Flyaway DC ... what could it be?

[Diverted]

Hi,

This is the strangest thing I think I've seen.
I had to put a new socket in a beater CBS champ I picked up cheap. I did that yesterday and all was well. Amp worked great.
Turned it on today, no volume. Checked B+ and got strange numbers on my multimeter.
Powered down, unsoldered all outgoing B+ leads from pin 8 of the rectifier, so no DC was getting out of tube to the circuit. Powered it up slowly on Sencore PR57 (iso transformer/variac). Started at 0 volts and all was normal until DC voltage started escalating. By the time I cut power at 50VAC, it was close to 800vdc. Yikes.

Suspected bad tube socket. Replaced it. No change.

Then started removing everything but the essentials. Took fuse, 6.3V supply and power swith out of circuit, so nothing was connected except AC to primary, 5V to pins 2 and 8, HV secondary to pins 4 and 6, HV center tap to ground. Still no change. Insanely high voltage.

Tried going through wall AC thinking there might be a fault with Sencore unit. I only had it on briefly but after a few seconds it had shot to nearly 1000V. Crazy. So it's not the power supply.
Multimetered AC at wall and at Sencore. Normal 120vac.

Tried two different rectifiers, Same.

Tried several different multimeters. Same.

I started thinking I might have a bad power transformer so I took a brand new unused one off the shelf (correct PT for this amp) and installed it in its place. With only 5v, hv secondary and hv secondary ct attached, I am still getting the same crazy high numbers.

So now I don't know what to do. I think I've eliminated every variable I thought there is, but obviously something is still wrong. Where next? Thanks.

Photo shows current state of amp. Rectifier and HV center tap visible. Apart from that the only thing currently in circuit is the AC in, switch and primary.

[Dave]

You can't get accurate readings if you don't have at least one filter cap connected and functioning.
Also, I agree that those voltages seem insane, but without a load, you still can't get an accurate reading.


Dave

[Diverted]

Thanks Dave. I know the number would be higher with no cap. But these numbers are at least 2x what I thought I’d see.

I just ran a 500v cap from rectifier 8 to ground. No change.


Without a rectifier in I’m getting 350vac on each plate at 120 wall ac. So that makes me wonder even more. But I have now tested and gone through three nos rca 5Y3s and a JJ. All test good.

[Dave]

It sounds like the PT is fine. I don't see how you could possibly get a real reading without putting it back together. I would definitely bring it up on the variac though. I can't imagine how a bad rectifier tube could cause your symptoms, but there's always the option of subbing in a couple of diodes just to make double sure.


Dave

[Diverted]

I agree it looks like the PT is fine too ... both of them.
The exact symptoms happen with each.

I tried using a few diodes to rectify per suggestion. This was with the B+ disconnected from the circuit, so I was measuring where the diodes coming off the HV secondary lines meet. Once I hit about 40VAC input voltage through Sencore variac/iso xformr the DC spiked super high... only saw it for a second before multimeter shut off at 1000V. This is exactly what happens with a 5Y3. It also happened plugged straight in to wall 120 so I am ruling out Sencore as problem.

I just don't see how it can be anything in the B+ rail causing this as it also happens with the entire circuit disconnected from B+. It also happens with the circuit connected as it should be.

I would love to be able to take voltages, but that's impossible. I can't increase incoming voltage close enough to 120vac to get relevant numbers. So I'm kind of stuck.

Unless both transformers read good (about 350 VAC to ground on either leg with rectifier pulled, and 5V on the heaters with rectifier pulled) but are bad, or both sockets I put in are bad, or all the rectifier tubes I've used are bad but read good on tester) ... it's got to be something else i'm missing. For the life of me I have no idea what else it could be.

Here are some pictures. I'm getting increasingly worried about what can be done about this. Thank you for looking and helping, if you spot something!

PS: Forgive the messy solder work. I have attached and detached power rail jumpers about 100 times today and they are getting raggedy. But the connections are OK from looking and continuity testing.

[shooter]

sounds like your cap filter circuit has lost reference to ground
got a schematic?

[Diverted]

Sure shooter, thanks. Here's the schematic. It isn't great. The original PT has the two winding primary as shown bottom right center in schematic, but currently I am using a different transformer to try to eliminate issue before I put original PT back in.

To your comment on failed/lost ground reference: Can you explain why you say that? I'll look for an issue there, but the problem exists even with no filter caps in the circuit — testing just the rectifier, reading DC at pin 8 disconnected from the rest of the circuit, it happens as well. So wouldn't that suggest the problem is elsewhere?
Thank you.

PS I did check all grounds in the amp ... electrolytics, cathode bypass caps and every other cap/resistor that goes to ground. All good. Also did not find any spots with continuity to ground that shouldn't have it.

[shooter]

problem exists even with no filter caps in the circuit — testing just the rectifier, reading DC at pin 8 disconnected from the rest of the circuit, it happens as well.

missed that, although in that "config" I would expect voltages to be as much as 2X "normal"

[Diverted]

I just put the original PT back in.
Again, disconnected everything from the rectifier, and all filter caps out of sheer desperation. The most recent reading I took yielded 680VDC at the rectifier, with about 30VAC coming in (I slowly ramped variac up from 0). Something is definitely very very wrong. HV secondary showing around 350 per plate with no rectifier.

Here's a photo of it I just took with original PT.
Anyone see anything amiss here? I have looked at it dozens and dozens of times and can't find anything that looks wrong.

[Diverted]

Can't find any answers looking at the amp so I start second guessig myself — did both transformers fail in the same way? Is that what is going on?

So I just swapped in a THIRD transformer, brand new Hammond pulled it right out of the box. With rectifier connected to nothing, I'm measuring that same insane voltage on 5Y3 pin 8.

I ran this PT from outside the chassis, AC wall voltage straight into the primary bypassing switch and fuse. I hooked up the 5V to pins 8 and 2 rectifier, HV secondary to pins 4 and 6, and HV center tap to chassis ground.

How can it possibly be that I'm getting these crazy voltages? The 5V heaters are only getting about a volt so I would think the rectifier wouldn't even work? But at about 1 volt, 40 or so VAC incoming from variac, voltages are off the chart.
My multimeters (I have used two) max out and beep at 1000VDC. I'm getting almost an immediate beep meaning it hits 1KV almost instantly. What the heck?

[shooter]

we'll hold off on operator error 


take ANY PT, bench wire it without an AMP, so 120VAC fused to PT, ALL secondary leads exposed for measuring BUT taped, glued, fixed somehow to a NON-conductive "test bench"  I use a chunk of 2 X 4


now measure VAC from ALL the lead "pairs"

[Diverted]

Operator error definitely a possibility but I did those tests with the first and second replacements. I can do it again no problem on either, both or all three.

Here is how I've wired everything up:

HOUSE AC CONNECTION:
1. white primary to line white (neutral)
2. black primary to line black
3. ground earth to chassis ground

Tests I did on each:
1. HV secondaries (red/red) to each multimeter probe. First replacement, got about 700. On the second, got about 660
2. Also metered HV secondary leads one at a time, to one probe, HV center tap to the other. First replacement, about 350 and second, around 320-330
3. I tested each PT's 5V winding as well: Yellow to one probe, yellow to another.  In each case, got about 5.6-8v or so.
4. Tested each 6V filament winding. Green to one probe, green to the other. Got about 6.6 each time unloaded

I have been getting the same readings on all three power transformers with the PT wires connected to correct spots (HV secondaries to rectifier pins 4 and 6; 5V winding to pins 8 and 2). I get proper readings in all three cases with no rectifier.
The numbers go haywire as soon as I plug the rectifier in. It does not matter whether or not I have the rectifier connected to first filter cap, output transformer primary and first dropping resistor, etc (the rest of the circuit). When I have a rectifier in, connected to the rest of the circuit or not, I get the crazy DC numbers that make both of my multimeters max out, beep and shut down.

The only time I get appropriate voltages, PT mounted or on its own, is when there is no rectifier inserted. This was also the case when I ran diodes off pins 4 and 6, connected the other (striped) ends to meter.

I can do any test anyone needs to see, again. I just cannot figure this out. Everything looks correct but obviously something is amiss. Definitely could be operator error but I think I'm pretty clear on where the power/PT wires should go. It just doesn't make sense.

Thanks for the help.

[Willabe]

Hook the fuse back up, that's not safe.

Stop testing it without a cap, that useless.

Do you have a light bulb limiter? I'd like to see what happens with that without using the variac.   

[Willabe]

I had to put a new socket in a beater CBS champ I picked up cheap. I did that yesterday and all was well. Amp worked great.

Did you take voltage readings after you installed the new socket?

[Diverted]

First off, thanks for everyone’s help.

Second of all, low batteries sure can mess things up.

I decided to take a known working amp I have and connect multimeter to rectifier only, and see what I got. Same nonsense mega high voltage I’ve been getting all day resulted.
So that pointed to the multimeter.

I tested again with a second multimeter. Same thing.

I had replaced batteries in both units earlier today, but I took one pair from our TV remote and the second from our junk drawer.

Got in the car, went to the store, bought new AAAs, went home, tried them out.

FINALLY normal voltages in the known good amp. Tested the problem amp with everything connected, voltages didn’t rise any higher than 480 before settling down to about 410 B+ as the tubes warmed up.

Hallelujah. What a fun way to drop eight hours going over phantoms over and over lol

Thanks for everyone’s willingness to help, and sorry for wasting your time. I thought I was going crazy.

[Willabe]

Glad you got to the bottom of things.

Could you measure the dcv on the bad batteries and post the numbers? I'd like to see them for future reference with guys here.

[Diverted]

Sure, but it will take a bit. We have an all day thing tomorrow, work Monday so after Monday night👍

Thanks.

[Dave]

Second of all, low batteries sure can mess things up.

That must have been a Fluke. LOL


Dave