[AMRadio] Re:Modulator Plate Voltage

Geoff/W5OMR ars.w5omr at gmail.com
Tue Nov 13 09:23:24 EST 2007

jeremy-ca wrote:
> ----- Original Message ----- From: "D. Chester" <k4kyv at charter.net>
> To: <amradio at mailman.qth.net>
> Sent: Tuesday, November 13, 2007 5:09 AM
> Subject: Re: [AMRadio] Re:Modulator Plate Voltage
>> Extra modulation headroom can be had two ways.  The modulation 
>> transformer turns ratio can be reduced (less step-down), or the plate 
>> voltage on the modulator can be raised, relative to the final.
>> Many transmitters run the same voltage to the modulator and final, 
>> and use a transformer with a turns ratio of about 1.6:1 to 1.7:1.  
>> This just barely allows about 95% modulation before the waveform 
>> flat-tops.  Often this is done intenionally to prevent 
>> "overmodulation".  The problem is, the flat-topping caused by 
>> modulator saturation produces exactly the same kind of splatter and 
>> distortion as overmodulation, so this is a bad idea.
>> Since my voice is asymmetrical, I need the extra headroom to 
>> accomodate the positive modulation peaks without splatter and 
>> distortion.  But the modulation transformer is fixed  ratio, so I 
>> raise the modulator plate voltage well above the final amp plate 
>> voltage.  I adjust to just below 100% modulation in the negative 
>> direction, per the oscilloscope, and let the positive peaks go where 
>> they may, making sure they don't  flat-top.
> What do you mean by well above? If the RF deck is at 2250 where would 
> you want the modulator to be?

You'd be better to reduce the plate voltage on the final.  813's -can- 
(and do) handle 2500v, just fine.

Raise the current in the final, reduce the plate voltage, run the 
modulator voltage higher.  Two separate supplies are required.

> Im in the process of building a pair of 813's with 810's audio and 
> have one big single PS that can get up to 3KVC@ 1A. How much can 
> 810's  be pushed in voltage or should I go to 572B's or even 250T's?
> Since Ive been unable to find a big mod xfmr so far the audio section 
> is still up in the air and flexible.

You're at the 'wonderful' stage of our hobby, Carl... experimentation!  
Slap it together, measure, clip tweek, peak, dip... take it apart, 
reconfigure and start again. Manipulation of electrons until they are 
doing what -you- want them to do!  Trial and error.  keep punching at 
it, man.

Ask yourself this... what impedances do I have to match?  Tube specs on 
tubes in modulator service are pretty clear about what the p-p impedance 
is, but still, impedance (AC Resistance) is nothing more than a function 
of voltage and current.

ie: the impedance of a final running 2000v @ 500mA = 4,000 ohms.  2000v 
@ 500mA is 1kW DC input.

The power output of the same final, running 4000v @ 250mA = the same kW 
of power, but what's happened to our impedance?  Why, it's now four 
TIMES as high (16,000 ohms).

Which, of the two, would be easier to modulate?

When you're trying to find a modulation reactor, the rule-of-thumb' is 8 
to 10Hy per 1000 ohms of Z in the final.  Is it going to be easier to 
find a 30~40Hy reactor, or would you rather try to find a choke near the 
160HY level?

Imagine 1000v @ 1amp.  Z = 1000 ohms.

I really think one would be better off, these days, running a good, 
stiff power supply for the modulator, and vari-ac the final power 
supply.  Yes, two separate supplies, to reduce the chance of common-mode 

With "Good Engineering Practices" is how rigs are to be built, correct?

Driving your AM Rig without a scope, 
is like driving your car at night, without headlights. (K4KYV)

73 = Best Regards,

More information about the AMRadio mailing list

This page last updated 22 Feb 2018.