Help! Calculating Custom PT voltages...

[FuzzFaceBuilder]

Hey all,
I need help calculating a PT voltage rating. The project itself very Dumble-esqe but I thought it might get more/varied responses here. The tube compliment for the amp is going to be a quad of 6550's (spec sheet here), (4) four 12AX7, (1) one 12AT7 in the preamp, and (3) three 5v non latching relays. It's gonna be solid state diode rectified.

Is there anything I missed, necessary to calculating the PT voltage?

Thanks so much in advance,
Hale

[stubbyfex]

Hello!
You basically just add up all the filament currents: 6550 1.6 amps, 12ax7 and 12at7 are .3 amps. So (4 x 1.6) + (5 x .3) = 7.9 amps. Of course you'll want a little more so the transformer will run cool, perhaps get a 10 amp filament winding. The 5v relays can work off the 5v rectifier windings, 3 amps is common but little relays will only draw 100ma or so. The 6550's will need lots of current, depending on your class of operation, and bias, you perhaps will need 500ma for four of them. All in all it will be a monster sized transformer. On another thread I showed pics of a monster bass amp I built with four KT88's. Here is a link to the monster sized toroid power transformer I used for it.

http://www.toroid.com/standard_transfor ... s_tube.htm

[FuzzFaceBuilder]

Stubbyfex,

You basically just add up all the filament currents: 6550 1.6 amps, 12ax7 and 12at7 are .3 amps. So (4 x 1.6) + (5 x .3) = 7.9 amps.

Thanks for that explanation! I've been staring at the spec sheets for the 6550's and 12AX7/12AT7, feeling pretty put out about not knowing what to do with all the info !

The 6550's will need lots of current, depending on your class of operation, and bias, you perhaps will need 500ma for four of them.

Of course you'll want a little more so the transformer will run cool, perhaps get a 10 amp filament winding.

Is there a standard for the extra filament current to keep things cool (i.e. 15-20% over)?

The 5v relays can work off the 5v rectifier windings, 3 amps is common but little relays will only draw 100ma or so.

Thank you for pointing this out!

I'll eyeball the layout and crunch the numbers using the formula above (for my education purposes). So I can do it myself, and hopefully help someone else next time!


Thanks much,
Hale

[stubbyfex]

Is there a standard for the extra filament current to keep things cool (i.e. 15-20% over)?

I don't think there is a set in stone rule, I like to add about 25% to the total figure.

Also to figure the 6550's current draw you must look at the characteristics (curves) that show the plate currents, plate voltages, etc, to really determine the amount of current draw. And you must add in the screen currents too. You can get a Typical amount per tube by looking at the specs, which for a 6550 I think is 140 ma for the plates, and 12 ma for the screens.
Yes, it's kind of complicated. I'm just glad that someone else did the math for the curves. That saves us lots of time!!

[FuzzFaceBuilder]

Hey all,
I just got a tip on a huge Fender style PT. Here are the specs: 120/240 primary, 335-0-335 VAC @ 600mA, a 60 bias winding. 2-6.3 fil windings one at 3 Amp, one at 9.5 Amp. Would this be a good PT for the project? I'll read up some more on PT applications too!

EDIT: what would I do to find the VAC @500mA? (How) does this tie in with Ohm's law? Can some one direct me to a chart or website to calculate VAC?

Thanks again,
Hale

[Phil_S]

The PT you describe is going to weigh in at about 8-12 pounds, so be prepared. 335-0-335 is going to give you plate voltage of more than 450vdc and less than 500vdc with sold state rectification.

If you want to confirm whether it can do the 500mA, the best thing I can think of is to build out the power supply with rectifier and filter caps. Then put a big dummy load on it and see what happens.

Let's assume you get 460vdc (335*1.4*96%) and I think this is reasonable. Ohms law: V=IR or I=V/R or R=V/I. We want the last one. V=460 and I=.5. 460/.5=920 ohms. Put a 1K load on the output and see if the voltage holds up. Actually, the thing to do is to start with a larger value resistor and work your way down. I would probably start this one at 5K and move down in steps of 1K. Read the voltage across the resistor and do the math. That will tell you the current flowing across the resistor. Let's say you get 500v across 2000 ohms, that's 250mA. You start reducing the dummy load, watching for the PT to get too hot and hope you don't see any smoke. You will hear a sizzle or smell something probably before you see smoke. Do not hesitate to kill the power at the first sign of trouble, really or imaginary.

Note, that for the sake of safety here, let's assume your potential is 500vdc at 600ma, and I don't think you'll be able to get quite that much, so 500*.6 = 300W. So, you'll need your 1K load to be rated to handle that. I have no idea where you get such a thing, and that is the problem with testing a monster transformer. Maybe you can find a bag full of 50 and 100 ohm 25W wirewound resistors?

To find the VAC at 500ma, your original question, you do the same thing, but without the power supply built out. You just hang the load across the two high voltage wires. 335+335=670vac. 670vac * 500ma = 335W. 670vac/.5A = 1340 ohms. That is your target test load, but start higher and work your way down.

The problem with doing this sort of thing is that you risk letting the smoke out of the PT and, as you can imagine, once you do that, you can't put it back in. Also, keep a fire extinguisher on the bench and hook it up to something that's got a fuse and a kill switch. I use a cheap surge protector power strip that will blow on excessive current draw.

You might try this simulation tool to get an idea of what your PT is capable of doing:
http://www.angelfire.com/electronic/fun ... grams.html
Look for the power transformer calculator, which is discussed, but you have to follow a link to get it.