choke input power supply help

[jbrew73]

 I need help.   I am  trying to use a hammond  power transformer that is rated 375-0-375 with a choke input filter . The transformer puts out 411-0-411so I was expecting about 370vdc(411 x .9) but im getting 450vdc. This is just a mock circuit with a 5h choke ,a 22 uf  cap and a 25k resistor for a load .    
Whats up? Why the high voltage?

[FYL]

 
Whats up? Why the high voltage?

Low load - 25K @ 400V is only 16 mA. Low losses - you're probably using a silicon bridge.

More than 400V is pretty normal here. Try to run simulations using Duncan's PSU Designer II (http://www.duncanamps.com/psud2/index.html) or do another test using a smaller value/suitable wattage resistor.

[jbrew73]

That is online with my thinking.
   It is to be used for a screen supply in a high power bass amp with 6 6550 or 8 6l6 tubes .   Im not sure how much current they will draw.  Any ideas on idle current
 for the screens ?

[HotBluePlates]

This supply will not feed the plates of your output tubes?

This makes some sense with 6550's, but very much less sense with 6L6. There's just not a lot to be gained by running 6L6's with the screen much lower than the plate voltage. With 6550's, however, it's typical to run 600v plate, 300v screen.

[jbrew73]

The plate voltage will be about 560 and I need the extra transformer for the heater circuit.  In one of Kevin oConners books he showed using a choke input supply to feed the screens for a svt  build so im  trying to implement it here.

[FYL]

   It is to be used for a screen supply in a high power bass amp with 6 6550 or 8 6l6 tubes .   Im not sure how much current they will draw.  Any ideas on idle current
 for the screens ?

I wouldn't use an LC filter for a screen supply. Currents vary a lot between idle and full, regulation will be quite poor.

You need a tight screen supply. I'd either use a CLC topology and maybe add a voltage regulator (a Mosfet and a few passives, suitably heatsinked). Your power tubes will thank you...

[jbrew73]

  It is to be used for a screen supply in a high power bass amp with 6 6550 or 8 6l6 tubes .   Im not sure how much current they will draw.  Any ideas on idle current
 for the screens ?

I wouldn't use an LC filter for a screen supply. Currents vary a lot between idle and full, regulation will be quite poor.

You need a tight screen supply. I'd either use a CLC topology and maybe add a voltage regulator (a Mosfet and a few passives, suitably heatsinked). Your power tubes will thank you...

I thought choke input supplies had good regulation ?
 This may not be the best power supply design and i have other options.   I was just trying to implement an unused transformer that I have.

{edit- quote-tag -PRR}

[PRR]

Radiotron Designers Handbook, Chap 30, Section 3.

For "Choke-input" action, L must be greater than R/(6*pi*f)

I'll ass-ume 60Hz.

> 5h choke ,a 22 uf  cap and a 25k resistor for a load

L greater than 25,000/(6*3.14*60)

L must be greater than 22.1H.

Conversely, if L is forced 5H, R must be less than 5.6K.

If L is smaller, action becomes more peak-catching, "cap input", and voltage soars from 0.9 to 1.4.

> I thought choke input supplies had good regulation ?

This was the claim when chokes were cheaper than caps and rectifiers were hollow

And _IF_ you observed the minimum load!!

In a world of solid rectifiers and cheap caps, put the choke money into a little more PT and good caps, regulation will be better than a practical choke.

In odd-parts "design", do what you can.

I do agree that choke-input is awkward for screen supply.

[HotBluePlates]

I thought choke input supplies had good regulation ?

Another way to view PRR's information:

Most of the old-school choke input supplies were shown feeding the plates of an output stage (or the whole amp); that means more current through the choke, lower implied "R" and therefore lower L needed.

Since the voltage output was Vin*0.9, the overall design tends towards lower B+, higher current draw, and class A (or very rich class AB). We know that means an overall design that has little average current variation.

Of course, this makes sense, because chokes inherently try to oppose a variation of current, while caps try to oppose a variation of voltage. Knowing this, it seems best to cast the choke in the role for which it's suited, and save choke-input for class A amps.

Considering all this, I really don't know if choke-input has good regulation because of its features, or because sensible overall design with choke-input tends towards class A, minimal current variation, a "pre-sagged" situation, and therefore little voltage variation.

I think it's a chicken-or-the-egg question, and no, I don't know which came first. 

[jbrew73]

Thanks prr and hbp you guys told me exactly what I needed to know.  The 5h choke was what i had in front of me. I swapped it out for a 20h and the voltage dropped to 380. 
This design is not set it stone .  Just playing around with the parts I have before I go out and spend more ..