The AM Forum
March 28, 2024, 08:08:35 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 4   Go Down
  Print  
Author Topic: Zorch, again on my 8-FET 75 meter transmitter  (Read 78636 times)
0 Members and 1 Guest are viewing this topic.
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #25 on: February 18, 2018, 10:57:55 PM »


Jon,

    Very good on the class E rig, and repairs done.

I see your Hantek scope. I have the 100Mhz version otherwise just like yours. There is a Hack to make these scopes 200Mhz:

http://www.eevblog.com/forum/testgear/hantek-tekway-dso-hack-get-200mhz-bw-for-free/msg448551/#msg448551
and
https://www.youtube.com/watch?v=Mtt-g5putLY

I have not tried this, but just passing it on.

Jim
Wd5JKO
Logged
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #26 on: February 18, 2018, 11:38:23 PM »

I raised my dipole ends and my tuner is now good to go.  Tomorrow I'll make tuning adjustments during the daytime.  

I don't think I'm going to touch the innards of my scope.  It's my first digital one, and I like them much better than the analog ones. I don't want to risk breaking it.  Besides, my highest measurements are at 7 MHz so it'll definitely handle that.    

One final thought.  In this new transmitter configuration, the VFO sits on top of the transmitter.  Since the 'enclosure' is made out of plexiglass, I thought that the RF fields might wreak havoc on the high quality Chinese construction of the VFO.  So, I put a copper plate underneath it and grounded the four corners to the threaded rods.  It seems to work well.  I didn't even want to try it without any shielding.  I'd probably have 8 blown FETs had I done so.

Jon


* IMG_0340.jpg (2213.84 KB, 4032x3024 - viewed 989 times.)
Logged

It’s not just values, it’s business.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #27 on: March 01, 2018, 03:20:08 AM »

There's a lot to be said for good housekeeping with wiring.  I thought that my coax run was as it should be, but things were switched everywhere and I couldn't tell what went where.  Throw in an MFJ tuner with a bad antenna selector switch, and a coaxial open and it got weird.  I couldn't get a good match with my new dipole, and the E rig drain waveforms wouldn't come into spec. 

I pulled all coaxes and started from scratch and I now get a 1:1 match on my dipole at 3.960 MHz.  AM is frowned upon up there, so I need to adjust my coax to move back down with 'you guys.'  Joke.

Jon
Logged

It’s not just values, it’s business.
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #28 on: March 01, 2018, 10:27:14 AM »



Well John, theres a lot to be said about the " zoro beat " when answering a CQ or checking into a net. Lots of fun, ' specially when one has a thick skin.

KLC
Logged

What? Me worry?
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #29 on: March 01, 2018, 11:26:37 AM »

True, I can zero beat and they’d probably never know. It’s an interesting fact that just by adding 2 feet of wire, IQ goes up by a factor of 10. Skin, eh, that’s another story.

Jon
Logged

It’s not just values, it’s business.
W2PFY
Contributing
Member
*
Offline Offline

Posts: 13312



« Reply #30 on: March 01, 2018, 02:35:41 PM »

Yo Jon, I read this thread from top to bottom. I think that since your are always building something maybe a new title for you would be something like;

"Class E Adventures With Jon!"

I also like the way you build and layout your class E stuff and the tube transmitters that you have build in the past. You get my Blue Ribbon Award Cheesy Cheesy Cheesy

 
Logged

The secrecy of my job prevents me from knowing what I am doing.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #31 on: March 03, 2018, 08:49:03 PM »

I now have a good match but the waveforms aren't quite there yet.  I've had them darn near perfect with this rig before, so something's up.  

Skimming the book of Proverbs again recently, it caught my eye that you can't have thin wire as a gate bus.  On my old configuration, I used short pieces of #12 solid copper.  This go around I used a piece of #18-ish gauge wire for the 2-FETs per gate bus.  I'm thinking that's where the problem is coming from.  

This time I'm going to add on flat copper to connect the two gates.  

In the meantime, here's a link for you plate modulated guys:

https://www.youtube.com/watch?v=C-nqMcWeTO4

Jon
Logged

It’s not just values, it’s business.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #32 on: March 17, 2018, 11:32:44 AM »

I finally got around to it. I put 1/4” x 1” copper strips in between each gate pair.

I’ll test a couple weeks from now. I’m redoing the shack with a new desk. I’m going to use Panduit as a way of wire management rather than the rats nest method.

More desk space, neatly routed cables... holy cow!

Jon


* B3E6469C-338C-40F3-AACB-8072AD64BF51.jpeg (3080.37 KB, 4032x3024 - viewed 1071 times.)
Logged

It’s not just values, it’s business.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #33 on: March 30, 2018, 08:25:06 PM »

Here's the start of the shack re-do.  I'm trying to simplify and make it a cleaner look.  As a result, I'm doing the office desk thing with a "class E corner" (otherwise known as Plexiglass City; like the song https://www.youtube.com/watch?v=Rbm6GXllBiw.

The rack stays the same with the DX-60/3-500 linear.  

There's an Alinco DX-SR8 with the remote mountable front panel on the desk under the lamp.  

On the other side of the room is a 3' high pile-o-wires that I removed from the original setup.  I'm trying to put back in as few as possible.

Jon


* IMG_0563.jpg (2387.48 KB, 4032x3024 - viewed 913 times.)

* IMG_0564.jpg (2681.29 KB, 4032x3024 - viewed 969 times.)
Logged

It’s not just values, it’s business.
W2PFY
Contributing
Member
*
Offline Offline

Posts: 13312



« Reply #34 on: March 30, 2018, 10:06:55 PM »

Neat oh, you should be on the cover of The Rolling Stone, er, I mean QST  Grin Grin As much as I love the old clunkers, I think I'll build a 25 watt class E rig one day. Why 25 watts, I'm listed in Dummy's 101 book with this stuff, but as QIX had said or I think he may have said that it's easier, to build a class E rig over a tuber radideyoo..........
Logged

The secrecy of my job prevents me from knowing what I am doing.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #35 on: March 30, 2018, 11:31:31 PM »

He's wrong.  No offense.  Tubes can take blasting and destroying until you get it right.  Mosfets, not so much and the tiny soldering spaces are a bother. 

BUT, and this is the best part, the smug factor is out-of-this-world!  I literally walk around sniffing my own farts after I built the 8-FET rig.  And now I'm doing the modern desk thingy, like I live in California. "Yeah dude... and it's got an E-rig."

That would make a good QST cover:  Sniff your own farts! 

I'm joking, but really, this E stuff is where it's at.

Jon

Logged

It’s not just values, it’s business.
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #36 on: March 31, 2018, 01:40:32 AM »

Transistors have ratings
Tubes have guidelines!


My Use net tag line, circa Y2K

Pretty much true, even today.

--Shane
KD6VXI
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #37 on: March 31, 2018, 10:31:41 PM »

I've literally built over 100 transmitters in my life - both tube and solid state, and hands down, it's easier (at least for me) to build a class E rig than an _equivalent_ tube rig - for sure - once you get over a hundred watts or so of carrier power.

Tube rigs are great and all, but there just are a lot more things to deal with - powering up the cathode/heater, the high voltage involved, possible forced air cooling, lots of metal work, special and often large components, the cost of parts, etc. etc. etc.

Even with experimental class E transmitters, failures are very rare if you pay attention to the details such as proper sequencing, good gate drive, no parasitics, etc.  and live by the discipline that you do not proceed to the next level of testing until all of the underlying, dependent tests meet the design criteria and pass.

I have built transmitters upon which I performed more than a week of testing and verifying every parameter (and bringing things up very, very slowly - rechecking everything along the way) before reaching full power.  All of this testing - because the design is experimental or unproven - and absolutely every possibility must be considered, verified, and even the smallest anomaly corrected before moving on to the next level or step.

This attention to details really does pay off, but doing so does require a lot of personal restraint - there is always SUCH a temptation to skip the tests, and go for all of the marbles right off the bat  Cheesy   In my experience, this almost always results in disaster.  I follow the same procedure with tube rigs - I've had plenty of failures on the tube side as well - when I got careless, but great successes when I refrained from moving ahead too fast.

Yea, it's a pretty conservative approach, but I'm an old person - what can I say ?   Wink
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #38 on: April 13, 2018, 02:13:58 PM »

Making the gate connections with flatter, heavier gauge copper seemed to have helped.  I made the change and finally got the rig wired up again on the new desk and the waveforms look better.  Here's both drains at 400 watts carrier.  I'm going to run it at 300 watts just to be a little more conservative (ha).  I'm tired of rebuilding the heatsink assembly whenever I do something that causes a catastrophe.  

I changed my dipole antenna to include a 1:1 current balun at the feed point.  I'm not the core winding expert, so I chose to go with the good people at Palstar.  Their 4000 watt rated balun (B4000-1:1) sells for $60 at HRO.  I mounted it to a teflon cutting board, and drilled holes for the wire and mount supports.  

Jon


* IMG_0638.jpg (2192.89 KB, 4032x3024 - viewed 965 times.)
Logged

It’s not just values, it’s business.
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #39 on: April 15, 2018, 09:23:39 PM »

One thing that may be of great value is a very good overload shutdown system.  This will save *lots* of time when you're experimenting.

Also make darn sure there is no connection between gates of devices - that is, if a gate to drain short occurs, that there is NO WAY for any gate voltage to get to other gates - like through drivers, back to the driver power supply and to other drivers, etc.  All driver power should connect to a common point, and that point should be protected with a transzorb.

Another thing about an overload system is that it shuts things down very quickly if there is an overload.  The idea is to shut things down faster than any device can fail.  For standard MOSFETs, this is about 100uS
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #40 on: April 15, 2018, 09:44:01 PM »

I do have a driver power bus, so adding a Transzorb definitely is a good idea.

I’ve looked at your overload protection board to see if it could be applied to a Heising modulated system, and it can. Now that I have my shack reconfigured exactly how I want it, I’ll probably get one from you at some point this year. I just need to add it to my to-do list (not all ham radio related).

Jon
Logged

It’s not just values, it’s business.
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #41 on: April 15, 2018, 10:41:46 PM »

You'll need a series element that will carry a lot of current, and will switch quickly.  I have used IRFP260N MOSFETs for this purpose, connected as a source follower - between the modulator output and the RF amplifier.

At some time, I designed a switch circuit for adding to existing analog modulators, to provide circuit protection for both the modulator and the RF amplifier.

I'll see if I can find it.  Been quite a while!
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #42 on: April 15, 2018, 10:55:50 PM »

That would be awesome!

I can make a source follower for the modulator. But Steve, can’t you do that using tubes?  Smiley

Jon
Logged

It’s not just values, it’s business.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #43 on: April 23, 2018, 10:36:47 PM »

What is ham radio without experimentation (and failures)? 

I heard George in Tucson call CQ on 5 watts last night.  I couldn't help but reply.  I spoke lightly so that my modulation peaks wouldn't go over a KW and everything was working fine for several rounds.  I'm not happy with the phase 'b' waveform and something is still amiss with it.  George was operating a Heil pine board transmitter that he built.  It sounds really good and I could hear him pretty well in Phoenix for 5 watts carrier.

Anyway, my 4 year old son doesn't understand moderate voice level.  He got excited talking into the mic, and on my last go around he did a close-talk shout.  You know the routine - fire shoots from a FET on phase 'b.'  George, if you're reading this, I had a crap out.  That's why I didn't come back to you.

My plan is to go back to the original design of the rig.  I was able to give my 2KW wattmeter a good test without any problem back then.  Each FET had it's own driver.  I had 8 FETs and 8 drivers. 

I'm also going to add an aluminum shield between the output network and the heat sink assembly.  I think that's causing some problems as well since 'b' side is closest to the transmitter output. 

Jon

Logged

It’s not just values, it’s business.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #44 on: May 06, 2018, 05:09:12 AM »

My advice to new class E builders is to make the heat sink assembly removable.  It has paid off immensely for me.  This is the second time that I've decided to do a total rebuild of the solid state area, and all it took was to unsolder a few wires and unbolt two nuts.  It was an afterthought when I built the rig.  Turns out it was an epiphany greater than the flux capacitor. 

Jon


* IMG_0790.jpg (2610.65 KB, 4032x3024 - viewed 891 times.)
Logged

It’s not just values, it’s business.
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #45 on: May 11, 2018, 12:13:36 AM »

The thing is, you shouldn't have failures.

Failures such as you are having almost always point to a parasitic, assuming the devices are rated for the job of course - voltage wise and current wise.

For standard MOSFETs, at carrier, the voltage should peak up to no more than 1/4 of the device's voltage rating.  At carrier, the current should be no more than about 1/10 the device's current rating.

Assuming all of this is met, look for parasitics.  Most readily observed by watching the drain waveform under (particularly very low frequency) modulation.  The parasitic will show up as "grass" rising above the otherwise normal class E waveform.  The parasitics show up better when observing the drain waveform as an audio "envelope" , rather than individual RF peaks.
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
N1BCG
Member

Offline Offline

Posts: 836


« Reply #46 on: May 11, 2018, 01:39:03 PM »

I spoke lightly so that my modulation peaks wouldn't go over a KW and everything was working fine for several rounds.

Sounds like your audio limiter needs attention
Logged
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #47 on: May 12, 2018, 11:26:03 AM »

I've been able to hit 2kw peaks with this rig, so the audio chain isn't the problem.  It is the waveform on side B.  With reconstruction and adding the shield from the output network, I think I'll be back in business. 

I'm building the rig the way I did the first time:  8 drivers/8 FETs, driver bus (even though I've heard they're bad), ground bus in between the driver bus and the drain bus. 

I had some planned building time today for this transmitter, but I ran out of mica insulators.  ...something so simple.  Anyway, I'll order more.

Jon


* IMG_0816.jpg (2582.12 KB, 4032x3024 - viewed 960 times.)
Logged

It’s not just values, it’s business.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #48 on: May 19, 2018, 03:13:30 PM »

I just finished the heat sink assembly this morning. I’ll install it back in the transmitter sometime soon.

Jon


* B8B20362-6328-4EF3-A212-064EE7765B9B.jpeg (3030.97 KB, 4032x3024 - viewed 873 times.)
Logged

It’s not just values, it’s business.
ka1tdq
Member

Offline Offline

Posts: 1509


Red part turned in for a refund.


« Reply #49 on: May 20, 2018, 01:02:07 PM »

Installed and ready for testing!

https://www.youtube.com/watch?v=YBC1Qob27sM

Jon


* IMG_0840.jpg (3012.26 KB, 4032x3024 - viewed 924 times.)
Logged

It’s not just values, it’s business.
Pages: 1 [2] 3 4   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 0.069 seconds with 19 queries.