Hello,
Over/undershoot specs are to be critical and need to be considered.
I have seen fails (not just reliability) due to this topic. In this case
it was not the area, but the peak limit, which was the reason, why some
electrons have been pushed into some bandgap reference circuit on the
DRAM, causing internal voltages to shift around...
Just to confirm the original statement: the absolute over/undershoot
voltage level is OK, but the Area is a problem ?
This is anyhow a specific topic (especially for measurements).
I know that some Scope vendors measure just a triangle, some the real
area between trace and limit. I think the latter one is the intention of
the spec, but drawn is only the triangle in the spec.
Some measure this over a long period of time, some just for a single UI.
There is no real length of the overshoot defined, so the first one
should be correct (according to the spec).
But this leads to the fact, that if you have some period where your
Signal is not driven it might drift to a low level slightly below VSS
and just adding up enough area to fail the spec.
This is also not intended by the spec.
I have seen in a DDR4 micron spec, that they defined the length of the
undershoot, but also not exact enough to really allow an implementation
in a measurement (or simulation).
In your case I guess you have an unterminated system. With termination
you will not run into this issue.
So I really suggest to address this issue (reduced drive Strength,
serial termination, parallel termination ...)
Hermann
EKH - EyeKnowHow
Signal Quality - Made in Bavaria
Hermann Ruckerbauer
www.EyeKnowHow.de
Hermann.Ruckerbauer@xxxxxxxxxxxxx
Itzlinger Strasse 21a
94469 Deggendorf
Tel.: +49 (0)991 / 29 69 29 05
Mobile: +49 (0)176 / 787 787 77
Fax: +49 (0)3212 / 121 9008
Am 20.06.2016 um 06:39 schrieb Bhagwath, Nitin:
Hi Fisayo,
Even if the signal never exceeds the absolute maximum rating (i.e. your
channel over/undershoot spec therefore passes), the signal might be violating
the overshoot/undershoot area by remaining beyond Vdd or Ground for too long
(i.e. the over/undershoot area spec therefore fails).
A quick way of checking this would be to approximate the region where the
signal goes above Vdd or below GND as the closest polynomial and getting a
rough estimate of the area.
By violating this area spec, you might still be able to bring up your system,
but the long term reliability of the part would be in question
Your Mentor AE should be able to help you with any questions about the
results from the DDRx Wizard.
Regards,
-Nitin
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On ;
Behalf Of Fisayo Adepetun
Sent: Sunday, June 19, 2016 2:45 PM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Overshoot and Undershoot area spec for DDR3L
Hello all,
The Jedec spec explicitly calls out overshoot and undershoot area spec. I
have observed that in hyper lynx, a ddr3L channel seems to fail this spec
even though the channel passes the overshoot and undershoot spec with
margins. Is the this OS/US area spec informational?
- Fisayo
------------------------------------------------------------------
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
