Blues Jr has Low B+ on V4 when phase inverter is plugged in.

[thelxinoe]

I recently repaired an amp with a couple of blown screen resistors and everything seemed good. Powered it up and measured the B+ at both CP3 and CP1 and at first all seemed good, it was even nicely biased. Then after a few seconds the B+ at CP3 (schematic for reference https://www.thetubestore.com/lib/thetubestore/schematics/Fender/Fender-Blues-Junior-Schematic.pdf ), the blue cable, drops from roughly 350v to about 80v. If I remove the phase inverter all is good and both B+ read at 350v. I'm a little stumped at this stage pin two on V5 is 5.34v and on V4 it's 5.13v. I think the amp plays fine, makes a fairly pleasing if thin tone, but I've never played a blues jr so I'm not sure.

{EDIT- space around URL so it clicks -- PRR}

[pdf64]

Probing an output valve anode can invoke oscillation, thereby invalidating the reading.
Pulling the valve in the phase splitter circuit breaks the positive feedback loop, stops the oscillation.
I think it’s weird how everyone perceives the output valve anode voltage to be a key metric. It only gets questioned if, as here, it’s way off.
I only ever measure the HT voltage at the OT CT, I avoid the anodes.

But why is the grid bias so low?
What’s the TP30 (bias) voltage?

https://www.thetubestore.com/lib/thetubestore/schematics/Fender/Fender-Blues-Junior-Schematic.pdf

[sluckey]

I'm a little stumped at this stage pin two on V5 is 5.34v and on V4 it's 5.13v.

That voltage should be -10V. Should never be positive.

But even if you really meant you have -5.34v and -5.13v, that's way too low. Fix that first.

[PRR]

......really meant you have -5.34v and -5.13v, that's way too low.

Might ask what meter was used. A 200kOhm meter would read half voltage here. A 10k/V@20V, or 20K@50V, needle meter could be in that range. They aren't all ten dollar junk.
https://www.amazon.com/Gardner-Bender-GMT-319-Multimeter-Tester/dp/B0002YUN4C

https://www.amazon.com/Hioki-3030-10-Manual-Ranging-Average-Sensing-Multimeter/dp/B008S0CCHI
https://www.sanwa-america.com/products/yx360trf-analog-multimeter-with-built-in-case

Also I have seen (may still have) a good brand DMM made *for* electricians which defaults to a rather low (by electronic technician standards) hundreds-kOhm impedance on Volts. This masks the small "ghost voltage" when you have an unterminated conductor in a cable with 120v or 240V conductors.
EDIT- no, the Klein CL2000 input impedance, ACV or DCV, is about 8.5 Megs. Not the 11Meg of classic VTVM or the 10Meg of most DMMs, but close. (And if you look for CL2000 on HomeDepot, they have it cross-linked with another product.)

[thelxinoe]

Probing an output valve anode can invoke oscillation, thereby invalidating the reading.
Pulling the valve in the phase splitter circuit breaks the positive feedback loop, stops the oscillation.
I think it’s weird how everyone perceives the output valve anode voltage to be a key metric. It only gets questioned if, as here, it’s way off.
I only ever measure the HT voltage at the OT CT, I avoid the anodes.

But why is the grid bias so low?
What’s the TP30 (bias) voltage?

I measured the voltage from the OT from P1, the blue wire, and it's about 80v often dropping to 0v when the PI is in. Does that also cause oscillation? I'm unsure where to measure to get a good reading if so.

TP30 measures at around -26V which looks correct according to the schematic.

[thelxinoe]

......really meant you have -5.34v and -5.13v, that's way too low.

Might ask what meter was used. A 200kOhm meter would read half voltage here. A 10k/V@20V, or 20K@50V, needle meter could be in that range. They aren't all ten dollar junk.
https://www.amazon.com/Gardner-Bender-GMT-319-Multimeter-Tester/dp/B0002YUN4C

https://www.amazon.com/Hioki-3030-10-Manual-Ranging-Average-Sensing-Multimeter/dp/B008S0CCHI
https://www.sanwa-america.com/products/yx360trf-analog-multimeter-with-built-in-case

Also I have seen (may still have) a good brand DMM made *for* electricians which defaults to a rather low (by electronic technician standards) hundreds-kOhm impedance on Volts. This masks the small "ghost voltage" when you have an unterminated conductor in a cable with 120v or 240V conductors.
EDIT- no, the Klein CL2000 input impedance, ACV or DCV, is about 8.5 Megs. Not the 11Meg of classic VTVM or the 10Meg of most DMMs, but close. (And if you look for CL2000 on HomeDepot, they have it cross-linked with another product.)

https://www.amazon.co.uk/gp/product/B01N35ZVKY/ref=ppx_yo_dt_b_asin_title_o04_s00?ie=UTF8&th=1

That's the multimeter sadly not sure if that would produce the effect that you're talking about but considering most other things seem ok maybe it's fine?

[thelxinoe]

Also as a general question if it is just oscillation why would this occur on only one power tube?

[pdf64]

Maybe it is, but not to such an extent?
If it isn’t, and the only way to be sure is to check with a scope, then the conditions weren’t suitable.
But that doesn’t preclude oscillation in other conditions.
Valve guitar amps are prone to oscillate.

[PRR]

> why would this occur on only one power tube?

Whatever you have there, one plate makes positive feedback, the other plate makes negative feedback.

But why are you measuring the plates at all?? The DC has to be essentially power supply voltage (transformer DC drop can only be a few Volts). The AC voltage can peak to twice the DC voltage, even exceeding the 600V limit on that meter.

> I think the amp plays fine

What are you trying to improve?

[thelxinoe]

> why would this occur on only one power tube?

Whatever you have there, one plate makes positive feedback, the other plate makes negative feedback.

But why are you measuring the plates at all?? The DC has to be essentially power supply voltage (transformer DC drop can only be a few Volts). The AC voltage can peak to twice the DC voltage, even exceeding the 600V limit on that meter.

> I think the amp plays fine

What are you trying to improve?

It was broken when I got it and during the repair I was checking the bias current as the amps often "run hot". Seems like the usual way to do this is to compare the voltage from CP2 and CP3 then divide by 100 to give a rough estimate of the currant but then ran in to this issue. 

[pdf64]

… Seems like the usual way to do this is to compare the voltage from CP2 and CP3 then divide by 100 to give a rough estimate of the currant but then ran in to this issue.

It would be accurate to divide the CP2-CP3 voltage by the CP2-CP3 resistance.
I suppose the typical CP2-CP3 resistance is about 100ohms?

Alternatively, use the CP2-CP1 voltage divided by the CP2-CP1 resistance, which may work better for you.

It’s worth noting the the latest 2018 onwards version 4 BJ seems to use cathode bias, 270ohms bypassed per EL84 cathode, which seems a far superior choice for EL84 in hot AB. It’s regrettable that wasn’t done much earlier.

[tubeswell]

So the phase inverter in this case is a LTP.


And when its pulled out, both power tube voltages are good, but when its plugged in, they're not.


So look at what's going on with your LTP. (Hint, are you measuring the plate voltages on your LTP?)