Hi Duffey Mike and all, This may be true in some cases, but only if the antenna system has a design error. There are three major mechanisms by which we may kill preamps. 1) Poor filtering/stabilization of the DC supply voltage. When lightening occurs the EMP might induce high voltages on the DC supply wire. A big enough capacitor will convert the electromotoric force into a current that will charge the capacitor in reverse proportion to its capacitance. 2) Floating inner conductors. (This is in my opinion where most problems originate.) Antennas that have a folded dipole should have the midpoint of the dipole connected to the screen of the feed cable. Antennas with a gamma match need a shorted 0.25 wl stub to provide a DC short cirquit between the screen and the inner conductor. A small coil is another alternative. In case the preamp has a capacitive input one should add an inductor or a shorted 0.25 wl cable to prevent the inner conductor from floating. Floating inner conductors may become charged by various mechanisms and when the relay operates a hughe wideband pulse may reach the preamp - and in this situation it is better to have the DC supply on. Clever operators do not have floating inner conductors so they never experience this problem. 3) Poor isolation in the relay. The signals we transmit are narrowband. That means that the signal is a sinewave with an envelope that changes slowly. The GAAS fet will work as a diode towards the channel and sinewaves that enter the preamp will see the FET as a low impedance because of the gate current that occurs. As a consequence the input SWR will be very large and the important factor is the cable length from the relay to the preamp. A shorting relay will also have a very large SWR and if the cable length is badly choosen the relay could become power matched to the saturated preamp. In this case it is best to have the DC supply off. The gate diode will limit the RF to a lower voltage with a higher SWR as a result (less power into the FET). A lower temperature is probably also helpful. The data sheet for MGF1801 specifies the max gate current to 1.5 mA. (I found no data for other devices) Quite a lot of power (in a 50 ohm system) is required to produce 1.5 mA considering the extremely poor SWR that will be present in the saturated preamp. I have succesfully used very high power with poor relay isolation by selecting the proper cable length: http://www.sm5bsz.com/pindiode.htm (read at the bottom of this page.) 73 Leif / SM5BSZ