The AM Forum
March 28, 2024, 09:23:27 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 ... 9 [10] 11   Go Down
  Print  
Author Topic: A Clean Canvas - 24 Pill Class E Rig Construction  (Read 166858 times)
0 Members and 1 Guest are viewing this topic.
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #225 on: May 24, 2010, 07:44:52 AM »

Well, Tom you certainly sound better than I've ever heard you on the air in the past - and that's saying something, because you usually sound quite good !!

I have documented construction of a 24 MOSFET RF amplifier (and schematics) on the class E web site including step-by-step pictures of each stage of construction at:

http://www.classeradio.com/24_fet.htm

for anyone who wants to check it out.
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 #226 on: May 24, 2010, 12:51:02 PM »

Well, Tom you certainly sound better than I've ever heard you on the air in the past - and that's saying something, because you usually sound quite good !!

I have documented construction of a 24 MOSFET RF amplifier (and schematics) on the class E web site including step-by-step pictures of each stage of construction at:

http://www.classeradio.com/24_fet.htm

for anyone who wants to check it out.


Tnx, OM.  Hey, except for a few "JJ mods" it's all your basic design. It was a long haul, but well worth it. This transmitter is a cut above any modulation transformer rig I have. There's no way to get this super fidelity, power and low heat when using tubes. Setting up the audio for this rig has caused me to rebuild my audio system and make big improvements. The audio chain now uses balanced cables, has less hum, is more low-end capable and has cleaner highs. The e-rig demanded this due to no blower noise masking the low level crud and cleaner response. Rico will make a great summertime complement to the harem though I will still be running Fabio at times…(4X1 rig)

Of course I would not have been able to pull this off without your trailblazing work offering a fully functioning system design – and even supplying the boards. That helped things immensely. Even standing on those shoulders, it still takes some time and creativity to implement it effectively. I’m sure the e-rigs will continue to evolve and become even better over time.  Thanks for the help, Steve.  

Also thanks to Frank/GFZ, Jay/VD, Mark/ QFZ and others for advice from time to time.

Brent -  You said you needed a sequencer and shutdown board interface control schematic. I have not seen anything out there and don’t even have one for my own rig yet. I plan to draw out what I have here over the next few days and post it. Mine uses a 4-step sequencer with five additional relays slaved to it to handle all e-rig related functions. I hesitated to make a schematic since the rig was still evolving and even had some sequencer changes yesterday to fix the pulse problem. But we’ll get ya fixed up soon, OM.

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.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #227 on: May 26, 2010, 07:40:18 PM »

I finalized my sequencing and control ciruitry, so wanted to post this schematic for Brent and future builders. I've not seen one out there yet, so maybe someone can draw this one out on the computer and put it into the E-rig site.

I did it from memory cuz the wiring harness is hard to trace in the rig at this point. But it shud be pretty close.

T

*Update - I see one minor error. The HV relay that gets controlled by three series of relay contacts should have an external +12V included in the circuit or whatever it takes to key the external relay which in turn keys the main HV relay.


* E-Rig Control.jpg (70.55 KB, 1123x816 - viewed 1000 times.)
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.
W1IA
Member

Offline Offline

Posts: 778



« Reply #228 on: May 28, 2010, 07:38:21 AM »

Thanks Tom. I solved some of the problems from previous post by keying my sequencer off the VFO RY1 relay. Also series the B+ through the overload board. Will let you know when mine will be on the air....few more days to go.

I finalized my sequencing and control ciruitry, so wanted to post this schematic for Brent and future builders. I've not seen one out there yet, so maybe someone can draw this one out on the computer and put it into the E-rig site.

I did it from memory cuz the wiring harness is hard to trace in the rig at this point. But it shud be pretty close.

T

*Update - I see one minor error. The HV relay that gets controlled by three series of relay contacts should have an external +12V included in the circuit or whatever it takes to key the external relay which in turn keys the main HV relay.

Logged

Run What Ya Brung!
W1IA
Member

Offline Offline

Posts: 778



« Reply #229 on: May 28, 2010, 09:10:16 AM »

Added additional screws to source buss and soldered legs of devices.
Yuuuuuz guyzzzs missed it? Look closely at my layout and see whats wrong with waldo?! My dyslexia got me again. Had to tear down and redo the rf deck.

The winner gets a membership to the Macaroni net! Shocked Grin


* Amp 345.jpg (290.07 KB, 900x1200 - viewed 1010 times.)
Logged

Run What Ya Brung!
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #230 on: May 28, 2010, 09:50:05 AM »

Hey Brent, that's pretty good.....

You hooked the drains of the MOSFETs up to the RF drivers.....  Tongue

Anyway, those things happen...... but with proper pre-flight testing, nothing will be lost except the time it will take to correctly wire the devices  Cheesy   But you know all about that sort of testing anyway  Wink

I have built a LOT of transmitters over the past 40 years, and I can tell you I make mistakes all the time.  But, there is never a component failure because everything is brought up on variacs first.  I even bring up low voltage and low power stuff that way - saves damaging chips put in backwards (which happens) and stuff like that!

Anyway, won't be too bad to move the gate leads to the driver outputs.  All in a day's "fun"  Grin

Regards,

Steve
Logged

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

Posts: 2599


Bap!


WWW
« Reply #231 on: May 28, 2010, 09:53:38 AM »

[Yuuuuuz guyzzzs missed it? Look closely at my layout and see whats wrong with waldo?! My dyslexia got me again. Had to tear down and redo the rf deck.

Oh yeah, blame the dyslexia  Cool  It's really that Hawaiian disease Lack-a-nooky starting to set in.  Turns the best brains to complete mush (Brent's wife is away for an extended family visit -and, well, Brent *is* French after all!!!!)..
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
W1IA
Member

Offline Offline

Posts: 778



« Reply #232 on: May 28, 2010, 09:54:14 AM »

Well that explains the problem I ran into Undecided Thanks for the parts yesterday.
I have 3/4 completed the rebuild and will start on the efficiency board later. Sometimes you have to take 2 step backwards to make progress Roll Eyes

P.s. your Macaroni membership card is in the mail Grin

Hey Brent, that's pretty good.....

You hooked the drains of the MOSFETs up to the RF drivers.....  Tongue

Anyway, those things happen...... but with proper pre-flight testing, nothing will be lost except the time it will take to correctly wire the devices  Cheesy   But you know all about that sort of testing anyway  Wink

I have built a LOT of transmitters over the past 40 years, and I can tell you I make mistakes all the time.  But, there is never a component failure because everything is brought up on variacs first.  I even bring up low voltage and low power stuff that way - saves damaging chips put in backwards (which happens) and stuff like that!

Anyway, won't be too bad to move the gate leads to the driver outputs.  All in a day's "fun"  Grin

Regards,

Steve
Logged

Run What Ya Brung!
W1IA
Member

Offline Offline

Posts: 778



« Reply #233 on: May 28, 2010, 09:55:25 AM »

[Yuuuuuz guyzzzs missed it? Look closely at my layout and see whats wrong with waldo?! My dyslexia got me again. Had to tear down and redo the rf deck.

Oh yeah, blame the dyslexia  Cool  It's really that Hawaiian disease Lack-a-nooky starting to set in.  Turns the best brains to complete mush (Brent's wife is away for an extended family visit -and, well, Brent *is* French after all!!!!)..

I will have to fly back to TX for a conjugal visit Cheesy
Logged

Run What Ya Brung!
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #234 on: May 28, 2010, 11:51:46 AM »

Sounds like you're making progress, Brent. Yes, your method of wiring the sequencer is FB - there's many ways to get the same results.


Yep, we all make wiring mistakes. I have some legendary ones that almost put me to the edge of tearing down the project.

I seem to do the worse damage during final testing and modifications. Adding to the other meltdowns, three days ago I accidently touched the +12V against the VFO driver board freq counter output. It blew out two chips that I had to reorder from Mouser.  Just got it fixed last night.

Here's a good test:  Last night I put the e-rig on at full carrier into the dummy load. I then used a fiberglass rod and banged around in the RF deck looking for bad connections. Wouldn't you know it - TWO of the chip bypass caps on the drains came loose!   They had plagued me for weeks as jumping final currents and slight noise in the AM monitor.   I had not used enuff heat on the bus. I resoldered them all and the rig is now rock solid. Clean and stable even with hard banging... Grin

I'm working on adding pretty blue LEDs to illuminate the meters. Also installed a temperature sensor that controls a digital programable kiln module to trigger the cooling fans at 95F or higher.  Most of the time the fans are off.

Also added full balanced audio cables throughout the audio chain. With the feed forward hum working and the new audio, I can't hear a trace of hum - but we'll see about the VFO phase hum when I get it on the air.  Almost ready for some new tests.

Brent, when wiring the FETs, remember the MOSFET pin-out as GDS -  Gate-Drain-Source   - "Gawd Damned Slopbuckets!"

T


** Look for some final pics of Rico Suave rolled into the corner soon. (All analog meters, digital meters, freq counter, temp sensor working and showing full strap parameters)
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.
WD8BIL
Contributing
Member
*
Offline Offline

Posts: 4409


« Reply #235 on: May 28, 2010, 01:01:33 PM »

I noticed it Brent but I didn't want the Macaroni Net Card!
Enjoy it Steve Grin
Logged
W1IA
Member

Offline Offline

Posts: 778



« Reply #236 on: May 28, 2010, 02:06:06 PM »

I noticed it Brent but I didn't want the Macaroni Net Card!
Enjoy it Steve Grin

HA! I just finished the rebuild....it was worth it. I paralleled up some .047's to make the bypasses for the input to the transformers and solder the source legs closer to the fets on the bus. I found the mistake in the fets cause of another mistake. I put the feedback diode in the filter output backwards Shocked So in conclusions 2 mistakes make a fix Roll Eyes HA!
Now onto building the new efficiency board.
I'll send you an honorable mention card Buddly Tongue
Logged

Run What Ya Brung!
WD8BIL
Contributing
Member
*
Offline Offline

Posts: 4409


« Reply #237 on: May 28, 2010, 02:47:31 PM »

Quote
I'll send you an honorable mention card Buddly 

Don't bother. I'll just take the cash equivalent!
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #238 on: May 29, 2010, 12:08:37 AM »

To maintain continuity to this thread, here's the posting of Rico's final pictures:

http://amfone.net/Amforum/index.php/topic,24276.msg179668.html#msg179668

Hasta la vista, baby!

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.
W1VD
Contributing
Member
*
Offline Offline

Posts: 401



« Reply #239 on: June 08, 2010, 10:00:42 AM »

A couple of folks have asked about the 0.1 uF chip caps used for bypassing...

Vishay/Vitramon VJ1825Y104KXEAT (Digi Key part # 720-1132-1)

CAP CER .1UF 10% 500V X7R 1825

Frank and I have been using these caps in MRI MRF150 amplifiers for coupling and bypass and I've been using them in the Current Mode Class D transmitters for output coupling and bypassing of the IXDD414 drivers. So far they've proven to be quite robust.   
Logged

'Tnx Fer the Dope OM'.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #240 on: June 08, 2010, 10:07:42 AM »

I have done extensive lightning testing using these parts and they take a beating. There is an interesting effect with XR7 dielectric though. Pulse transients can reduce the effective C to almost 1/2 value during the pulse but the cap recovers.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #241 on: June 08, 2010, 11:59:12 AM »

Yes, those chip caps appear to work quite well. I use three 0.1's in parallel (0.3uf) for each module's modulated B+ transformer bypass point. So far they remain cold and FB during operation.

These chip caps can be soldered directly to a PC board creating zero lead length. The perfect choice for the job.


It's not a well focused picture, but see the PC copper board with a gap cut with a Dremel tool. The caps are soldered over the gap. The grounded end is aluminum angle that is bolted to the chassis. Keep the transformer leads short and you'll have minimal stray inductance.   The short transformer connections are on the back of the PC board - not visible, but short. The long red leads are B+ and can be any random length. The RF path is thru the short leads from the transformer, (in back)  thru the wide copper path, thru the caps and to gnd thru the aluminum angle.

Tnx for the cap suggestion Jay.

T


* 4X1 Rig 779.jpg (330.9 KB, 1280x960 - viewed 949 times.)
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 #242 on: June 09, 2010, 12:47:10 PM »

Tom, Where do you have those transorbs connected in the driver circuit?
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #243 on: June 09, 2010, 01:35:02 PM »

Tom, Where do you have those transorbs connected in the driver circuit?

Frank,

I don't use transorbs in the driver circuit.  I once added them at the input to the RF deck 414 drivers (at the junction of the four cable split), but found they rounded out the edges of the driver pulse due to capacitance.  I didn't try them at the final fet gates to see if it was any better.

I figure that the digital drive circuit with a self-contained VFO is pretty idiot proof, so no ricebox overdriven accidents will happen anyway.  There are transorbs in all the other obvious places, though.

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 #244 on: June 09, 2010, 03:07:02 PM »

Put them in the driver VCC line. Internal IXDD414 substrate diodes clamp transients to VCC on positive and 1 diode drop below ground on the negative side. Digital drive you take advantage or internal part die protection designed in and you won't load the output signal. As I said before any FET failure and drain voltage appears on the driver VCC line. This will cascade and take out all drivers and FET gates. This is why I thought each driver chip vcc should have a series diode. Feeling lucky......sucker.
This is why I run squarewave drive on the 160 meter rig transformer coupled.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #245 on: June 09, 2010, 04:28:04 PM »

Yes, we had talked about this before.

I already have an 18V transorb across EACH 414 VCC bus module to heatsink ground. (four transorbs as per Steve's schematic)

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 #246 on: June 09, 2010, 04:49:41 PM »

Good, but any FET failure will conduct back through the 414 substrate to the transorbs. Wanna bet who wins if the power supply gets across the 414 VCC bus. If you are very lucky the vcc bond wire will fuse in 1 414. Worse case everything blows. So a buck worth of diodes or load one round in the magazine and giver her a spin.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #247 on: June 09, 2010, 05:07:26 PM »

So what diode ratings and where exactly would you put them... in series with the VCC  to each 414  +lead? One for EACH 414 = 12 diodes?

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.
KB2RSK
Member

Offline Offline

Posts: 7


« Reply #248 on: June 09, 2010, 05:28:17 PM »

 Nice clean project.
 BTW  thanks for the comments on the signal from my power SDR Flex 3000
 I still need to do perform more tweaking on the EQ
 Thanks again to you and the group on 75 AM.
  73's
KB2RSK


* PICT0161 (3).jpg (152.44 KB, 800x600 - viewed 949 times.)
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #249 on: June 09, 2010, 07:56:49 PM »

Tom,
You want at least 200 volts PIV for a diode. Yes one in series with each 414 power lead. Transorb and bypass caps right at the 414. This way if a FET blows and puts HV on the driver output the diode will be reverse biased isolating the HV from the driver power bus. I bet 1N4007s would work fine, 3 amp diodes would be over kill but fine. This way if something goes wrong you take out 1 FET, 1 driver.
My 160 meter rig with transformer coupled square wave drive, when a FET blows, yes I'm blown 3 or 4 since '95 doing stupid things the CT of the driver transformer shorts the DC to ground. Most times the FET just explodes and the rig keeps on transmitting. This is due to the current mode feedback modulator. Imagine replacing all the drivers and FETs after ascrew up....That would send you back to tubes.
Logged
Pages: 1 ... 9 [10] 11   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 0.071 seconds with 19 queries.