TPG2102 Phase shifter

RF Solutions RF All Other (switches, controls, attenuators, ..)
1

I have a couple of questions related to the operation of the TPG 2102 phase shifter:

  1. Is there an upper bound to switching (on/off) the phase elements.
  2. Can I switch at a rate above a 1 MHz.
  3. Is the switching speed and the RF through performance independent of each other
2
  1. Do you mean an upper bound in terms of switching frequency? If so then yes, this will be determined by the phase shifter switching time which should be <50ns.
  2. With <50ns switching time then a little above 1MHz should be ok. Depends what you are trying to achieve in the application.
  3. Depends what you mean by this. The RF performance will obviously not be met when he phase shifter is changing state and in transition. So when you are changing states very quickly there will be a higher % of time when the phase shifter is in transition between states.
3

HI Chris,

Thanks for your prompt reply.
-I am using the phase shifter as basically a BPSK modulator and would like to employ a modulation frequency of approx. 10MHz. It appears from your answer that the phase shifter may be limited to 1MHz.
-Would you expect any difference in performance if we were to use the other 4 values, (totaling 165 deg) instead of the 180 degrees.
-Does the employed state impact the phase shifters tolerable transition time.

With regards to the RF performance. The phase shifter is introducing a change in phase: I understand during the phase transition the amplitude of the RF signal is impacted either side of this the amplitude should recover?

4

OK, thanks. These phase shifters are normally used in radar applications, I’ve never seem them used for a BPSK modulator before. The switching time I quoted would be for 50% control to phase settling, so in your application you could assume faster as you’re more worried about the time taken to change 180Degs. With this in mind 10MHz may be possible. The switching time does vary dependent on states, and combinations of states changed. If several bits are changed it will be dominated by the slowest. I don’t have this fully characterized for the TGP2102. The <50nsecs quoted was taking into account the worst case. It’s also worth taking into account phase error of the TGP2102, see plots in datasheet. Amplitude will be affected very slightly during the phase transition, you can also get a small amplitude difference (or error) between different phase states.

5

Hi Chris,

Thanks for your response. I understand the TPG2102 employs air bridges in its construction so the parts have to handled with care. Having said that we just received parts in a waffle package is that acceptable?

6

Yes the TGP2102 is shipped in waffle packs. This app note on die handling and assembly may be useful.