Appropriate switch to implement triode/pentode mode

[mresistor]

I've been reading the posts here about the triode/pentode mode switching and want to implement it on a super rev build.. The ground switch hole is wide open.. I just tried it out on a Marshall DSL-40c and liked the sound of it. I saw that HPB was going to use a pair of mini switches in one post.  But I want to only use one switch and I think it will be dpdt  ON-ON  or even ON-OFF-ON would work.  I have the latter here, and older Carling rated at 15a 250v.  I was looking at this switch though too..



https://www.ebay.com/itm/Toggle-switch-DPDT-Toggle-Switch-ON-ON-20-AMP-125-VAC-K202/192476849039?hash=item2cd083a38f:g:apsAAOxymiVQ9Qnm


would somthing like this or the Carling work?  What do you guys use? I see some of you have implemented this and wonder what switches are being used.


Thanks

[mresistor]

Here are the specs of that switch on that e web site


 
TEMCo ID: CN0262Quantity: 1Nameplate Markings: ON - ONPoles and Throws: DPDT - Double Pole, Double ThrowNumber of Terminals: 6Amp Rating:AC Rating - 20A at 125 VAC, 15A at 250 VACDC Rating - 30A at 12 VDC, 20A at 24 VDCContact Resistance: 50 mΏ maximumInsulation Resistance: 500 VDC 1000 MΏ min.Dialectric Strength: 1500 VAC, 1 minuteOperating Temperature: -77°F to 185°FElectrical Life: 10,000 CyclesWarranty: 2 Year Warranty »

[sluckey]

I would use that big DPDT that Hoffman sells. The ebay switch looks like cheap "Radio Shack" quality. I would not trust any mini toggle switch.

[jjasilli]

+1.  All the literature states to use a reliable, full size DPDT SW (for 2 power tubes). That necessarily implies On-On.  Do not use a SW with a Center-Off position.  The screen needs to have proper voltage even in triode mode.  If the screen is deprived of voltage I think the tube will blow swiftly & catastrophically. 


Also, proper screen resistors need to be used for triode mode operation.  And you may want to find a bias compromise that works for both modes of operation; or have switchable bias.


Lot's of info & circuit diagrams in the web.

[mresistor]

Thanks Sluckey - JJ  I read that if the screen is turned off  that the tube will just stop conducting, no harm to the tube. Sort of like a standby mode.

[pdf64]

...The screen needs to have proper voltage even in triode mode.  If the screen is deprived of voltage I think the tube will blow swiftly & catastrophically...

I suspect the above is a common misapprehension.
No damage will be directly caused if the screen grid loses its voltage supply.

The screen grid needs some voltage for the tube to conduct plate current; all else being equal, as screen grid voltage increases, plate current increases (until plate current reaches saturation). Likewise, as screen grid voltage is reduced, plate current will reduced.
Below a certain screen grid voltage, plate current will tend to cut off.

Note that signal voltage on power tube plates may be well >1kV. A switch connected there needs sufficient voltage rating between its terminals. Damage may be caused if the switch is actuated whilst the power tubes are passing signal.

[jjasilli]

Ok, thanks for that info!

[kagliostro]

Ciao JJasilli

If the screen is deprived of voltage I think the tube will blow swiftly & catastrophically. 

I don't think so

look to the Standby switch

http://www.chambonino.com/construct/constwire1.html





--

http://www.chambonino.com/construct/constwire2.html





Franco

[jjasilli]

Ok, that's been established. 

But the rather unusual standby SW in your schematic still feeds the Screens +/- 40VDC, not 0 VDC, in Off position.  And given the above posts, I don't understand how this Standby SW works; and why the power tubes don't operate in triode mode (or near triode mode) with this Standby SW turned Off???  Probably, per pdf64's Reply, the bias level voltage on the screens is not sufficient for the plates to conduct current.  But then why connect the screens to any voltage supply in standby mode?


(I'm also suspicious of running fans and heaters, both of which may be noisy, off the bias supply.  But if it's quiet then fine.)
(And, I get the soft start relay, but not the use of 2 relays to accomplish one purpose???)

[kagliostro]

If you decrease the voltage on G3 you decrease the output power of the tube (a way to control the power using a VVR instead to lower plate voltage)

and if you give no voltage no power will come out

About the use of a negative voltage I asked some time ago and PRR explained something about it

but I don't remember exactly what he told and I'm not able to find the thread ( )

I only remember that doing that may be good practice

--

About the use of two relays I haven't an explanation for you, sorry

Franco

[jjasilli]

Ok, thanks for the input.  The unusual circuits you often present are definitely food for thought, which generate productive mental exercise!

[PRR]

Reducing Vg2 reduces Ip. Vg2= zero is nearly Ip= zero. No blow-up.

The monster 813 amp applies -130V in standby. It may be that "near zero Ip" was not "silent" on such a monster (1000Watts??) and it took a little more bias to shut it up.

[92Volts]

On the other hand, a plate with no voltage and a screen WITH voltage will kill tubes. Screen current rises at low plate volts. Pentodes/beam tetrodes decrease this in normal conditions but it's still true with zero plate voltage.

Even worse, with cathode bias no plate current means you get less bias voltage which would otherwise limit screen current.

I have forgotten to hook an OPT center tap to B+ and the screens of my EH 6V6s glowed like a lightbulb! They actually survived, but I was surprised they did.

You can definitely kill a tube by disconnecting the wrong part from the HV supply, that part just isn't the screen.

[mresistor]

Interestingly - the recommended switch is also made in China.

[DummyLoad]

Interestingly - the recommended switch is also made in China.

electronic parts aren't made in china? 

--pete

[mresistor]

You know what , that switch is very large... and I don't think it will work.. there are terminals literally hanging over a 6L6 socket with around 1/16" clearance Guess I could use some shrink tube on the contacts though.

[sluckey]

I predict that before you flip that switch 100 times you will be pulling it out.