First PSU schematic, not sure how I did

[Rybow]

So I have plans on doing a Royer mic mod, and I have a lot of parts laying around. I thought I'd try to come up with a PSU that would work using just what I have on hand. The big issue is the transformer I have is only a 2 x 9V, so I employed a cascade multiplier for the B+, and used the heater portion of the G7 PSU schem. I know it's not ideal, but i can't spend money right now,  I'd like to see if this will work, and if I've learned anything. Got a big over seas move coming soon.

Schem attached. I put a Zener on the end to clamp the voltage at 100V, but I don't think it's needed here. I am not sure if the 10uF caps in the multiplier will be sufficient. Still investigating.

Any advice would be greatly appreciated!

 

[PRR]

Lower-left diode in the bridge rectifier is shorted out (grounded on both ends).

You also have an apparent short on C5, probably a drafting slip.

Ignoring that fault: I don't think you get anywhere near 100V.

 

[PRR]

This is what I get for the given transformer. Yes, it is ugly. You can't find any little 120:12 to use backward off the 9VAC to get a decent HV directly?

 

[gyraf]

If you have 2x9V, why not have secondaries in series to get 18V AC into the tripler, and use centre tap for the heater voltage?

10uf is probably too small a capacitor for the tripler chain - I'd use at least 100u or so (and they do not need to be rated 350V, 35V would probably do fine)

edit: this is exactly what Paul suggests.

Jakob E.

 

[Rybow]

Thanks for the help guys! The shorted diode, doh! C5 was a drafting slip. Not sure how I missed that. I don't know why I thought a bridge rectifier was the way to go. I thought that I would high voltage caps, well not 350V, I just threw that in there, but as I think about it, that high voltage won't be across those caps, so thanks for clarifying that for me. I'll build it up and see what happens.

I was actually thinking about why I hooked up the transformer parallel, and I think I was more focused on the heater supply than the plates. PRR's schem certainly makes more sense.

I did find a small 120:12 transformer locally for about $15, so if I can't get what I need from this PS, I'll redo with that. I wanted to see if I could get a working PSU from all the parts I have on hand. Gotta use them up before I move.

Thanks again for the advice Jakob and Paul. You guys are the best!

By the way Jakob, I read the CMOS document from your site. Very interesting stuff. Looks like a lot of applications for that chip. Possibly a DC to DC converter board for a multi pattern phantom powered mic? The document talks about getting a self oscillating circuit out of a CMOS voltage multiplier.

 

[abechap024]

.... (and they do not need to be rated 350V, 35V would probably do fine)

Is that for all of them? I always assumed if the double/tripler makes X amount of voltage they would have to be rated for at least X. But even now that I write this, they only need to be rated for the maximum difference of the voltage they see huh? Ahhh.....sweet!!

 

[PRR]

>> they do not need to be rated 350V, 35V would probably do fine
> Is that for all of them? I always assumed if the double/tripler makes X amount of voltage they would have to be rated for at least X.

Did you loook at the picture that I posted?

BTW: I didn't, just noticed and fixed a typo, you should re-load it.)

It is like building a pyramid. Or a stone church with steeple. The base layer is short and wide. The top is narrow and tall.

Voltage on my C4 rises to 24.5V; I rounded-up to 35V reasonably-safe standard value. .

C5 C6 C7 C8 get to 49V (63V rounded).

Ripple is high. C10 C9 filter that; rise to 91V-98V (150V rounded).

The input cap has BIG current at low voltage, so must be BIG uFd but can be low voltage. C5 is the first "doubled" cap and must be twice the voltage. (Supply is a bit more solid with large-uFd here.) C6 C7 C8 track and must be same voltage rating as C5. The output filter caps must be full voltage.

Calculated: 25, 49, 98 (1:2:4)
Rounded-up: 35, 63, 150

This with negligible series resistance. A real transformer will give less.

The big ugly is the half-wave heater supply. Not only does it use most of the spare flux of the proposed 30VA core, the RMS current is a full One Ampere, and will soon fry your standard 1N400x rectifier.

 

[abechap024]

Ahh! Thanks as always PRR for explaining that. It makes more sense now.

 

[Rybow]

So I guess a 54XX rectifier is what I am looking for to handle that current. Looks like this will push my toroid pretty close to it's limit. Geting a cheap 110/12V transformer is looking like a good idea.

Well, I'll give it a shot and see if it works out. Thanks again!