• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Pentode to Triode Switch

Today I "finished" soldering up my SE EL84. It is pentode connected and I think it sounds pretty good though I haven't tried it with any good speakers yet.

Since this is the first amp I've "designed" and since I've used almost all spare parts for it I'm planning on doing some experimenting. One thing that seems easy to do would be to put in a switch to go from pentode to triode connection. It appears I can simply wire to poles of a DPST switch to the B+ and two to the anode and run the center two to G2.

Is that really all there is to it? Putting it into SE Amp CAD it appears that G2 is connected directly to the anode.

Thanks,
 
Yes, that is usually all that you need, but I usually add a 100 ohm resistor from the switch to G2. Mount the resistor right at the tube socket. Remember to NEVER flip the switch with the power on. Trust me bad things will happen. I know that the old Chinese KT88's will spark out big time. I was dumb enough to do it twice, with the same result both times.

If your transformer has ultralinear taps you can use a 3 way switch to go from triode to UL to pentode mode. I put a switch like this in the guitar amp that ate the previously mentioned KT88's. 38 Watts in pentode, 30 Watts in UL and 18 watts in triode mode with about 400 volts B+.
 
Sorry, I'm trying to post during a hurricane, and I lost the connection just as I hit the submit button. Hey the cable is out, there is nothing better to do. I am sure the power will go out before the night is over.

You need a DPDT switch for an SE amp. The center goes to G2, one end goes to B+, and the other end goes to the plate. The second set of poles is for the other channel. I usually add a 100 ohm resistor from the switch to G2. Mount the resistor right at the tube socket. Remember to NEVER flip the switch with the power on. Trust me bad things will happen. I know that the old Chinese KT88's will spark out big time. I was dumb enough to do it twice, with the same result both times.

If your transformer has ultralinear taps you can use a 3 way switch to go from triode to UL to pentode mode. I put a switch like this in the guitar amp that ate the previously mentioned KT88's. It was a push pull amp, 38 Watts in pentode, 30 Watts in UL and 18 watts in triode mode with about 400 volts B+. You can't push your luck with those tubes since they are known for fireworks.

Flipping the switch with the power on causes the tube current to drop to almost zero and then return as soon as the switch makes contact. You will at the least get a loud pop in the speakers. Not good for your tweeters.
 
tubelab.com said:
Yes, that is usually all that you need, but I usually add a 100 ohm resistor from the switch to G2. Mount the resistor right at the tube socket. Remember to NEVER flip the switch with the power on...

If your transformer has ultralinear taps you can use a 3 way switch to go from triode to UL to pentode mode...


Thanks for the reply! I think I'm going to go ahead and wire in the switch. This is a sort of test amp. My first ever with my "own" schematic and values and I would like to get as much info and different listening experiences as I can from it.

Unfortunately I'm using Hammond 125ESE OPTs so I don't also have an ultralinear tap or I would wire in a rotary switch to let me pick any of the three modes. On the plus side the 125ESE OPTs sound amazingly good for $37.

I sort of assumed I shouldn't flip the switch with the power on. I didn't know if it would be bad for the tubes but I thought the 425V B+ would arc across the switch eventually either burning it out or fusing it.

I also have a pair of SE KT88 monoblocks which are running in ultralinear mode. I may put switches in those to switch between the three modes as well.
 
I have flipped the switch with the power on while using a "sacrificial pair" of cheap 6L6's and nothing bad happened. I did however get a loud pop in the speakers. This won't hurt my "500 watt" guitar speakers, but is not good for delicate tweeters. Those 6L6's are still alive after years of abuse, but 2 different Chinese KT-88's died a fiery death from flipping the switch with the power on. Only one tube (not matched pairs) in the push pull amp died each time. Since I abused this amp with my guitar on a regular basis, I never put expensive tubes in it.
 
tubelab.com said:
You need a DPDT switch for an SE amp. The center goes to G2, one end goes to B+, and the other end goes to the plate. The second set of poles is for the other channel. I usually add a 100 ohm resistor from the switch to G2. Mount the resistor right at the tube socket. Remember to NEVER flip the switch with the power on. Trust me bad things will happen. I know that the old Chinese KT88's will spark out big time. I was dumb enough to do it twice, with the same result both times.

If your transformer has ultralinear taps you can use a 3 way switch to go from triode to UL to pentode mode. I put a switch like this in the guitar amp that ate the previously mentioned KT88's. It was a push pull amp, 38 Watts in pentode, 30 Watts in UL and 18 watts in triode mode with about 400 volts B+. You can't push your luck with those tubes since they are known for fireworks.


George-
I'm just finishing wiring my SE KT88 amp, and I'd like to clarify the switching arrangement you describe above.
Is the 100 ohm resistor always in the circuit? I was under the impression that it's only needed between the plate and the grid when in Triode mode. It sounds like you have it between the switch and the grid, which means it's always in the circuit. Did I misinterpret your explanation?
Glenn
 
Gents -
For the sake of simplicity; let's consider a SPDT switch for one tube. This switch would be an 'On-On' sort, and it would have three solder tabs on it. You would wire it in something like this:

--
| |--- Wire from UL Tap
| |--- Wire to Grid
| |--- Wire to 100ohm resistor & 1N4007 soldered to the cathode
--

So, when the switch is in the #1 position, the amp would be in UL mode (so the switch connects the grid to the UL tap). When the switch is in position #2, it connects the grid to the cathode via the 100 ohm resistor and diode.
More on the diode - wire it in series with the resistor, having the cathode (the stripe end) on the grid pin

Hope that helps !
Steve
 
Why would you connect the grid to the Cathode? Do you mean Anode (Plate)? This answers my question about the resistor though. It's only connected to the grid when in Triode mode.
I've also seen people use the diode in similar applications.

I also want to run this amp in Pentode mode, so I need an additional position on the switch to connect the grid to the B+.

Thank you for the diode suggestion, I'll probably add that as protection.

Glenn
 
In my implementations I have the 100 ohm resistor in the grid lead . It is in the circuit all of the time, and functions as a "stopper" resistor. It can help prevent oscillation.

The diode in the screen grid lead has been discussed before, and many people are of the opinion that it could cause harm with some tubes, or it can cause distortion. Others claim that it prevents damage and cures distortion.

My testing showed that it didn't make much difference in normal HiFi applications, but did cause some weird sounding effects in a guitar amp operating well into clipping. It could cause some damage in an RF tetrode operating at high power since the screen grid may actually source current (caused by secondary emission) but is likely harmless in an audio application.
 
That makes sense, I've seen this in guitar amps before.
I guess the 100R is low enough a value that it doesn't change the voltage much. Probably worth putting it in the circuit all the time if it helps with oscillations. 5W wirewound okay? I just happen to have those around.

I think I'll leave the diode out for now.

Thank you for the explanation.
Glenn
 
A wirewound resistor may cause the oscillation that you are trying to cure since it is by definition also an inductor.

Many people tell you (and the math confirms) that this resistor will dissipate low power so you can use a 1/2 watt resistor. I can tell you from experience that the 1/2 watt metal film resistors sold by Mouser or DigiKey will fry if the amp is overdriven. I have toasted 3. I now use 1 or 2 watt carbon film or carbon composition resistors and have not killed any even when using my SimpleSE as a guitar amp.
 
This was interesting reading on using the diode:
http://www.webace.com.au/~electron/tubes/oes.html

The diode is in series with the grid stopper, cathode towards the grid, but is the order important? I assume the diode goes between the grid stopper resistor and the plate because the object is to keep the grid stopper soldered close to the pin of the tube, right?

[EDIT]
I just realized that you probably also leave the grid stopper resistor in the circuit because it's soldered right on the pin of the grid.
 
Ouch ...
I may not be popular for posting this, but in fairness to the less experienced, a note of warning must respectfully be sounded with regard to the "Optimised Electron Stream" article given in the internet link above. While a lot of very valid basics are admirably covered, the author is in error with some of his reasoning a.o. the physical electrode distances, function of G3 and the effect of the G2 diode for dc and ac (basic electronics, not an opinion).

It is probably unfair to simply say this without a more complete discussion, but the latter will take a whole lot of space, and has been covered before as said ... but I apologise (can't seem to find that thread now).

As an aside, a wire-wound resistor as a loudspeaker replacement load is nearer a loudspeaker equivalent impedance than a totally non-inductive load, in as much as any inductance does play a role.

I also agree with Tubelab regarding the wattage of the G2 stoppers. But then again, I had tubes saved by the burning out of G2 stoppers when the anode circuit failed, so it worked both ways. But Tubelab's argument probably occurs more frequently, so.
 
But Tubelab's argument probably occurs more frequently, so.

Not everyone is crazy enough to plug a guitar preamp into their HiFi amp and overdrive it by 10 db either. I do this to all of my designs to find their weak points. In this case I zapped a few screen resistors, so I started using bigger ones.

As an aside, a wire-wound resistor as a loudspeaker replacement load is nearer a loudspeaker equivalent impedance than a totally non-inductive load, in as much as any inductance does play a role.

This is true, and for simple amp testing and burn in I use a pair of 8 ohm 500 watt wire wound resistors that I got from Ebay. A forum member much wiser than me pointed out that these should not be used to load an amp for critical measurements like frequency response and power output. I built a non inductive load box and it did give different results.

I do not put much faith in the "Optimised Electron Stream" theory either, but that was covered in the thread mentioned.