[AMRadio] RE: negative cycle loading

Tue Apr 6 22:40:50 EDT 2004

John,

Good point about the screens on tetrodes. I had that problem when trying 
to modulate a pair of 1625's. When using the normal voltage dropping 
resistor to feed the screens, the screen voltage would cut off long 
before the plates went to 100% negative. Sounded terrible.
I ended up with a voltage divider for the screens so that they did not 
receive as much modulation. Part of the screen voltage was obtained from 
B+ and part from the modulated B+. That worked fine once the proper 
values were found.

I monitored the screen voltage with a dual trace scope to see when it 
got cut off.

73
Gary  K4FMX

John Coleman, ARS WA5BXO wrote:
> Bret,	
> 	What Dennis is saying about the clipping is correct.  It would
> seem that it is all a matter of degree.  That is, less abruptness, in
> the clipping or limiting, means less higher order harmonics distortion
> and vise versa.  There or other things that may cause splatter during
> high modulation.  I have taken the liberty of making a list of check
> points that I use.  You may be familiar with these, but here they are
> for the record any way.  I have also used shaping circuits in the low
> level preamps of the modulation system.  High capacitance Varicap diodes
> can do some really neat stuff in high Z low level circuits.
> 
> Check points:
> 
> 1. The Class C circuit that is being modulated may go into spurious
> oscillations at the positive peaks of the modulation and this may not
> occur until a certain voltage at a certain audio freq appears on the
> plate.  It is more likely to happen if a circuits positive peaks or
> allowed to go higher than normal.  The gain of the tube begins to
> increase a lot on positive modulation peaks.
> 
> 	 Cures 
> 		1. Make sure that the tube is capacitively neutralized
> well, and that there is no inductive feed back path.
> 
> 		2. Make sure that you have enough RF grid drive for the
> high positive peaks.  As you raise and lower the plate voltage the plate
> current should follow linearly and so should the output RF.  (no change
> in efficiency).
> 
> 		3. On tetrode tubes the screen should be modulated as
> well as the plate circuit but when the plate reaches double the
> quiescent positive voltage, the screen should only have reached 150% of
> its quiescent voltage.  The same is true for the negative going audio.
> When the plate voltage has reached 0 volts on the negative audio peak,
> the screen should only have gone down to about 1/2 of its nominal value.
> The positive peaks here are the ones that will cause the tubes gain to
> increase causing any RF feedback to break into oscillation.  The
> incorrect negative peak ratios may cause the tube to shut of quicker
> than normal.  The exact correct ratio of plate modulation to  screen
> modulation may vary from one circuit to another.
> 
> 		4.  Reducing the load on a tank circuit so as to draw
> less plate current may cause the Q of the tank circuit to be increased
> and this will increase the possibility of spurious oscillations if any
> feed back can be found.  The better way to reduce power is to lower the
> plate voltage (and screen supply if a tetrode) keeping the Q of the tank
> where it should be.
> 
> 	2. I have heard that it is possible that some of the splatter
> from over modulation (carrier pinch off) occurs when the plate current
> shuts off and the tank circuit is left to ring.  I can see this to some
> extent and more so in high Q tanks but I would suspect this to be at a
> very low level only.  Others may have more info on this? I think the
> majority of the splatter is audio harmonic distortion imposed on the
> modulated envelope.  One thing is for sure.  It is a good idea to
> provide a load for the modulation circuit after the final has reached
> below the zero plate voltage point on the negative peaks of the audio
> cycle.  This can be done with a single HV diode in series with a
> resistor of the same resistance as Ep/Ip.  This series network is place
> from the plate modulation point to ground so that the diode is reverse
> biased except when the voltage at the plate goes negative.  This will do
> nothing for the ratio of positive to negative peaks of RF but it will
> help provide a load for the modulator during this period and keep the
> modulation XFMR from acting like a spark coil.
> 
> Good luck to you on your experiments. And keep doing them, that's where
> everybody learns, especial when you share the experience.
> 
> 73, John, WA5BXO
> 
> 
> 
> 		
> 
> -----Original Message-----
> From: amradio-admin at mailman.qth.net
> [mailto:amradio-admin at mailman.qth.net] On Behalf Of W7QHO at aol.com
> Sent: Tuesday, April 06, 2004 4:17 PM
> To: amradio at mailman.qth.net; brett.gazdzinski at mci.com
> Subject: Re: [AMRadio] RE: negative cycle loading
> 
> Brett,
> 
> Are you using the three diode circuit at   <
> http://www.amwindow.org/tech/htm/3diodeka.htm>?
> If so, this thing will clip all negative peaks that cross the "keep
> alive" 
> level.   How abruptly would depend on the internal impedance (i.e.,
> stiffness) 
> of the keep-alive supply but would give essentially the same effect as
> clipping 
> in the PA by crossing the zero line. This is where your splatter is
> coming 
> from.   Give the circuit at
> <http://www.amwindow.org/tech/htm/3diode.htm> a try. 
>   I've had good results with this setup adjusted to give up to 200% in
> the 
> positive direction without overmodulating the negative side.   For the
> resistor, 
> I've found a value around the load impedance presented by the PA works
> well. 
> You must use a scope to adjust this scheme.
> 
> Good luck.
> 
> Dennis D. W7QHO
> Glendale, CA
> 
> 
> --- StripMime Report -- processed MIME parts ---
> multipart/alternative
>   text/plain (text body -- kept)
>   text/html
> The reason this message is shown is because the post was in HTML
> or had an attachment.  Attachments are not allowed.  To learn how
> to post in Plain-Text go to: http://www.expita.com/nomime.html  ---
> _______________________________________________
> 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
> 