Bosoz psu problem

[vit]

Hi folks.

I changed my trafo to a smaller one, but with 2x60VAC. Now I have a problem. Loaded with dummy load (15kohms) everything is OK, -59,9V and +59,8V.
It's when I attach the amps when the problem occures. With one channel attached I get -5V voltage drop over -V. When I attach both channels I get -10V voltage drop. End up at - 50V.
This couse the 220R-resistor to burn up after some time.
+V-side works fine.

What's wrong?



cheers,
Vit

[vit]

I get -54V with 750R dummy load.

[GRollins]

It sounds as though the new transformer is too small. It may have the right voltage, but it doesn't sound as though it can deliver enough current.
You didn't say whether you were using the 15k resistor across one rail (meaning from + to GND or from - to GND) or across both, but assuming that it was hooked from the + rail to the - rail, that would draw 8mA. A quick scratch calculation I just did indicates that the BOSOZ draws something on the order of 80mA per channel, or ten times what your dummy load was drawing. (If I got that wrong, I'm sure someone will give me a quick kick in the pants.) When you hooked the circuit to the power supply, it drew current faster than the power transformer could supply it and the rail collapsed. Two channels will draw 160mA, which is twenty times what the test load was drawing. If you can get a 750 ohm power resistor, that would be a better dummy load, but keep in mind that it will need to dissipate around 20W without burning up. A 50W resistor would be a good idea.
The regulators are of no help here. Their job is to keep the rail from going above a certain voltage, but they cannot supply extra electrons to help boost it back up if it falls below that level.
You said you changed your transformer...perhaps you should consider going back to the original one if it was working properly.

Grey

[vit]

Dear Gray

The new transformer should be more then enough. 100VA! The old ones are 2 x 120VA (2x30VAC). I want them for my xgc-project.
If the transformer is too weak, then the positive side would be affected, wouldn't it?

Hmm...now I recall that I swiched the IRF9610 for a better looking one. I'll try to replace it tomorrow.

Cheerios,
Vit

edit: 50W resistor? I made it with two 1K5R 2W. But only for a short time

[wuffwaff]

Hi Vit,

did you use the right isolation resistors between power supply and caps (27R I think)? Wich 220R are you speaking of?

William

[vit]

Hi William

The resistor I'm speaking of is R104 (221R) in the original schematics (http://www.passdiy.com/pdf/balzenpre.pdf)
Yes, I'm using 22,1R after the PSU.

Cheers,
Vit

edit: If the gate resistor is fried, wouldn't the gate of the irf9610 also be fried?

[wuffwaff]

Hi,

well if it´s the gate resistor no current should be flowing through it as the fet is a voltage controled device. If this resistor get´s hot the fet is probably faulty.

William

[GRollins]

vit,
You're right, 100VA is plenty. And, yes, both rails should behave pretty much the same. That's the downside of trying to deal with this stuff during off moments at work--I'm always on the run and miss details.
I agree with wuffwaff, no Gate resistor should ever burn. Something is likely wonky in the region of the negative rail pass device.

Grey

[vit]

New FET and 750R load:

-V=-58,9V

acceptable

Thx for your help guys

[GRollins]

Presumably both positive and negative rails are roughly equal, yes?
You can expect some variation from the nominal 60V figure due to parts tolerances on the Zener diodes, etc. If you want to get crazy about it, you can rig things a little differently and use a pot to trim the voltage in a little closer, but I wouldn't worry about it.
Go forth and listen. Troubleshooting is hard work. You deserve a break.

Grey