• AN313 App note (pdf)
• Communication Concepts AN313

8/20/07 -- Still struggling to find or build a wideband transformer that can work from 1 - 150 Mhz.  After discussing the problem with a few guys at work, the idea finally sank in, use two transformers, one for HF, and one for VHF.  Mount them on a relay board and use the band switch to switch between them.  This relay board can sit between the PA strip and the low pass filter right on top of the FET. 

8/10/07 -- I ordered the MRF275g from Henry radio and will be etching the board.  I'm unhappy with the way the T/R switch board is working, for some reason whenever I keydown with a little exciter power, the relay chatters.  Increasing the exciter power makes the chatter go away...very weird.  Maybe there is RF getting into the T/R circuit, or I think more likely some weird grounding issue since I etched the board single sided and it doesn't have a good ground plane. 

7/18/07 -- More and more progress.  The T/R board is etched, populated, and operational.  The low pass filter board is also etched and mostly populated.  At least two of the relays work.  Both boards an mounted in the enclosure.  The coax is routed from the T/R board to the hf/vhf pa strip, to the LPF, and back to the TR board.  The zener voltage regulator is installed on the PA strip to provide constant bias so the bias voltage doesn't need to be re-adjusted with varying power supplies.   I did some quick and dirty testing into the dummy load, 1 watt yeilds about 15 watts out with a 13v or 26v supply.  15 watts input gives about 110w out with 12v supply and about 195w output with 26v supply.  These were just quick "fun" tests to see that everything was working.  Final power in/out with voltage and current measurements will be done nearer to the project end.

7/5/07 -- No, it hasn't been this long since I've worked on the project, but sometimes it feels like it.  To date I've gotten the filter board layout done and the T/R board layout done.  I've also mounted the heat sink to the spreader board and setup most of the enclosure.  The power switch and input/output and key connectors are also mounted.   At this point, the only big remaining unknown is to make the input/output transformers work lower than the 10 Mhz that was specified.  Unfortunately, my quest for type 43 E-cores has proved unsuccessful.   I'm going to attempt to switch to using binoccular cores with a trifilar winding. 

2/19/07 -- More parts!  The UPS girl is getting to know my address.  She dropped off 3 more boxes of parts from Digikey, Mouser Electronics, and Gateway Electronics.  I have all the parts for the T/R circuit, and most of the parts for the low pass filter board.  All that is missing for the low pass filter board are the torrid cores, and a few capacitors that I have to track down or "make up".  I even have all the switching relays for the filter board.  Tonight I was able to prototype the T/R board.  Everything went well, it functions really well, seems to even work for QSK, at least it could keep up with my Kenwood TS-680 doing high speed CS.  I think I can even use this T/R circuit on Project 3, the high power amp.

2/10/07 -- I settled it last night to use the MRF275G for the UHF PA strip.  This part is rated at 150 Watts output up to 500 Mhz.  Not quite the bang for the buck of the MRF141G, but surely acceptable for what I'm looking for.  What is really nice is the test circuits and application notes within the data sheet have references for 225, 400, and 500 Mhz amplifier strips.  Also, the MRF257G is physically the exact same package as the 141G, so I can do my machine layout using the 141G as a reference.  See the mock up of the HF/VHF section and the UHF section assembled on the copper heat spreader.

2/1/07 -- Working below 7Mhz...NOT!  I called communications concepts today, I couldn't figure out what the core material of the input and output matching transformer was.  I was thinking that it must not be the app note, for operation to 2Mhz, a permeability of over 600 per cm sq. would be required with Curie temperatures of 130 to 145C.  Communications concepts confirmed that the ferrites were type 67 material, so the plan is to use the "experimental method" and order type 43, and type 77 material as large, or larger but just enough to fit around the hardline coax inductors. 

1/27/07 -- PA test strip.  The PA test strip is assembled and on the heat spreader.  Initial testing with a 12 volt power supply and the FT-817 look good, the am is outputting between 50 & 100 watts on all bands above 7 Mhz, exactly how it's spec'd.  Testing on 50 Mhz with the Kenwood TS-680 putting 10 watts into the amp is acting strange.  It might have something to do with not having enough voltage and current to supply the amp.  WOW! it gets that heat.