Back in the 1950s and 1960s a few vacuum tube PAs were sold that did not use
the conventional plate-tuned circuit, but used broadband tranformers. They
died an early death, thank heaven.

Push-pull tube amplifiers with link coupled coils etc were very difficult to
TVI proof. The pi-network (and pi-ell) with tubes in parallel was a big
improvement.

Bill W0IYH

"Highland Ham" wrote in message
...
" I just noticed in this ad: http://hamstation.com/ybstkftmk5.htm that
Yaesu's FT-1000 finals are MOSETS in push-pull. I don't ever recall
seeing a pair of final tubes in an xcvr or linear in push-pull except
perhaps way back in a very old QST. Can someone explain why push-pull is
used in the Yaesu and why we don't see it in "modern" tube amps?
=============================
Tube amps with 2 output valves can be used in push- pull but then for 1
band
only unless you would switch multiple inductors.
A single valve or multiple valve in parallel can readily output into a
multi-band PI filter arrangement.
In QST -March 1967 , pages 11 - 15 there is an article on a 1kW amplifier
for 50 MHz with 2 pcs 4-125A or 4-250A or 4-400A push pull . This is a
single band amplifier with inductive coupling on in- and output.
Unless you would perform band switching by changing the inductors through
plug-in , switching for various bands would be very complicated.

Frank GM0CSZ / KN6WH.

On Sat, 16 Apr 2005 09:47:25 +0000 (UTC), "Reg Edwards"
wrote:

Bipolar and MOSFET amplifiers are by comparison (with tubes) low
impedance
devices

==============================
Actually they are not.

Please expand upon your remarks.

==============================

For example, the internal resistance, Ra, of a triode tube is -

Ra = Mu / Gm

where Mu is the amplification factor and Gm is the mutual conductance.

Typical values for a triode are Mu = 10 and Gm = 5 milliamps per volt
which gives Ra = 2000 ohms.

Corresponding values for a silicon NPN bipolar transistor are Mu =
1000 and Gm = 40 milliamps per volt which gives an internal resistance
of 25,000 ohms.

Typical values for a beam tetrode are Mu = 200 and Gm = 5 mA/volt
which gives Ra = 40,000 ohms.

For a field effect transistor Mu = 1000, Gm = 10 mA/volt and an
internal resistance of 100,000 ohms.

Typically, transistors have higher internal resistances than tubes.
The volts and amps at which they work, the power ratings, are
coincidental and have nothing to do with it except that when comparing
one device with another they should have similar power ratings.

Take a look at their DC characteristic curves. Internal resistance is
the SLOPE of the anode (collector) current versus anode (collector)
volts characteristic for constant grid (gate) volts (or constant base
current).

OK?

No.

All of this is very interesting, but you are referring to the
*devices* that are used in the amplifier, while I said, "Bipolar and
MOSFET *amplifiers* are by comparison low impedance.

Reg Edwards wrote:
Bipolar and MOSFET amplifiers are by comparison low impedance

devices

==============================
Actually they are not. But everything else you say is quite correct.


Well their INPUT IMPEDANCE can be high, but their
OUTPUT IMPEDANCE when used as a power amp is just as
low as bipolar transistors, so the broadband transformers
are just as valid. And as far as their input impedance,
well these are class AB or B power amps, so they take
some driving power. In this case because of the required
driving power, the input impedance isn't very high.

"Ken Scharf" wrote in message
. ..
Reg Edwards wrote:
Bipolar and MOSFET amplifiers are by comparison low impedance

devices

==============================
Actually they are not. But everything else you say is quite
correct.


Well their INPUT IMPEDANCE can be high, but their
OUTPUT IMPEDANCE when used as a power amp is just as
low as bipolar transistors, so the broadband transformers
are just as valid. And as far as their input impedance,
well these are class AB or B power amps, so they take
some driving power. In this case because of the required
driving power, the input impedance isn't very high.
=======================================

This little discussion is about the relative internal resistance of
tubes and transistors. It has nothing to do with driving resistance
which, in the case of tubes and FET transistors, are both very high.

Somebody said the internal resistance of transistors, in general, was
much lower than that of tubes. He was incorrect.

Of course, it is possible to find an example of a high power
transistor which has a lower internal resistance than a given low
power tube.

But, in general, transistors have high internal resistances, greater
than tubes.

To base conclusions on the greatly different working voltages and
currents of devices is entirely incorrect. But it is easily done and
leads to misunderstandings when novices and learners attempt to
analyse circuit operation.
----
Reg, G4FGQ

"Reg Edwards" wrote in message
...


To base conclusions on the greatly different working voltages and
currents of devices is entirely incorrect. But it is easily done and
leads to misunderstandings when novices and learners attempt to
analyse circuit operation.

It is not incorrect. It is quite correct, and everyone who designs
solid-state linear RF power amplifiers (including me, not a novice) knows
that it is correct. The truth in this matter is perfectly obvious to any
really knowledgeable individual.

Bill W0IYH

William E. Sabin wrote:
"Reg Edwards" wrote in message
...


To base conclusions on the greatly different working voltages and
currents of devices is entirely incorrect. But it is easily done and
leads to misunderstandings when novices and learners attempt to
analyse circuit operation.


It is not incorrect. It is quite correct, and everyone who designs
solid-state linear RF power amplifiers (including me, not a novice) knows
that it is correct. The truth in this matter is perfectly obvious to any
really knowledgeable individual.

Bill W0IYH


Are we confusing load impedance with the device equavalent resistance?

Highland Ham wrote:
" I just noticed in this ad: http://hamstation.com/ybstkftmk5.htm that

Yaesu's FT-1000 finals are MOSETS in push-pull. I don't ever recall
seeing a pair of final tubes in an xcvr or linear in push-pull except
perhaps way back in a very old QST. Can someone explain why push-pull is
used in the Yaesu and why we don't see it in "modern" tube amps?

=============================
Tube amps with 2 output valves can be used in push- pull but then for 1 band
only unless you would switch multiple inductors.
A single valve or multiple valve in parallel can readily output into a
multi-band PI filter arrangement.
In QST -March 1967 , pages 11 - 15 there is an article on a 1kW amplifier
for 50 MHz with 2 pcs 4-125A or 4-250A or 4-400A push pull . This is a
single band amplifier with inductive coupling on in- and output.
Unless you would perform band switching by changing the inductors through
plug-in , switching for various bands would be very complicated.

Frank GM0CSZ / KN6WH.



There were dual band tank circuits for tube amps that worked on
both 6m and 2m without changing coils. The circuit was made in
both single ended and push pull versions. The push pull circuit
had two separate pick up link coils and basicly was series tuned
on 2 meters and parallel tuned on 6. The 6 meter tank coils were used
as rf chokes on 2 meters. The single ended version used a similar
trick, but was link coupled on 2 meters, and could either be link
coupled or a Pi network on 6 meters.
Source: old ARRL handbooks and VHF manuals from the 50's and 60's.

On Tue, 19 Apr 2005 20:22:10 -0400, Ken Scharf
wrote:

William E. Sabin wrote:
"Reg Edwards" wrote in message
...


To base conclusions on the greatly different working voltages and
currents of devices is entirely incorrect. But it is easily done and
leads to misunderstandings when novices and learners attempt to
analyse circuit operation.


It is not incorrect. It is quite correct, and everyone who designs
solid-state linear RF power amplifiers (including me, not a novice) knows
that it is correct. The truth in this matter is perfectly obvious to any
really knowledgeable individual.

Bill W0IYH


Are we confusing load impedance with the device equavalent (sic) resistance?

You may be confused, I am not. The OP wanted to know why solid-state
amps were push-pull without tuned matching networks while tube amps
are usually single-ended and have tuned matching networks. Bill and I
offered accurate explanations for this and Reg, as often happens,
wanted to inject comments designed to obfuscate the discussion.

Reg also opined, "Somebody said the internal resistance of
transistors, in general, was much lower than that of tubes. He was
incorrect." I believe that he was trying to put these words in my
mouth when of course, I didn't say that, nor did I see anyone else
saying that in this thread.

A vacuum tube amplifier operating at Ep = 4 KV, Ip = 500 mA and a 50
Ohm load impedance almost demands a tuned matching network.

An FET amplifier operating at Ed = 50V, Id = 40A is more readily
matched using broad-band transformers.

When comparing *one to the other* I contend that the vacuum tube
*amplifier* is a relatively high impedance device and the FET
*amplifier* is a relatively low impedance device.

As Bill already stated correctly, "The internal impedance of the
device is irrelevant."

On Tue, 19 Apr 2005 20:27:57 -0400, Ken Scharf
wrote:
[snip]

There were dual band tank circuits for tube amps that worked on
both 6m and 2m without changing coils. The circuit was made in
both single ended and push pull versions. The push pull circuit
had two separate pick up link coils and basicly was series tuned
on 2 meters and parallel tuned on 6. The 6 meter tank coils were used
as rf chokes on 2 meters. The single ended version used a similar
trick, but was link coupled on 2 meters, and could either be link
coupled or a Pi network on 6 meters.

Only someone who never used one of thse would claim that they
"worked." [g]