The AM Forum

Looking for any information (schematics etc) on Progressive Series Modulation circuits.  Basically a series modulator with a second device (two transistors in parallel), very interesting, simple and much more efficient than a regular series modulator.  Looks promising, less power dissipated by the series devices.

Also, anyone ever work with a Diode Modulator circuit using a single diode?  RF and Audio are fed through resistors to the Diode, and then into a tank circuit.  Low level AM but looks interesting for something to feed a small linear.

Isn't than basically what the so-called class-H modulator is? 

Not quite so efficient as the pulse-width modulator, but much more efficient than the classic class A series modulator where one device serves as a variable resistor between the DC power source and the rf amplifier.

I have heard of something similar called a "totem pole" circuit.

The basic idea is to have several analogue devices wired in steps in such a way that all but one are either in saturated state or cut off, and each one serves as a variable resistance over a small limited range between 0 and 100% modulation. The more "steps" in the circuit, the more limited the range of each device, the more efficient the modulator, but also the more complex the configuration.

The late Hilmer Swanson of Gates Radio/Harris in Quincy had a design using PSM for the MW1 1 kW transmitter. I have the manual/schematic on another computer, you can also sign up with Harris as a customer and download manuals for their rigs, although they quit supporting the MW some time ago. I am shipping out an old MW1 on Friday via freight to an engineer for reuse in a commercial station.

The MW1 was Harris' first foray into solid state AM, before they did PDM. It followed the BC1H model that used 833s.

Hilmer Swanson of Harris Broadcast Division also invented a successor of PSM: digital amplitude modulation.  The technique digitizes the incoming program audio, and uses it to control the output powers of separate r-f modules which are then combined to directly produce the final, amplitude modulated r-f output signal.  There is no modulator, per se.

The design is highly efficient.  The 10 kW version needs only about 17.4 kW of a-c input power to produce a 10 kW carrier with +/-100% sinewave modulation.

The total r-f output power of a 10 kW carrier modulated +/-100% by a sinewave is 15 kW, so the a-c input to r-f output conversion efficiency of the complete transmitter is about 86% -- and this includes the power needed to cool it in ambient air temps to 122 degrees F.

This method also permits assymmetrical modulation capable of positive peaks to 145%, although the FCC limits AM broadcast stations to +125%.

Here is a link to a data sheet for the 10 kW version...

http://www.broadcast.harris.com/media/DX-10-15_25-8215.pdf

RF

We cannot overlook the neat W1FYO FET series modulator:

http://amfone.net/Amforum/index.php?topic=27856.0

RE Harris MW-1A:

Yep, as I recall, the MW-1A had a bipolar "Progressive Series Modulator" consisting of a bi-polar transistor audio driver driving a Class-D RF module.

There were 12 such modules with each module providing about 100 watts whose power was added before the output network. The power supplies were 70 and 140 Volts for the modulator/PA.

It was an innovative circuit arrangement at the time.

Phil - AC0OB

here's a tube circuit that i guess uses what you described.

http://amwindow.org/tech/htm/wb9eckseriesmod.htm (http://amwindow.org/tech/htm/wb9eckseriesmod.htm)

and the single diode circuit sounds like some kind of balanced modulator, i've always seen them with 2 or 4 diodes, they can be used to create the signal in a ssb tx, the output of a "balanced" balanced modulator is single sideband suppressed carrier, then that was fed through a filter to lop off one of the sidebands. what you could do with one is "unbalance" it enough to allow the correct amount of carrier for full carrier AM through, or use it for reduced carrier operation. the three diode balanced modulators i know of are the shunt type, ring type, and the modified ring.

I thought the multi-diode circuits would be either a balanced modulator (2 diodes) or ring modulators (4 diodes), but the single diode job is supposed to provide AM (Carrier & both Sidebands). It seemed so simple a setup, that for a low level rig, it would be sweet.  There must be something about it that makes it impractical, possibly low power handling? Or does it produce some type of distortion?

The WB9ECK series modulated PW rig is purely a series modulator. The two 6LF6 tubes are in parallel, and seem to be biased the same.  Most likely done for the current handling?  A progressive modulator would have one biased to cutoff before the other to limit power lost in the modulator. 

ED,
     In that series mod circuit, it looks like the 6LF6s were paralleled not for their current handling capacity, but to keep the plate resistance low for proper cathode mod operation. (note that he is applying screen voltage to them instead of tying the screens to the plates)

A single 'LF6 is probably capable of more current than the 807 that it is modding. A pair of 6Y6s or a 6AS7 would probably be a better choice for the modders.

In series cathode modulation, a very low plate resistance in the modders is crucial for good operation.

Progressive series modulation is Harris's "trademarked name" for Class H, which is a well defined class.

This is similar to Harris calling pulse width modulation Pulse Duration Modulation (PDM), another trademarked name for a common technology (pwm, in this case).

I can provide several class modulator schematics if you like.  It works very well for transmitters under a couple of hundred watts or so.  Above that, I'd use PWM because very little heat will be produced.  Class H modulators can be up to 80% efficient at carrier, but the efficiency of the modulator drops significantly under modulation.

you just answered a question i was asking myself the other day, what the difference between PDM and PWM was.

I thought PDM was Pulse Density Modulation.


                          _-_-bear

I'd like you to publicly provide your modulator schematic for the digital amplitude modulation technique invented by Harris Senior Scientist Hilmer Swanson, as covered by its patent holder:  Harris Corporation of Melbourne, Florida.

Note that the a-c line input power to amplitude-modulated r-f output power of the complete transmitter at all modulation levels when operating under that patent exceeds the 80% efficiency you report for Class H modulators producing zero modulation of the carrier (and less when they do modulate the carrier).

RF

Steve -QIX  if you have a simplified two device class H modulator schematic that would be great!

80% (or better) efficiency in-to-out is typical of modern, solid state broadcast transmitters of any design.  Pulse width modulators can exceed 95% efficiency or more (this is very typical), and class D / class E RF amplifiers operating in the broadcast band are well over 90% efficient. 

I have no schematics of so-called "digital modulation" transmitters, although I have seen one of a 50kW unit.  It was VERY complex.

There are advantages to keeping everything in the analog domain, however the down side is lower efficiency.  A class H (or class G) design is sort-of the best of both worlds - much higher efficiency than a standard series modulator, but entirely analog.

Personally, I tend to use analog modulators at low power, and pulse modulators above a couple or few hundred watts.

It is completely possible to make an analog modulator more efficient under modulation by using more steps.  By doing so, the voltage drop across any single device is reduced and the overall efficiency is subsequently increased.  This adds more complexity, and eventually it is just easier to use a PWM than bother with all the extra circuitry associated with a multi-step class H design.

You can find the paper on digital modulation and the DX-50 here:

IEEE TRANSACTlONS ON BROADCASTING, VOL.35, N0.2, JUNE 1989
DIGITAL AM TRANSMITTERS
HILMER SWANSON
Senior Scientist
Harris Corporation, Broadcast Division
P.O. Box 4290
Quincy, IL 62305-4290


Abstract: Digital amplitude modulation is a new system for generating
a high quality AM signal. This new modulation system has an overall
efficiency of greater than 80%. Digital AM was introduced in March
1987 with a 10 kW solid state transmitter. Also completed are a 25 kW
and 50 kW transmitters with plans for transmitters up to 2000 kW. The
Digital Amplitude Modulation system utilizes a high speed 12 bit
analog-to-digital converter (AID), a digital modulation encoder, and a
power multiplying digital-to-analog converter. The power multiplying
digital-to-analog converter reconverts the digital information back to
analog to form an amplitude modulated signal. Use of this system
results in an AM transmitter with outstandingeffciency, audio performance,
and reliability.

I have the complete electronic paper but didn't know if I could upload it or not.

Basically, the circuit comprises a 12-bit flash converter with DC bias and some extra circuitry, controlling Class-D
H-bridge RF modules. I believe there were a total of 128 Class-D H-Bridge RF modules.

Phil - AC0OB