The AM Forum

As I look back through old handbooks, I see that tetrodes are commonly used with either Pi network or "conventional" plate tank circuits. Triodes on the other hand, do not seem to be used with Pi network tank circuits unless they configured as grounded grid amplifiers.  Is it possible to use a grid driven large triode (say an 8000 or 810) with a Pi network plate tank? If so, why don't we see these circuits in the old handbooks?

Thanks,  Ron W8ACR

Ron,

I think it is mostly a matter of timing.  Tetrodes transmitting tubes became available and popular about the same time the pi network was "discovered".  Since tetrodes required less drive and were generally easier to neutralize, they quickly replaced triodes in the output stages of transmitters. 

Using Grounded grid class B triodes as high powered linear amplifiers solved the dual problems of neutralization and what to do with the large amount of drive available from most exciters/transmitters.

 

It's more a problem of neutralisation.  With the triode, either the grid or plate tank must be tapped to give the 180° out-of-phase rf sample to feed back through the neutralisation capacitor.  This makes it more complicated to bandswitch.  So the triode circuit is more likely to be designed to use plug-in coils.

A triode can be used with a pi network if the tapped grid tank is used.  Sometimes the National multi-band turret tank assemblies are used with the grid tank, and a pi network is used with the plate tank, for easy bandswitching.

With grounded grid, neutralisation is usually not an issue, so the pi network can be easily employed.

The first pi network circuits appeared in RADIO magazine in the mid 1930's.  It was originally designed by Collins, and was sometimes referred to as the "Collins Coupler".  The original circuit used a parallel tuned plate circuit, with the pi network fed from a tap on the main tank coil and used to match to the antenna or transmission line.  It was considered more an antenna tuner than the tank circuit itself. Art Collins used the circuit in all his mid 30's ham transmitters. The old RADIO magazines and RADIO Handbook published  circuits very similar to the pi networks used to-day.  They called it the "simplified Collins coupler" or "Simplified PI network, in that the pi network WAS the tank circuit, without the tuned circuit that fed the pi network.

Im running parallel 250TH's to a bandswitched pi-network 160-20M. With a bit of series R in each grid lead, a single neutralization setting on 20M holds for the others; the R also forces equal power sharing to a sufficient degree. I hate plug in coils and open wire line ;D.

A HK-354 amp is under construction which will be similar.

Still looking for some HK-1054, 750TH/TL, 1000T, 6C21 or similar big glass.

Carl
KM1H

Hi Carl,

There was a time in the early 70's when I owned a Collins TDH-5. It was a 3-bay rack with a pair of 750TL's modulated by a pair of 450TL's.   It was from a 1942 WWII ship. One cabinet alone, empty, weighted 440 pounds.  Black krinkle.

Anyway, at one time I had six 750TL spares. Two were brand new. I ran a pair in the final for many years.

But one by one all of them started dropping their filaments into the bottom of the glass. Even in storage I would find them broken. The old story of not having enuff tungston during the war made the fils brittle.

Anyway, the last one I gave to Derb who put it in a museum.

They are really trippy tubes.  Something a glassblower wud be proud of.

I'll bet a 1000T  or especially the 2000T are big $$ if they are to be found anywhere but in a museum.

T

Carl, Im interested in how you have your neutralization set up on your
pair of 250's in parallel..  I run a pair of 304tL's in parallel with pi net output, on 160 and 80m
I use a  center tap grid coil, and tapped the coil with a  bandswitch to change bands. Is that
what you are doing?. Also what value of series grid resistors are you using?
Thanks-Chuck

That is exactly what I did Chuck.

After looking at a B&W input assembly I built one to cover 160-20M that could handle the Viking I or HT-9 drive power.

For neutralizing I did it cold using a SA and a mica trimmer. Then measured the value and trimmed a piece of RG-142 to match. Works fine.

The grid resistors are 50W Carborundums I found at a flea market. Still new in the military foil package, marked 50 Ohms, and measured 68 Ohms.



Tom, when I was at NAS Olathe in 59-60 the base ham shack, K0NAB, had a 32V2/BC-610 and 75A3. When the QRM on 20M got bad for the phone patches we used the amp on the back wall; a pair of 1500T's with 1000T modulators. It was surplus from a shore installation but Ive no idea of the model #. No idea of the output but the 6el Telrex at 90' never complained; thinking back it must have been beefed up but there was no reason to climb and look.

There were also Telrex 6el for 10/15M however the big amp didnt tune that high, but the BC-610 had been modified for those bands. This was during the peak of Cycle 19 and the reports were great from all over the world.

Carl
KM1H

I also conclude that it is just a matter of timing.

In the past two years, I have built a single 8000, a single 811, dual both and now a single 810, all using balanced grid and PI output. 

They all follow the same basic plan as the 811 rig schematic at http://schmidling.com/radio.htm

One other misconception brought out by tetrodes is the anal attention to shielding between grid and plate.

You will note also in all the old handbooks that simply making the grid coil perpendicular to the plate coil was all that was required.

The price one pays for all that gain in a tetrode in the pain in the butt of shielding and sometimes having to neutralize anyway.

Neutralizing a triode is trivial by comparison.

I had pondered building a tetrode rig and a friend kicked it off by giving me a 4-400 but I just can not come up with a good reason to start it.

Triodes are nice.

js