Do you think this would work

[Leevi]

I have wired an SE amp to which I have added a power tube selector.
The selector works but there is a very loud popping when I switch between the tubes.

I attached the current circuit (Alternative 1) and my proposal for the new circuit (Alternative 2)
Before I start to wire it I thought to ask your opinion.
Do you think I'll get rid of the popping in the Alternative 2?

/Leevi

[Leevi]

does the amp have a standby switch

Unfortunately not, there is no place for stand-by in the panel

Does it make sense to turn off and on the heaters?

Yes, this could be an option

Make sure you have enough 6.3 v amperage to supply both tubes

The PT supports the higher power draws. Of course the "carbon footprint" would be less
if I turn off the heater of the inactive tube, which is a good idea.

/Leevi

[kagliostro]

Hi Leevi

yoy can try to add two large resistors (500k - 1M) between ground and the two (other) contact of the switch (option 1)

K

[PRR]

It's going to pop, whichever way you switch the cathode networks, or add bleed resistors.

Agree with drgonzonm. Switch the heaters. One tube fades out, the other tube fades in. There may be a few seconds of no sound. So you can't use this trick to jazz-up three notes of your solo, but certainly switch between songs, and maybe (with planning), in a rest before/after your solo.

The alternative is two complete always-hot power stages and a dummy-load for the unused stage.

[Leevi]

yoy can try to add two large resistors (500k - 1M) between ground and the two (other) contact of the switch (option 1)

I have tried this with pretty low resistor values but it didn't help.

/Leevi

[stratele52]

Two switch in one for standby / power .

I use in many amps ;

www.tubedepot.com/p-dpst-3way.html

Kagliostro are right with resistor

[tubeswell]

What about something like a 250k pot with the end lugs connected to each cathode rk/ck and the wiper connected to ground?

[Leevi]

Which tube is your O/T selected for (Primary impedance)?

It's closer to EL84 i.e 5000/8 Ohm. I use Hammond 125DSE where you can select different primaries.

/Leevi

[Leevi]

Agree with drgonzonm. Switch the heaters. 

I added this change and the popping disappeared.

One tube fades out, the other tube fades in. There may be a few seconds of no sound.

The operating tube doesn't fade out since also the cathode is disconnected. I think it is not good to feed a cold tube.

Thanks for your tips
/Leevi

[HotBluePlates]

I think it is not good to feed a cold tube.

It won't hurt anything. With the heater cold, the tube is just bits of metal in a glass case.

[kagliostro]

What about to use a 3PDT Switch ?

When the 1M resistor charges via B Switch the capacitor

C Switch disconnect the parallel resistor

so the resistor can't act as a bleed resistor

EDIT: Posted image for 3PDT solution was wrong, look to this schematic

K

[Leevi]

It's going to pop, whichever way you switch the cathode networks, or add bleed resistors.

Kagliostro, I refer to what PRR mentioned (above). There is in your proposal still a switch that breaks the cathode
which will probably cause a pop there.

In order to be 100% sure about its behavior it should be tried out.

/Leevi

[Leevi]

Are you saying that the cathodes may be poisoned with the heaters off?

I was worried about the situation where I feed the plate of a cold tube (no heaters connected) with high voltage
and the amplified signal from preamp goes to the grid. That's why I thought to disconnect the corresponding cathode.

According to HotBluePlates "It won't hurt anything". Maybe I should switch the heaters only?

/Leevi

[DummyLoad]

perhaps this might work?

--pete

[Leevi]

I have tried this with even lower resistor value but it didn't work
/Leevi

[Jack1962]

Which tube is your O/T selected for (Primary impedance)?

It's closer to EL84 i.e 5000/8 Ohm. I use Hammond 125DSE where you can select different primaries.

/Leevi

this may be a very stupid question but here goes anyway. are both tubes plates connected to different windings on the primary of the OT ? If so why couldn't you just switch the grids ?

Ok never mind I just looked at the wiring diagram

[Leevi]

are both tubes plates connected to different windings on the primary of the OT

No, there is only one winding on the primary side. The primary impedance is set on the secondary side
by selecting right load.

If so why couldn't you just switch the grids

Could work

/Leevi

[kagliostro]

If you switch grids tubes are always on the circuit, so you must remember the effect that it has on the impedance required for the primary

as to switch between grids may be you can try to do it via a pair of LDR, one quickly fade out the other quickly fade in, may be a very small part of the signal is anyway present at the output of the LDR that is out, but I think that in this situation this problem will be of very low importance

K

[DummyLoad]

I have tried this with even lower resistor value but it didn't work
/Leevi

make-b-4-break switch then... 

--Pete

[Leevi]

If you switch grids tubes are always on the circuit, so you must remember the effect that it has on the impedance required for the primary

Yes, If I now use 5000 Ohm/tube when they are not operating parallel. In the grid switching solution it should 10000 Ohms since
the tube are operating parallel.

Here is one solution. The pop might exist here as well?

/Leevi

[Leevi]

I made the changes according to your proposal i.e. I switch the heaters only.
I measured cathode voltages of the non-operating tubes in order to check if the tube is active or not.
The results were 0V which means at least according to my understanding that HotBluePlates's statement is correct.
/Leevi

[kagliostro]

I made the changes according to your proposal i.e. I switch the heaters only.
I measured cathode voltages of the non-operating tubes in order to check if the tube is active or not.
The results were 0V which means at least according to my understanding that HotBluePlates's statement is correct.

OK, and which was the result ?

K

[Leevi]

OK, and which was the result ?

So the measured voltage on cold tube cathode was 0V, no popping.

HotBluePlates' statement:

It won't hurt anything. With the heater cold, the tube is just bits of metal in a glass case.

/Leevi

[kagliostro]

I was missing "no popping"

K

[PRR]

> perhaps this might work? (separate cathode networks with cross-bleed)

The cathode caps still must charge-up, POP.

It *might* be possible to devise high-resistance networks to pre-charge the caps to near-correct voltage, then divert the tube current with low-resistance bias resistors.

Unless he needs "instant" switching, heater-switching is simple, clean, and ecological.

[DummyLoad]

It *might* be possible to devise high-resistance networks to pre-charge the caps to near-correct voltage

this may work...

[Leevi]

If I have understood that correctly your theory is based on keeping the voltage level on both cathodes.
And when you switch a charging of the cathode cap is not needed which keeps the switching smooth? Right?
/Leevi

[DummyLoad]

If I have understood that correctly your theory is based on keeping the voltage level on both cathodes.
And when you switch a charging of the cathode cap is not needed which keeps the switching smooth? Right?
/Leevi

in theory, yes.

there is a delay at first power on. if you switch before the one unused RC combination is charged it may still pop when switched. the time-constants are approximately 10K*70uF and 10K*220uF depending which valve is active at turn on.

--pete

[kagliostro]

I find interesting DummyLoad's idea

and I would like to know the result if you try it

---

In order to maintain charged the caps avoiding the bleed action of the cathode resistor

I've think to this, but I'm not sure what will happen, if you try, please report

K

[Leevi]

and I would like to know the result if you try it

Unfortunately the amp is no longer at my place so further tests with this amp
are not anymore possible.

I ended up to the heater switch that worked well and is ecological.

Thank you for ideas. Of course new ideas are still welcome.

/Leevi