The AM Forum
March 28, 2024, 12:36:22 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Old Harris PDM paper  (Read 16466 times)
0 Members and 1 Guest are viewing this topic.
W1DAN
Contributing
Member
*
Offline Offline

Posts: 899



« on: October 01, 2017, 07:44:49 PM »

Folks:

Some fun old info on PDM:

http://steampoweredradio.com/pdf/gates%20harris/pdm%20manual.pdf

Dan
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #1 on: October 03, 2017, 10:54:54 PM »

Very exciting article.

Just for fun thoughts, what size tube would it take to PDM a class C 4-1000 assuming a 6KV power supply and wanting to be able to run a 1390W carrier @3KV/(the lowest shown in the data sheet)?

Would another 4-1000 really be needed or would something smaller like a 3-500Z work? No, I think the 3-500Z can't take the current. Two might not either.

Another 4-1000 perhaps. I don't intend to ever run over 1500W peak on the air, but the only thing here that this thought experiment can be done on, where there is knowledge of the existing transmitter, would the the Tucker with the 4-1000 stage.

I do happen to have a couple of good and big 75KHz resonators from such PDM thing.

The supply of 4-1000s is limited, I believe they are no longer made? Something smaller maybe? 3-500Zs? just speculating wildly. I have to finish up the 3CX3000 amp first.

Where could I find manuals for PDM transmitters?

Or a rig itself? -save it from the scrapper! I'm OK with a higher powered unit, It can always be run at lower power to some degree by reducing plate voltage if these work the same way as usual switching supplies and class C amps. I saw a 5KW unit that was just a wide rack 6 FT tall at an AM station north of Dallas. Continental 315-R1 on 1310 I think.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #2 on: October 07, 2017, 03:34:05 PM »

Interesting historical document, but don't build it  Smiley

First, I will preface this:  I have built a fair number of TUBE pulse width modulators and their respective final amplifiers (PDM = Pulse Duration Modulation = PWM = Pulse Width Modulation:  PDM is a marketing term).

It can be done !  It is more complex than doing a solid state PWM modulator, but don't be daunted.

1) Choose LOW IMPEDANCE tubes (lower voltage, higher current tubes) whenever possible.  This will simplify the implementation greatly.  Consider a pair of 813s or 810s or modern Russian tubes or something like that for the RF amplifier.

2) Modulate the RF amplifier on the cathode side (this is not cathode modulation - it is high level modulation that happens to be applied to the cathode side of the RF amplifier rather than the plate side).  You can couple RF the drive through a DC blocking capacitor, and use pure grid leak grid bias.  Works like a charm.

3) The power supply voltage will be 2.5x the DC voltage you want to run on your RF amplifier.  Example: 813s.  You run 2000V @ 400mA (a pair), and your power supply voltage will need to be 5000V.  This will give you almost 150% of positive peak capability.

4) The filter, while it is big, is not overly hard to construct.  It may be possible to use a toroid in the 2nd inductor, shrinking its size considerably.  If anyone want to actually build one of these, I will assist in the filter and other aspects of the design if you wish.

5) Use a modern PWM generator.  The ones used in the class E rigs are 100% applicable to a tube implementation.  The PWM generator needs to be as clean as possible.

6) Use modern  MOSFETs and other components in the PWM driver.  This will make your life so much easier!

7) Use tried and true methods:  Drive the modulator control grid directly. It's just easier :-).

Cool Remember, you _MUST_ use analog compensation in a tube PWM design.  Tubes are very imperfect switches.  Analog compensation means you drive the modulator grid harder when modulating in the positive direction, and drive it less hard as you approach 100% negative modulation.  This linearizes the tube.  In fact, as the output from the modulator drops to about 15% or 20% of the carrier value, the modulator tube is pretty much operating in linear mode.  This will make things much cleaner.  Harris does the same thing in their designs.  Analog compensation is not overly complex to implement, but it has to be done right.  I can give you pointers.

Here is another link to a historical document.  It contains much good information, but the implement is somewhat dated.  Take the good ideas and use modern components.  There is a discussion about building a triangle wave generator for the PWM generator.  You don't have to worry about any of that - a superior implementation is available in the PWM generator board, available for about $50.00.    http://www.classeradio.com/pdm_article.html

Larry NE1S has the only tube PWM transmitter on the air that I know of.   He built the circuit shown in article, above.


Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #3 on: October 07, 2017, 04:15:38 PM »

Beginning of 6146 PDM rig:
http://amfone.net/Amforum/index.php?topic=35443.0

Evolution into 4D32 PDM rig:
http://amfone.net/Amforum/index.php?topic=35596.0

Pair of 4X'1's PDMing pair of 4X1's:
http://amfone.net/Amforum/index.php?topic=15816.0


Opcom:

Tube PDM rigs can be built with a lot of effort and are worth the time.  In the links above you will see info on the 4-1000A X 4-1000A PDM rig I built some years ago, as well as my current "working and on the air" Dual Quads 4D32 tube PDM rig.  The 4D32 quad PDM finals modulated by 6LF6 PDM tubes will put out an easy 500 watts of carrier at 180% and is one of my cleanest tube high level modulated rigs. I had it on the air last week shaking out the cobwebs for the season. I built another 4-1000A X  3CX-2500F3 PDM tube rig in the early 90's with tips from Frank / GFZ, as well as a third 4X1 version linked above in 2008.  Those were fine rigs, though they can be very dangerous because of the floating grid/screen supplies at full HV as well as the HV on the cathode, etc., etc. Yes, triodes are easier than tetrodes for both the final and PDM modulator. But 4X1's are what I had available and are my favorite tube.

Good luck with the project!

And Steve/ QIX, you have a short memory that Larry is the only person with a tube PDM rig. You and I  have talked many times with my 4D32 tube PDM rig as well as your comments within the old threads above.

T




Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #4 on: October 07, 2017, 04:48:56 PM »


And Steve/ QIX, you have a short memory that Larry is the only person with a tube PDM rig. You and I  have talked many times with my 4D32 tube PDM rig as well as your comments within the old threads above.


Hi Tom, I did forget about your rig - wasn't sure if you still had it since it has been SOOOOOOOO LOOOOOOOOOOOOONG since I've heard you on the air  Smiley Smiley Smiley Smiley Smiley   But I have heard it and as I recall, it sounds quite good !

It is possible to build a high power PWM rig with no floating supplies.  My pair of 450TLs modulated by a single 4PR1000 didn't have any floating supplies (that rig is documented in the article I linked in an earlier reply in this thread).

So Tom, it sounds as if you're going to be back on.  Very good!  Look forward to hearing you.  SC
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #5 on: October 08, 2017, 06:23:47 PM »

Taking the good advice here and looking at the projects, I think it best for me to do QRP experiments.. just a few watts to learn without destroying anything. 

Only then if I'm satisfied I could maintain one, look for a take-out commercial made transmitter. The 1KW power rock looks like it has room for changing bands. The 5KW version might be a bit much with tube prices what they are, and it may not tame down so easily to ham levels.

I can see that much more time and effort went into building successful high power units at home than I have available. 8-)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #6 on: October 09, 2017, 10:13:10 PM »

Taking the good advice here and looking at the projects, I think it best for me to do QRP experiments.. just a few watts to learn without destroying anything. 

Only then if I'm satisfied I could maintain one, look for a take-out commercial made transmitter. The 1KW power rock looks like it has room for changing bands. The 5KW version might be a bit much with tube prices what they are, and it may not tame down so easily to ham levels.

I can see that much more time and effort went into building successful high power units at home than I have available. 8-)

Yeah, I wouldn't start with anything too high power for a first project.  However, why waste all the effort for a PISSweaker?  With about the same effort, you could build a 50 to 100 watt PWM tube transmitter using some 6DQ5s - nice, low impedance tubes.

Just a thought..
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #7 on: October 13, 2017, 02:39:32 AM »

I have two cases of NOS 12BQ6s. Kinda small but low enough impedance and are free/on-hand. I thought about an 'am broadcaster' for use around the house. That tube would work there. Even that would need an attenuator on the output. A driver/PWM converter used there could drive the bigger tubes later.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Steve - K4HX
Administrator
Member

Offline Offline

Posts: 2722



« Reply #8 on: October 19, 2017, 10:56:56 AM »

I recall seeing the large air-wounds coils for your PDM rig many years ago Tom. It was quite a message.  Wink

Also GFZ PDM'ed a Viking II back in the 90s.



Beginning of 6146 PDM rig:
http://amfone.net/Amforum/index.php?topic=35443.0

Evolution into 4D32 PDM rig:
http://amfone.net/Amforum/index.php?topic=35596.0

Pair of 4X'1's PDMing pair of 4X1's:
http://amfone.net/Amforum/index.php?topic=15816.0


Opcom:

Tube PDM rigs can be built with a lot of effort and are worth the time.  In the links above you will see info on the 4-1000A X 4-1000A PDM rig I built some years ago, as well as my current "working and on the air" Dual Quads 4D32 tube PDM rig.  The 4D32 quad PDM finals modulated by 6LF6 PDM tubes will put out an easy 500 watts of carrier at 180% and is one of my cleanest tube high level modulated rigs. I had it on the air last week shaking out the cobwebs for the season. I built another 4-1000A X  3CX-2500F3 PDM tube rig in the early 90's with tips from Frank / GFZ, as well as a third 4X1 version linked above in 2008.  Those were fine rigs, though they can be very dangerous because of the floating grid/screen supplies at full HV as well as the HV on the cathode, etc., etc. Yes, triodes are easier than tetrodes for both the final and PDM modulator. But 4X1's are what I had available and are my favorite tube.

Good luck with the project!

And Steve/ QIX, you have a short memory that Larry is the only person with a tube PDM rig. You and I  have talked many times with my 4D32 tube PDM rig as well as your comments within the old threads above.

T





Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #9 on: October 19, 2017, 08:43:42 PM »

Hey Tom, still waiting to hear that tube PWM back on the air !!  Fall conditions are here - the bands have been good.  C u soon OM  Smiley
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #10 on: October 19, 2017, 08:50:52 PM »

So.....


I have some 1.6 kv fets.

And I have a pdm modulator.

And I have a Viking II.

Would it really be as simple as using my hv fets in place of the irf640 I currently use and a larger potential power supply as well as that pesky damping diode to make a tube based pdm TX?

Something says no.

--Shane
KD6VXI
Logged
Steve - K4HX
Administrator
Member

Offline Offline

Posts: 2722



« Reply #11 on: October 21, 2017, 07:09:12 PM »

Separate fil tranny for the 6146s.
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #12 on: October 21, 2017, 07:13:43 PM »

Would it need to be one of the special hv rated fil xformers?

Oh, the possibilities!

--Shane
KD6VXI
Logged
Steve - K4HX
Administrator
Member

Offline Offline

Posts: 2722



« Reply #13 on: October 21, 2017, 07:19:57 PM »

Yes.
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #14 on: October 27, 2017, 07:54:58 PM »

If you're building a solid state PWM type of modulator, even for a tube output, you don't have to float the RF amplifier, so no special fil transformers would be needed if you did it that way.

With solid state, you can much more easily float the modulator output, which in fact is what I did when I build the 6 6DQ5s modulated by a pair of 1600V IGBTs.  Now, this rig was a little out of control - I used a 2000 Volt DC power supply, and had a sophisticated arrangement to put 2 of the 1600V IGBTs in series.  Fairly complex, but it actually did work quite well.

So, if you use a single device as the modulator, most of the complexity disappears.  You can use an opto-isolator to get the PWM signal to the floating modulator output board.  From there, it's just fairly regular circuitry.  The only other thing that's the least bit unusual is that you'll probably want to have a little on-board (on the floating board) power supply for the driver and the opto-isolator.  This would involve winding a small toroidal transformer and feeding some high frequency signal through it.  Rectify the signal with fast recovery diodes, filter out the switching frequency and apply to a linear voltage regulator - 7812 and 7805 - to power the driver IC and the opto.  You only need an ampere or so to run everything on the board.

All of the current class E PWM implementations that use a standard buck regulator (my designs are examples) float the modulator output stage, and the RF amplifier runs at DC ground potential.
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #15 on: October 27, 2017, 08:53:05 PM »

Very interesting, and thank you for the reply, Steve.

When I designed / built the base pdm board, I did take a lead off the UCC12550  and run that into a homebrew  toroidal xformer (I don't remember the turns ratio, but it makes +/- about 20 volts from the pwm out pin.  Used a 43 mix)   That's fwb rectumfried with some BAT85 schottkeys, and run into a 3 terminal 5 and 12 volt regulator (and I also use that for - 5 and - 12 for my assymetry board!  Multi purpose!).

My 500 Watt rf deck runs at just about 100 vdd.

The pwm runs at 150 kHz, so it doesn't take much to filter the DC.

Thanks for the sanity check.  I have to change out my filters (I used thhn originally, only good to 600 volts supposedly) wire, and find something for the damping diode at these voltages.  Any ideas?

--Shane
KD6VXI
Logged
Pages: [1]   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 0.067 seconds with 18 queries.