[SI-LIST] Re: Calculating rise time from IBIS
- From: "Muranyi, Arpad" <Arpad_Muranyi@xxxxxxxxxx>
- To: SI-List <si-list@xxxxxxxxxxxxx>
- Date: Thu, 27 Nov 2014 17:33:35 +0000
From: Muranyi, Arpad
Sent: Thursday, November 27, 2014 11:26 AM
To: SI-List
Subject: RE: [SI-LIST] Re: Calculating rise time from IBIS
Ferhat,
Regarding “real systems have capacitive load(s)”, true, but…
Most faster systems use (resistive) on-die terminations these days.
But even if there are no resistive terminations, the PCB traces act
as a transmission line. Unless the capacitive loads are connected
directly to the driver, the driver will first “see” the T-line, not
the capacitance at the end of the T-line. If the buffer switches
faster than the signal travels a round trip on the T-line (which is
the case in most modern designs), during the time the signal travels
from the driver to the load capacitance and back, the driver sees the
impedance of the T-line. During this time the T-line looks like a
resistive load to the driver, and the driver has no clue that there
is a capacitor at the end of the T-line until the reflection arrives
bat at the driver. So the waveform you will see at the driver’s output
during this time will be the same as if you connected a resistor to
it with a value equal to the T-line impedance.
This is actually reason IBIS deals with resistive loads when it
comes to extracting the data for the waveform tables and the [Ramp]
keyword.
I hope this helps,
Arpad
======================================================================
From: Ferhat Yaldiz [mailto:ferhatyaldiz@xxxxxxxxx]
Sent: Thursday, November 27, 2014 2:57 AM
To: Muranyi, Arpad
Cc: SI-List
Subject: Re: [SI-LIST] Re: Calculating rise time from IBIS
Hithesh,
Actual rise time depends on loading. IBIS contains for resistive load, but real
systems
have capacitve load(s). You can use your simulated waveform for comparing
measured
waveforms. As initial you can use it for estimating bandwidth of signal, but
lastly you must use real
topology for real spectrum of signal.
Ferhat Yaldiz
On Thu, Nov 27, 2014 at 7:05 AM, Muranyi, Arpad <Arpad_Muranyi@xxxxxxxxxx>
wrote:
Aubrey,
You are correct, but note that he wants to scale to 10-90% of
3.3 V, not the swing of the buffer as the waveform data in the
IBIS file might have it. We have yet to find out why he wants
to do that...
Also note that in his calculation he "scaled" both the time and
voltage values. A buffer doesn't do that. The switching time
of a buffer stays pretty much the same, no matter how many
ohms the load resistance is, so for that reason he shouldn't
"scale" the time value of the [Ramp] (other than recalculating
from 20-80 to 10-90 which for me would give 253 ps, not 720 ps
that he got. (190/0.6*0.8=253)
Thanks,
Arpad
===============================================================
-----Original Message-----
From: Aubrey Sparkman [mailto:asparky@xxxxxxxxxxxxxxxxxxxxxxxx]
Sent: Wednesday, November 26, 2014 10:25 PM
To: tom@xxxxxxxxxxxxxxxxx
Cc: Muranyi, Arpad; SI-List
Subject: Re: [SI-LIST] Re: Calculating rise time from IBIS
Hitoshi,
If your model has waveform data, you can get the 10/90 time from that. Better
than scaling.
Aubrey
SparkRight Solutions
Sent from my iPhone
> On Nov 26, 2014, at 12:11 PM, Tom Dagostino <tom@xxxxxxxxxxxxxxxxx> wrote:
>
> Be very careful about scaling from a 20-80 % risetime to 10-90%. Scaling
> assumes the risetime is linear from the 10 to 90% points on the edge. It is
> very unlikely to be the case.
>
> Tom Dagostino
>
> Teraspeed Labs
> 9999 SW Wilshire Street
> Suite 102
> Portland, OR 97225
>
> tom@xxxxxxxxxxxxxxxxx
> www.teraspeedlabs.com
>
> 971-279-5325 office
> 503-430-1065 cell
>
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
> Behalf Of Muranyi, Arpad
> Sent: Wednesday, November 26, 2014 7:51 AM
> To: SI-List
> Subject: [SI-LIST] Re: Calculating rise time from IBIS
>
> From: Muranyi, Arpad
> Sent: Wednesday, November 26, 2014 9:47 AM
> To: SI-List
> Subject: RE: [SI-LIST] Re: Calculating rise time from IBIS Hitesh,
>
> What are you trying to do with the 10-90% of 3.3 V?
> The reason I am asking is because the “scaled” value might be wrong depending
> on what you are using it for…
>
> Thanks,
>
> Arpad
> =======================================================
>
> From: Hithesh [mailto:hitheshn@xxxxxxxxx]
> Sent: Wednesday, November 26, 2014 12:59 AM
> To: Muranyi, Arpad
> Cc: SI-List
> Subject: Re: [SI-LIST] Re: Calculating rise time from IBIS
>
> Thanks.
> So for 10-90% of 3.3V, I just scale the 700mV/190ps.
> This will be 2.65V/720ps.
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>
> List forum is accessible at:
> http://tech.groups.yahoo.com/group/si-list
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
>
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>
> List forum is accessible at:
> http://tech.groups.yahoo.com/group/si-list
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
>
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
