[AMPS] About triodes, tetrodes and IMD

[ilkka.kontola@nokia.com]

Some people are very persistently claiming tetrodes being 
superior to triodes in IMD sense. Some tube data sheets
do support this claim. I do no think that this is
always true, however. For example 3CX800A7 and
4CX250B are an example of one triode being
much more linear than one tetrode.

As W8JI pointed out some time ago some amateur HF
amplifiers using (Russian) tetrodes seem to have worse
IMD than other amplifiers that are using
3CX800A7 or 3CX1500A7 triodes.

My interest on the subject arises from the facts that I 
own Acom 2000A (2 x GU74b / 4CX800) and 
that I have started a 144 MHz amplifier 
project with GU84b (close to 4CX1600 ?).

To the utmost disgust of some QRP-fanatic PSK-31 
operators I have run 1 kW (Finnish legal limit) when
comparing MFSK-16 and PSK-31 on OH-W paths during
slight auroras. (MFSK is superior to PSK, BTW).
Before committing that "crime" I did make sure
that my signal is clean. Acom 2000A clearly is
much cleaner than my IC-746. With a professional
spectrum analyzer I could not see any difference 
on the relative IMD levels of barefoot 30 or 40 W and 
Acom driven with 30 or 40 W.

According to the QST May 2000 test of Acom the
3rd and 5ft order IMD products are -37 dB and
-60 dB from PEP at 1.5 kW output, respectively.

Some modern 2 x 3CX800A7 HF amplifiers seem
to have slightly lower 3rd order but higher 5th order
IM product levels.

According to QST September 2001 test of Ten-Tec 
Titan II (4CX1600) worst case 3rd and 5th order
IMD levels are -29 dB and -31 dB from PEP at 1.5 kW.
Even the 11th order IMD was only about -50 dB rel PEP!

According to articles on 

http://www.ifwtech.com/g3sek/index.htm

at least 4CX250B amplifiers must have good screen regulation
in order to achieve even acceptable IMD performance

On the FCC database you can find quite a lot of material
about the amplifiers that are type-accepted in USA.
Both Acom 2000A and Ten-Tec Titan II schematic
diagrams can be found there.

It seems to me that Acom has a well-regulated screen
supply that has a low output (DC) impedance for both
normal and reverse screen currents (not a rare phenomenon
on tetrodes!). As far as I understood the operation
of the Titan II screen supply it has a pretty high
output impedance that is even different for normal and
reverse screen currents.

Since the tetrode types used are different one can
not draw any definite conclusions about the issue.
In my opinion, however, this is yet another hint
that G3SEK is right on the screen supply
regulation (or DC impedance) issue.

73 
Ilkka OH3NJC




--
FAQ on WWW:               http://www.contesting.com/FAQ/amps
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com