Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
Keying a 400 watt rig




 
The AM Forum
May 25, 2019, 03:47:46 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]   Go Down
  Print  
Author Topic: Keying a 400 watt rig  (Read 2940 times)
0 Members and 1 Guest are viewing this topic.
ka1tdq
Member

Offline Offline

Posts: 1208


Massachusetts Democrat


« on: January 07, 2019, 05:30:06 AM »

I'm still toying with the idea of CW keying a class E rig. Specifically, I'd like to key my 75 meter 400 watt 8-FET rig.

It runs about 10 amps at 45 vdc. I was thinking about using two PNP transistors in parallel (MJ4502G) and keying the base with an opto-isolator. I know we've talked about key clicks before, but the opt-isolator has a slight rise and fall time on its output NPN transistor. It might be just enough when keying the base of the two MJ4502G's. A sketch is provided for what I'm thinking.

The PNP transistors will each handle 100 volts at like 30 amps, or something like that.

Jon


* IMG_0376.jpg (2690.5 KB, 3626x2716 - viewed 130 times.)
Logged

Just follow the instructions.
http://www.ka1tdq.radio
PA0NVD
Member

Offline Offline

Posts: 530


Nico and Chappie (Chappie is the dog...)


« Reply #1 on: January 07, 2019, 08:04:56 AM »

The base current of each transistor will be in the order of 100 mA. The opto via the two resistors will not supply that Jon. You better use a FET and click filter in the gate.
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 1996


Making AM GREAT Again!


« Reply #2 on: January 07, 2019, 08:32:45 AM »

Here's where all that automatic stuff of PWM comes into play, Jon 🤔


Sorry bud, I couldn't resist!  Love just having to ground a pin in the modulator to go full bore TX!

--Shane
KD6VXI
Logged
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 253


« Reply #3 on: January 07, 2019, 09:39:32 AM »

There are 2 problems to solve here. Enabling the B+ from a morse key input and then shaping the leading and trailing edge to soften the pulse to avoid click harmonics.
A SI8234AB driving a pair of IRFB4227 FETs in a half bridge will solve problem 1. Key the logic input pin with +5V.  Adding a small series inductor and shunt capacitor ahead of the RF deck will solve problem 2. Since the deck probably already has (or should have) a bypass cap on the B+ only an inductor would be needed.
Note that this is really a low freq Pulse Width Modulator for about $15 worth of parts.  Time to roll up your sleeves and make some SMT PCBs. Its really not that hard and will open up many exciting new possibilities including a high power audio PW modulator for about $20.

73s  Nigel
Logged

A "Tube Free" zone.
W1DAN
Contributing
Member
*
Offline Offline

Posts: 831



« Reply #4 on: January 08, 2019, 12:56:42 PM »

Hi Jon:

In the Jan/Feb 2001 QEX magazine (and a related QST article), there is an article "A Keyed Power Supply for Class E Amplifiers" by Jim Backwater KF6SWC and others from Caltech. It is basically a line switcher power supply where the control is keyed and shaped.

Unfortunately, the ARRL search page does not reveal this article. Here is a questionable web site that has the article (be careful with downloading here)

http://docplayer.net/34167147-A-keyed-power-supply-for-class-e-amplifiers.html

Dan
W1DAN

Logged
ka1tdq
Member

Offline Offline

Posts: 1208


Massachusetts Democrat


« Reply #5 on: January 09, 2019, 06:22:54 AM »

All good stuff, thanks! Yeah, I guess I知 not keying a 2N3906 transistor and these pull more base current. I like my idea about the PNP痴 and I値l incorporate a series inductor/shunt as mentioned before.

To key the bases, I値l use a regular typical big FET/IXDD pair and key the enable/input lines of the FET driver.

I知 really looking forward to belting out 400 watts CW and tell the other OP that I知 using a class E rig. They池e going to be so jealous!

Attached is a picture of the overall heatsink assembly.

Jon


* IMG_0379.jpg (3144.07 KB, 4032x3024 - viewed 113 times.)
Logged

Just follow the instructions.
http://www.ka1tdq.radio
VE3ELQ
Contributing
Member
*
Offline Offline

Posts: 253


« Reply #6 on: January 09, 2019, 08:13:57 AM »

All good stuff, thanks! Yeah, I guess I知 not keying a 2N3906 transistor and these pull more base current. I like my idea about the PNP痴 and I値l incorporate a series inductor/shunt as mentioned before.
Jon
Using a single pass transistor through an LC filter will shape the leading edge only. It will need a big fast flyback diode to provide a return discharge path for the filter to shape the trailing edge. A FET will work much better and is cheaper.  Thats why I recommend a half bridge.  Happy building.

73s  Nigel
Logged

A "Tube Free" zone.
ka1tdq
Member

Offline Offline

Posts: 1208


Massachusetts Democrat


« Reply #7 on: January 09, 2019, 05:16:40 PM »

I didn't think about that. A recovery diode rated for 300 volts at 20 amps should do the trick. I added an SDUR2030 to my DigiKey order.

Jon
Logged

Just follow the instructions.
http://www.ka1tdq.radio
ka1tdq
Member

Offline Offline

Posts: 1208


Massachusetts Democrat


« Reply #8 on: January 14, 2019, 10:00:27 AM »

I placed an order to DigiKey for all the parts. Here's the heatsink that I'll mount everything to, including the terminal strip.

On the side of the transmitter, I've mounted a terminal strip with a 120v / 20a house switch to transfer drain voltage from either the moduator or this CW interface (an AM / CW switch).

I'm adding a few turns to the output coil on the transmitter too. Four turns should get me in the CW portion of the band. When I want to go back to AM, I just need to jumper out that portion of the coil.

Care needs to be taken when moving from one part of the band to the other. If the rig is mistuned, lots of drain current can be drawn.

Jon


* IMG_0399.jpg (2807.13 KB, 4032x3024 - viewed 96 times.)
Logged

Just follow the instructions.
http://www.ka1tdq.radio
ka1tdq
Member

Offline Offline

Posts: 1208


Massachusetts Democrat


« Reply #9 on: January 21, 2019, 09:16:18 AM »

I'm getting close to finishing, but I have one concern. I'm using a single fast recovery diode to compensate for the flyback effect pre-filter. It's rated 300 volts at 20 amps. I don't have it mounted to the heatsink because I figure the duty cycle will be low enough with CW that it won't generate much heat.

If that diode should fail, the FETs will feel voltage below zero volts. That can't be good.

Maybe I should use a couple of FREDs in parallel like on Steve's modulator? They're rated at 600 volts and 37 amps (and mount to the heat sink).

I'm also using a single PNP power transistor for switching rather than using two like I had originally planned.

Jon


* Sparky.jpg (3361.5 KB, 4032x3024 - viewed 92 times.)
Logged

Just follow the instructions.
http://www.ka1tdq.radio
ka1tdq
Member

Offline Offline

Posts: 1208


Massachusetts Democrat


« Reply #10 on: January 21, 2019, 02:35:30 PM »

I had to use a pull-up resistor on the big PNP to make it work.  For testing I used a 2.2k, but I'll put in a proper value (it's removed from the picture).

It works on the bench.  I'll throw two FRE-diodes under the coil mounted to the heat sink when I get them from DigiKey.  Then, throw it in the rig and see if I can generate some sweet high efficiency dots and dashes! (maybe click free?)

Jon


* Sparky2.jpg (1962.38 KB, 4032x3024 - viewed 115 times.)
Logged

Just follow the instructions.
http://www.ka1tdq.radio
ka1tdq
Member

Offline Offline

Posts: 1208


Massachusetts Democrat


« Reply #11 on: January 25, 2019, 07:02:57 PM »

All done except for the testing!

Jon


* 854B7313-4A2E-4649-B6D7-7C67CF70B6DC.jpeg (3351.57 KB, 4032x3024 - viewed 111 times.)
Logged

Just follow the instructions.
http://www.ka1tdq.radio
PA0NVD
Member

Offline Offline

Posts: 530


Nico and Chappie (Chappie is the dog...)


« Reply #12 on: January 25, 2019, 10:43:41 PM »

What is your affection for a transistor Jon? A FET is much more easy to filter at the input and doesn't require input current
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone ゥ 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.051 seconds with 18 queries.