The AM Forum
March 28, 2024, 01:38:53 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: PDM rig construction updated 2/7/05  (Read 17236 times)
0 Members and 1 Guest are viewing this topic.
W1IA
Member

Offline Offline

Posts: 778



« on: February 06, 2005, 05:29:36 PM »

Making progress...many thanks to young Bill KE1GF for his technical assistance.

https://home.comcast.net/~w1ia/
Logged

Run What Ya Brung!
wavebourn
Guest
« Reply #1 on: February 06, 2005, 06:41:31 PM »

Looks very impressive!!!
Logged
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10057



« Reply #2 on: February 06, 2005, 11:42:24 PM »

Keep a close watch on those PVC coil forms for any sign of overheating.
Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #3 on: February 07, 2005, 12:36:42 PM »

Vely Vely nice, Brentina!   I like the custom welded frame enclosure. Oughta be a nice looking unit when finished - and strappingly built too.

A few comments...

The PDM coils will be A-OK on the PVC pipe since they are running only 110 kc through them. I use the exact same set up as you are - even the red #6 wire.  They run cold.

But, the coupling between the two coils placed side by side is going to create some feed-thru problems. It will show up as "pilot carriers" up and down 110kc from your frequency. Probably down only -60 db or so.  I isolated my coils farther apart, at right angles, and even placed them outside the cabinet away from obstructions and initially only achieved about 60db of switching freq attentuation. I ended up putting in a 110kc resonant series trap right at the final B+ input which sucked it down to an acceptable -130++ db. It cud barely be heard on my own RX afterwards.

Also, consider adding some more drain bypass caps. I see only one per module. (500pf?)  I have the same layout as you, including the plexiglass holding the output transformers.... and found I need about 1500pf total to get a nice output waveform. Even 2000pf/per might be OK, depending.  Also, with the high current they draw, you need to spread the heat out amongst a few more, physically, if your intention was to use only one with a higher value...

Other than that, it looks great .  Post more pics as you progress, OM..

Yours will be the exact same config as mine - I'm so glad I built it.  Powerful, efficient, cool running, quiet - everything you want for AM. Just missing the glow, caw mawn.

The JJ PDM Rig meltdown last year:



And, the completed RF deck with the extra fet module spare:

http://home.comcast.net/~k1jj/wsb/html/view.cgi-photo.html--SiteID-1471619.html


Later -

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.
W1RKW
Contributing
Member
*
Offline Offline

Posts: 4410



« Reply #4 on: February 07, 2005, 03:43:23 PM »

T,
Those FET's look like a pile of spent roaches in an ashtray.
B

Quote from: K1JJ


The JJ PDM Rig meltdown last year:



And, the completed RF deck with the extra fet module spare:

http://home.comcast.net/~k1jj/wsb/html/view.cgi-photo.html--SiteID-1471619.html


Later -

T
Logged

Bob
W1RKW
Home of GORT. A buddy of mine named the 813 rig GORT.
His fear was when I turned it on for the first time life on earth would come to a stand still.
wavebourn
Guest
« Reply #5 on: February 07, 2005, 03:48:06 PM »

Quote from: W1RKW
T,
Those FET's look like a pile of spent roaches in an ashtray.
B



I had exactly the same impression... Dead bugs... Wink
Logged
W1RKW
Contributing
Member
*
Offline Offline

Posts: 4410



« Reply #6 on: February 07, 2005, 03:50:18 PM »

Actually, I didn't have dead bugs in mind.
Logged

Bob
W1RKW
Home of GORT. A buddy of mine named the 813 rig GORT.
His fear was when I turned it on for the first time life on earth would come to a stand still.
W1DAN
Contributing
Member
*
Offline Offline

Posts: 899



« Reply #7 on: February 07, 2005, 05:31:04 PM »

Hi Brent...

Good to chat last night.

I looked at your pix and it looks very good. It will probably work just fine as you are building it.

A couple of thoughts...

On the pink output transformer secondary ground...consider adding copper flashing to the center of the 4 groups of copper source busses and take the output ground from the center of the copper. The location of where you are taking tis from the heatsink may not be optimum.

On the PDM filter coils, if you do not have enough 100kc attenuation, maybe add a shield between the two coils that are paralell. I still like the idea of a series trap like K1JJ did.

Your caps should be in the mail this week. 3 4.7uf 1KV mylars and two 0.15/2kv drain bypass caps.

73
Dan
W1DAN
Logged
kc2ifr
Guest
« Reply #8 on: February 07, 2005, 05:50:35 PM »

Isn't there a web site dedicated to the class e stuff?Huh Do we have to see it here?Huh
Only kidding Brent........ :-P
Logged
W1IA
Member

Offline Offline

Posts: 778



« Reply #9 on: February 07, 2005, 06:49:11 PM »

Quote from: K1JJ
Vely Vely nice, Brentina!   I like the custom welded frame enclosure. Oughta be a nice looking unit when finished - and strappingly built too.

A few comments...

The PDM coils will be A-OK on the PVC pipe since they are running only 110 kc through them. I use the exact same set up as you are - even the red #6 wire.  They run cold.

But, the coupling between the two coils placed side by side is going to create some feed-thru problems. It will show up as "pilot carriers" up and down 110kc from your frequency. Probably down only -60 db or so.  I isolated my coils farther apart, at right angles, and even placed them outside the cabinet away from obstructions and initially only achieved about 60db of switching freq attentuation. I ended up putting in a 110kc resonant series trap right at the final B+ input which sucked it down to an acceptable -130++ db. It cud barely be heard on my own RX afterwards.

Also, consider adding some more drain bypass caps. I see only one per module. (500pf?)  I have the same layout as you, including the plexiglass holding the output transformers.... and found I need about 1500pf total to get a nice output waveform. Even 2000pf/per might be OK, depending.  Also, with the high current they draw, you need to spread the heat out amongst a few more, physically, if your intention was to use only one with a higher value...

Other than that, it looks great .  Post more pics as you progress, OM..

Yours will be the exact same config as mine - I'm so glad I built it.  Powerful, efficient, cool running, quiet - everything you want for AM. Just missing the glow, caw mawn.

The JJ PDM Rig meltdown last year:



And, the completed RF deck with the extra fet module spare:

http://home.comcast.net/~k1jj/wsb/html/view.cgi-photo.html--SiteID-1471619.html


Later -

T


Tom....I used 1000pf doorknob shunts, same as young Bill. Perhaps a piece of aluminum as a shield between those parallel filter coils? Otherwise I could run them vertically down the back with more separation....this is going to be tough if I need proximity separation and isolation.


Brent W1IA

p.s.   I finished the RF driver and the new photos are on the site
Logged

Run What Ya Brung!
W1UJR
Guest
« Reply #10 on: February 07, 2005, 07:01:58 PM »

Quote from: K1JJ
The JJ PDM Rig meltdown last year:


T


Hi Tom,

Did all those FETS crap out at once, or is that the result of some trial and error design?

Not being a Class E or Christine might argue "Classy" guy, what did those puppies cost? Looks expensive.

-Bruce
Logged
wavebourn
Guest
« Reply #11 on: February 07, 2005, 07:02:14 PM »

Quote from: W1IA
Perhaps a piece of aluminum as a shield between those parallel filter coils? e


Box or can, for better magnetic shielding.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #12 on: February 07, 2005, 07:50:53 PM »

Hi Bruce,

That crap-out was all at once.  It was my own fault that it got so bad.  The protection circuitry was shutting it down, but I couldn't see where the arcing was coming from. So, just like tube gear, I kept keying it wanting to see it blow out so I cud find the cause easily.  Two bad fets turned into a big handfull very quickly...  :lol:

But up to that point the four modules were running at different currents - no balance. When I replaced them the rig ran MUCH better and balanced. So perhaps I had a problem in there left over from initial testing that was ironed out the hard way. Cost, cheap - about $1.50 per fet.

Brent, I tried shielding between coils while watching the feedthru on the scope and it had no effect at all. Not sure why, but by better placement I got it to -60db down. That is still not great. Other guys did not need a series trap like I did to achieve -130 db , but whatever does the job...

For bypass caps, I ended up using 1500pf caps /module and Frank even suggested 2000pf. But since Bill/GF got away with 1000pf for the same design, it tells me that it is a custom thing and you just add them until the waveform and power eff is optimum.

If you run that rig at max QRO, I'll bet the single 1000pf gets too hot for comfort.  I wud use two 500's, one at the top and one at the bottom of each module. I added one in the center too. But, then again, take a look at the picture above and decide if you wanna take my advice, caw mawn... :lol:

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.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #13 on: February 07, 2005, 08:14:35 PM »

I agree more shunt C Brent. Remember drain bypass becomes part of the output filter C so you may have to change the output cap to a lower value. looks good!

Blowing FETS: The later the hour the louder the bang!!!!
Logged
KM5TZ
Guest
« Reply #14 on: February 08, 2005, 08:06:34 PM »

Quote from: K1JJ
Hi Bruce,

For bypass caps, I ended up using 1500pf caps /module and Frank even suggested 2000pf. But since Bill/GF got away with 1000pf for the same design, it tells me that it is a custom thing and you just add them until the waveform and power eff is optimum.

If you run that rig at max QRO, I'll bet the single 1000pf gets too hot for comfort.  I wud use two 500's, one at the top and one at the bottom of each module. I added one in the center too. But, then again, take a look at the picture above and decide if you wanna take my advice, caw mawn... :lol:

T


Tom,

Do you have any numbers? I'm curious how much of an improvement you got in efficiency with the additional shunt cap - and the change in pk drain voltage also?

I'm building a 3 band Class E tranmsitter (160/80/40m) so some compromise is expected - I switch in/out shunt caps with high current relays.

David.
Logged
KM5TZ
Guest
« Reply #15 on: February 09, 2005, 01:45:48 AM »

Quote from: W1DAN
Hi Brent...

Good to chat last night.

I looked at your pix and it looks very good. It will probably work just fine as you are building it.

A couple of thoughts...

On the pink output transformer secondary ground...consider adding copper flashing to the center of the 4 groups of copper source busses and take the output ground from the center of the copper. The location of where you are taking tis from the heatsink may not be optimum.

73
Dan
W1DAN


The output is transformer coupled. It does not need to share the same DC ground as the FET's. I'd recomdend the output ground be bolted to the ground in the vicinity of the output tank network. Just my 5c worth - looks very nice!

David.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #16 on: February 09, 2005, 08:56:07 AM »

I agree with the secondary to avoid RF current loops.
Additional shunt C will allow higher output power that is stable.
I'm running about 5000 Pf per 11 FETs which is 1 phase. I can
easily get 1200 watts of stable carrier  at  46 volts.
Output coil pretty hot so efficiency is only 83% I figure I could gain 3 or 4
% with a better inductor. This agrees with NU2B software.
The RDS on hurts efficiency at high power running 47 volts
so 1000 watts is better efficiency.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #17 on: February 09, 2005, 11:23:09 AM »

Quote from: KM5TZ
Tom,
Do you have any numbers? I'm curious how much of an improvement you got in efficiency with the additional shunt cap - and the change in pk drain voltage also?

I'm building a 3 band Class E tranmsitter (160/80/40m) so some compromise is expected - I switch in/out shunt caps with high current relays.

David.



Hi Dave,

I don't have any actual numbers, just observations on the scope and wattmeter in general.

The caps really cleaned up the waveform, not sure if it did anything for efficiency - never correlated it.

What I did was tune up the rig for the best looking class E waveform. I tried for a textbook look with the nice clean peak and the lower base hugging the bottom as shown on the class E site.  

Then I checked the power in vs: power out.  I adjusted the tap on the tank coil , thus changing the L/C ratio.  I added output bypass caps or subtracted them.

Eventually you will find a sweet spot where everything comes together... the power out  eff is high, the waveform looks perfect, the tank coil does not get too hot (optimum Q for situation) and the positive peaks are clean and high, etc.

It took me a good day of fussing, but it was worth it. I don't even look at the E waveform anymore, just the modulated RF envelope like any other AM rig.

I think everyone needs to go through this since all rigs are a little different even if they use the same number of fets, etc.

Also, if you go PDM, check your 100kc sidebands and if needed, add a series trap across the modulated B+ to ground. I needed it and now the supression is down about -130db.  I can post values for the L and C if needed.

Hope this helps.

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.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #18 on: February 09, 2005, 11:41:27 AM »

Tom and I did this effort in parallel with two different configurations.
I arrived at a higher shunt C because my power supply was only cranking out about 47 volts and I wanted to make a clean stable 1 KW carrier.
This meant my Drain Z was a bit lower than Tom's.
Tom ran a bit higher voltage so didn't need to run his drain Z as low meaning he didn't need to increase his shunt C as high.
We both arrived at the same conclusion that more shunt C is required for stable operation.  I also found the load cap value needs to be above 1500 pf in my case. I used a 1000 pf vac cap with 1000 pf fixed in parallel.
Series cap is 500 pf vac at almost 400 pf. Tom's cap values a bit lower if I remember as his Z is a bit higher. I thought we both tracked well in our conclusions. Tom's efficiency was a bit better as my load Z was only 10 times RDS on so the resistance of the FETs were coming into effect. fc
Logged
W1IA
Member

Offline Offline

Posts: 778



« Reply #19 on: February 09, 2005, 03:04:52 PM »

Quote from: K1JJ
Vely Vely nice, Brentina!   I like the custom welded frame enclosure. Oughta be a nice looking unit when finished - and strappingly built too.

A few comments...

The PDM coils will be A-OK on the PVC pipe since they are running only 110 kc through them. I use the exact same set up as you are - even the red #6 wire.  They run cold.

But, the coupling between the two coils placed side by side is going to create some feed-thru problems. It will show up as "pilot carriers" up and down 110kc from your frequency. Probably down only -60 db or so.  I isolated my coils farther apart, at right angles, and even placed them outside the cabinet away from obstructions and initially only achieved about 60db of switching freq attentuation. I ended up putting in a 110kc resonant series trap right at the final B+ input which sucked it down to an acceptable -130++ db. It cud barely be heard on my own RX afterwards.

Also, consider adding some more drain bypass caps. I see only one per module. (500pf?)  I have the same layout as you, including the plexiglass holding the output transformers.... and found I need about 1500pf total to get a nice output waveform. Even 2000pf/per might be OK, depending.  Also, with the high current they draw, you need to spread the heat out amongst a few more, physically, if your intention was to use only one with a higher value...

Other than that, it looks great .  Post more pics as you progress, OM..

Yours will be the exact same config as mine - I'm so glad I built it.  Powerful, efficient, cool running, quiet - everything you want for AM. Just missing the glow, caw mawn.

The JJ PDM Rig meltdown last year:



And, the completed RF deck with the extra fet module spare:

http://home.comcast.net/~k1jj/wsb/html/view.cgi-photo.html--SiteID-1471619.html


Later -

T


Well...after some pondering I have mounted the filter on the back of the rack cabinet....now aprox 1 plus foot of separation between the inductors.

I will move the secondary ground to the rf deck when it gets mounted inside the rack. Thanks for the tips guys!

Brent  W1IA

p.s. new pic on website

http://home.comcast.net/~w1ia/
Logged

Run What Ya Brung!
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #20 on: February 09, 2005, 03:32:51 PM »

Brent
The Secondary lead has a lot of RF current on it. You want both ends going right to the tank circuit with short leads.
I really think you need more tie wraps on the driver transformer
Hi Hi FB OM  WX fine here after school but I an a nerd in the basement
and my mom is yelling at me. "Egor come up you need light"
LOOKING GOOD!
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 0.082 seconds with 19 queries.