<html><body><div style="color:#000; background-color:#fff; font-family:times new roman, new york, times, serif;font-size:12pt"><div style="RIGHT: auto">Hi All,</div>
<div style="RIGHT: auto">I spent many years building receiver circuits and also loved Ulrich's articles over the years. </div>
<div style="RIGHT: auto">I have been on an IMD quest and find most information out there a copy of poor designs.</div>
<div style="RIGHT: auto">I have a little simulation going and quickly found. 1. The coupling betweeen drain phases must be tight better than 98%. 2. Same thing in the DC shunt inductor. 3. Take that little core used in most shunt inductor designs and throw it in the trash. I got a good IMD change by replacing it with a T106-2 with 10 turns bifilar (50volts). Iron core takes a DC offset better. 4. TLT transformers are the best way to go if you want to cover 160 through 6M. </div>
<div style="RIGHT: auto">Transformer core selection is key. Check ON9CVD's site on transformer design. Bob sure set me straight.</div>
<div style="RIGHT: auto">Hint, IMD degrades long before the core gets warm.</div>
<div style="RIGHT: auto">I have my modified 300 watt MRI strip doing -30dB IP3 at 200W PEP. When I get it where I want it to be I will</div>
<div style="RIGHT: auto">do work on my 1200 watt SD2932 boards. I built a heat exchanger that will hold 4-1200 watt boards that will be run at reduced power. Be careful with FET specs. Most degrade IMD above 2/3 power rating due to gain roll off. That new Feeescale 1200 watt part doesn't have great IMD above 600 watts. Freescale sent me a linear amp APP that showed it. </div>
<div style="RIGHT: auto"><VAR id=yui-ie-cursor></VAR>Frank WA1GFZ</div></div></body></html>