This may be a futile attempt if the intermediate gain stages (pre-driver and drivers) fall off with frequency too. I have not done this myself yet… still on my list… to actually publish a table showing the output at each stage with frequency to determine what needs attention BEFORE the PA. Then a change to the PA makes full sense.
Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ
Owner - Operator
Big Signal Ranch – K9ZC
Owner – Operator
Villa Grand Piton – J68HZ
Soufriere, St. Lucia W.I.
Rent it: www.VillaGrandPiton.com
Like us on Facebook!
From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of Alex Buurlage
Sent: Thursday, January 25, 2018 3:37 AM
Subject: Re: [BITX20] uBitx TX performance with frequency
Yes, I've seen similar behaviour on my homebrew uBitx. Increasing the drive will overload the PA at 80 meters but will hardly increase the power output at the high bands. I even made a change in the driver circuit that decreases the gain below 10MHZ so I could crank up the overall gain of the TX chain more trying to get more output at 14MHz and up. No success. After extensive experiments I've decided to replace the IRF510's with RD16HHF1 mosfets (not pin-compatible!).
I am now optimizing the output transformer and impedance levels but so far the TX power ranges from 10 watts at 80m to 8 watts at 17m. The 15m and up LPF is not finished yet, but reviewing the specs of the RF16HHF1 I don't expect much further power drop at 28 MHz.
Secondly I like the idea of the housing of these mosfets being the source, so a direct grounding to the case is easy. No insulation material needed and a better RF ground.
A slight disadvantage is that the bias current of the RD16HHF1 is a little higher, but that's ok for me.
If you should make such a change with the commercial PCB version, beware of the pin incompatibility.