Triode Trick

[jkeny]

Has anybody tried this trick to convert pentode to triode operation ala Bill Perkins from Pearl Inc and shown here: http://t-linespeakers.org/tubes/triode-trick.html

This seems to be an alternative to ultralinear operation - am i right?

Any impressions?

John

 

[Tweeker]

Only insofar as triode connection is an alternative to UL operation. Its an alternative to a ~100 ohm resistor.

 

[Jaap]

yes

the reports are that this solution sounds better than a resistor.

 

[ray_moth]

An alternative to UL connection is not what is intended here. All he is saying is, if you want to connect a pentode/tetrode as a triode, it's better to do it his way. The arguments make sense to me, but how much better it will sound I wouldn't know.

 

[Geek]

That's also known as "boosted triode".

You'll also see that used on some Lafayette and Hallicrafter tubed CB radios on the RF final.

 

[jkeny]

Ah! now that I know the name for it I can find past refernces to it in here.

I had a quick search and found refs but nobody seems to have tried it and reported on the sonics. I'm only new to tubes & seem to be ploughing the same furrow many have before - Have I missed something?

Thanks
John

 

[Luanak]

I have tried a similar aproach. But I do it with a programable tl431
shunt regulator at around 25V , with a Elna 450uf cap across it.
The power is a little bit smaller, but the sound is bit better to me, more relaxed or open.

Regards

Cosme

 

[Klimon]

Hi,

I've incorporated this 'trick' in my el84 se and must say that it makes an excellent improvement. My grievances before the mod were a somewhat slow, romantic but also muddled sound lacking attack; after the mod the amp sounds much more detailed and tighter. I wouldn't consider building another triode-strapped amp without using the trick. BTW the schematic contains a few small errors, ask me if you're really interested than I'll take a peek into my amp and see how I did it. fwiw Eddie Vaughn made a few suggestions for optimising the circuit (he uses something similar in his carina), namely using 5,6V zeners IIRC (same number would do the job) + one 1000µF electrolytic instead of the two 470µ and resistors. He also explains the benefits on his website (think he calls it black box).

Simon

 

[pescalune]

TRIODE TRICK

Hi,

Simon,

What are the few small errors in the schematic ?

Bien à vous .

Robert

 

[jkeny]

Hey Cosme,
I would be interested in a schematic for the tl431 version if that was possible?

Simon,
Ditto with your corrected schematic - would love to hear more.

John

 

[Klimon]

10 ohms resistor should be connected to the plate; all the rest is as Bill Perkins prescribed in the original joelist thread. Less complicated than I remembered🙂

Good luck

Simon

 

[jkeny]

Cosme,
TL431 version would be very interesting as it requires so few easily available parts and could be quick to implement

Thanks in advance
John

 

[Luanak]

This is the schematic

 

[jkeny]

Thank you sir,
Very kind - I presume this provide for the seperation of AC & DC pathways as per the original zener method?

John

 

[Svein_B]

This "trick" sounds interesting - but a bit confusing

There is this article in EDN about Get more power with a boosted triode , but they it are running g2 at a higher voltage than the plate

SveinB

 

[Tubes4e4]

Hi,

to me, that doesn´t make much sense from a technical viewpoint:

What is done there, is lowering Vg2 while maintaining the same AC swing at g2 as being present at the anode.

That means, theoretical possible total AC swing of g2 before catastrophic distortion havoc is lowered by 2 times the DC voltage drop from anode to screen grid introduced by this circuit.

In turn that means when using that circuit you limit usuable (not severely distorted) anode swing by the same factor, since AC swing still is the same.

This _may_ work for lowish gm pseudo-triodes to some extend and within limits, because the g2 DC voltage drop introduced will not intefer with the possible AC anode swing at its lowest possible value along the load line, but one would have to check each case in detail to make sure it actually doesn´t.

Anyway, this is not a good "general" recipe to use, since for correct triode-operation of pentodes you want the screen grid to be able to swing the full swing the anode may swing - in perfect synchronicity.

Besides that, it escapes me how a "g2 swing limited pseudo triode" can sound better than a "non g2 swing limited pseudo triode" - there is not the least traceable hint in this article why it should.

Actually, shifting Vg2 a little bit lower DC-wise while maintaining AC swing at anode and g2 only will shift the plate family of curves a little bit "horizontally" towards higher Va values; you can obtain pretty much the same effect by simply increasing the bias voltage slightly towards more negative values.

Regards,

Tom Schlangen

 

[Klimon]

Besides that, it escapes me how a "g2 swing limited pseudo triode" can sound better than a "non g2 swing limited pseudo triode" - there is not the least traceable hint in this article why it should.

If the title is well chosen the aim of the circuit is to get higher output power, I didn't find any reference to sounds quality in the article. Also this looks like a very different animal than the g2 conditions as per Bill Perkins.

Simon

 

[Tubelab_com]

I experimented with the "boosted triode" idea that was proposed in the EDN article. They used a 6L6 type tube in their article, so that is what I tried. The result, a dead 6L6GC. Their boost idea applies a positive (G2 higher than the plate voltage) offset to the screen grid. This does allow the plate saturation voltage to drop down to about 20 volts allowing slightly higher power on a lower plate voltage. What the article did not mention (and the authors may not have noticed) is the screen grid dissipation will be exceeded at idle and it will shortly melt. Mine lasted about half an hour. Do you want to run any 6L6 type tube with 400 volts on the plate and 500 volts on the screen, not for long. Their simulator (Microcap) will not melt but a 6L6 will! Been there tried that.

The Triode Trick pointed out in this thread is highly useful for operating sweep tubes in triode SE. Many sweep tubes have a low G2 max voltage which limits their usefulness in triode mode.

 

[aletheian]

I experimented with the "boosted triode" idea that was proposed in the EDN article. They used a 6L6 type tube in their article, so that is what I tried. The result, a dead 6L6GC. Their boost idea applies a positive (G2 higher than the plate voltage) offset to the screen grid. This does allow the plate saturation voltage to drop down to about 20 volts allowing slightly higher power on a lower plate voltage. What the article did not mention (and the authors may not have noticed) is the screen grid dissipation will be exceeded at idle and it will shortly melt. Mine lasted about half an hour. Do you want to run any 6L6 type tube with 400 volts on the plate and 500 volts on the screen, not for long. Their simulator (Microcap) will not melt but a 6L6 will! Been there tried that.

The Triode Trick pointed out in this thread is highly useful for operating sweep tubes in triode SE. Many sweep tubes have a low G2 max voltage which limits their usefulness in triode mode.

I saw that article a while ago, and that was my thought... using the grid OPPOSITE it's intended purpose. i will eventually try it on a rainy day with a with a more rugged-screened tube just for kicks and giggles. I did consider trying it on small signal pentodes, but never got around to it.

 

[jkeny]

Thanks for that Tubelab.com - that saves a number of us lessor experienced inmates from killing our tubes and also thanks to Tom for dispelling the confusion & misinformation that seemed to be contained in the EDN article.

John