Bias supply Problem, NOT TYPICAL

[isaac]

Hey, I'm building a 50W two el34's and I have Power transformer that doesn't have a bias winding, just a standard high voltage 290-0-290. So I used a standard kind of bias thing by putting a resistor off of the main HV winding right before the normal rectifier like-unto Almost ALL vintage examples, ;-) but its not working right for some reason. I drew up a schematic of what I have, and then I drew up a schematic of what I'm thinking of. I've got a bridge rectifier for the main HT throughout the amp, and since it doesn't have a grounded CT on the transformer, I measured the voltage on the CT (not connected to anything now) and it was about 129, so I am wondering if I can draw bias voltage off there. See attached schematic.

Number 1 is what I have, and number 2 is what I'm thinking of. I wrote the voltages by the wires. I have no idea whether it would work or not, but I measured the voltage there and it looks like it might work. I just don't know if the way that these things work would allow for this. so #1 is what I have that won't work for some reason... and #2 is what I think might work but need help with.



Also, that is a standard chassis mount pot for the bias, and it is overheating a little, what kind of pot, except a trim pot hopefully, should I use? I might just find the average amount of bias i need and put a fixed resistor if I have to.

Thanks guys,
Isaac

[loudthud]

Neither 1 or 2 will work because the transformer winding never swings negative (more than 1 diode drop) to turn on the 1N5400 diode. FYI you should use a diode rated at 1000PIV. Look for a Marshall 9005 schematic. It uses a capacitor in series to get the bias.

Link: http://www.el34world.com/charts/Sche...eramp_9005.pdf

[isaac]

Thanks. I'll try that. What exactly does the capacitor do, though? I wondered why #1 didn't work. I haven't had a problem in using the IN5400 vs IN4007, but I have more of the IN5400's so I try to use them more. But thanks. Will try that.

[J M Fahey]

Dear Isaac, please do not design, but faithfully follow some established commercial circuit.

The PSU you drew creates 820V DC.
As soon as you turn it on, those poor capacitors will explode as grenades, not kidding.

PS: as a side note, please do not post 6500x5000 pixel drawings.
The Internet Browser will show it too small or unreadably large.
For general use try to draw screen size or smaller.

[isaac]

It was doing 620VDC right on the positive lead of the bridge rectifier when I had the standby off. Once I turned it on, it went down to 497VDC, so I didn't worry. If I have to I can use a half-wave (or do I have my names mixed up... ?) two diodes on the windings and CT grounded. I had a transformer kinda like this that had, probably 325vac on the windings, and it did 560 with this same power supply design. and I lied, I just realized in both of those drawings I drew in a 32uf+32uf capcan, I really have 50+50.
I'm not "really" designing here, as these things have been done many times, I just gotta find which ones have this and how they are done, but I had a similar design before, except on the older power transformer I had with the higher voltages, It also had a bias winding on the transformer, so since I replaced that one with this one, I don't have a bias winding anymore. And will keep drawing size in mind for the future. Sorry about that. Thanks.

[nickb]

(1) Your meter is misleading you. In DC mode it will indicate the average voltage. So with the standby open you are not seeing the peak value which Juan correctly points out is 820V.
(2) Don't worry about it? Think about what happens if you have the standby switch on when you apply power. The heaters are cold and it will be some time before the voltage will fall from 820V. I for one would not want to be in the same room when that happens if your caps are not rated for 1000V.

[isaac]

My meter is definitely not broken, so Why would it display a wrong voltage. Its a very nice Fluke meter. So I should revert to this? Last time I tried this style power supply, I had about 325VAC on each side of the winding, but I only had 215VDC TOTAL at the diodes. Why would that be. I think it must have been that transformer, as it just always seemed to be problematic. Thanks for your help guys, as always it is very helpful and knowledgeable.

[isaac]

Also, that bridge rectifier I have, is a bridge rectifier module, its not me putting together diodes and caps. its a little black box with four leads comin gout of it. probably about 3/4 inch square and 1/4 inch deep or so. With one corner lopped off for teh positive, although they are all marked, +, -, and two of these ~ (AC). I haven't had any problems with that before though.

[isaac]

OK last post for now, (three in a row, sheesh... ;-)) I just re-read what you said about average voltage, like on a wave, the peak is 820VDC, but average would be what the meter reads. That makes sense. Thanks.

[nickb]

OK last post for now, (three in a row, sheesh... ;-)) I just re-read what you said about average voltage, like on a wave, the peak is 820VDC, but average would be what the meter reads. That makes sense. Thanks.

Exactly!

[isaac]

Ok thanks, I got the last diagram right, correct? That's what I should try? I will once I'm done working on a highly modded other single channel marshall 18w based amp. Which could be a while, so maybe I should sneak this in sooner or later. THanks guys.

[nickb]

Ok thanks, I got the last diagram right, correct? That's what I should try? I will once I'm done working on a highly modded other single channel marshall 18w based amp. Which could be a while, so maybe I should sneak this in sooner or later. THanks guys.

That last version will give around -~55V for bias.

But their are other issues. Under load you'll get 300V or so on the HT. You can't expect to get 50W out of a pair of EL34s with that low a voltage. You need nearer 400V. We don't know what the pre-amp consists of so the 100K resistor might might not be right - it's kinda high. Lastly we know nothing about the transformer - does it have sufficient power rating for the job?

The best advice has already been given earlier - go with a known proven design. You'll be much happier with the end result.

[isaac]

OK yea, I was wondering about that. How can I get more voltage from it, so as to be able to do enough to do 50W? And just since I'm wondering, what happens when you lift the ground on a bridge rectifier? Would you then want to re-establish the Center tap to ground. Also, could I put a big resistor in series with the rectifier, after the bridge rectifier to lower the voltage, but still have enough for 50W. Because it needs to have at least 50W for me to be happy. ;-)

[J M Fahey]

You have what you have, it will put out what it will, it's not positive thinking what will bend laws of Nature.

290V AC will give you solid 410V DC, perfect to make a fine amplifier.

No, it won't put out 50W RMS (almost no "50 W RMS" labelled does, by the way, with very few exceptions) but certainly >40W which is perfect by a drummer or onstage.

Just build and enjoy

EDIT: by the way, soon you'll post asking about attenuators because "it's too loud )
Not kidding.

[nickb]

.. what happens when you lift the ground on a bridge rectifier? Would you then want to re-establish the Center tap to ground.

I think you are asking what happens if you disconnect negative terminal of the bridge. Yes, you will need to connect the center tap to establish a return path for the current. OTOH, half of the bridge is not being used.

Also, could I put a big resistor in series with the rectifier, after the bridge rectifier to lower the voltage, but still have enough for 50W. Because it needs to have at least 50W for me to be happy. ;-)

No - that won't work as you need to have current flowing in the resistor to drop the volts and you will have zero with cold heaters so you'll still face the 820V cap breakdown problem.

As JMF says, 290VAC will give 400V and you will get 50W on peaks - just not continuously.