How does this look?

Could any of you fine gentlepeople tell me if this looks OK.
Would it be better to use a transformer with dual secondaries and use two bridge rectifiers going into what are essentially two separate supplies and ditch the doubler, or is what I have now OK? Did I do the math right for the LM317? (side note: the HV is spec'ed to output up to 57V max, so I figured 48V wouldn't be a problem)
[Edit]
Thanks a bundle!

Ethan, that won't work as I look at it. You have that 1500µF cap in parallel with a single diode and no winding driving it

The bottom half is right at a cursory glance, and everything after the 6th diode looks good on the top half... but that multiplier looks wrong to me.

Keith

...Picture Updated... (Click on the pic for the full view)
Ah! I missed that!
How's it looking now?

[quote author="Ethan"]the HV is spec'ed to output up to 57V max, so I figured 48V wouldn't be a problem[/quote]

Not so. These integrated regulators don't have a maximum voltage spec, but rather a maximum input to output differential. In other words, as long as the voltage difference between input and output doesn't exceed 57V, you're fine. This is how people are using these things for HT supplies in tube circuits.

I don't have the datasheet for the 317HV, but if it's like the other regulators out there, I'm getting around 50.5V at the output of the 48V supply. And the voltage doubler still looks wrong here.

Peace,
Al.

Oh, also: What's R1 and R10 for?

And you may want to consider changing your voltage-setting resistors to make them sink enough current to keep the regulators regulating. There was a post about that a few days/hours ago.

Peace,
Al.

Should be fixed now.

Those resistors are supposed to be a load of yet to be determined value, to be determined based on what transformer I get. I just thought a "pre-load" might be needed if the rectified secondary voltage is a wee bit high, so the V-reg doesn't have to dissapate too much heat. The 240R as recommended by the datasheet should keep them conducting, no?

[quote author="Ethan"]Should be fixed now.[/quote]

That doubler still makes no sense to me... You're hooking up the (-) leg of the 1500µF cap to AC!

[quote author="Ethan"]I just thought a "pre-load" might be needed if the rectified secondary voltage is a wee bit high, so the V-reg doesn't have to dissapate too much heat.[/quote]

Nah, forget about that. First, the way they are right now, those resistors aren't dropping any voltage. Second, let's stretch it and say you're feeding the regulator 100V rectified and filtered DC, and pulling 50mA from the phantom supply. That's an extreme, never-gonna-happen scenario... but it's just 2.5W of heat dissipated by the regulator - a medium-size heatsink will take care of that.

[quote author="Ethan"]The 240R as recommended by the datasheet should keep them conducting, no?[/quote]

I just got a copy of the datasheet and you're correct: these things only need 3.5mA to maintain regulation. 240R is just fine.

Peace,
Al.

[quote author="alk509"]That doubler still makes no sense to me... You're hooking up the (-) leg of the 1500µF cap to AC![/quote]

That cap and the first diode work like a half-wave rectifier (at least that's what I've read).

Thank for all the continuous help guys! The constant questioning really helps me rethink, revise and learn! :grin:

*Schematic Updated*

hmm i third that notion that something is wrong with the cap to the AC input.. shouldn't that be to ground? are you thinking of "charge pumping" that cap by running pulsing DC across it? I would think that you need your cap referenced to ground, the + from the bridge through a diode and across the + side of the cap and through another diode. just a thought..

It should work fine the way you have it now. Looks the same as +24V and +48V here: http://www.forsselltech.com/gerbers/MikePrePS1.PDF

I have built a variation of Fred's power supply using the 317AT and 317HV and it works great.

I think maybe D4 in the doubler should be referenced to ground rather than the rectified 24V from the lower circuit. Floating it like this may make the voltage too hogh for the 48V side.

Hi,
what about something like this?


You coul put a diode or two in series with the ground pin of IC2 if you need 25V
The value of C1 and C2 depend on the current requirement of your circuit but 1000uf should be ok for 100mA load.
C3 -C8 are 100uF to be placed near the regulators.

Alessio