• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 16 subscribers
  • Views 263 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TS5V330: Recommendation for 3:1 mux for composite video?

Tom Schultz
Expert 3865 points
Part Number: TS5V330
Other Parts Discussed in Thread: TS3V330

Team,

I have a customer looking to switch between 3 different NTSC/composite video channels to a single output.  I see you have the TS5V330 series which is a 2:1 MUX for composite video, but what would you recommend to achieve a 3:1 mux?  Can any 3:1 analog mux be used?

Thanks,

Tom

  • 0
    TI__Mastermind 46327 points

    Hi Tom,

    So the TS5V330 has a spec that other switches we have don't. That is differential gain and phase which is pretty specific to video signals. That being said - other video switches that we have don't have this spec'd, so its not necessarily a impassable barrier.

    This device is still a standard switch and if another device has similar performance or better it should work fine. 

    With that in mind - we don't have many 3:1 options as typically most multiplexers are powers of two (so 1:1, 2:1, 4:1, 8:1......) so a 4:1 is probably the best bet. 

    That being said I have a few questions to see if we can narrow down a good part for your application:

    1. Do you know what the max current through the switch could be? The current video switch has a decently high current for a signal switch so if this can be lowered it opens up our options a bit. I think you should be able to get away with less than 128mA if I am understanding the NTSC spec correctly, but I want to double check with you first. 

    2.  How is the termination set up for the signal - this will influence the resistivity of the switch and what bounds that can be in. 

    Please let me know and I will see what parts will be the best fit for your application.

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    TI__Expert 3865 points

    Parker,

    Here is the reply from customer.

    Thanks for the reply.  I believe that composite video is 1Vp-p into 75 Ohm.  If you can do the straight math, that's 13.3mA DC current.

    So...

    1. The DC current is low, but I am guessing the high current capability is to allow a high slew rate into high capacitance?  I am really not sure what the AC current is, that might have to be measured.

    2. Everything that I see is straight 75 Ohm DC termination to ground.  Not sure what the input capacitance is specified at.

    That said, I think that I would be fine starting with the Video switches as the cost is low enough for me to prototype with.  I'm also happy to also socket another analog 3:1 or 4:1 mux if you have a recommendation.  It might take some testing.

    As far as the video muxes, I'm looking at the TS3V330 since it's available in QFN, any reason you would recommend staying with the 5V part?

    I look forward to your reply.

    -Tom

  • 0 in reply to Tom Schultz
    TI__Mastermind 46327 points

    Hi Tom,

    Thanks for the information!

    So first off - this switch cannot pass negative signals - so if the 1Vp-p is centered at 0V there could be issues passing the signal without issue either with quality of signal or damage to the switch as it has a ESD diode from ground to the I/O that will conduct at inputs < 0V. So the signal when it comes to the input of the mux its voltage w.r.t. ground needs to be positive or 0V for it to pass through okay.

    I thought the current would have been pretty low - but a lot of our switches can only handle 30mA - that's why I wanted to check. With a 75 Ohm load lower on resistances are definitely necessary. 

    With that being said from what I could find capacitance should be pretty low - it depends on the length of the signal pathway and the material of the trace/cable. However there could be possibly in-rush current that is pretty high  - but the peak should be pretty short with lower capacitance. However I think keeping the 128mA is probably best to avoid issues.

    As for the 3.3V versus the 5V part - not really a huge difference ; they have similar specs so its whatever voltage supply is most beneficial for you to use in the application should guide selection. 

    So with all those considerations I think these line of parts is our strongest option  to move forward with, the have low on resistance, sufficient bandwidth, as well as some headroom for inrush currents. A 4:1 can be made using 2:1 devices (it would be 3 2:1 devices and the on resistance would double to 6 Ohms typically and 14 Ohms max due to 2 switch pathways). This is still low pretty low resistance as its typically still under 10 Ohms. The biggest concern is that cascading multiplexers can degrade bandwidth of the signal pathway - so if the signal frequency is already near the top the bandwidth limitations of the part you may see significantly more attenuation than with only one device. To prevent issues from cascading a high impedance buffer stage would also be required to keep signal integrity high. There is a "C" version of the 5V part that also includes powered off protection + undershoot protection as well. This is https://www.ti.com/product/TS5V330C.  

    Another option however is still a 2:1 device - but it does have a few more channels than this device/ :

    TS3V712E 

    or

    TS3V712EL 

    or

    TS3V713EL (currently out of stock - but is just a pinout variation of the 712EL) 

    They all have pretty similar performance, are offered in WQFNs. 

    I think those are our best options for these applications - our video portfolio is comprised of various 2:1 switches and they are unique enough spec wise that the parts I listed are the best suited for the application.

    If you have any other question please let me know.

    Best,

    Parker Dodson