DS90UB925Q/926Q with High-Speed Differential Multiplexer / Demultiplexer

Luca Tosetti

Other Parts Discussed in Thread: HD3SS3412, TS3L301

Hi,

I've a question.

In my application I'm using ser DS90UB925Q and des DS90UB926Q, to drive a 640x480@25MHz display true color(24bit) with touchscreen, speed of the link is about 875Mbps. Now I need to switch signals from ser DS90UB925Q to two different des 926Q in two different position, but only one of the two is active.

I just read this topic: http://e2e.ti.com/support/interface/high_speed_interface/f/138/p/209684/745774.aspx#745774

and I understand than FPD-Link III chipsets only support point-to-point applications., no "active" switch or repeater.

But, is possibile with a High-Speed Differential Multiplexer / Demultiplexer like TS3L301 ("with no buffer or repeater inside") or similar to split the signals from SER without interrupt the bidirectional signaling of the FPD-Link III interface?

Thanks

Best regards

Luca

over 9 years ago

0 Jason Blackman over 9 years ago

TI__Genius 14015 points

Hi Luca,

Yes, using a passive mux with sufficient bandwidth should work. One part that we've tested in this sort of application with 925/926 is the HD3SS3412.

http://www.ti.com/product/HD3SS3412

Thanks,
Jason

0 Mark Sauerwald1 over 9 years ago

Prodigy 30 points

Luca

It may work, but I would not recommend it. The link between the serializer and the deserializer has signals operating for the control channel in the ~1-5MHz band, as well as the forward channel which is going to be at several 100s of MHz in your application. Reflections or attenuations in either of these bands will compromise the link. The TS3L301 has about 4Ohm on resistance, so there will be an impedance discontinuity at the switch which is liable to create issues, as well as the bandwidth limitations of the switch which are liable to adversely impact the channel. The FPD III devices have been characterized to operate with a continuous link between the serializer and deserializer, and operating them with something else may not allow for the expected performance of these parts.

Mark Sauerwald

0 Luca Tosetti over 9 years ago in reply to Jason Blackman

Prodigy 90 points

Thanks to both for the quickly and clear answer

I understand that this operation may be critical for the link between the serializer and deserializer for the reason listed by Mark, but maybe if we use the right component with sufficient bandwidth(no TS3L301 at this point) is possible to have a reliable and working solution

In this regard I’ve two short question to Jason:

-Is possible have more information or details about the test that you have made on application with 925/926 and HD3SS3412? Maybe there an Application Note or another technical document.

- Is there a demo board or tool for HD3SS3412 or similar component to do a quickly test?

Thank you

Best Regards

0 Luca Tosetti over 9 years ago in reply to Jason Blackman

Prodigy 90 points

Forgive me if I insist,

But I need more information about HD3SS3412 or similar component

-Is possible have more information or details about the test that you have made on application with 925/926 and HD3SS3412? Maybe there an Application Note or another technical document.

- Is there a demo board or tool for HD3SS3412 or similar component to do a quickly test?

Thank you

Best Regards

0 Jason Blackman over 9 years ago in reply to Luca Tosetti

TI__Genius 14015 points

Hi Luca,

Unfortunately we don't have any test document which can be shared publicly. I also don't know of an evaluation board available for any of those switches. I am not in that group though, so I am not sure. To find out for sure, you can make a new post on the Consumer and Computing Interface forum here:

http://e2e.ti.com/support/interface/digital_interface/f/130.aspx

Thanks,
Jason

0 Luca Tosetti over 9 years ago in reply to Jason Blackman

Prodigy 90 points

Thank you for you answer.

I try to make a new post on on the Consumer and Computing Interface forum.

Regards

Luca