The AM Forum
November 10, 2024, 09:17:12 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1] 2 3   Go Down
  Print  
Author Topic: 350W 3 band Class D FET AM Transmitter  (Read 83320 times)
0 Members and 1 Guest are viewing this topic.
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« on: February 09, 2017, 12:02:20 PM »

After countless hours of fun experimenting over several years, my pw modulated FET AM transmitter is now completed. It is a 350W carrier 1500W PEP band switched (160/80/40) high efficiency desktop design running in class D.  The RF deck employs 6 Cree C3M0280090D SiC FETs in single ended push pull running at 70V carrier at 5.3A driven by a single IXDD614 per phase. The output transformer is a T200A-2 core with opposed windings. The 3 relay switched output filters are fixed tuned 3 pole butterworth low pass using 2 layer air core inductors and silver mica capacitors. The modulator pulse generator is based on the LTC6992-1 chip fed by 2 audio stages from a TL072 op amp with feed forward ripple cancellation. The output is a half bridge using a pair of IRFP260 FETs driven by a IR2110 with high side bootstrap feeding a 4 pole butterworth modulator filter using a 6 layer aircore 1st inductor and a ferite 2nd inductor. A DDS VFO followed by an amplifier/buffer and co-phase driver provides the RF source. The case is wood construction with aluminum front and rear panels (I hate metal work). Tests have shown very high efficiency on 160 and 80m, the RF deck barely gets warm and on 40M just slightly warm.  Likewise the modulator runs just slightly warm. Convection cooling is all thats needed, no fans.

Still to do:  Make new meter scales 0 to 10A and RF power/SWR. (looking for a PC app)
Experiment with the LTC6992-4 variant PW mod chip which limits the negative peak PW to 5% instead of zero which may (or not) work out as a negative peak limiter.
Re-design the negative peak clip LED circuit which seems prone to RF or something.

Look forward to working some of you with this transmitter.  If you hear me on please say hello.

73s  Nigel


* 20170209_113835.jpg (552.49 KB, 2048x1536 - viewed 3879 times.)

* 20170209_113726.jpg (563.5 KB, 2048x1536 - viewed 2308 times.)

* 20170209_112535.jpg (892.64 KB, 2048x1536 - viewed 5870 times.)
Logged

A "Tube Free" zone.
WB5IRI
Contributing
Member
*
Offline Offline

Posts: 117


« Reply #1 on: February 09, 2017, 12:15:29 PM »

Absolutely beautiful, Nigel. Hope to work you soon!

Doug
Logged
KQ6F
Member

Offline Offline

Posts: 124


« Reply #2 on: February 09, 2017, 01:10:51 PM »

Nice work, Nigel.

Here's a meter graphics program that might work for you.
http://http://tonnesoftware.com/meter2.html
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2597


Bap!


WWW
« Reply #3 on: February 10, 2017, 12:36:24 AM »

That's fine construction - looks great.  Near layout of the modulator.  Do you have any pics of the RF deck?
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #4 on: February 10, 2017, 08:19:50 AM »

That's fine construction - looks great.  Near layout of the modulator.  Do you have any pics of the RF deck?

Steve,
In my haste to get this done I neglected to take enough pictures. The RF deck is the same one I used for fixed tuned 40M class E experiments only I changed the transformer to the double thick 200A-2 and added 4 relays, 2 per phase, and additional drain caps for band switching.  Attached is a picture of that deck before the changes.  Hope that helps.

BTW how is the high power GaN FET project progressing??

Edit:  Added a close up which shows the band switch relays and caps, the bottom half looks the same.

73s  Nigel


* 20161019_113456.jpg (776.96 KB, 2048x1536 - viewed 3321 times.)

* 20170211_142812.jpg (837.35 KB, 2048x1536 - viewed 3241 times.)
Logged

A "Tube Free" zone.
pw fallon
Contributing
Member
*
Offline Offline

Posts: 10


« Reply #5 on: February 12, 2017, 08:50:53 PM »

Hi Nigel,

Can you post a schematic of the transmitter?

Thanks,

Joe (aka PW Fallon)
WA1IWQ
Logged
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #6 on: February 13, 2017, 01:37:29 PM »

Hi Nigel,

Can you post a schematic of the transmitter?

Thanks,

Joe (aka PW Fallon)
WA1IWQ
Joe,  
The schematics are currently hand drawn in several notebooks with lots of scribbles and test notes, so need to be re-drawn. I will be happy to post them but it will take some time.  
I have a kind offer from forum member and fellow builder Rod KQ6F who is skilled with CAD to draw up the schematics from hand drawn PDF scans. We can certainly do this if there is any interest.
In the meantime I can post any part of it with a Word format narrative that you or others may be interested in.  It consists of VFO buffer/driver, RF deck, 3 band output filters, Power/SWR bridge and TR relay, modulator audio and pulse generator, modulator output and filter, and the basic low and high voltage power supplies.

In the past few days after some intitial modulator pulse ratio and mic gain tweaks, I have had some nice QSOs with this transmitter receiving good signal and "broadcast audio" reports.  Think I will take the rest of the day off.  Cheesy
73s  Nigel
Logged

A "Tube Free" zone.
W1AEX
Un-smug-a-licious
Contributing
Member
*
Offline Offline

Posts: 1481


Apache Labs SDR


WWW
« Reply #7 on: February 14, 2017, 02:04:17 PM »

Looks beautiful Nigel! Bet it sounds as good as it looks too! I look forward to finding you on the bands again.

Rob
Logged

One thing I'm certain of is that there is too much certainty in the world.
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #8 on: February 15, 2017, 02:49:43 PM »

Had a few Emails regarding the modulator design.  Attached is a PDF scan of the hand draw schematic. Perhaps KQ6F Rod will make it pretty.

Its a "Plain Jane" design, no peak limiter or audio band passing. I use a RE-27 mic and a W2IHY 8 band EQ externally which nicely tailors the audio and rolls off pretty quick above about 3200 hz. Several output FETs were tried but the old reliable IRFP260s worked best. The Gate Rs of 4.7 and 16 ohms are critical to prevent shoot through and overheating. My carrier PW is at 45% to allow for my own natural positive peak asymmetry, I see just under 400V modulated RF peaks which is almost 1600W PEP. A "Key Up" circuit is included to key the TR relay and also provide a 25ms transmit delay with zero turn off delay from a single switch.  This timing is important to allow the TR relay to settle before RF is applied. Mine is spec'd at 15ms so the delay was set to 25ms.  Open with Adobe, click view rotate CW.   Hope this is useful.

73s  Nigel

A revised copy of schematic attached which corrects some errors.

Reduce the 27k pull up R on pin11 to 10K and connect it to +12V

Also a pic of the modulator output board. Amazing how much audio power this little thing can produce at about 99% efficiency.

* Modulator.pdf (491.22 KB - downloaded 3004 times.)

* 20170220_102511.jpg (831.85 KB, 2048x1536 - viewed 4855 times.)
Logged

A "Tube Free" zone.
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #9 on: February 19, 2017, 07:22:40 PM »

When trying to draw a schematic from an assembly of notes in different locations the inevitable happens, mistakes occur.
I have replaced the schematic above with a revised one which I think is now correct.  It does work, honest.

73s  Nigel
Logged

A "Tube Free" zone.
M0VRF
Member

Offline Offline

Posts: 177


« Reply #10 on: March 18, 2017, 01:38:33 PM »

Here's my version!

Same design but I'm using TC4422's as drivers, same C3M Cree Fets in O/P.

You can just alter the turns on the toroids to change the O/P impedance.

I'm using 8T and getting 300W O/P @ 60V, efficiency is around 90% Will use my PWM modulator for this one.

I've also built one for 24V and used (4T on toroids) and using a Modtranny and 100W class D amp gives a 100W carrier @ 85% efficiency.

The PCB is my first attempt, have made a few minor changes in a new version.


* amp setup.jpg (89.92 KB, 784x724 - viewed 4015 times.)

* 300W amp.jpg (343.87 KB, 2159x1456 - viewed 3269 times.)

* amp pcb.jpg (132.76 KB, 808x970 - viewed 2496 times.)
Logged
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #11 on: April 08, 2017, 08:57:46 AM »

Some updated improvements to my 3 band RF deck.  The drivers were changed to 6  NCP81074BDR2GOSCT-ND drivers one per FET, running at 9 volts 1 regulator per phase with 6V input drive. They will NOT toggle with 5V input. These are very fast drivers with matched delays which allow the FETs to operate at their full fast switching speed minimizing the transition time through the linear range thus improving efficiency and upper freq limit. I have not been able to accurately measure efficiency but at 350 watts on 40/80/160 M the deck runs cold in class D so its way up there.  At a $1.25 each plus $5 each for the FETs makes for an economical high performance configuration. Couple pics attached.

73s  Nigel

Edit:  20/05/2017  Suppliers are now stocking the "A" version which toggles at logic level input (5V) independent of Vdd making driving them a bit easier.  Part #  NCP81074ADR2GOSCT-ND


* 20170328_152150.jpg (565.88 KB, 2048x1536 - viewed 4710 times.)

* 20170328_201604.jpg (793.83 KB, 2048x1536 - viewed 3743 times.)

* 20170328_162541.jpg (693.16 KB, 2048x1536 - viewed 3005 times.)
Logged

A "Tube Free" zone.
W3GMS
Contributing
Member
*
Offline Offline

Posts: 3048



« Reply #12 on: April 24, 2017, 06:33:46 PM »

So great to work you on 3.705 with your new rig Nigel.   It sounded fantastic! 

73,
Joe-W3GMS 
Logged

Simplicity is the Elegance of Design---W3GMS
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #13 on: April 25, 2017, 10:07:48 AM »

So great to work you on 3.705 with your new rig Nigel.   It sounded fantastic! 

73,
Joe-W3GMS 
Likewise Joe, thanks for a very nice QSO. Thoroughly enjoying the AM experience, builders, experimenters, restorers, all know which end of a soldering iron to hold. Smiley    Compliments appreciated.
73s  Nigel
Logged

A "Tube Free" zone.
ND8D
Member

Offline Offline

Posts: 8


« Reply #14 on: April 30, 2017, 01:31:39 PM »

Nigel,
What are you using for an HV supply on this?
Logged
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #15 on: May 01, 2017, 07:40:23 AM »

Nigel,
What are you using for an HV supply on this?

Andy,
The B+ is from a 120V primary multi tapped isolation transformer that I was fortunate enough to acquire in trade from a friend VE3GSI. Am using the tap that gives +160V into the modulator. The filter caps are 6  1,000mfd 200V electrolytic. Note that 160V is only about 10V shy of rectified mains so am about 1/2 done on a new 80M mono band TX based on this design that will operate with "off line" B+ and get rid of that big, heavy, hard to find, expensive power transformer. It should work, will know in a couple weeks.
73s  Nigel
Logged

A "Tube Free" zone.
ND8D
Member

Offline Offline

Posts: 8


« Reply #16 on: May 01, 2017, 02:26:52 PM »


Quote
"off line" B+

Similar to the K7DYY Class D transmitter, It would be convenient but I would be lying if I said I wouldn't be nervous about the safety aspects. What precautions need to be taken when rectifying straight out of the wall?
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2597


Bap!


WWW
« Reply #17 on: May 22, 2017, 05:16:36 PM »

I built an "off line" power supply / modulator, and have it at the Rattlesnake Island shack.

However, this one is implemented in a manner similar to a switching power supply - there is a high frequency pulse transformer between the rectified/chopped AC and the load side.  This completely isolates the output from the AC mains.

The whole thing weighs about 10 pounds - no power transformer.  Definitely more complex than a modulator using a power transformer, but much lighter.

Here is the circuit:






Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
KQ6F
Member

Offline Offline

Posts: 124


« Reply #18 on: May 22, 2017, 07:44:56 PM »

Steve -

I've been toying with the notion of trying something like what you have done but have the same safety concerns expressed by ND8D.  Can you share with us how you went about playing with the circuitry on the "hot" side of the pulse transformer?  Temporary isolation transformer?  Floating scope?  One hand in the pocket? Or what?

Rod
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2597


Bap!


WWW
« Reply #19 on: May 23, 2017, 12:35:57 AM »

For testing and development, I used an isolation transformer (1:1).  Once the development and testing was completed, I didn't need the isolation transformer anymore, and all has been well ever since (quite a few years at this point).

The circuit, as shown, works quite well.  I have also configured an overload shutdown board into the completed modulator, and use the standard PWM generator that I use for all of the class E rig modulators.

The PWM filter can be made entirely from toroids, which keeps the size down.  In fact, now I'm using filters made from only toroids at the KW level.  Makes for a much smaller filter for sure.
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 258


« Reply #20 on: May 23, 2017, 07:22:00 AM »

Steve -

I've been toying with the notion of trying something like what you have done but have the same safety concerns expressed by ND8D.  Can you share with us how you went about playing with the circuitry on the "hot" side of the pulse transformer?  Temporary isolation transformer?  Floating scope?  One hand in the pocket? Or what?

Rod
Rod,
My new TX project is intended to run off a 240VAC source wired according to the 2011 electrical code.  It starts at the power company transformer outside as a center tapped secondary. White is the neutral center tap and black and red are the two 120V hot wires running 180 degrees out of phase. The neutral wire is at ground potential and is indeed grounded either at the breaker box or by the power company at the transformer. This can be confirmed with an AC voltmeter from neutral to the extra (for safety purposes) ground wire at the outlet, or the separate station ground, mine reads zero volts. Wiring in older houses prior to the 2011 code change is anyone's guess but probably the same.

Using two 50A rectifier diodes from both hot leads, black and red, feeding the filter cap with the white as ground, full wave 170VDC (depending online voltage), will be seen at the cap. An inrush current protector device and shorting relay should be included to protect the diodes.  That's the plan.  I have NOT yet done this. 

My project is in the final stages of bench testing and not yet in a box. Its currently powered from a 120V isolation transformer fed from a Variac as its prudent to bring things up slowly when testing new designs. I'm currently experimenting with different output transformers, different low pass filter designs, and a higher voltage modulator, and so far its all working FB. The last step will be the power supply as above.  The only stuff in my house using big iron 120V power transformers is my HAM gear and I hope to change that.

73s Nigel

Logged

A "Tube Free" zone.
M0VRF
Member

Offline Offline

Posts: 177


« Reply #21 on: April 17, 2018, 01:52:56 PM »

Anyone else built Nigel's Modulator?
Logged
WD5BJT
Member

Offline Offline

Posts: 4


« Reply #22 on: April 17, 2018, 10:22:43 PM »

Nigel,

Very sexey unit indeed. May I be placed in the queue for a unit?
Keep up the great work. You give hope to us construction  challenged
hams.
73,
Charlie
WD5BJT
Logged
M0VRF
Member

Offline Offline

Posts: 177


« Reply #23 on: April 30, 2018, 08:38:21 AM »

I ask as I can't get the modulator to work and repeated mails to Nige have gone unanswered, NIGEL! Roll Eyes

It's behaving a bit wierd as I cant get the bottom FET to switch.

At least I think that's what's happening?

If you set the duty cycle to <50% it will NOT mod upto 100%.

I'm sure there's a mistake on the schematic and quite possibly a mistake in my layout altho' I have gone over it a 1000 times!

If I can get it to work properly I'll release the design so you can make your own PCB's (with Nigels permission of course!).

It's quite neat, about an inch square.

J.



Logged
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 599



« Reply #24 on: April 30, 2018, 10:07:01 AM »

John,

As I mentioned in my email, I am pretty good with pcb work. I would be glad to take a look at your schematic and layout and give it a second set of eyes. Could be a translation error on inputing the schematic into whatever tool you use or could be a layout error as well.

Just before I saw your post, I was looking at the hand-drawn schematic and thinking how easy it would be to make a pcb for it. I sometimes find it easier to do the pcb first, build it, then revise rather than building it up dead-bug or on perf board...

John - W9JSW

on edit - did you see the change where pin 11 is changed to a 10K pullup to +12? Is the enable line being properly controlled on the IR2110?
Logged
Pages: [1] 2 3   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.065 seconds with 19 queries.