Merlin's Standby w/ Choke Input

[hylaphone]

Would a "soft start" standby be possible for choke input?
The Valve Wizard

Voltage at startup definitely exceeds the cap ratings before it settles, and I'd rather not run series caps through the whole supply..

Also, amp in question is using a 5R4 rectifier -- after a choke input, is it safe to bump up capacitance way beyond the "input cap" specs?

[R.G.]

Don't ever put a cap in a position where it will be regularly over-volted. It may survive for a while, but it's long term life will be affected. I have not done any serious thinking about the resistor setups recommended on the link you provide, but I have designed a soft-start for coming out of standby.

My soft start was intended to clamp the current going to the cap when switching out of standby, and a fancier version includes standby. See
Return of the More MOSFET Follies
and
A Tube Amp Current Clamp

These approaches of course involve semiconductors and are appropriately shunned as blasphemy by all True Tube Amp Believers. About the only thing good about them is that they work.

It's worth thinking about what happens when you connect voltage to an L-C filter. Initially, the voltage is all dropped across the inductor, as the current in the inductor cannot change from the initial 0ma instantly. The current in the inductor ramps up as dI/dt = V/L. The voltage across the cap is zero, and rises as current fills it. It only rises above the supply voltage if the circuit rings, and this in turn depends on the load on the cap "damping" the ring.

[hylaphone]

So in other words, B+ will approach full rectified value until the amplifier begins to draw current...?
And Merlin's resistor-across-the-switch won't do the trick here.

[R.G.]

OK, so I went off and read Merlin's send-up on standbys.

Would a "soft start" standby be possible for choke input?

Yes, it's possible. Merlin's comments are largely intended to convince you that you don't need a standby, but if you're determined to have one, there are some less-bad ways to do it.

So the first question is - after reading that, why do you want a standby?

The second question is - what does a power standby do for you that a mute switch on the signal doesn't do?

Also, amp in question is using a 5R4 rectifier -- after a choke input, is it safe to bump up capacitance way beyond the "input cap" specs?

Probably. The problem with tube rectifiers is that after a certain peak current, the cathodes get damaged. The inductor will limit currents in most cases. After a certain critical current, the inductor makes the current into the cap -constant- at the average DC current value, so it's not running peaks at all.

But is there a good reason not to use one of Weber's solid state replacements for the 5R4? They're quite highly thought of by people who (1) have open minds (2) have tried them and (3) don't sell rectifier tubes.

The approaches to standby and current limiting that I suggested are the direct approach: use a big honking MOSFET to control the current. They directly prevent excessive currents. They're outside the morally-correct mainstream of "do it all with tubes", which is why I call the genre heresy and folly. But they do work.

As a side light, I'm guessing that you went to choke input because your power transformer has too high a secondary voltage for the amp you have after it. If that's right, wouldn't it be simpler to find a right-voltage PT?

[hylaphone]

Ok, the context is this PA amplifier:

(Had another thread going about that R25 screen supply resistor, hope it's not indecorous to consolidate my questions here)

I am re-purposing the amp and would like to squeeze out a bit more power. I replaced that interesting screen setup with more standard RC node to bump up screen voltage, and output power increased accordingly. Would like to change to fixed bias as well.

Problem is, voltage now soars at startup until the tubes start pulling current. Did that screen supply resistor provide some instantaneous load or something for the power supply? The theory is beyond me.

In any case, I CAN run series caps for the whole supply, but if there were a way to implement a "soft" startup to allow the caps to charge without exceeding their ratings - then I'd save parts and real estate.

I don't care about the standby per se, just want to do whats practical for the filter caps.

[g1]

Problem is, voltage now soars at startup until the tubes start pulling current. Did that screen supply resistor provide some instantaneous load or something for the power supply? The theory is beyond me.

That resistive path (R24,25,26) put a constant load on the supply. In the other thread Gingertube mentioned a bleeder current of 35mA. If you changed the screen supply, you probably no longer have that load on the supply. You could make one with power resistors so you get the same kind of bleed current partially loading down the supply.

[catalin gramada]

I don't care about the standby per se, just want to do whats practical for the filter caps.

It will efective protect you filter caps at starting and the protection in HT circuit can become more effective using a fast fuse against a standard "T" one. I use all the time a 10K/3W across the stb. switch between secondary winding and rectifier position

[hylaphone]

That resistive path (R24,25,26) put a constant load on the supply. In the other thread Gingertube mentioned a bleeder current of 35mA. If you changed the screen supply, you probably no longer have that load on the supply. You could make one with power resistors so you get the same kind of bleed current partially loading down the supply.

Ah!
Did some math this morning, and with a tap on the voltage divider for heater elevation, I ended up with nearly the same values as the original schematic.
It's almost as if RCA knew what they were doing

Thanks for pointing out something obviously obscured by flux smoke