[AMRadio] RE: GB> Update On Supply for PP 304TLs

[John E. Coleman]

Hi Donald:
	I used to run a pair of 304TLs Push Pull on 75 meters.  It was a
real brute of a rig and really had lots of head room for modulation.  As
I recall for Class C Plate Modulated you will need about -500 volts of
bias at about 150 ma of grid current (75 ma / tube).  I used a
combination of fixed and grid leak to get the -500 volts.  I think it
was about -100 from a shunt regulated bias supply and the rest was from
a grid leak resistor.  The exciter was a Push Pull pair of 812s.  I ran
about 1600 volts at around 600ma on the 304s.  The thing would put a
solid 800 watts into a dummy load (read with many different meters).  I
wanted to run the lower voltage of 1600 Volts because I didn't want any
arc over on the plate tank capacitor.  It was a split stator from a
BC610. The spacing on it didn't look like it would handle a lot of
voltage and modulation too.  I used shunt coupling to the output tank
with a couple of 500pf large ceramic capacitors and a special hand wound
center tapped RF choke.  I used another Christmas tree style choke from
the center of the home brewed choke to the HV supply point.  This kept
the HV off the tank capacitor and the coil at the same time.  I recall I
had trouble getting the thing to load to enough current at first so I
took some turns off of each end of the B&W plug in tank coil until it
took almost all of the capacitance of the split stator to resonate it on
3850 MHz (it would not go to the CW end of the band).  I then parallel
tuned the link with a .001 ceramic (a large one).  Now I could get some
current to flow when I pushed in the link.  The 1600 Volts / 700 ma also
made a nice match for the modulator which was 4 - 813s AB2 (2750 Volts
on the plates) conventional Push Pull Parallel circuitry.  That
modulator would easily take the plate of the 304s from their static 1600
volts to 4200 volts on a positive peak.  And the 304s head room would
follow the modulation right on up to the peak at 7200 Watts PEP output.

	But alas, I moved and just didn't have the room for a driver
that was as big a Globe King 500 plus a 6 ft rack of power supplies and
final.  So I switched to a pair of 250ths.  Not quite as high on the
plate efficiency but a lot easier to drive.  
	Good luck on yours. I hope you have as much fun as did.      


-Original Message-----
From: amradio-admin at mailman.qth.net
[mailto:amradio-admin at mailman.qth.net] On Behalf Of Merz Donald S
Sent: Thursday, November 07, 2002 12:00 PM
To: amradio at mailman.qth.net; glowbugs at piobaire.mines.uidaho.edu
Subject: [AMRadio] RE: GB> Update On Supply for PP 304TLs

Well, I have discovered the errors of my own making here. Your comments
helped me realize where I had gone astray.

First, let me say a few words in my own defense. Most homebrew rigs that
are passed along come with no documentation. A few come with some
documentation. And one in a hundred comes with good documentation. This
"W9TB Desk Kilowatt" is in that middle group. In this case, the partial
builder's notes that I have were ambiguous and I mis-interpreted them.
The power supply notes that I have are for the 4E27-based exciter--not
for the 304TL's final amp. I have almost no information on the final amp
supply that W9TB used. But I screwed up and thought his cryptic notes
were for the final amp. 

Sooo...I only discovered my mistake when your comments started me
thinking about the use of the word "screen". Finally it dawned on me
that the notes I was looking at were for the exciter. Screen actually
meant screen, not grid. Now it all makes sense. But it means that I am
worse off than I thought for information on his original 304TL supply
design--there is basically none.

At the risk of boring you, let me back off here and describe this rig so
you have a complete picture. In a word--incredible. Imagine the finest
mechanical and electrical engineering skills pumped into a late-40's
homebrew CW kilowatt. It's a complete station with receiver, frequency
meter, monitor receiver, exciter, and final. It has provision for
antenna rotor control on the front panel. And it has provision for a
remote control console attachment. Based on the notes, the rig was built
in 1948-49, updated frequently over the years and used on the air until
about 1985. I will take some pictures and pass them along. 

The control console, receiver and exciter are built on top of a
removable desktop. Each has its own compartment that it fits into. The
desktop itself appears to be commercial. But everything else is built
from scratch--sheet aluminum with aluminum bracing. The receiver has
been extensively solid-stated. I have not even opened it up yet. But I
believe it is 100% solid state. I have opened up the band switching
exciter and it is awesome. The construction quality is nothing short of
intimidating. It uses a 4E27 final and a National MB-150 tank circuit.
It is otherwise partly solid-stated. The transmitter and receiver both
use National PW dials for tuning.

The two pedestals that hold up the desk are packed with gear. The left
hand pedestal holds the final amp--push-pull 304TLs with the biggest
plug-in coils that B&W ever made and a variable center link operated by
a turns-counter lever on the front panel. One side of the pedestal is a
safety-interlocked door that opens to provide easy access to the final.
The other pedestal has a Meissner HF receiver built into the top
section, followed by a home-made heterodyne frequency meter and a drawer
for holding the final coils. I have not looked at the Meissner receiver
and freq meter at all. 

The power supply for this rig was built into a closet in W9TB's house.
The parts of the power supply that I have are whatever was salvaged from
that closet including the pole pig. It's hard to make sense out of what
is here. 

I think I have HV, screen and filament covered for the exciter from my
extensive "junk-box" power supply collection. But the junk box let me
down on the final supply. I have no transformer capable of replacing the
pole pig. It looks like w9TB was running the HV for both the exciter and
the final from the pole pig. But there are no chokes in the "closet
parts" that I received with the rig. And some hefty chokes will be
needed. So I am only going to run the final from the pole pig and use
the junk box supply as-is--unchanged (a lucky break) for the exciter.

So that's where this stands. I have no web site. But if someone has
someplace I can post some pictures, I would like to hear from you.

Thanks.
73, Don Merz, N3RHT


-----Original Message-----
From: Merz Donald S [mailto:merz.ds at mellon.com]
Sent: Friday, November 01, 2002 3:28 PM
To: amradio at mailman.qth.net; glowbugs at piobaire.mines.uidaho.edu
Subject: GB> More On Supply for PP 304TLs 


"Screen" is what the original builder's notes show instead of "grid" or
"bias". So I was propagating the mis-wording when I used "screen." The
304TL is of course a triode. Otherwise the audiophools could not love
it.

It looks like he was running the grid bias at about 80 volts. He had the
primary of the HV transformer connected to a motorized variac that
slowly brought up the HV. This sits behind a mechanical timer assembly
of WWII surplus vintage that only starts the variac moving after the
filaments have been on for a while. He's got the relay sequence set up
so that you can't get HV if you don't have bias--er, I mean "screen"
voltage. 

It's all a bit overly-clever if you ask me. But W9TB ran it for almost
40 years from the late 40's to the mid-80's. And he's got a pretty
impressive log here to show for it. I've got all his old parts and his
original diagrams so I'll probably duplicate the whole thing--except for
that doggone HV supply. That is going too far for me. 

The thing with the HV supply is that he converted to solid state diodes
at some point. But the lower voltage drop of the new diodes left his HV
and "screen" supply too high. But instead of building a new supply, he
took transformer filament windings and hooked them up backwards to buck
down some of the voltage. When that wasn't enough, he inserted these
huge ceramic resistors in the line in banks of 3 or 6 in a haywire
parallel/series lash up to lower the voltage to where he wanted it. 

The rest of the rig is built like a fine Swiss watch. But I guess he
sure didn't want to buy a new power transformer. So he made what he had
work. But I'm not going to heat the house with all these resistors or
fool around with bucking transformer windings. And the pole pig is just
too damn big.

I have some perfectly good HV supplies here that are not in use--sitting
in my garage. I want to adapt one of those somehow. That's my goal. 

73, Don Merz, N3RHT


-----Original Message-----
From: George Pritchard [mailto:gpritchard at comtechpst.com]
Sent: Friday, November 01, 2002 2:52 PM
To: amradio at mailman.qth.net; glowbugs at piobaire.mines.uidaho.edu
Subject: RE: [AMRadio] Re: Supply for PP 304TLs ??


Use 3000 Volts max. I use a pair in AB1 Audio. No... I did not pay
$85,000
for it! I would not sell it for that either!!! Having to much fun.
Anyway,
No screen supply needed since they are triodes. Neutralization is a big
must. For class C, I would use 304THs since the bias requirement is
higher,
and will give better efficiency and class C linearity when plate
modulated.
They do require plenty of drive however. The TLs versions are more
linear
for audio. Use a pair of THs for the class C, and a pair of TLs for the
audio... if you have them. Audio phools are driving the price up on
these.
Glad I bought spares years ago! Have fun,
George AB2KC

-----Original Message-----
From: amradio-admin at mailman.qth.net
[mailto:amradio-admin at mailman.qth.net]On Behalf Of Brian Carling
Sent: Friday, November 01, 2002 1:33 PM
To: amradio at mailman.qth.net; glowbugs at piobaire.mines.uidaho.edu
Subject: [AMRadio] Re: Supply for PP 304TLs ??


Here you go DOn.

This outfit sells transformers for building a 304TL amplifier.

You can even buy the complete amplifier kit for $85,000

Hehe - yes guys, that IS $85,000

http://www.alumrocktech.com/prices.shtml

Of course, it's a stereo audiophool amplifier.

Looks like you could get a set of transformers for a mere $14,000 or so.


On 1 Nov 2002 at 11:59, Merz Donald S wrote:

> By any chance is anyone running an HF amp or transmitter with
> push-pull 304TL tubes in the final? I have this homebrew here that
> uses this configuration. But the power supply it came with is in
> pieces and it is a mess. It was modified many times over the years by
> the builder and has been patched more than a Microsoft software
> product.
>
> So this thing needs a filament, screen and HV supplies. I can use some
> of the original parts. But for the HV, I don't want to use the
> original design because it is built around a 230V pole pig transformer
> and has lots of quirky design anomalies that I can live without
> (Anbody need a 125 pound pole pig? It'll give you 1700 to 2200 volts
> all day. The only drawbacks are that it is huge and weighs a ton).
>
> So I'd like to cheat and see how someone else is powering their rig to
> get some ideas.  References to complete designs in any of the popular
> literature would be useful too. Can anyone help?
>
> 73, Don Merz, N3RHT
> _______________________________________________
> AMRadio mailing list
> AMRadio at mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/amradio


_______________________________________________
AMRadio mailing list
AMRadio at mailman.qth.net
http://mailman.qth.net/mailman/listinfo/amradio

_______________________________________________
AMRadio mailing list
AMRadio at mailman.qth.net
http://mailman.qth.net/mailman/listinfo/amradio
_______________________________________________
AMRadio mailing list
AMRadio at mailman.qth.net
http://mailman.qth.net/mailman/listinfo/amradio