De-Clip leaded, simple 400W. Xmtr (put into an enclosure!). Pics & Schematic

[steve_qix]

Well, after um-teen months of using the ClipLead 400 (r)    transmitter unshielded with the tank sitting on the bench, I finally got 'round to packaging it up (at least a bit).

The simple RF amplifier is complete, including the RF drivers, and fits very nicely into a 7 inch high 14 inch deep, 19 inch wide rack-mount enclosure.  Shown with the cover off, so reveal the components within.  A cane metel top is normally used.

The unit operates nominally at 45 volts, 10 amperes (450 watts input), and puts out 400 watts (carrier).  Max modulation percentage is 200% positive.

Coupled with a pulse width modulator, occupying another 7 inch rack enclosure, the entire 400 watt (output) station will fit nicely into a small table rack, and is light enough to be EASILY moved.

This RF amplifier and the accompanying pulse modulator will become the recommend-to-build class E setup.  The design is virtually foolproof, and has been successfully reproduced many times.  The only tuned circuit is the output network - drive is broadband, digital drive accomplished using inexpensive ($3.00 typ) IXDD414 drivers.  The VFO described on the class E web site is connected directly to the RF amplifier.  No other drivers, etc. are needed.

Here is the RF amplifier with the cover off.


Here is the schematic:  (click http://www.classeradio.com/class_e_8_fet.pdf here for a PDF schematic)

[nu2b]

There has to be at least 1 clip-lead hidden in there somewhere!

Regards,
BobbyT

[steve_qix]

There has to be at least 1 clip-lead hidden in there somewhere!

Regards,
BobbyT

Ahhhhhhhhhhhhh.... if you look VERY carefully, you will find one    (hooked up, that is!).

[w8khk]

Looks like a yellow cliplead on a white wire to the fan.  Without it the mosfets would zorch!

Also an extra green cliplead not doing anything obvious, just laying there wondering how long he will be unemployed!

But I confess, I cheated.  It is not visible on the photo in the cropped AMfone view.  I had to save a copy and open it to view the right edge of the photo.

[steve_qix]

Looks like a yellow cliplead on a white wire to the fan.  Without it the mosfets would zorch!

Also an extra green cliplead not doing anything obvious, just laying there wondering how long he will be unemployed!

But I confess, I cheated.  It is not visible on the photo in the cropped AMfone view.  I had to save a copy and open it to view the right edge of the photo.

Bingo!  The fan is it      The fan operates only the transmitting; stops during receive.

Ok on the cropped view of the pic....hmmm.... I think you should be able to scroll the picture around.  If I reduce the resolution of my browser, I get scroll bars around the post, and can scroll to all parts of the picture and the schematic.  But, not all browsers are the same, so I don't know if you get the scroll bars around the pictures or not!

Fans.... not crazy about fans -  I don't actually need the fan on the heat sink, but without it and after a REAL old-buzzard transmission, the heat sink gets very, very warm.  I just don't like heat, so I put the fan in there to keep the sink COOL all the time.  It barely breaks temperature now - and the fan is a low RPM, quiet fan that really doesn't move too much air.   So, the fan seems like the prudent way to go - then nothing is stressed, heat-wise. 

I'm dissipating about 60 to 70 watts or so on the heat sink - about 20 watts of drive and driver power (all of the drive power is lost to heat), and 40+ watts in the R  D-S on when the MOSFETs are conducting.  The drain efficiency is around 90% (probably a bit better, actually, but I am losing a small amount of power in the tank circuit and output transformers).  The overall efficiency tests out at around 90%, but that includes everything, not just the MOSFETs themselves.

Regards,

Steve

[w8khk]

Very nice rig, Steve.  You got me interested in solid state and class E.  But I need to finish the projects I have started.  The tube rigs will keep the shack cozy warm in the winter, but come spring I guess I will order up some mosfets, etc.  I have the caps needed for the output circuit, and lots of heatsinks, so I guess I will start learning some things that cannot be found in the '40s handbooks.

Kewl on the quiet fan, only on xmit.  Quiet and unobtrusive.  That is the reason I prefer the 833As instead of the 4-400s, etc.  Convection and radiation, no forced air needed if ventilated properly.

I was wrong about the scrolling, the scroll bar was under the schematic, at the bottom of the post.  Works fine with firefox and internet exploder here.

[Mike/W8BAC]

I am green at this Class E stuff Steve. I'll admit that up front. How in the world can you have a 400 watt carrier transmitter and not have a huge heat sink?  Maybe everybody here already understands this but I can not comprehend how that works.

You say "overall 90% efficiency? That is astounding. It's  almost unbelievable. My experience with solid state finals in vhf and uhf amp's meant testing the heat sink often for warning signs.

If this is true I'm going to have to cross over to class E sometime soon. This is snake oil or goat gland radio, right?

Mike

[steve_qix]

I am green at this Class E stuff Steve. I'll admit that up front. How in the world can you have a 400 watt carrier transmitter and not have a huge heat sink?  Maybe everybody here already understands this but I can not comprehend how that works.

You say "overall 90% efficiency? That is astounding. It's  almost unbelievable. My experience with solid state finals in vhf and uhf amp's meant testing the heat sink often for warning signs.

If this is true I'm going to have to cross over to class E sometime soon. This is snake oil or goat gland radio, right?

Mike

Hi Mike,

88% to 92% efficiency is pretty much what you'll get with class E if everything is right - on 75 meters.  Better than that on 160, not quite as good on 40.  Caltech documented efficiencies of around 90% on 40 meters in their experiments.

This particular RF amplifier is about as good as it gets.  The drive is digital, providing an almost square wave at the gates.  The shunt capacitors are very high Q multilayer ceramics(no heating).  The RF bypass caps are 2 orange drops in parallel (no heating) on each module.  The output transformers use 8 cores, keeping things well out of the saturation range - so again, low power loss due to heat.  The output tank design is very important.  Too much coil and too little capacitance will result in high circulating tank current, and coil heating.  This one uses a lot of C and not much L, so the coil does not heat up very much.

Proper adjustment (VERY important) is facilitated by the "efficiency meter" - a device I created to show when the RF amp is correctly adjusted.  It compares the power input to the power output.  There is a board for this. 

The input waveform duty cycle is controlled at the VFO - set to around 40%.  Experiments have shown that a duty cycle of around 40% to 45% gives the best efficiency.  At 50% and above, the efficiency drops off RAPIDLY.

I know of other folks who have achieved high efficiencies with their class E transmitters as well, so my experience is not unusual.  Every class E transmitter I've built since 2000 has achieved at least 88% efficiency on 75 meters.

Interestingly enough, class E is used for some cellular telephone output amplifiers, both in the phones and at the cell sites.  I don't have particulars, but I have read about this when reasearching class E.

I generally use 88% as my "design value" for efficiency, when figuring power input V/S output, and for calculating heat sinks.  If it's a little better - great; if it's worse - I look for problems.

A note here:  making measurements at these high efficiency numbers is challenging.  The calibration of every measurement device has to be checked.  I have some calibrated volt meters and 2 calibrated ammeters that I use to verify the metering in the transmitters.  I use a known value (measured) military dummy load and a calibrated oscilloscope to make power output measurements.

Anyway, if you (or anyone reading) wishes to build one of these transmitters, there's lots of help available 

Regards,

Steve

[KF1Z]

Yup,
No extra heat for the shack with these rigs! 

My heatsink is 5" x 16" with shallow (1") fins.

I don't buzzard on as  much as some do (usually...)
But I have never felt the need for a fan on this thing.
Once in a while, the heatsink gets just warm to the touch, never got it hot.

Of course it's 12 fets, and I only run at most 300 watts carrier.
Could go lots more, but never seems like I need to ..


Now, of course I have to build this new modulator, and transmitter....
Just because it's there!


That is after I rebuild my VFO that went up in flames the other day, while I was twiddling around with my 12 volt class-D rig.   

[steve_qix]

I have recently updated the class E web site with complete documentation on the 400 watt (output) transmitter, VFO, and modulator/power supply.

There are also more detailed pictures, and construction details on every aspect of the project.

Go to the class E web site at http://www.classeradio.org for more info 

Regards,

Steve

[WA1HZK]

Awwwwwwwwwwww
You guys will Freeze To Death in the dark!!