Safe to use 6L6/5881 with 120mA rated PT?

[elipsey]

I have a Thordarson T13R14 which, as far as I can tell, is rated for 120mA HT (350-0-350@117Vac), and a 4.7A 6.3Vac filament supply. There is a 5V winding as well, but I cannot find the complete specification for this. I would like to use this transformer with 2 6L6WGB/5881, a 6K6 ohm Deluxe Reverb OT, and a solid state rectifier. I would like to be able to bias to about 16W per tube at idle for 5881 or 6L6WGB. If I wild guess about 450Vplate, that would imply 70mA (35mA x 2) at idle.

Does this sound safe?

[nickb]

It depends a bit on actual filter caps values and the DC resistances of the windings, but as a rough rule of thumb, for a full wave rectifier the rms current is about twice the DC current. In your case of a center-tapped secondary that means there will be half of that on each side i.e 70*2/2 = 70mA. At idle that is fine.

As you crank it up there will be something like 40 watts out and 40 watts dissipated in the tubes i.e. 80W == 180mA. Of course you won't be doing that continuously. I use a 20% duty cycle for an estimate, others may have different opinions . Thus I come up with 70*0.8 + 180*0.2 = 92mA for typical use. So you still have a nice margin in that case too.

If you are not using the 5V winding and the 6.3V sounds like it may not be to heavily loaded then that is also all for the good. Go for it.


PS: The 22W Deluxe Reverb OPT might be a bit wimpy. You might find that it saturates at low frequencies.

[big_teee]

Calculate Tube Amp Power Transformer Current
It shows the below.

Selected transformer voltage: 350-0-350

Calculated voltage at first capacitor (B+): 479.5V. Subtract 6V, if using choke.

Calculated filament current (typically the 6.3v secondary): 2.79A

Calculated current: 108.43mA at 9997R calculated load (10% plus factored in). Preamp valves current draw is estimated at typical 12AX7 max dissipation of 1.2W at 330V, i.e. 3.6mA per triode

[elipsey]

Thanks guys, I think I'll give it a try and see what happens.

P.S. If anyone knows where to find authoritative specifications for this transformer, I would be grateful.

[big_teee]

I think the Tung-Sol 5881s are only 400v rated.
I guess there are other alternatives?
GL,
T

[DRH1958]

Just out of curiosity, how does SS rectification of 350-0-350 get down to 450v for the plates? 350 X 1.4 = ~490 VDC out of the rectifier. Where does the 40 volt drop come from? Power supplies are next on my list to lean more about, so why not here?

EDIT: I see Big Tee has a link but I'd like to learn, not just plug numbers into a calculator.

[elipsey]

Your math looks more reasonable than my "wild guess".

I probably don't want more than about 450V, and I might use a tube rectifier, a sag resistor, or a zener to get there. Honestly I haven't really thought this all the way through yet. I'm still trying to decide if these parts want to be friends.

Edit: If you click "Show Formulas", you can see the rectifier coefficients used in the calculator. I don't really know what I'm doing in this area, but I have measured a bunch of different rectifiers at home out of curiosity, and those numbers look similar to what I found. I always imagined that the internal resistance of the rectifiers could be treated as simple and linear, more or less, and back calculated from these measurements. I don't know if this is really true; maybe someone else will chime in here?

[DRH1958]

Yes, I did see "wild guess". That's why I was wondering about this and agree with Big Tee as the data sheet says 5881/6L6GB is a 400 volt tube. I have found that those JAN Sylvania 6L6GBs are pretty sturdy. I know this rating is exceeded a lot and power dissipation is mostly the key but having a large difference in voltage for calculating current will make a difference. Thanks for your thoughts on dropping the voltage as that's where my mind was headed, zeners, tube rect.,etc. but "SS rectification" stopped me. Yes, a 5U4 should help a lot with dropping voltage.

EDIT: Thanks for your "edit" response, elipsey. I didn't see the "show formulas" link in the calculator. I guess their 1.37 is pretty close to my 1.4 for SS. I've got a rectifier tube data sheet for anybody who's interested to calculate voltage drops for tubes without that calculator.

[Gingertube]

DRH
The 1.414 x RMS Voltage thing doesn't work underload.

That gives you the PEAK voltage you get with no load on the power supply.

Once you start drawing current you will get more like the AVERAGE voltage rather than the PEAK voltage. Use x 1.2 or even X 1.1 for heavy loads.
To be pedantic, Average voltage is 0.637 x 2 = 1.274 of the RMS voltage

So 350 - 0 - 350 will give 350 x 1.414 = 495 Volts unloaded and if using a SS rectifier you filter caps need to be able to handle this voltage (so I would NOT use 450V rated caps)

AND

about 350 x 1.274 = 446V when loaded.

It may drop to say 350 x 1.1 = 385V under heavy (too much) load.

Rectifier diode drops ignored.

Cheers,
Ian

[Enzo]

In the absence of ripple, isn't that drop due mainly to winding resistance?

[Gingertube]

Enzo,
No - On each mains half cycle
When the instantaneous supply voltage is above the cap voltage then the diode switches on and cap charges toward peak voltage value.
When the instantaneous supply voltage drops below the cap voltage then the diode switches OFF and the cap voltage is partially discharged by the load.
This repeats every half cycle of the mains and the cap will typically settle at around the AVERAGE voltage with the ripple symetrical about that voltage.
Winding resistance losses then need to be added on top of that "averaging" function.

Have a play with PSUDII to see.
PSUD2

Cheers,
Ian

[Enzo]

Oh I believe you, and have seen the phenomenon, I just never gave it any thought before. Winding resistance was in my head as "good enough". I guess that was a place holder waiting for the real story.

You see, even after 60 years of this stuff I still learn things every day.

[Dave H]

Before the advent of programs like PSUD2 I used a quick and dirty way to guesstimate the effect of ripple. Say the load current is 100mA, Capacitor 47u and mains half cycle time 10mS (UK). Calculate the voltage the capacitor will discharge at 100mA for 10mS. Q = CV and Q = it therefore V = it/C which is (0.1 * 0.01)/47E-6 = 21V this is the p-p ripple and if it's riding on a peak voltage of 350V it will look like 350V discharging down to 330V on a scope which will average to about 340V on a DMM.

[nickb]

Oh I believe you, and have seen the phenomenon, I just never gave it any thought before. Winding resistance was in my head as "good enough". I guess that was a place holder waiting for the real story.

You see, even after 60 years of this stuff I still learn things every day.

Well, you're not wrong either.

Practically speaking the winding resistance does matter since when operating there is always significant current draw, even at idle. BUT, the transformer is spec'd for 350V at full load so the unloaded voltage will be higher, perhaps 10%. Thus the worst case, no the load voltage could be as high as 350*1.1*1.41 = 543V. Therefore you need to ensure than the filter caps you pick can withstand this surge voltage. A typical figure for surge is 25% over rating for a maximum of 30 seconds with a least 5 minute between surges. So a 500V cap will take a 625V surge and so should do the job nicely.

[g1]

Thus the worst case, no the load voltage could be as high as 350*1.1*1.41 = 543V. Therefore you need to ensure than the filter caps you pick can withstand this surge voltage. A typical figure for surge is 25% over rating for a maximum of 30 seconds with a least 5 minute between surges. So a 500V cap will take a 625V surge and so should do the job nicely.

I'm not sure if you have taken standby (for more than 30 seconds) into consideration or not?