10 Meter AM Transmiter

[W9LCE]

Solid State AM Transmitter for 10 Meters of about 100 watt output

What would be the best way to AM modulate a medium power Solid State Transmitter for 10 Meters?  I'm considering several approaches, lets look first at the traditional modulation transformer approach:

I picked up the RF amplifier board from a Heathkit HW 5400 (includes the heat sink).  The final amp uses a pair of MRF 454 transistors (or equivalents) in push-pull, which are rated at 80w each (?Class AB).  They have a peak voltage (CEO) of 25v and 20 Amps each (rated: 100w each). (The Collector-Base voltage is 45v)   Dissipation is at 250w each (500w).   At the considered AM voltage levels  - 4-1 peak to operating voltage - Vcc at 12v

To operate the board as Class C, the final transistors (and probably the second driver stage, a pair of MRF 479 transistors, also in push-pull) could be modified by removing the base resistor and replacing it with some form of RFC (pair of ferrite beads on wire).  (The first stage amplifier - pair of MRF 476 pp - likely could be left Class A or AB.)

The AM modulation transformer approach would match the power input to the Final, with an additional half power added from an audio amplifier.  Again the established concept is that the transformer matches the audio output resistance with the transmitter input resistance.   If the MRF479 transistors are driven at 3.5w (spec chart), the output (MRF454) is about 70w each (or pp - 140w), that means it is running at just under 12A.   At E=IR, the transmitter impedance is 12v/11.6A = 1.03 ohm   Is my math correct?  Is that an air-wound toroid?  Or maybe wound on a small plastic ring?  In my readings, you recommend a turns ratio of 4-1. 
Lets see – 140w/2 = 70w   -a 75w amplifier – use the 4 ohm speaker contacts.  [That would give 17.32v at 4.33A]   4 ohms to 1.03 ohms is 4-1 [actually 3.8-1 - that is OK] - and then there is the figuring of the number of turns.    
[Np /Ns = √(Zp/Zs)]   [√(4/1) = √4 = 2]  therefore – primary is 2x turns of secondary -
Do I just use any number of turns? (not!) Do I use a ferrite core?  Advise here is needed - -

[M0VRF]

You need IRON not air for audio (low) frequencies.

Audio frequencies don't have the magic of Radio (high) frequencies.

They need a magnetic medium to travel thru'

 

If you really have to use a mod tranny, then you will have to wind your own and it's going to need lots of turns so get a decent low end response.

Something like a T400-26D with min 150T on the secondary.

However as you'll be running a 12V carrier, the mod tranny will double this (nearly as there are losses) to 24V which is dangerously close to the 25V max CEO of your devices.

 

You'd have to run the amp from 6V to stay safe.

[W9LCE]

Thanks - I don't think I will try - with 15Amps - my table lists that as #16 wire - that would be tremondus! - -
I was just trying the idea out.

Is this where the 4-1 voltage that I vaguely remember comes in?   Does that also apply on the #2 message - on Series Modulation?

[M0VRF]

Hi, it's 2:1 and NO as with series modulation there's NO TRANSFORMER.

Just think of it as a variable resistor especially if it's a 3055!

Forget using a mod tranny, they're not good!
Lossy, non linear, heavy, a PITA to wind etc etc.

Stick with PWM, there's a chip that does ALL the hard work (UCC3570....)

Shame you cant (really) use class E at 10m so you're stuck with class C.

You're going to need a class C amp with 50V devices and run a 12V carrier with 24V peaks.

The amp you have will not work unless you run a 6V carrier and 12V peaks even then it's not designed for A.M. service and is purely an amp, probably quite linear? and for SSB?

I'd do a bit more research.

BR

Stretchy.