TMUX1511: What is the power off isolation over frequency?

Hi Bryan,

We don't have that data so I won't be able to provide it. Why do you need isolation up to 1GHz if your signal is 100Mbps?  Are all 4 signals data at 25MHz each to make 100MBps or is it 100MHz per?

3GHz is what the switch can support when it's on. When it's off there were always be worse isolation at higher frequencies. Typically off channels aren't also seeing frequencies as high as what the ON channels can support. 


-Rami

Thanks for your reply Rami.  My signal is 100Mbps NRZ, so it has frequency content well above 100Mhz due to the square edges.  Maybe I don't need isolation up to 1Ghz, but probably up to 500Mhz. 

Regarding the power off isolation, I understand that you don't have a plot to share.  Do you have a circuit model that represents the power off impedance?

Hi Bryan,

I see, for the SI you want some headroom there. 
So you could do a discrete model using VISO/IOFF as your resistor and the CON values. But for I-OFF it isn't a consistent value so it may be tricky to do properly. But with a constant voltage and frequency it could be done.

You can see an example here:
https://e2e.ti.com/support/switches-multiplexers-group/switches-multiplexers/f/switches-multiplexers-forum/1064288/faq-what-if-my-analog-switch-multiplexer-is-missing-a-pspice-model

After talking to my design team, the other issue would be that the off-iso would be worse in powered-off than powered-up. So if the current Off Iso values aren't acceptable then expect to get worse in the powered-down mode. We unfortunately don't have that tested though to provide an ballpark number of what to expect. 

-Rami