Estimated Power of a Pair of EL503 runned out of spec ............... Question

[kagliostro]

As I wrote on one other thread I got an FBT amp that runs EL503 (out of spec) as Power Tubes

http://www.mif.pg.gda.pl/homepages/frank/sheets/010/e/EL503.pdf

the EL503 tube datasheet, for a pair in PP report 40W output with 265V B+ - Load 2.4K - Bias Resistor 56R

but on the FBT the conditions are overspec, 365V B+ - Load 2K - Fixed Bias 20.5V

and I would like to know an approximated Output Power value for this conditions (not able to do math by myself)

There is a Geloso amp, the G1 1070-A

that uses the same power tubes at 330V B+ (?) (assuming the data I've find are correct), a Mixed Bias (Fixed + Cathode) and is published to reach 70 - 90W of power











Thanks

Franco

[sluckey]

but on the FBT the conditions are overspec, 365V B+ - Load 2K

P = V² ÷ R_L, so 365² ÷ 2000 = 66.6W

[kagliostro]

Many Many Thanks Steve

So, no matter the Bias voltage ?

Franco

[92Volts]

Many Many Thanks Steve

So, no matter the Bias voltage ?

Franco

It might matter, but not within that formula. That describes the maximum power that could go through that transformer with that voltage, but tubes aren't able to pull that much power.

If you look at the current required to swing 365v across 2000ohms, it's 182ma. That's a lot for the tubes to pull, idk if the EL503s could do it, but they definitely can't pull it as plate voltage approaches zero. So actual power will be lower. Typical pentode currents drop off as plates drop below 100-150v so without really looking at datasheets, a good ballpark is to do the same math at 100v lower.

A voltage swing of 265v through 2000ohms gives 35w and requires the tube to max out at 130ma while plate voltage is 100v, which is more realistic than 182ma with plate current near zero.

Bias comes into play because effective current through one side of the transformer is actually the difference between idle vs. maximum current. So if your tubes can do 160ma max and you bias at 40ma you lose 1/4 of that. But practical constraints-- avoiding crossover distortion at low currents and keeping tube bias safe on the other hand-- keep it in a sensible range without huge variation between most common designs.

[kagliostro]

Many Thanks also to You

Franco

[sluckey]

That was only a rough estimate.

[kagliostro]

And was what I was asking for

Franco

[PRR]

> no matter the Bias voltage ?

Bias is like the Idle Speed setting on a car. (In days before computers took that job away from us.) Turning it to 300RPM or 1800RPM has nearly NO effect on what Power we get when we slam the go-pedal to the other extreme.

Claimed power output is probably correct. Note the transistors driving the EL503 grids. This is a class AB2 stage. Grids are driven positive! Compare with early 6L6 (no GC) data. AB1 is maybe 24 Watts out, but AB2 is 47 Watts output.

Exact computation must include a lot of plotting and a very difficult (by hand) estimate of dissipation, both plate and screen. You can arbitrarily increase drive and current, but it may melt. It may melt on sustained test-tone (maybe at full, maybe only near half power) but not on speech/music. And maybe for over-driven guitar.

But McIntosh took the 6L6 up to 50W in AB2 with extreme reliability in tough work (shaker tables). Reading 6L6 as 19W Pdiss and EL503 as 29W Pdiss, 76W does not seem impossible. If the EL503's "29W" is as brave as the 6L6's original 24W rating, we get 60W output.

Do you *need* 66.6 Watts? Are EL503's cheap? I'd really think about putting 8 Ohms on the 4 Ohm tap and accepting say 40 Watts output, less stress on everything.

[kagliostro]

Many Thanks PRR

As I told, I was looking for an aproximation of the power, so the data I've now are enough

Do you *need* 66.6 Watts? Are EL503's cheap? I'd really think about putting 8 Ohms on the 4 Ohm tap and accepting say 40 Watts output, less stress on everything.

No, really don't need 66W, and those tubes are really difficult to be find and expensives

unfortunately the amp (the FBT Bass Organ) has only an 8 ohm tap, so to have a load adequate to EL34 tubes (4K) I can only use a 16ohm speaker

--

I asked about the estimated original power of the amp because my complicated brain had a seed of an idea ....

to parallel to each EL34 tube a smaller tube (like EL84) to try to avoid the use of a 16 ohm speaker and continue to use an 8 ohm speaker

a smaller tube will add only a fraction of the power of the EL34 tubes, but I doubt that I can find a small tube that paralleled with the EL34

will require a load of only 2K (the OT primary impedance of the FBT OT with 8 ohm  speaker load) a russian 6P14P can solve the problem

of the high B+ level, but seems not adequate as impedance, I think the same is for 6V6, also if I'm not 100% sure

Franco

[92Volts]

EL503s might be better with 2k output impedance, but it's not ideal for them either. Higher B+ usually calls for higher impedance, the parameters of this amp are both higher-voltage and lower-impedance than the datasheet example.

A 6L6GC might be a better modern replacement, not because they'll handle low impedance better, but because heater current is 0.9a compared to the EL503's 1.05a. EL34s require more, 1.5a per tube.

Considering the cost of EL503s, a new output transformer might be a good investment if it lets you avoid buying even 1 pair, it would certainly pay off after an additional tube replacement.

[kagliostro]

On this thread some info are missing, if you want, read the other thread about the FBT amp


Franco