Pentodes and Choke Loading

[ianc13]

Hi,

I was looking around and have never seen an example of a pentode driver tube being loaded by a plate choke.

I have searched and seen a couple of posts here and there that allude to the fact that this should never be done but I have never seen an explanation of why in electrical terms.

I was wondering if someone on this forum could enlighten me as to why this would be a bad idea?

Many thanks and I hope this is not too stupid a question!

Ian

[MQracing]

Hi Ian:

RE: choke loading of small signal pentodes. This is an area that I have some interest in exploring as well.

In the following thread you can find several circuit schemas that Kevin Carter and I drew up as thinking pieces. There are several different strategies shown in the thread referenced below as well as several pointers to some even earlier posts showing yet more circuits\ideas;

http://www.audioasylum.com/forums/ma...ages/7998.html


a UL tapped plate choke was discussed in the following paper written by Herb Keroes;

http://www.pmillett.com/file_downloads/ultralinear.pdf

see pages 16 and 17 as I recall.


I haven't had time to follow-up much with these ideas--- in the future I want to build several of the different type inductors shown in our schematics and give the arrangement the ole college try.

MSL

[ianc13]

Hi MSL,

Both those options would be fine if I had a tapped plate choke to do UL - which alas I do not at present.

I do currently have a 20mA 150H plate choke which I would like to play around with but this would only allow me to do straight pentode operation.

Assuming that I were to hook this up to a c3m/c3o running around 225V on the anode, choke loaded and around 150V on g2 from a fixed potential divider would this work?

Thanks,

Ian

[MQracing]

Originally posted by ianc13
Hi MSL,

Assuming that I were to hook this up to a c3m/c3o running around 225V on the anode, choke loaded and around 150V on g2 from a fixed potential divider would this work?

that's the 64 million dollar question (or about 128 million pounds if British).


How much inductance would the plate choke need? As a first approximation--- take what would be the magnitude of the resistor
is you were R loading the anode---- and then calculate backwards
how much L you might need at the lowest operating frequency of interest.

typically---- a lot of pentodes have extremely high plate loads when operated conventionally----

the second portion of the 64 million dollar Q is does UL operation change this---- and how would we go about deciding on what effective load impedance the anode of that pentode wants to see--- both if only being straight plate choke loaded (as you propose) as well as if we use a UL tapped plate choke.

I don't have an answer for you---- on how large that plate choke should be. My gut level hunch---- a WAG (wild butt guess) is that you would need substantially more than 150 H sitting on top of the anode.

But... with some care.... why not wire it up with the plate choke you have and give it a try? worse that could happen (I suppose) is that it just sounds and measures terribly---- in which case you unsolder everything and punt.

MSL

[Bandersnatch]

Originally posted by MQracing





How much inductance would the plate choke need? As a first approximation--- take what would be the magnitude of the resistor
is you were R loading the anode---- and then calculate backwards
how much L you might need at the lowest operating frequency of interest.

Look at required gain, and the approximation that gain is gm*Load. Keep in mind the choke's non-ideal behaviour, eg self resonance which makes it a very high impedance and usually in low-to-midband (1-2kHz ). The approximation will become deviate once the choke impedance gets near the plate resistance.

Originally posted by MQracing


typically---- a lot of pentodes have extremely high plate loads when operated conventionally----

This is a constant, resistive load, usually carefully drawn across the curves at a particular g2 voltage. Vary g2 or the load and you'll need to develop new assemptions

Originally posted by MQracing


the second portion of the 64 million dollar Q is does UL operation change this---- and how would we go about deciding on what effective load impedance the anode of that pentode wants to see--- both if only being straight plate choke loaded (as you propose) as well as if we use a UL tapped plate choke.

In U-L the lowered plate resistance means that something more like the triode model can be implemented. The gain is now a function of the spacing of the now-tilted plate lines since g2 is now varying with signal. Can you derive U-L curves, or at least their outter envelope with the tube's data provided?

Originally posted by MQracing


I don't have an answer for you---- on how large that plate choke should be. My gut level hunch---- a WAG (wild butt guess) is that you would need substantially more than 150 H sitting on top of the anode.


MSL

that would depend on the pentode, and use of the circuit you're designing. How much gm do you have at your OP? How much gain do you need? How much can you tolerate the gain varying with frequency? I suspect it will be the last bit stopping you from building with a straight pentode.

Now if you've got a few U-L taps to experiment with, one can tailor the pentode's new response. A 20% tap will build a new set of plate curves far different from a 40% tap or a 60% tap. I'd suggest all three for maximum room to wiggle in.
cheers,
Douglas

[cerrem]

Pentodes have not been the first choice for choke loading due to the poor damping factor to step load response....
At least with the triode, the low plate resistance dampens the response so you have a reasonable Q......with triodes the mid-band gain with a choke load is simply the mu....
With a pentode a R would need to be placed "across" the choke to stabilize the response....This R would dominate the load impedance the pentode would work into and the L would be choosen so that it's reactance to be equal to the R at the -3dB POLE of choice.... A suitable C may be placed in series with this R to create a ZERO in the response if need be to help boost the low frequency ......

Chris

[dtut]

As a down and dirty experiment, I put a 125ESE on an EL84 in UL and used that to drive several different OP tubes. Mostly the results sounded and measured, as best I could, pretty good. Certainly well enough to make me wish some winder might come up with something a bit more elegant which would work with small pentodes.

Doug Tuthill

[Bandersnatch]

Originally posted by dtut
As a down and dirty experiment, I put a 125ESE on an EL84 in UL and used that to drive several different OP tubes. Mostly the results sounded and measured, as best I could, pretty good. Certainly well enough to make me wish some winder might come up with something a bit more elegant which would work with small pentodes.

Doug Tuthill

See Heyboer in MI, they are very responsive to custom stuff, and have very good payment terms( pay once you get the part )! I haven't had to go anywhere else, power or signal...
cheers,
Douglas

[rdf]

Originally posted by cerrem
With a pentode a R would need to be placed "across" the choke to stabilize the response....This R would dominate the load impedance the pentode would work into and the L would be choosen so that it's reactance to be equal to the R at the -3dB POLE of choice.... A suitable C may be placed in series with this R to create a ZERO in the response if need be to help boost the low frequency ......

Sims performed on that arrangement suggest, with a variable R in series with the choke to set the following stage grid voltage, very promising performance as the front end of a two-stage direct coupled SE amp. The R in the RC string sets the stage gain.