The worst signal which DS110DF111 re-timer can handle?

hung dang

Other Parts Discussed in Thread: DS110DF111

Hi all,

I need to verify the application which is used DS110DF111 re-timer. The detail is below:

The connection between CPU's XFI (TX) and DS110DF111 re-timer (RX)

CPU's XFI (TX) ==>Transmission line on PCB==> SlimSAS connector ==>SlimSAS Cable (350mm)==> SlimSAS connector ==> Transmission line on PCB ==> DS110DF111 (RX).

To verify that re-timer can handle above design, I had done below:

I export the transmission line to free-form on Hyperlynx, connect to S-parameter model of SlimSAS connector and cable to extract S-parameter of the whole channel which is described above. And then, I read the insertion loss and return loss to determine that re-timer can handle the signal on the whole channel.

Is this method correct ? If not, please help to show the other method? IBIS-AMI analysis can help in this design ?

Thanks all.

over 8 years ago

0 Loren Siebert 1 over 8 years ago

TI__Genius 13935 points

Hello Hung,

We do not support that part in High Speed Amplifiers. I will move this thread to the High Speed Interface forum.

Regards,

Loren

0 Lee Sledjeski over 8 years ago

TI__Mastermind 43795 points

Hi,

Using IBIS-AMI and s-parameter modeling is a good start. The DS110DF111 can handle up to ~ 30 dB of loss at 5 GHz. To achieve 0 BER with a lossy channel will require that the crosstalk and other noise sources are under control.

For any IBIS-AMI modeling make sure the DS110DF111 link is AC coupled.

Regards,

Lee