[AMRadio] Push-Pull Parallel Tubes In Modulator

[Brett Gazdzinski]

There was an article in the AM press exchange a long time ago,
about feedback using the screen resistors.
The idea was to regulate the screen supply, run the voltage
slightly higher than normal, then put resistors in each
screen circuit to introduce neg feedback.
I tried some experiments, and maybe improved things a little,
but I think much depends on the screen current curves.

A good idea, since it also helps balance tubes in push pull parallel.

I don't remember who wrote the article, but it was many years ago,
about a DX100 I think.

Brett
N2DTS




 
> 
> 	Carbon resistors or placed in the plate wires of PP and
> especially PPP circuit as a preventative for VHF and UHF oscillations.
> When placed in parallel, triodes have a tendency to oscillate 
> because of
> the high plate to grid capacitance.  The short lengths of 
> wires that are
> used to connect the grid, plates, and filaments as well, may 
> become high
> Q VHF resonate circuits and it is easy to have a tuned grid 
> tuned plate
> (TGTP) oscillator when the tubes or wired in direct parallel. 
>  Placing a
> carbon resistor in one plate wire unbalances the circuit and 
> lowers the
> Q of the plate resonates hopefully to eliminate TGTP 
> oscillation.  Some
> folks place a resistor in each plate wire.  I have heard pros and cons
> about this.  I lean towards the theory of one resistor in 
> series with a
> single plate on each side of the XFMR and the other plate connected
> directly to the XFMR.  These resistors may be any where from 
> 10 ohms to
> 100 ohms (what ever works).  They are generally carbon or some type of
> non reactive resistor.  It is the skin effect that makes it 
> really work.
> The resistors have a lot more resistance at the VHF 
> frequencies than at
> audio frequencies.   
> 	Resistors in series with the screens of each of the four tubes,
> is a good idea.  This provides a self balancing inverse feed back to
> help equalize the gain of each tube dynamically.  Some one 
> has expounded
> on the virtues of screen equalizing resistors here on the 
> reflector some
> months ago as to the increase linearity even with just two 
> tubes in push
> pull.  I'm sorry I don't remember who it was.  I'll have to 
> look it up.
> Resistors should not be used in the control grids of a circuit where
> dynamic grid current will flow.  That is grid current that is 
> not linear
> with the input signal.  I would think this would include most 
> all class
> B and AB2 circuits.  We have enough trouble keeping the signal voltage
> up on the control grids with out placing resistor there.
> 
> There I go again with another long lengthy gol-darn explanation.
> 
> Sorry bout that.
> John, 
> WA5BXO 
> 
> -----Original Message-----
> From: amradio-admin at mailman.qth.net
> [mailto:amradio-admin at mailman.qth.net] On Behalf Of
> brett.Gazdzinski at mci.com
> Sent: Wednesday, December 03, 2003 7:36 PM
> To: amradio at mailman.qth.net
> Subject: [AMRadio] Push-Pull Parallel Tubes In Modulator
> 
> Don,
> When running tubes in push pull parallel, you can add a small 
> amount of
> resistance to the plate, value depends on plate voltage and current.
> 
> On screen grid tubes, you can add resistors in the screen and/or plate
> circuits.
> 
> To much resistance will cause problems, as will tubes that are much
> different.
> 
> In my ppp 100TH mod deck, I have something like 25 to 50 ohm 2 watt
> resistors in line with each plate, I don't know the exact 
> value as they
> have
> heat
> shrink tubing over them.
> All the tubes glow with resting current, and light up more under
> modulation, and all are equal.
> 
> The modulator runs 2000 volts, 120 ma (30ma per tube) resting current.
> You can likely discount peak current, as its short, and does not add
> much
> to the resistor dissipation.
> 
> Push pull parallel using triodes is not a great way to go.
> It works, but is never as clean as something running AB1.
> Even driving the grids with a very low impedance source presents
> such a variable load on the driver, very hard to keep clean.
> With AB1, its only voltage, so you can drive lots of grids without
> problems.
> I don't think you gain much with triodes, substantial driving
> power is needed, at a low impedance, harder to do than a stiff
> screen voltage supply in AB1.
> 
> With high power solid state driver devices, the triode method likely
> works
> very well.
> 
> Brett
> N2DTS
> 
> >
> > I have had this modulator chassis sitting around here that I
> > was fiddling with over the weekend. It came to me with no
> > tubes. It uses 4 tubes in push-pull parallel. The mod
> > transformer is a UTC VM-3 rated at 125 watts. The VM or
> > "Varimatch" series of UTC mod transformers was sold pre-war
> > and evolved into the later "CVM" series that is more commonly
> > seen today. This chassis may have come from a commercial
> > amateur rig because the construction standard is strictly
> > professional.
> >
> > Through some artful guesswork using only the finest
> > assumptions, I have concluded that the modulator tubes should
> > be Taylor TZ-20's. The TZ-20 is (quoting Taylor) "a high-mu
> > triode designed for zero bias class B audio operation". The
> > Taylor catalog even suggests this modulator design under the
> > TZ-20 listing, saying "4 of them in push-pull parallel
> > operation will form a most economical 140 watt modulator",
> > though no actual design for such a unit is given.
> >
> > Fortunately I have some TZ-20's (though I can use a few
> > spares if anyone wants to part with a couple). But I know
> > nothing about running tubes in PP-parallel. Are there any
> > special considerations? Should the tubes be matched in any
> > sense? And lastly, I suspect this should be obvious but I'll
> > ask anyway, how critical is the resistor between the 2 plate
> > caps of the tubes in parallel?
> >
> > Comments?
> >
> > 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