[SI-LIST] Re: Logic level at unused DQS

  • From: Hermann Ruckerbauer <Hermann.Ruckerbauer@xxxxxxxxxxxxx>
  • To: Balamanikandan K <kbmanick@xxxxxxxxx>
  • Date: Sun, 20 Oct 2013 14:39:54 +0200

Hello,
I guess it will not affect your lower channel Signals, but Cross Curent
in your Receivers can add some additional power consumption.
It should not be too much per Receiver, but I have no detailed numbers
how much this can be.

Hermann

EKH - EyeKnowHow
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

schrieb Balamanikandan K:
> Hi
>
> Thanks. But I could not understand "Cross Current". Also will it
> effect the lower byte signals?
>
>
> On Thu, Oct 17, 2013 at 11:03 PM, Hermann Ruckerbauer
> <Hermann.Ruckerbauer@xxxxxxxxxxxxx
> <mailto:Hermann.Ruckerbauer@xxxxxxxxxxxxx>> wrote:
>
>     Hello,
>
>     if you don't have anydefined high/low level you might end up with some
>     midlevel while receiver are switched on. This could cause some
>     Crosscurrent through the receiver that you might not like.
>     This happens only during Writes to this device, so it is not
>     normal ODT
>     that needs to be considered, but dynamic ODT. Are you have control if
>     this is enabled and which value is selected, or is this a flexible
>     feature in your system ?
>
>     Just as some theoretical thinkig: It depends on the logic
>     implementation
>     of the devices what can happen. Usually the logic of the DRAM will
>     count
>     DQS edges to get a "Write complete".  The DRAM does not know that it
>     will get a non valid DQS (e. g. no edge at all or but many edges
>     due to
>     midlevel). So if the logic in the DRAM will wait for completion of
>     both
>     write bursts, or get one completion too early, what will happens ?
>     I don't think this is defined in the JEDEC spec, so you might get a
>     different behavior for different devices.
>     ... I know such a configuration was used in the past for x16 ECC
>     implementations, but I have to admit I never thought about this ...
>
>     Hermann
>
>
>
>     Am 17.10.2013 08:47, schrieb Balamanikandan K:
>     > Hi
>     > In a x16 DDR3 device, DQ bus is comprised of two bytes. But I am
>     using only
>     > one byte...let us say lower byte and unused DQ signals are left
>     unconnected.
>     >
>     > In an ODT enabled condition, Why should I  connect the unused DQS to
>     > ground through some resistor and  unused DQS# to  VDD through some
>     > resistor? Why should we bother about the logic level at unused DQS?
>     >
>     > My understanding is whatever it may be the logic level at unused
>     DQS, it
>     > will not affect lower byte signals.
>     >
>
>
>     --
>
>     EKH - EyeKnowHow
>     Hermann Ruckerbauer
>     www.EyeKnowHow.de <http://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
>
>
>
>
> -- 
> Best Regards,
> Balamanikandan.K



------------------------------------------------------------------
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: