A problem about S11>0!!!

Hi all,

I have designed a LNA and a PIN attenuator separately. But when I made them in the same PCBoard, A problem appearred.In some frequency range, S11>0.

1) LNA:maybe its S11 is not good, seems below -4dB from 950MHz to 2150MHz, the best is -7dB. its S22 is below -5dB, in some frequency range it can reach -12dB.

2) PIN attenuator: When this attenuator work in large attenuation mode, its S11 is bad.

3) When LNA is connected with PIN attenuator, S11>0.

I think it probably oscillates, so I series a resistor to transistor's collector, but it cut no ice. and I have try some other methods, the results can't satisfied. so I need your help!

Thanks a lot!

You post is confusing. S11 is always, without exception, greater than zero in the real world (unless you mean 0 dB?)

When LNA’s S11 is positive in dB at some frequencies is not necessarily to oscillates, but at that frequencies is potential unstable. The reason could be the layout design (external feedback), internal feedback inside the active device, the emitter (source) degeneration, the load presented at the output, excess gain at that frequencies, etc.
A better characterization of instability is K-factor (Rollet Stability Factor). When K-factor is greater than unity, the circuit will be unconditionally stable for any combinations of source and load impedance.
Another stabilization method can be realized with a simple emitter feedback inductor. A small inductor can make the circuit more stable at higher frequencies. But if the source inductance is increased, the K-factor at higher frequencies eventually falls bellow 1. This effect limits the amount of source inductance that can safely be used.

Hi vfone,

Thank you!

Here I'll paste the layout picture, and wish you check it to find the reasons. Because I design the LNA and Attenuator separately, they work well, and also they work well when I used SMA connectors to connect them together. But when I design them in this pcboard, the problems occur.

1) The performance of LNA is different from the old.

2) Test the total performance of LNA+PIN Attenuator, dB(S11)>0 in low frequency when attenuator work in large attenuation.

About adding an emitter inductor, I realize it with strip and two ground vias, so it is difficult to tune it.

Check LNA's K factor, it might be already less than one.

maybe low frequency oscillation.

check DC

Seems that the ground of emitter is connected to the bottom ground layer only, through vias. Even if you will lose inductance try to connect to the top ground also.
Between LNA output and PIN attenuator try to use a low-pass topology matching circuit and tune this one for a compromise of all impedances of the attenuator (from low to high attenuation).