Vt40 bias problem, redplating output tubes

[JMMP]

Ampeg vt40, schematic calls for -65v on r49/c21. Most I can get is -51v. Tubes redplate within 5 mins of being on. Measured all components (in bias circuit) and all within spec. R38 is where I plan on putting bias pot in series. Can someone shed some light plz? Thanks in advance.

[The Dude]

Do you have a scope? If not, use a DVM and check for AC on the bias supply. I'm wondering if your bias filter caps are shot.

[JMMP]

Wondering the same, read 191v at r33-r32 at some point. Turned off then back on,... Back to 50ish. What's most frustrating are the 6 attached fasteners on that board, secured by tin spring threads. Need to get to the bottom of that board to check/replace components there.

[AL30]

I had a similar problem not long ago. Bias would be fine one minute and then jump and cause the tubes to red plate. It turned out to be a bad shorting jack - one of the ext. amp jacks (J3/J4). It wasn't grounding 100% of the time and would inject a signal back through the amp - whenever it felt like it. That was a nightmare to track down.

Those amps can be tricky. I've found an adjustable bias to be really useful. Check your B+ also, it's high to begin with but with newer wall voltages it can go much higher.

And keep an eye on the 3k/5W resistor in the power supply. I've had several Ampegs pop those - some blew up and some un-soldered themselves. Every time someone had cranked the amp.

Apologies I realize most of this won't help your issues but it may give you some insight in the future. Best of luck

AL

[g1]

C15 is critical. Also the bias filter caps The Dude mentioned, C14 and C21. Those two must have (+) ends going to ground.
Also, coupling caps from PI (C10 and C12) could cause issues.

[The Dude]

Good catch on C15, g1! I hadn't even noticed that it's a cap coupled supply. That definitely could be a problem.

[JMMP]

Thank you and sorry for not checking in sooner. Been struggling a bit with this, so took all the measurement's, reads fine except the bias drop/red plating and the resistor mentioned below.
Here are the 7027a tube pin readings.

Pin
1- 580 v
2- 3.3.v
3-. 586 v
4-. 585 v
5-. -48 to -70 v (this is what I'm chasing)
6-. --6 v
7-. -65 v
8-. -3.6 v
The large 10 watt 1k resistor reads 2.6k when hot and drops very quickly to about 1.2 or so while cooling.
Please advise if you folks have any knowledge of what this may be.
thanks in advance.

[Helmholtz]

You can't measure the resistor in circuit. Disconnect one leg and measure again.

Do you use real 7027s?

A 7027 has pins 5 and 6 internally connected, so should have same voltage.
Pin 8 is connected to ground and should have 0 DCV.
Pins 2 and 7 are heaters and should have 0 DCV- or are your numbers actually ACV?
What does the dot before the minus sign mean?

Either your meter is bad or the black lead wasn't well connected to ground.
Some cheap meters give a false DC reading when connected to an AC voltage (like with heaters).

As long as power tubes tend to redplate, pull them and cure the bias supply first.

[g1]

Did you check the caps from post #5 ?

[JMMP]

Some you can, some not. What I measured in circuit was same as out of circuit, I know many you can't. Using original USA Amperex 7027's, JJ's not as robust, red-plate earlier. Meters are 2 fluke 8060's, a Beckman lm22a LCR and diode, an Atlas ESR 70 cap checker, 2 hand held dvm with scope plus 440's, Simpson analog meter. All are calibrated yearly. Yes helmholtz, faulty original meter lead. So what I did was replace c25 with .1uf, bias came in pretty quick and solid. 50k trimpot in series with r38, bias now fully adjustable between -56v to -76v. Dots are just periods, mean nothing. Yep pin 8 is 0vdc. Not done with it yet, taking a bunch of readings to better familiarize myself with these models, have a VT22 on deck after this one. Again thanks to everyone who responded. I'm sure I'll have questions with the VT22 so stay tuned. Thanks all.

[The Dude]

......So what I did was replace c25 with .1uf......

I must be blind. Where is C25?

[JMMP]

Sorry, C15.

[Helmholtz]

Make sure its voltage rating is high enough.

[JMMP]

C15, fat fingered it

[JMMP]

Yes, what's strange is that all the parts coming out are 450-500 volt. Just measured 603 volts on c19!