[PCB_FORUM] Re: Length matching for DDR

Can you send me the link url.

Thanks 

-----Original Message-----
From: icu-pcb-forum-bounce@xxxxxxxxxxxxx
[mailto:icu-pcb-forum-bounce@xxxxxxxxxxxxx] On Behalf Of Alan Williams
Sent: Thursday, February 28, 2008 8:35 PM
To: icu-pcb-forum@xxxxxxxxxxxxx
Subject: [PCB_FORUM] Re: Length matching for DDR

Hi,

Unfortunately I deleted all the previous messages from this thread when
I came in this morning (there were 27 or so!!)....

....so I don't really know what you were discussing...

However if I get a gist and people are looking for guidelines for DDR
length matching (not DDR2, or QDR..) Then I just did a board and spend
lots of time and effort trying to get it right. I constrainted my design
to the guidelines put out by micron in TN-46-14.. You should find it on
their website, if you don't let me know I'll hunt out the url for you. 

I must mention the board is being manufactured as we speak, so for all I
know it won't work, but I'm confident it will. 

If I'm completely off topic just ignore me!

Cheers,

Al. 

-----Original Message-----
From: icu-pcb-forum-bounce@xxxxxxxxxxxxx
[mailto:icu-pcb-forum-bounce@xxxxxxxxxxxxx] On Behalf Of Austin Franklin
Sent: 28 February 2008 14:53
To: icu-pcb-forum@xxxxxxxxxxxxx
Subject: [PCB_FORUM] Length matching for DDR - was - RE: Re: Allegro
V6.01 Constraint Manager

Hi Richard,

> > I'd like to send datasheets, but I can't.  See  
> > http://focus.ti.com/lit/an/spraac5f/spraac5f.pdf , page 19.  Not 
> > exactly our designs, but the same idea.
> > It is to force reflections to cancel.

> I looked at that page/diagram.  It doesn't say a thing about making 
> the
length
> matching from the T to the destination tighter than from the source to

> the destination, at least that I saw. They only provide source to 
> destination length match numbers.  Did I perhaps miss something?

Actually, I did miss something.  Their diagram confused me, they used
rectangles to indicate the segment lengths (yet they say "length A to
C", when they really mean A + C), instead of lines.  But, none the less,
I believe I understand it now, and they specify:

"Length B should match length C within 100 mils."

Which is apparently the T to destination...so they do specify
this...but...

"Length A to C and A to B should match within 100 mils within ADDR_CTRL
net class."

The A segment is the same for both A to C and A to B, which means it is
only the B and C segments that length match to the 100 mils...so stating
that B should length match C to within 100 mils is redundant.

The proof:

A to C = A + C, A to B = A + B, so they want the following:

100 = ABS((A + C) - (A + B))
100 = ABS(C - B)

So:

ABS((A + C) - (A + B)) = ABS(C - B)

for simplicity let's remove ABS for now...and since no order of
precedents exist in addition/subtraction:

Let's substitute - (A + B) = - A - B in the equation:

A + C - A - B = C - B

A - A makes A drop out so we are left with:

C - B = C - B

...and means the two constraints are equal, which means they aren't
constraining the T to destination tighter than the overall length match.
Unless I made a mistake somewhere...but I think I got that right.

Best Regards,

Austin

-----------------------------------------------------------
To subscribe/unsubscribe: 
Send a message to icu-pcb-forum-request@xxxxxxxxxxxxx
with a subject of subscribe or unsubscribe

To view the archives of this list go to
http://www.freelists.org/archives/icu-pcb-forum/

Problems or Questions:
Send an email to icu-pcb-forum-admins@xxxxxxxxxxxxx
-----------------------------------------------------------

__________ NOD32 2908 (20080228) Information __________

This message was checked by NOD32 antivirus system.
http://www.eset.com


-----------------------------------------------------------
To subscribe/unsubscribe: 
Send a message to icu-pcb-forum-request@xxxxxxxxxxxxx
with a subject of subscribe or unsubscribe

To view the archives of this list go to
http://www.freelists.org/archives/icu-pcb-forum/

Problems or Questions:
Send an email to icu-pcb-forum-admins@xxxxxxxxxxxxx
-----------------------------------------------------------
-----------------------------------------------------------
To subscribe/unsubscribe: 
Send a message to icu-pcb-forum-request@xxxxxxxxxxxxx
with a subject of subscribe or unsubscribe

To view the archives of this list go to 
http://www.freelists.org/archives/icu-pcb-forum/

Problems or Questions:
Send an email to icu-pcb-forum-admins@xxxxxxxxxxxxx
-----------------------------------------------------------

Other related posts: