100 W amp. build progress notes

[Nick Kennedy]

I'm excited about building this amplifier and hoping for good results.

Most of the BOM parts were available on Mouser. Some have long lead times but I was generally able to find substitutes.

I purchased a Meanwell 48 VDC, 4.2 A power supply from eBay that looks good.

Oh, those 0805 parts!  I used to think that was the perfect size. Now two cataract surgeries, one retina surgery and a Fuch's dystrophy away, I'd love to see some 1206 parts in places where they fit.

I've got a nice heat sink about 2.5" x 3.5" with a 1/4" thick copper mounting surface and lots of aluminum radiator fins on the other side. The last time I tried using one, I was breaking taps and getting frustrated but this time I managed four mounting holes and the MOSFET mounting hole with minimal trouble. I took the WD40 advice.

With everything fitted up and using 1/4" sleeve type standoffs, I have a 0.010" gap between the MOSFET body and the thermal compensation diode, so I'll file down the standoffs about that much.

I've been mounting the low profile parts first, so most of the dreaded tiny ones are done. I check each one with an ohmmeter or L/C meter after installation to be sure I've got it in right.

I printed out large images of the boards and completed amp for additional guidance, but it turns our they are for the 'A' revision.  An x-ray image of the amplifier board similar to what was recently put out for the coupler board would also be nice. I will be attempting the coupler after I get the amp going.

Before connecting the power MOSFET or output transformer I'm going to energize the board and verify a proper adjustment range on the bias and its regulated voltage.  Also I should test the bias _off feature. I assume you apply +5 V or so to pin 1 to turn the bias off?

That's it for now. I'm proceeding with caution but excited about getting it going.

73

Nick, WA5BDU

[Nick Kennedy]

Answering my own post, here's another progress report. 

I'd read about C2 getting hot in this forum so I looked at some options for a physically larger capacitor. I settled on a leaded 220 pF dipped silver mica.

Before installing the MOSFET I connected the power supply and verified the adjustment range of bias voltage. It was 0.93 to 5.57 V. 

After installing the MOSFET and output transformer, I adjusted the bias to give 100 mA drain current which occurred with 2.73 V gate bias.  I let it run for ten minutes to see if the current inched up and it did about 15%. Of course the MOSFET doesn't get very hot with no RF drive.

Next I connected my KX3 set to 1 W output on 40 meters to the RF in terminals and the RF out terminals to a 40 meter LPF followed by my homebrew Arduino wattmeter and then a dummy load. Also I had my oscilloscope connected to the load.

My wattmeter showed 88 W while my scope showed 202 Vpp for 100 W. So I'm in the appropriate range. I put a 0.1 ohm shunt in the supply lead and measured 3.44 A, for 172 W at 50 VDC input. Assuming the 100 W output figure is correct, this is 58% efficiency.

Time for a QSO?  Or start building the T/R switch & SWR bridge board?

73

Nick, WA5BDU