New project calculation help

[Phil_S]

Now that I have a chassis, I need to make holes in it and you know where that leads. I have some stuff I'd like to use if I can, but think that I am fumbling the calculation in Duncan's PSU, so I'm here asking for help.

Someone (here?) gave me a UTC R-8 (375-0-375 @ 100mA). The problem is high B+, low current. I am thinking I can solve both problems with FW choke input, and I'm better off with a tube rectifier because it will help to drop the B+. Depending on just how low I can go, that will help determine the power tubes. However, I don't want to cut octal holes when I might end up with noval power tubes. The socket choice will affect the spacing. Otherwise, I'd probably just build the power sections and see what happens. I'm hoping I can get into a lower power affair, maybe with a pair of 6V6, or if B+ goes low enough, 6AQ5 or EL84. I also have some 6CM6 (noval 6V6).

(According to what I see from Hammond FW Choke input will supply VDC = .45 x secondary VAC, and IDC = 1.54 x. This is the basis for my questions and also the reason it won't be reasonable to use 6L6 or EL34.)

When I plug in this PT, unloaded secondary is ~855VAC. I figure it is running above spec due to the modern primary voltage supply and the rest is due to no load. All this seems to be consistent with a primary input differential of 125/110. Primary DCR is 4.6Ω and secondary DCR is 219Ω.

I have a Drake choke meant for a Vox AC30. It is supposed to be 20H 100mA. The DCR is 446Ω, which strikes me as high, but I don't have a lot of experience with chokes. Is this choke going to be up to the job? Do I need a choke rated for more mA?

For, let's say a 5Y3 (a real one) and a GZ34, is someone here willing to estimate B+ for me or confirm what I modeled?

Here's what I got from Duncan's PSU, voltage at C1:
5U4GB, 535VDC
5Y3GT, 520VDC
GZ34, 552VDC

The reason I believe these results are wrong is that simple FW cap input would give a result of 375 x 1.4 x 125/110 = 610. The numbers above are not sufficiently below this. It wouldn't be all bad if my problem is that I used the voltage from both legs together in the PSU, so that the above results are simply 2x. Then I'd skip the tube rectifier and simplify things.

I recognize that I also have to deal with high filament voltage, but I think that is manageable. I'm going to need to drop about 0.8 or 0.9V.

Thanks very much.

Phil

[strelok]

This is my best guess from looking at the datasheets for the various rectifiers:

5y3gt:
Choke input: ~300VDC
Capacitor input: ~420VDC
http://www.r-type.org/pdfs/5y3gt.pdf

5u4gb:
Choke input: ~300VDC
Capacitor input: ~450VDC
http://www.r-type.org/pdfs/5u4gb.pdf

GZ34:
Choke input: ~325VDC
Capacitor input: ~450VDC
http://www.drtube.com/datasheets/5ar4-ge1959.pdf


Just did some interpolating of the curves (as theres no curve drawn for 375VAC) by sight so these are just ballpark numbers. Also did it based on the assumption that the full 100ma was being drawn from the PT, so the numbers will be higher for lower currents obviously.

As for the choke I'd say if you're not getting close to 100ma it would be fine (as you shouldn't since that's what the PT is rated for as well.) Though it is a very large value for a choke input. The datasheets all specify 10H though they don't seem to say what the max is. I'd expect that large of a DCR for a choke of that value. If you do use its probably going to get fairly warm. Between it and the tube recto it will make for quite a saggy PS which may or may not be what you're looking for.

[ToneMerc]

Phil, assuming that PT is 375 @110 mains, I work that out to 426.14 @125 mains. Thus, 426.14 * 1.4= 596.6 unloaded. Now of course there's lot's of variables, but when plug in my rule of thumb rectifier numbers, I'm within 4-5 volts of the Duncan number.

TM.

[xtian]

How I stopped worrying (about high voltages) and learned to love the bomb (MOSFET B+ reducer). Seriously, Phil, the MOSFET thing is super easy, and you can dial your voltage to suit. Forget the tube rectifier.

PM me if you want one, preassembled.

[ToneMerc]

I also have some 6CM6 (noval 6V6).

I picked up about 20-25 of these about 2 years ago to experiment with some 12-15 watt Plexi like builds. The spec max plate voltage is not much above 305 or so, but folks have been running them successfully as high as 350-360. I had a few small PT's wound just for these tubes to run them around 315 or so.

TM

[sluckey]

I just did a conversion that uses a 430-0-430 PT, GZ34 and choke input. I don't know the winding resistances or the current rating but I would estimate 150mA based on iron size. I do have accurate voltage readings. This is for a P/P 7591 amp. Take a look at page 4 of the linked pdf. It 'may' give you some idea what to expect from your PT.

http://sluckeyamps.com/hammond_2/AO-63.pdf

[TUBEDUDE]

How I stopped worrying (about high voltages) and learned to love the bomb (MOSFET B+ reducer). Seriously, Phil, the MOSFET thing is super easy, and you can dial your voltage to suit. Forget the tube rectifier.

PM me if you want one, preassembled.

Very nice, mounting the FET on the edge of the perfboard! Makes a compact efficient unit, kudos.

[jazbo8]

Here's what I got from Duncan's PSU, voltage at C1:
5U4GB, 535VDC
5Y3GT, 520VDC
GZ34, 552VDC

All the above voltages seem high. Could you please post a screen shot of PSUII?

I just did a conversion that uses a 430-0-430 PT, GZ34 and choke input.

Would you know the inductance of the chokes in the AO-63? Also was the imbalance on the output tubes' bias done on purpose for the tone?

[sluckey]

Would you know the inductance of the chokes in the AO-63? Also was the imbalance on the output tubes' bias done on purpose for the tone?

I don't have any more info on the chokes other than the part numbers stamped on them. They are about the size of a Fender reverb transformer, smaller than you might expect from just looking at a schematic. Here's a link to the conversion project if you have more interest...

http://sluckeyamps.com/hammond_2/hammond2.htm

The imbalance on the power tubes is all due to poor tube matching. I converted the original balance adjustment to a bias voltage adjustment. Also, the 10Ω sense resistors on the cathodes are only 5% tolerance so I haven't been concerned with the imbalance.

My whole purpose for posting the link was to give Phil a rough idea of how much lower the B+ 'can' be when using a choke input. I know my numbers will not be his numbers but seeing some real world numbers may at least show that pursuing a choke input supply is viable. You don't see a choke input filter very often in guitar amp circuits.

[Phil_S]

Hi everyone. Thanks very much for all the input!

I'm not able to reproduce the PSU results. I guess I don't remember the exact inputs. I have new numbers for the 5U4GB. It says 609 at VC1. Is that 2x because I input 855VAC? If so, then 305VDC at C1 is consistent with what Strelok posted. Have a look?

[jazbo8]

Hi everyone. Thanks very much for all the input!

I'm not able to reproduce the PSU results. I guess I don't remember the exact inputs. I have new numbers for the 5U4GB. It says 609 at VC1. Is that 2x because I input 855VAC? If so, then 305VDC at C1 is consistent with what Strelok posted. Have a look?

Ok, I see what happened, there were a few errors made for the PT parameters. The secondary voltage should be half of the off-load voltage since it has a center-tap, if you hit the Help button, you will see what I mean. Also, by entering a very low regulation figure, it resulted in a very high effective resistance, which also threw off the calculations.

So, if you change the voltage to 855/2 = 427.5V and the regulation to say 20% or 15%, then the estimated voltage should be more realistic.

[jazbo8]

I don't have any more info on the chokes other than the part numbers stamped on them. They are about the size of a Fender reverb transformer, smaller than you might expect from just looking at a schematic.

From the voltages shown on p. 4, it seems the inductance of the chokes in AO-63 are quite low. Since Phil is using 20H, the B+ voltage at C1 is likely to be much lower.

[jazbo8]

double post

[Phil_S]

Aha
This was all very instructive and much appreciated. Re-running PSU for choke input, 5U4GB, using 375V/100mA input and upping the regulation to 14% forced the off-load voltage to 427.5 to be consistent with what I saw on the meter, and set the source impedance to 525Ω. This produces VC1 = 308.

Using the graph in the GE data sheet for 5U4GB, assuming 100mA and Va = 427.5, this finds input to the 1st filter cap at about 350VDC. (Cap input yields around 510V!)

These numbers are not far apart at all, IMO.

It looks like the 5Y3 will drop about 20V more. Solid state FW result is 313.

So, I think it is safe to conclude that the choice of rectifier won't make that much difference in B+ result, and I can get this down into a safe/reasonable range to allow a choice of smaller bottle tubes.

I really appreciate the help!

Phil

[strelok]

Using the graph in the GE data sheet for 5U4GB, assuming 100mA and Va = 427.5, this finds input to the 1st filter cap at about 350VDC. (Cap input yields around 510V!)

Yeah I did the reading from the graph based off of 375 not 427 as I somehow overlooked that part of your post, sorry if that added to the confusion. Glad you got it sorted out though. Quite surprising how much the choke input drops the voltage, eh? I just learned that recently, may be able to take advantage of it to use some iron I have laying around in lower powered amps as well.