are visible when the chassis is dropped and slid out slightly, without any further dismantling). Simply meter the voltage between this pin and the chassis.

**If you're not happy/experienced with working with dangerous voltages, it might be a good idea not to do this.**

It's not particularly risky, if you're careful, but you need to know what you're doing. The bias voltage itself isn't dangerous, but there are other ones nearby that definitely are - and you do need to have the amp fully on to measure it.

If the bias voltage is off, you're going to need to find out why, and which component needs replacing. This is more of a pain on this amp, and will need the board out.


There's a schematic for it in Aspen Pittman's 'Tube Amp Book Volume 4', BTW - which is a good book anyway - provided you ignore the overall Groove Tubes hard sell! - and contains a lot of good info and useful diagrams (the Mesa ones are out of alphabetical order, you have to search for them BTW).

 

[justonwo]

John, very useful and interesting information here. Thank you for taking the time to explain it.

I measured the voltage on pin 2 with respect to chassis ground on both output tubes and got -11.1 volts (i.e. negative 11.1 volts below chassis ground). I suppose this value has dropped (or increased, I suppose, depending on how you look at it) from the schematic value of 11.6 Volts. Is that 0.5 Volt difference sufficient to generate that large a jump in cathode to plate current?

 

[John Phillips]

I think it could be. The tube current is very sensitive to the bias voltage - a variation of 10% is enough to take it from one end of the useful range to the other, often.

(That's actually why bias trimmers were fitted to amps in the first place - to correct the problem of 10%-tolerance resistors in the bias supply causing large variances in performance. The idea of fine-tuning it for tone only came much later.)

In other words, if the amp is running too hot at .5V too low, you might find it was too cold at .5V too high (12.1V). Mesa specify that voltage very precisely, to three significant figures (ie they don't say "12V"). This amp is on the limit of what you can achieve with two EL84s (the plate voltage and output power are very high), which probably makes it even more sensitive than usual.


Having looked at the schematic again, there's something that might be important too - there is a very unusual feature, which is that the phase inverter tube is operated fixed-bias as well, driven from the same supply circuit. If the PI tube is running a bit hot, I think that could actually affect the main bias voltage. Try changing the PI tube (the one closest to the power switch end) before getting deeper into component values and other things which would involve taking the board out.

 

[RussB]

I currently run a 5751 tube in the PI of my 22+. I have used a 12AT7 with good results as well.

 

[justonwo]

John, I changed the PI tube with no luck. Still got the same current reading. However, I called Mesa today and spoke with a technician about what I observed. He told me 38 mA at 390 Volts is a perfectly acceptable measurement. I guess since this is a class A amp that maybe it's supposed to run hot. Although this amounts to 14.8 Watts. Anyway, per their suggestion, I'll install the new tubes at this current and see what happens. Thank you very much for running through this project with me. I appreciate the advice.

 

[John Phillips]

If that amp is Class A, I'm Eric Clapton .

It's a good example of a hot-running Class AB - and it's not even cathode-biased, which is the usual excuse for calling them Class A.

I actually think it's too hot, since it's above the rated power even at idle, and I wouldn't expect normal modern tubes to last very long at that. Most or all of the ones I've worked on (whch is actually very few, since they rarely break, like all Mesas) ran cooler, from memory.

I would definitely think about getting some JAN/Philips - they were (I think) the stock Mesa EL84s at the time it was built - or GE 6BQ5s for it. Even at that, I'd be inclided to lower it a bit.

 

[justonwo]

"Class A Power" is printed on the back of the amp in quotes. Maybe the quotes actually mean something - class A . . . but not really.

I agree with you, John; the Mesa technician I talked to actually didn't know much about biasing himself. He had to go ask a co-worker and then call me back. Even if the amp were true class A, it doesn't make sense to me to be running it that far above the rated power. So, unless there's some simple modification that can be done to change the bias current, I guess I need to take the amp to my tech.

 

[John Phillips]

You may need to change a resistor in the bias supply. The bad news is that this will involve removing the whole circuit board, which is a pain.

Looking at the schematic, you will need to change either R111 (33K) or R272 - which is labeled as 170K, a very odd and nonstandard value, so I'm not sure if this schematic is correct. In either case, you will need to reduce the value in order to increase the bias voltage (ie make it more negative). Changing R272 will have more effect; but changing R111 may give finer control - you could even replace it with a trimmer, probably 22K since you already know 33K is too large a value. That would be going against the spirit of Mesa amps though .

However, before you do that I'd suggest changing C10 (220uF 63V) - the bias supply cap. If this has started to fail with age (electrolytics do, and this amp is probably getting old enough for it to be a possibility) it will drag down the bias voltage.

 

[RussB]

All this talk got me interested in my 22+...

I checked the bias in it today, 36mA @ 396v, Damn, that's at 14 watts, 120% dissipation!

So, I swapped R272 with a 100k (was 120k) and put a 250k trim pot in place of R131 (33k). I now have the draw down to 25mA @ 402v, 10 watts, 80% dissipation. I put in a new set of JJ EL84's as well (rated "37" by Euro-Bob) They should last a while.

An 82k in R272 would be just about perfect, but I don't have any 82k's around (must make out an order to Mouser!) I could've switched the pot to the R272 position I suppose, but I wanted to be able to "fine tune" the bias voltage.

The rosin in the solder makes ot all look sloppier than it is...really!

 

[RussB]

OK, I went with an 82k resistor in series with the 250k pot, the bias range is 11mA to 18mA...curse it all, a resistor in the 90k range will bring me to a more usable range, I rekon

I'd like to able to set the bias "too hot" to "too cold"

 

[Cees van Bavel]

I have a studio 22+ and did not find a need to rebias as well. Bought my tubes at www.thetubestore.com.

Anybody have any suggestions as to how to change the reverb in a more fenderish reverb (installing a new tank or so)? I do not like the reverb on the 22+.