[Lowfer] Ultimate LOWFER Transmitter?

[W2MXW]

Hi everyone,

A Part 15 'colleague' of mine, Keith Hamilton of Hamilton Printed Circuit 
Board (PCB) Design, who manufactures FCC Certified Part 15 100 mW AM 
broadcast transmitters recently expressed the idea that he would like to 
manufacture a LOWFER transmitter based on his very successful AM 
transmitter design. Basically he would (within reason) like to make The 
Ultimate LF Transmitter.
His decision to do this would be predicated on being able to sell enough 
units to at least recoup the cost of FCC Certification (Type Acceptance) 
and so I am presenting this here so that you all might comment on what you 
would like to see in an LF transmitter that you would consider purchasing. 
Remember, unlike any other presently-available commercially-made LF 
transmitter this unit would be fully FCC Certified, which gives a measure 
of legal security over non-Certified equipment in that being tested for 
compliance it would be most likely to pass muster without nitpicking in the 
event of an FCC inspection.
Of course aside from legal issues it would be a full-featured, plug-n-play 
standalone LF unit which in and of itself would be desirable to many 
operators I'm sure.
The present features available on the current AM unit are as follows (not 
all of them would perhaps carry over to an LF unit and the LF unit we 
expect will have others not listed here - that is what we want your input for):

1. AM capability of 120% upward modulation (AM capability would remain in 
the LF setup because we anticipate some broadcast applications for this as 
well as LOWFER use, and because there isn't any other AM-capable LF rig 
besides the North Country kit available "out there").
Modulation is via transformer - as the present units are designed for Part 
15.219 100 milliwatt use even though the trans. is purposely over-rated at 
400 mW for extra safety factor its value would need to be increased for the 
1 watt input level.
2. On-board final tank/antenna tuning network (originally designed to 
resonate with the 8-1/2 foot whip antennas we use in AM service) which 
consists of an Amidon T-106-2 toroid RF transformer (I'm thinking this core 
would have to be changed to #3 material for LF if this design were 
desired??) which is multi-tapped on its secondary with the taps 
shunt-jumper selectable for tuning; the secondary is hi-Z naturally for 
tuning the whip; the primary is fixed at 4 turns (although he could make it 
tapped and jumper-selectable as well) and is lo-Z for matching to the 
paralleled MOSFET finals. Fine tuning is accomplished with a 10 pF sapphire 
dielectric high-voltage trimmer across the coil secondary to ground. I 
suspect in an LF version this value would need to be made roughly up to 100 
pF or so, perhaps using a parallel bank of fixed high voltage NPO ceramic 
caps which are jumper selectable. Something like a parallel bank of 
50-20-10-10-5-5-2-2 pF perhaps where one can add-in as many as needed (in 
other words select more than one cap simultaneously, in fact all if needed, 
with the jumpers).
LOWFERS may also want to use their own big external loading coils so 
perhaps the present design with the on-board lo-Z to hi-Z toroid 
transformer may not be wanted (it serves the same purpose) although leaving 
it out would detract from the 'plug-n-play' philosophy somewhat, and while 
perhaps not *the* most efficient design possible compared to the HUGE 
Lowfer coils is still very efficient nonetheless, and a good deal more 
legal in the face of strict rules interpretations. That toroid coil has a 
pretty high Q.
Would you like to see a very lo-Z output from the finals directly instead? 
A match from the very-lo-Z of the finals to 50 ohms nominal? Or some other 
design suggestions that would be popular?
3. The antenna output has a 1.5 kV gas tube arrestor to ground. This may be 
okay for LOWFER CW/QRSS/WOLF etc. use @ 1 watt but may need to be higher 
rated for AM modulation peaks at this level. There is also a static bleeder 
resistor across the ant. and a HV blocking cap to keep DC off the final 
tank. Volt rating of the latter is 3 kV and may need to be increased somewhat.
4. The finals are presently 3 paralleled Zetex MOSFETs with a combined 
dissipation of 3 watts (yes, Keith likes to over-engineer, his products are 
very reliable as a result) but for LF at a full watt may need to be made 
larger dissipation-wise or more added in parallel for extra safety margin. 
They are protected in the present AM design by a fast 40 volt Schottky 
diode across the drains. He can probably design the finals to operate in 
your choice of classes (C? D?)
Are there any 'pet finals' you would like to see?
5. There are on-board metering jacks for final current and voltage to make 
final input power compliance measurements, but an on-board meter itself is 
also possible.
6. The oscillator is normally fundamental xtal-controlled (10 ppm xtals) 
but these are expensive at LF and many LOWFERs might not want to be 
rockbound; a PLL design, or PLL/divider, or something else, are also 
possible. The osc. chip is presently a 4000 cmos. There is an optional 
trimmer for setting xtal freq. exactly. Any special taps or inputs/outputs 
desired in the osc. stage?
7. Buffer stages are presently 4000 cmos. Basically, the design of the 
whole unit is all-digital in the RF train.
8. Modulator is LM386, with on-board mod. level pot, and as mentioned the 
mod. transformer. This IC may need to be upgraded to a huskier model along 
with the transformer. Audio input is 600 ohms balanced also via a separate 
transformer. The audio quality of these units has gotten rave reviews from 
many of Keith's customers. This of course may not be an issue for strictly 
LOWFER use but we are keeping broadcast functionality too, remember.
9. There is an optional on-board NE570 compressor stage (which can be 
jumpered).
10. Power regulation for finals and all other circuitry is via separate 
on-board regulators; the final volts reg. is adjustable to make it easy to 
trim power input. The voltage for all the rest of the circuits is fixed. 
Regulators may need to be beefed up a tad, and/or have larger heat sinks. 
Power input is protected by a PolySwitch (tm) auto-resetting fuse and an 
MOV, and a reverse-polarity protection diode. Nominal power input in the 
present model is 11-15 VDC.
11. The unit is built on a double-sided plated-through-hole FR4 board. I 
assume that won't change. :-)
12. It comes in a nice fiberglass-epoxy weatherproof enclosure with 
gasketed lid and SS hardware  -but this adds to the cost. I assume most 
LOWFERS will want just a bare board to use with their own choice of enclosure.
13. These units in the original design are capable of multi-unit 
synchronization- in other words, you put up a cluster of 4 or 5 units at 
one location, one has the xtal and is the 'master' and sends it's RF signal 
to the others which are "slaves", each is a separate transmitter in its own 
enclosure with its own antenna but all units share RF drive from the 
"master", power and audio. Since each unit shares a common freq. source 
there are no heterodynes. Each unit incl. the master has on-board provision 
for both coarse and fine phase delay adjustment to assure the relative 
phases of the radiated fields match thereby summing the radiated power 
without mutual cancellation. This allows one to get a 6 dB increase (double 
the range) over one unit and is perfectly legal. So in essence the cluster 
functionally acts as if it were one unit of 4 times the power.
Now, this feature is expensive and the phasing tricky to adjust w/out a 
calibrated FIM so I'm assuming it won't be wanted in LOWFER circles (and 
impractical for LF hobby broadcasting because of the much larger antennas 
compared with mediumwave), but if I'm wrong please correct me!
14. Power/audio input/ant./gnd./sync in/out are all via captive PCB screw 
terminal blocks. Is this good or would other connections be preferred?
Are there any special features (which could be made standard or optional) 
that would "put you over the top" so to speak if you were "on the fence" 
about purchasing this unit? Any special considerations for LOWFER 
modulation/keying  requirements, IDers, etc?
Remember this is going to be a stand-alone unit not a PC card type thing 
and there is no direct interface with a computer.
I cannot comment on the price as it is dependent on the features we wind up 
with except to say that it probably will be a couple to a few hundred 
dollars. This may seem expensive but remember this will be a professionally 
engineered, highly reliable stand-alone rig, not a kit, and will be FCC 
Type Accepted, (which ain't cheap! :-)
Keith will naturally try to keep the cost as low as possible, but as you 
can see from the specs of the present AM broadcast unit he does not 
sacrifice quality and reliability.
Sorry to go on for so long here but remember, YOUR input will help design 
this rig!

73, Jon W2MXW