• 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.

Triode vs. Pentode in the same amp?

I only kinda understand this, but my understanding was that Triode and Pentode tubes were different. How are some amps able to switch between these modes on the fly? A quote regarding the Rogue Audio Cronus Magnum:

"The design’s tube complement (two 12AX7, three 12AU7, and four KT120 output tubes) now allows users to switch “on the fly” between triode or “ultralinear” (pentode)"

What is it about the design's tube complement that allows this? Is there any compromise implied in having tubes that can run in both modes, if only one is desired?
 
Member
Joined 2004
Paid Member
The pentode second grid (G2) switchable
  • to output transformer ultra linear trap: UL
  • or via few hundred Ohm resistor to anode: Triode mode.

In triode mode the tube's anode "resistance" is fraction than in UL, so loading (tube output resistance versus output transformer impedance) is more better. In this mode (apart from other changes) the output power is decreasing (reduced by about half), and distortion also changing (usually decreasing too).

This switching is the compromise, UL and triode mode operating points are usually not the same.
 
AX tech editor
Joined 2002
Paid Member
How are some amps able to switch between these modes on the fly?
A pentode has an couple of additional electrodes compared to the triode.
So if you connect those extra electrodes to the other elements, you can actually wire a pentode as a triode. ("Triode-wired pentode").
With some switches you can do that on the fly.

Jan
 
SoaDMTGguy,

Caution:
Do not ever switch a warmed up amplifier from any one of these modes, to any other of these modes . . .
Pentode; Ultra Linear; Triode.
Your speakers, your output transformers, might be damaged; and your ears may not like it either.

The KT120 (Kinkless Tetrode) is actually a Beam Power tube, sometimes referred to as a Beam Pentode or Beam Tetrode.
It has Beam Formers, instead of a Suppressor grid (Pentode G3).

Pentode mode and Ultra Linear mode are not the same.
I am guessing your amplifier uses the KT120 in both Ultra Linear mode, and in Triode mode, Right?
Neither of those modes are Pentode mode; and neither of those modes are Beam Power mode.
A Beam Power tube can be operated in Beam Power mode; Ultra Linear mode; and Triode mode (but your amplifier switch only allows Ultra Linear and Triode modes.

An EL34 is a true Pentode, it has a Suppressor grid (and no beam formers).
An El34 Pentode can be operated in Pentode mode, Ultra Linear mode, and Triode mode.

Quite a bit of similarity of modes for Beam Power and Pentode tubes.

No, you can not just re-wire the connections from a KT120 socket to new connections to use a True Triode tube.
Is that what you want to do?

I am pretty sure that both the Ultra Linear mode, and the Triode wired mode of your KT120 tubes are fairly optimal settings for that tube.
Just leave your amplifier switch in Triode mode, that seems to work best for you.

Generalizations:
An amplifier that can properly be switched to work in two different modes has a few advantages:

Allows the owner to get more power (Ultra Linear mode).

Allows the owner to get a little less power, but with these benefits:
Lower distortion at low and medium power out.
Global Negative Feedback does not have to 'work as hard' (Triode mode intrinsically has lower distortion; and Triode mode intrinsically has higher damping factor).

The other advantage of a switch is to sell more amplifiers; many customers want to try and use both modes.

Finding a True Triode tube type to replace your KT120 tubes, would require a complete redesign of your amplifier.
If what you want is a True Triode amplifier, you would need to purchase a different amplifier.

Relax, sit back and listen to your favorite music, set the switch to Triode mode, and I am pretty sure you will be pleased.
 
Last edited:
  • Like
Reactions: 1 user
An amp that has a triode/ultralinear switch uses the same output tube for both conditions. It’s the switch that changes the circuit between the two modes not putting a different tube in. Triode tubes like the 300B are entirely different animals than the KT120 kinkless tetrode. I built all my SE amps with switchable modes. Triode has less power but I like the way it sounds as compared to the UL. Some people want the 50% more power so they use UL. Take a look at the Mikael Abdullah LT88 schematic to see how the switch is implemented. The distortion spectrums are different between the two modes and that is what determines the sound of the amp mostly.
 
The pentode second grid (G2) switchable
  • to output transformer ultra linear trap: UL
  • or via few hundred Ohm resistor to anode: Triode mode.

In triode mode the tube's anode "resistance" is fraction than in UL, so loading (tube output resistance versus output transformer impedance) is more better. In this mode (apart from other changes) the output power is decreasing (reduced by about half), and distortion also changing (usually decreasing too).

This switching is the compromise, UL and triode mode operating points are usually not the same.

Another compromise in switchable amps is that many of them do not adjust the feedback compensation when making the change from UL to triode. Since the overall gain of the amplifier is reduced in triode mode, you reduce the feedback by switching to triode mode. This may be why some people like triode mode, though the amp may not perform as well with sub-optimal feedback.
 
  • Like
Reactions: 1 user
SoaDMTGuy,

mdpaudio's SE amplifiers:
All use either Pentode output tubes, or Beam Power output tubes. (His Post # 7)
All his SE amplifier builds do Not use True Triode output tubes.
That is because he says they all have Ultra Linear / Triode mode switches (a True triode can not be/do Ultra Linear).

grovergardner:
Alludes that the lower gain of Triode mode causes a lower ratio of open loop gain to closed loop gain . . . True.
But the actual negative feedback is the same.
A millivolt of distortion at the output, [usually] still causes the same 10uV of distortion to be applied to the negative feedback node.

I would not worry about the lower open loop to closed loop ratio of Triode mode versus Ultra Linear mode (closed loop is with negative feedback applied; which they are in both cases of your amplifier).
Remember, Triode mode usually has Intrinsically lower distortion and absolutely intrinsically has higher damping factor before negative feedback is applied.
Remember, Ultra Linear mode usually has Intrinsically higher distortion and absolutely intrinsically has lower damping factor before negative feedback is applied.

The most often cause of different square wave shapes, if any, with the switch in Triode mode, versus the switch in Ultra Linear mode is because the output transformer primary is driven by the different plate impedances of the two modes.

The amount of plate impedance in Ultra Linear is mostly caused by the Ultra Linear Tap percentage (often somewhere between 25% to 50%).
In Triode wired mode, the percentage is 100%.
A typical pentode or beam power tube in triode wired mode might have 1/2 (0.5) of the plate impedance, rp; versus the higher plate impedance of ultra linear mode.
Example:
rp triode wired mode 1500 Ohms
rp ultra linear mode 3000 Ohms

Have fun listening to your amplifier.
 
Well, in terms of a true triode output tube, your options are limited. The 45, 2A3/6B4G, 300B, and big transmitting types like 845 and 211...there are a few others. But these have very different parameters from a triode-wired pentode or beam tube, and require some real gymnastics to employ effectively in an amp. In terms of what you currently have, there is no "true triode" you could drop into the Cronus that would be remotely compatible. You'd basically have to redsign the amp. ;-)

That said, a well-designed push-pull 300B amp is a thing of great beauty, and you can achieve 20 watts per channel. A push-pull 2A3 amp is also lovely, but at best you can get 15wpc. But again, they have to be well-designed, and there's more iron and complexity involved. But yes, true push-pull triodes can offer the sort of realism and immediacy that a single-ended triode amp can offer-- with double the power or more.

Personally, I gave up on "true triode" amps a while ago. They are expensive and complicated. I greatly prefer either a triode-wired or UL Williamson amplifier with KT66 output tubes. Excellent bandwidth, better control of the speakers, lower distortion and a different kind of realism that comes from wide bandwidth and uncolored representation of the recording.
 
Member
Joined 2010
Paid Member
Thank you all for these great replies! The final thing I’m trying to understand is:

Is there an advantage to an amp using true triode tubes, compared to one that is switchable?
Accepting the fact that the KT120 is not a 'true' triode, then no, for all intents and purposes, there is no difference between a triode amp designed with those tubes and an amp using those tubes that is switchable to triode mode. Yes, some will go down the rabbit hole of minute differances, but if you have the amps, like how they sound in triode, use them.
 
True triode amps can be very very good if and only if they are well designed and their limitations are respected. Many are not well designed and people expect way too much out of them. Triode wired amps can be very very good under the same restrictions. The triode wired amp is usually less expensive all things considered and gets very close to the pure triode in sound quality. My amps give me lots of versatility in the output tubes used when a true triode is designed around a single type of tube. A 300B amp is only a 300B amp. I can use most of the KT tubes with little effort. I’ve even done switchable bias resistors for some people to enable different tubes to be used. Do a lot of reading and buy a truckload of parts to experiment with. Remember that these voltages can kill! Dont over stress on perfection the first time, no one is judging you. Most importantly, have fun!
 
SoaDMTGguy,

You seem to have started with a very powerful stereo amplifier, that uses four KT120 tubes.

For a low power push pull True Triode amplifier, here are some Indirectly Heated tube types (not DHT).
6CK4 I designed a self inverting push pull amplifier using a pair of those.
6BX7-GT
6BL7-GT, GTA
6AH4-GT
6S4 (I have a few of those to try).
12B4
None of these tubes are current production, but there are good NOS out there.
The advantage of an Indirectly Heated Triode, is there is a true cathode, and a true filament (much easier to deal with than the DHT filament).

Current Production tubes you can use in Triode wired mode;
Most of these will put out more power than the True Triodes I listed above.
These are easy to deal with 'Indirectly Heated'.

As mdpaudio said, you can use the KT tubes (Kinkless Tetrodes).
I have used KT66, KT77, and KT88
And . . .
EL84, 6BQ5, EL34, and 6CA7 Pentodes
Beam Power tubes: 6L6GC, 5881, 7591, 6V6, and my least favorite, the 6550
I have used most of these in Triode mode.

Just for the tube rollers:
These are generally Plug and Play compatible types:
EL84 and 6BQ5
EL34 and KT77

Design, build, and listen.
You may be surprised.
 
Last edited:
Never really liked the idea of switchable triode-ultralinear operation. You can not optimize the design for both modes simultaneously, especially if feedback is involved. If you optimize it for UL, when you switch to triode, loop gain falls, feedback is now overcompensated and you could end up with compromised frequency response and higher distortion. Optimize for triode, and when you switch to UL you can end up with instability.

Besides, pentodes and beam power tubes should be operated in pentode mode with regulated screen voltage considerably lower than B+ for lowest distortion (with plenty of feedback to lower output impedance). This requires excellent, wideband output transformers; if you are not willing to invest in those, go with either triode or ultra-linear and optimize the design for the chosen mode.
 
obseedian,

I never said a UL / Triode wired switch made the amplifier performance optimum for one mode, the other mode, or both modes.

Please re-read my Post # 10.
Start reading at: grovergardener

Of course no switch is optimum.
You can have dual transmissions in your car, and an"input switch" and an "output switch", to select which transmission is engaged.
The two ratios will not be optimum at the drive wheels (one will be higher top speed, one will be higher torque).
But . . . as you can see, there is an Advantage to Both Transmissions.

Not all output transformers are created equal; insertion loss, primary inductance, turns ratio, phase, leakage reactances, resonances, etc.
Not all amplifier circuits that drive that output transformer have the same gain, phase, frequency response, output tube plate impedance, etc.
Global Negative Feedback has to deal as best it can, given the two above complex details.

Some amplifiers are designed for the very best performance numbers (into a non-reactive load resistor). That is Marketing and Sales at their 'very best', to convince you to purchase their amplifier.
Some amplifiers are designed to work with a wide variety of loudspeakers.
Some amplifiers fall in between those two extremes.

If you are looking for the best music sound, you probably should attend more live concerts, and listen less to your Hi Fi Stereo.
And, if you can not enjoy listening to your Hi Fi Stereo, I feel sorry for you.

If you are unsure of your amplifier performance, you can take it to your local audio repair center, and have them test it into a resistive load,
speaker load simulator, etc.
Or send it to me, and I will drive it through its paces, and send it back to you. (in some cases, I will have to bring it to a friend and associate of mine for some higher power tests, and certain higher dynamic range analysis).

obseedian, thanks for giving me the idea . . .
SoaDTMGguy should take his amplifier to his local audio repair center, and have them test it into a resistive load, including square wave tests in both UL and Triode mode, frequency response, distortion, damping factor, etc.
It might perform better than both you and he expect.

Just my opinions.
 
Last edited:
Some time ago I built a Class AB1 push-pull amplifier with 807 type tubes and a 25:1 turns ratio output transformer. It had a switch of g2 between the UL tap (40%) and the anode. It did not have any global negative feedback. There was not as much difference that would made this worthwile. A triode connected beam tetrode has lower anode internal impedance and would probably need a 3k:8R transformer (20:1 or so), while the same tube in UL needs 5k:8R trasformer (25:1). Beam tetrodes and pentodes usually have lower g2 voltage rating than anode voltage rating. For 807 it is something like 300V vs. 600V. At triode and UL strapping one has to find a compromise somewhere around 400V. They work best in real pentode mode, with g2 stabilized at 300 to 350V. But then it needs some amount of global negative feedback, otherwise the output impedance would be high, and loudspeakers do not perform well.
Recently I modified this old design, eliminated the triode/UL swith, set B+ to 350V, connected the tube as UL with a 20:1 transformer, and applied 12dB GNFB. I think this is a good compromise for home listening where one does not need high power. If you do so, go for pentode mode. For triode mode I would choose a different tube, i.e. a real triode.
 
You can not optimize the design for both modes simultaneously, especially if feedback is involved. If you optimize it for UL, when you switch to triode, loop gain falls, feedback is now overcompensated and you could end up with compromised frequency response and higher distortion. Optimize for triode, and when you switch to UL you can end up with instability.
This part makes very little sense.

The amount of feedback of the entire system stays the same.
Even the loadlines stay the same.
Only the amount of gain changes. (although usually also not more than about 3-5dB between UL and triode mode)
So you can just design the whole amplifier accordingly.
Which is very manageable with a feedback of around 6-9dB

The only argument that can be made is optimizing for lowest amount of distortion.
(or at least some nice compromise that works according to taste)
Which sometimes can be (slightly) different between triode mode and pentode mode.
But with UL you fall kinda nice in between, so that difference isn't so shocking.

The biggest difference beside the difference in power, is that de Ri of a triode is much lower.
So the lower -3dB point shifts down. (assuming that the primary inductance of the OT is low enough)
When this is not being considered in de designing stage, you could maybe boost (compensate) way to much.
Which can either clip the signal in preamp (feedback) stage or saturate the output transformer.

But again, between UL and triode mode, this difference isn't that shocking either.