Strange plate/screen supply issue

[fdesalvo]

Hi, all.

Another day, another strange amp issue. I've been working on many old amps for a currently defunct amp co, whose og owner has passed on. The new owner has supplied me with a bevy of old prototypes to fix and it seems every 1/20 presents an interesting issue I've not yet encountered. This particular amp is working well with one exception.

During warmup (no standby switch) the plate voltage will increase to 380vdc, then drop down to 320vdc, when there when an audible pop will accompany the voltage instantly rising to 384vdc - theres a .x volt wiggle here. Meanwhile, the screen supply (at cold start) jumps to 400vdc, then steadily drops to it's destination of 380vdc.

All of these amps are built in true proto fashion and it’s become obvious the original owner/engineer worked for Leo (which he did). The amps are a tac-soldered hodgepodge of re-appropriated eyelet boards and old school astron, ajax, and blue/brown turd caps from the 60-70s. It's as though the vintage component storage facility blew up next to this amp. All of these which have failed have been replaced. Resistors are great. I'd suspect the JJ kt77's at this point, but both are on the warmup roller coaster. Have any of you seen this type of behavior before? TIA for looking.

Best,

fd

[Jon Snell]

How is the HT derived ... valve rectifier or SS rectifier?
If the rectifier delivers voltage before the output valves draw current, the HT will start high, drop quite a way and as the rectifier warms up, gradually recover. If silicon rectifier the HT will start high and gradually drop, as the output valves start to conduct.
If the output valve bias is not designed correctly, (too high a value smoothing capacitor or feed resistor) the -ve voltage may take too long to reach the desired point that correcly controls the current flow in the output valves.

Who knows with no schematic or quality ptotos, just guess work.

[fdesalvo]

How is the HT derived ... valve rectifier or SS rectifier?
If the rectifier delivers voltage before the output valves draw current, the HT will start high, drop quite a way and as the rectifier warms up, gradually recover. If silicon rectifier the HT will start high and gradually drop, as the output valves start to conduct.
If the output valve bias is not designed correctly, (too high a value smoothing capacitor or feed resistor) the -ve voltage may take too long to reach the desired point that correcly controls the current flow in the output valves.

Who knows with no schematic or quality ptotos, just guess work.

Two 5408s rectify AC. The bias cap is a 60uf cap. I routinely use 100uf caps in this position without issue. Ty for posting.

[Helmholtz]

The bias cap is a 60uf cap. I routinely use 100uf caps in this position without issue.

Too large capcitance in the bias supply will slow down bias ramp-up and thus may lead to excessive power tube current during warm-up.
Consequently B+ will drop/sag temporarily.

[g1]

Can you monitor the (power tubes) idle current during the cycle, as well as the bias voltage at KT77 grids.

[Helmholtz]

To exclude a temporary oscillation during warm-up see if pulling the PI makes a difference.

[fdesalvo]

Can you monitor the (power tubes) idle current during the cycle, as well as the bias voltage at KT77 grids.

Power on - plate volts hit 400vdc;
10 seconds later, plate current rises and stays steady. at 340vdc @ 10ma
15 seconds later - "Pop" occurs and values suddenly rise to 380vdc @ 40ma. Heater voltage increases by .2vac at this time, as well.


Helmholtz - no change in condition without PI tube installed.

TY all.

[Helmholtz]

As this gives no clue I would monitor also the other power tube(s).

[fdesalvo]

As this gives no clue I would monitor also the other power tube(s).

Conditions are identical across the other power tube. Thank you.

[Helmholtz]

Conditions are identical across the other power tube. Thank you.

So if its not power tube related, pull them and watch B+ ramp-up without them.

[fdesalvo]

So if its not power tube related, pull them and watch B+ ramp-up without them.

It could be! I don’t have any el34’s or 77’s on hand to rule them out. The issue doesn’t occur without them. I apologize for not including that within my first post.

[Helmholtz]

But then at least one of your power tubes must behave different during warm-up, i.e.drawing excessive current.

[fdesalvo]

But then at least one of your power tubes must behave different during warm-up, i.e.drawing excessive current.

just returned home and tested bias rise in both during warmup. You NAILED IT. One tube hits over 100ma and then crashes down to 40 as the other just climbs to 40. The excess current was pulling down the plate voltage. Thank you. Such a woodshed working on these amps. Thank you guys. Learning so much now.

[The Dude]

Your picture in post #3 shows what looks to be a screen resistor that's been hot- possibly burned up or out of tolerance. I'd certainly check or replace it. That is also a clue indicating a potentially bad tube.

[fdesalvo]

Your picture in post #3 shows what looks to be a screen resistor that's been hot- possibly burned up or out of tolerance. I'd certainly check or replace it. That is also a clue indicating a potentially bad tube.

It’s a shadow cast by the filament string. It jumped out at me too on initial inspection.