why is my cap busting?? care to speculate??

[cooldude666]

Hello friends, I wired up this bias supply, but for some reason the first 22uf cap coming off the rectifier, the reservoir cap, keeps bursting. Has happened to me twice now. The cap is not in backwards (at least not according to the diagram). Is the schematic in the book wrong? Should it be connected the opposite way?

This started happening when I began trying to adjust the voltage divider to give me my desired range for my 6550s. Right now I've got a 22k resistor in series with lug 1 of the 10k pot. Seemed like it was working too! Until it popped. I'm using 10uf caps because that's what I had on hand but I don't think it's related. Stinks, was my last appropriately sized electrolytic too (250v, orange and black Sprague atom, for Halloween )

Thank you!

[pdf64]

The schematic shows the caps with opposing polarities - one of them is going to be in trouble.
Their positive terminals should be connected to ground.
Pete

[cooldude666]

So why isn't my voltage divider working this way? The pot seems to do absolutely nothing now. Its just sitting at the full DC value of ~84v

[pdf64]

I don't see the purpose of the diode off the pot wiper; how about if it's removed?

[cooldude666]

Wow, yes that did it. Is it possible to alert Merlin to this thread somehow? The diagram is from his book I'm curious what he has to say.

[Jazz P Bass]

Removing the (what is that doing there) diode & now it works makes no sense.
The diode provided a 0.6 volt drop.
From the drawing it appears that the anode of the diode is attached to the 100K resistor.
That is wrong.
The wiper should be all by it's lonesome.
If the anode was connected to the 100K resistor, that is why removing the diode worked.

[cooldude666]

The anode was connected to the 100k resistor, as in the diagram

[trobbins]

The diode on the wiper allows for 'normal' charging of the bias voltage cap when power is applied, but slows down the discharge of that cap when power is removed.

If you wire up the circuit with just the first cap after the bridge, then you should get a negative voltage on the end of the cap that is not connected to ground (0V). Adding the divider network of resistor and pot will lower the voltage on the first cap. Adding the cap from 0V to wiper will generate a neg bias voltage across that cap. The 100k is a failsafe resistor if the wiper contact is poor, and allows the bias cap to discharge when mains is off.

[potatofarmer]

Merlin has some notes on the errors in this circuit here: The Valve Wizard -Cathode Follower

[Jazz P Bass]

Here is the goods on the revision.

[cooldude666]

ah! thanks. good stuff.