1

I have a need to implement RF switch for HF circuit (~10 - 30 MHz) and I've been browsing options for that. Electromechanical switch would of course work, but I want to consider other possibilities too. Researching the options I have determined that PIN-diode based switch could be an option (need for low <2 pF OFF-state capacitance and low <5 Ohm ON-state resistance), but the low frequency sets some limitations for the diode.

I have been reading and in my understanding the width of the intrinsic layer is what limits the lowest usable frequency (based on this), but I do not understand how the minority charge carrier life time affects the PIN-diode characteristics. Seems like it's the main contributor for intermodulation distortion, but does it somehow relate to the I-layer width or are those independent of each other? Does it limit the lowest usable frequency? And is there some other parameter which does limit the frequency?

I can of course read from datasheets what's the frequency range for a specific PIN-diode, but I'd like to know what's limiting it.

EDIT
According to this the thickness of the intrinsic region determines the minority carrier lifetime. The wider the I-layer the more time the charge carriers have before they recombine. So, wider I-layer width leads directly to longer charge carrier lifetime. To use a PIN diode at relatively low frequency, a wide I-layer width is needed.

I did some measurements with BAR6403WE6327HTSA1 PIN diode (I-layer width of 50 micrometers), with a TEE compound switch topology, here's the used circuit:

TEE compound switch circuit

It worked quite nicely as a switch, insertion loss of less than 0.7 dB and isolation of 60 dB with up to 30 dBm input power at 13.56 MHz.

Don't know if this edit was really necessary, but hopefully somebody can find the information useful.

SamGibson
  • 17,231
  • 5
  • 37
  • 58
RautSa
  • 51
  • 6

1 Answers1

1

My thinking is that as long as you keep the PIN diode properly reverse biased the signal frequency should be irrelevant for conducting / blocking a signal.

I found this article where on page 2 is says:

At frequencies much lower than f τ , the capacitance characteristic of the PIN diode resembles a varactor diode. Because of the frequency limitations of common test equipment, capacitance measurements are generally made at 1 MHz. At this frequency the total capacitance, CT , is determined by applying a sufficiently large reverse voltage which fully depletes the Iregion of carriers.

Which seems to confirm my idea that as long as the reverse biasing is present and the signal doesn't momentarily forward bias the diode, the diode will behave as a (variable) capacitor.

In conclusion I think that the lowest signal frequency isn't a "hard" limit by itself. What does happen is that, as long as the PIN diode is kept in reverse bias, the reverse voltage will change with the signal and as the diode behaves as a varicap the capacitance will also change (with the signal).

Perhaps the lowest frequency is the frequency above which this effect (changing capacitance) becomes less relevant so that the diode can be seen as a (somewhat) constant capacitance.

Bimpelrekkie
  • 80,139
  • 2
  • 93
  • 183
  • Thank you for your answer. I think what you say is true, and that's pretty much what I've also been thinking. My problem is that the PIN-diode should be zero biased when not conducting, that's why the I-layer of PIN-diode is a must (so although the AC signal would forward bias a normal diode, PIN-diode will not be forward biased). I have a low frequency and at least few volts peak AC signal going through, which seems to limit the available PIN-diodes to those "few" which have thicker I-layer. But I'd also like to understand what PIN-diode characteristics the charge carrier life time affects. – RautSa Feb 22 '19 at 11:33