Re: [SI-LIST] : Flight time measurememt for Mobile processor

About this list Date view Thread view Subject view Author view

From: Suchitha.V@smartm.com
Date: Wed Feb 14 2001 - 12:32:51 PST


Hello,
Thankyou for your reply.
I did measure the propagation delay from the Vmeas of the Driver(processor)
to the Vmeas of the receiver(440BX)
The delay i measured was negative because of the receiver appearing befor
the driver (for reasons explained)

I have calculated The Tflight_max and Tflight_min equations for the GTL+ bus
using timing equations.
I'm using these equations to calculate the Tsetup and Thold from the
measured flight-times in XTK.
This will help to determine if timings are met.

In XTK, pin to pin delay measured at Vmeas is the propagation delay, and not
the Tflight.
How do i confirm if the Tsetup and Thold timings are met for the GTL+
signals?

 looking forward to your replies and suggestions
Regards
Suchitha

---------------------- Forwarded by Suchitha V/Apex Data Inc/01 on
02/14/2001 11:46 AM ---------------------------
From: Tadashi ARAI@arap on 02/13/2001 03:10 PM
To: si-list@si-list@silab.eng.sun.com@SMTP@EXCHANGE
cc:

Subject: Re: [SI-LIST] : Flight time measurememt for Mobile processor

Hi Suchitha,

I met this issue and was confused too, but now it was a legend.

Receiver waveform sometime appears before the driver waveform has reached
to its end of swing. Note that you have observed not only driven
waveform but also reflections. It was reasonable result.

So that you should consider about substitution of reflections from
observed waveform, but it must be difficult.
You can measure the propagation delay instead of the flight time.

To measure propagation delay, you should know the driver delay, in XTK
words "TIME_TO_VM". You can measure it from simulation using a test
fixture, or it was automatically calculated in ibis2xtk.
Prop. delay was measured from when driver waveform reached to
measurement voltage (Vmeas) with test fixture, to the time receiver
waveform reached to Vmeas.

I hope this helps you.

Regards,
//// /// // / / / / / / / / / /
Tadashi Arai//Platform Developing Dept.,Desktop Prd Div. Fujitsu Limited
arap@pc.fujitsu.com TEL:+81-42-370-7624 Inagi-shi, Tokyo, Japan
/ / / / / / / / / / / // ///
On Sun, 11 Feb 2001 15:09:16 -0800
Mail from Suchitha.V@smartm.com described as below:

>
> Hello, All,
>
> I have a question related to the flight time measurement of GTL+ signals.
> on the FrontSideBus between the Mobile Pentium III processor and the
440BX
> chipset.
>
> Since the GTL signals are very short and have no termination, there is
> ringing
> seen at the Chipset end when the Processor is driving.
> I have simulated this in XTK tool, with the processor as driver and the
> chipset as the receiver.
> I have measured the flight time on the rising edge. I have taken the
> impedance to be 55ohms +/- 10% for simulations
> After the processor reaches 1.0 volts or so, at the rising edge, the
> waveform at the processor pin dips, and then rises.
> Therefore after around 1.0 volts, the chipset appears to be leading the
> processor which is wrong.
> Is this happening due to change in impedance? can anyone explain.
>
> In a case like this, how should the flight time be measured?
> For measurement, should i consider the falling edge including a cycle of
> ringing at the receiver end?
> Can anybody explain?
>
> looking forward to your replies and suggestions.
>
> Thanking you
> regards
> Suchitha
>
> **** To unsubscribe from si-list or si-list-digest: send e-mail to
> majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
> si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu
> ****
>
>

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Tue May 08 2001 - 14:30:50 PDT