This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

RF/ Video Switch ON Resistance vs current

Other Parts Discussed in Thread: SN74CBT3253, TS3V340

Trying to find an economical video/RF switch to select RF bandpass filters from 1-50 MHz with ~ 50 Ohm impedance. I've noticed that the on resistance seems to go up with lower channel currents. Are there any suggested video or analog switches with low on resistance for low current signal levels? What is the mechanism that makes the CMOS switches increase on resistance with lower channel currents?

Thanks, Bill

  • have you had a look at the Peregrine parts? i have not found any TI devices which operate well in this frequency band. 

  • Thanks Josh,

    I have looked at Peregrine and Skyworks... But they are not cost effective for this project as I would like to be under $0.10 per channel. So I was looking into bus switches and video muxes for a multi-channel alternative. Chips like the SN74CBT3253 and the TS3V340. I can have a DC bias on each channel and the AC current will be 1~10mA (0.05~5mW). I can also absorb the channel capacitance in the tuned filters but the on resistance needs to be ~4 ohms or less...

    -Bill

  • the Peregrine switches can get that low in cost, in volume - but thanks for pointing out the TS3V340. that thing looks pretty interesting for what i do, too. 

    other device i have also used that is in TI portfolio is the CD4066, and specifically the military version for best performance, but you still have to watch the input level and pay close attention to routing to avoid crosstalk.

  • Oh that's right, the 4066 has been around quite a while too and I believe it is a complementary MOS switch. I did find some good white papers on the R-on changing with signal voltage. NMOS R-on goes up with signal voltage and PMOS does just the opposite.

    To reduce distortion due to R-on changing in a NMOS device, we can run VDD as high as possible, AC couple the RF and use as low a DC bias as possible. So on a 5V device, run a DC bias of about 1.5V to remain in the flattest R-on area... Provided the RF signal is < 3V pk-pk...

    I guess it is all about weighing the compromises!

    Regards, Bill