[SOLVED] ATF-35143 misbehaving?

Hello all,

Newbie here. I am trying to build a 7GHz amplifier using the ATF-35143 pHEMT with matching that I designed myself. I don't have a lot of experience with discrete transistors and amplifiers in general, my previous work was more in passive RF.

When I apply a 7GHz tone to the input of the amplifier, after biasing it with V_DS 2V and I_DS at 5mA, I get a lot of strange "tones" at the output, as in the photo:


I made some I/V curves on the same transistor (without a source or a load attached) and this was the result:


What could make a transistor behave like that? I have a package of them, this is the third one I've tried all with fairly similar results. I don't know if there's something that I'm doing wrong to break the transistors, I have taken all reasonable precautions against static.

Thanks in advance.
James

Your amplifier is oscillating. Could be from a poor PCB layout design or due to the bias setting and design.
ATF-35143 it is a high gain device, so the first step to fix this issue would be by reducing a bit its gain.
Adding a low value resistor (~10 ohms) in series with transistor drain would also help reducing oscillations, and not affecting much the noise figure.

Just to follow up - this is a good point, I didn't think about stability. I had expected an oscillating amplifier to produce a single strong tone, but never having done this before I guess you live and learn.

I have managed to stabilise the amplifier by adjusting the bias point, have not tried a resistor in series because there wasn't really any place in my board design for it, it was pretty compact.

As a result, my amplifier is quite rubbish but at least it amplifies a little bit and doesn't oscillate anymore. I think that by adjusting the bias-point, the S-parameters of the transistor are no longer valid and hence the gain characteristics of the amplifier will change.

On to round two I guess!