100W Power Amplifier from DB9MAT : real RF output

358 views

Skip to first unread message

Julien NICOLAS

unread,

Sep 24, 2021, 9:37:15 AM9/24/21

to Hermes-Lite

Hello,

I have built a 100W Power amplifier from DB9MAT. I hacked the bias to avoid a thermal runaway that cause magic smoke,
and I changed the 6dB attenuator to a 3dB attenuator.

I noticed that I didn't get 50dBm from this amplifier,
I did several test with different power supply voltage : 45V to 50V,
and I only get 48dBm from it at 50V (130W) with the TX level of the band set to maximum (255) in quisk.
(Mesurent on an Agilent spectrum analyser)

Interestingly, with a 45V power supply, I get almost the same RF out : 47dBm.

I always set the bias to 100mA. This is maybe not enough ?

May I know how much power you get from this amplifier ?

Thank you for you help,
Julien, F4HVX

Matthew

unread,

Sep 24, 2021, 12:17:21 PM9/24/21

to Hermes-Lite

I'm not aware of too many of these boards having been built up so it is nice to hear of one being built.

Some questions to try and help understand your problem a little better:

- Which PCB revision have you built/populated?

- I'm presuming you have seen the issues list here https://github.com/mathisschmieder/HL2-MRF101/issues ? Did you make the bias mod against my notes in the issues list there?

- What "number" did you settle on for the 100 mA bias? Mine is around 190 I think (after the bias mods I describe in the link above).

- You mention getting 48 dBm/130 W, I presume the later is a typo?

- You mention 255 as the band specific PA level for each band. Can you post, per band, PA current and resulting output with 48 V?

- I am a little concerned that if you are driving the PA at full power out (from the HL2) and only using a 3 dB attenuator, you could be over driving the input to the FET? Can you confirm with an N2ADR board installed (instead of the MRF101 board), the power output from the HL2?

- Are you confident you have the J16/OC settings correct for the MRF101 PCB (as it is different from the N2ADR board)?

I have the bias set at 100 mA and I have to turn down the drive level a fair bit to maintain a steady 100 W on each band. Something doesn't sound right with your setup.