Blowing fuses

[martin manning]

Well, that eliminates that. Another idea: In standby mode put your meter on the bias tap and then switch to run and see where that voltage goes. A few seconds will probably be enough to see if it stays at the expected ~60 VAC. If it does then the transformer is (probably) not at fault.

[gui_tarzan]

Goes from 61v to around 58v as it warms up and as usual, blows the fuse. Transformer is probably ok, bias board eliminated, I'm going to disconnect the B+ from the main board and try it again.

[gui_tarzan]

Nah, that's not gonna help, it can't get power to the tubes that way. I'll get back at this tomorrow night. Starting to get very tired and that doesn't mix well with high voltage.

[martin manning]

Can we see where you are now? With the power tubes removed, power up and switch the standby to play. Check and post voltages at the plate and screen nodes, at the bias filters, and at the power tube grids.

Is there any evidence of carbon traces on the power tube sockets?

[gui_tarzan]

Ok, here's what I have without any tubes. And the sockets all look fine, no visible tracks.

V7-V10 pins 3 & 4 all have 478v. Screens all have -20 to -22v. The negative end of the bias cap still shows -51v. The picture below shows voltages on the filter caps.

[martin manning]

By screens you mean grids, pin 5's? For the grids, -20V is not low enough.

Per the schematic, the first bias cap (on the rectifier board) should be at -52V, and the second bias cap (at the balance pot's wiper) should be at -42. What do you get there?

What is the voltage at the outer lugs of the bias balance pot? Is the pot centered? If the voltage goes down at the balance pot then that suggests a conductive board or some low impedance path to ground.

I believe your drawing is missing a brown wire from the 2k2-10k junction. It should go out with the black, red, and yellow leads and connect to the PI plate load resistors.

Out of curiosity, where are the orange and green leads grounded?

[Stevem]

Add another wire and pull off that -52 volts and feed it to all the grids, does it still pop the fuse, I bet not!

[gui_tarzan]

Orange wire goes to the brass plate at the front of the chassis, the green wire is grounded next to V7.

And yes, I forgot that brown wire, just imagine it's there.

Screens, grids, I always mix those terms up. Sorry.

The first cap has always shown around -50v. The front wiper and the wiper off the back of the bias pot is reading -2.8v. Hm.

The pot is centered, leads 1 and 3 show -43.5v.

[martin manning]

So from -43V at the 1 and 3 lugs it goes up to -22 at the grids. Do you still have the HV leads off the board?

[gui_tarzan]

Nope. Everything is hooked up as normal right now.

[gui_tarzan]

Something just occurred to me - the PI plates have 245(ish) volts of DC that go to the .1 caps that filters DC out and only lets the AC signal through, correct? So then from the bias pot we have -50(ish) volts DC coming from the bias supply to the output side of the .1 caps and going into the grids.

If what Steve suggested by taking those bias leads off the grids and providing the bias voltage straight from the bias board to the grids that effectively disconnects the PI connection to the grids. No signal flows but neither does any stray DC that might be getting through the .1 caps, correct?

[gui_tarzan]

So I moved the grid bias leads to the bias board at the junction of the negative end of the cap. I get as I should, a solid -50v on all four tubes. No surprise there, but the fuse isn't blowing either.

Then I checked the voltage on the output side of the .1 caps from the PI and get varying readings of 20v-40v from pin 1 and -8v to 4.0v from pin six and it jumps all over the place in those ranges. Leaky caps? Sure looks like it to me, but why would that cause the fuse to blow?

[JerryFJA]

This makes no sense. The outer terminals of the pot feed the power tubes and you said those have -43v. The wiper on back is the raw bias voltage and you say that is -22v or -2.8v not sure which one is correct there.

Did this amp blow fuses before you changed the caps?
Is the bias cap on the pot installed with the correct polarity, positive to ground?

Didn't you say further back you changed those PI coupling caps already?

[martin manning]

Could be leaky caps, or the eyelet board or the backer board is leaking DC into those eyelets from the several points where the high voltage is connected. You could lift the ends of the coupling caps and check for leakage by connecting a 1M to ground and measuring DC voltage. I don't know why you don't see bias voltage at the bias pot's tap. Are you sure that's right?

Try jumping over the bias pot by shorting all four of its lugs together.

[gui_tarzan]

This makes no sense. The outer terminals of the pot feed the power tubes and you said those have -43v. The wiper on back is the raw bias voltage and you say that is -22v or -2.8v not sure which one is correct there.

Did this amp blow fuses before you changed the caps?
Is the bias cap on the pot installed with the correct polarity, positive to ground?

Didn't you say further back you changed those PI coupling caps already?

This came to me with the symptom of blowing fuses. Two of the filter caps were splooging out the end and two were wrong, so I replaceded all of them with new caps.

The latest voltage readings are what I posted a little while ago. I know that some of the voltages have varied, that's still a mystery to me.

Yes, that cap is positioned correctly, it's the original. And yes, I did change the cap on the PI pin 6 but not the other one.