[SI-LIST] Re: The question for HDMI compliance spec & Debuging method
- From: "T.K. Jeon" <tkjeon@xxxxxxxxxxx>
- To: 김창희 <ch80.kim@xxxxxxxxxxx>, "si-list@xxxxxxxxxxxxx" <si-list@xxxxxxxxxxxxx>
- Date: Fri, 16 Dec 2011 18:25:22 -0800
Hello Chang,
1) Rise time
When the link needs to create the noise(or jitter) budget, you should consider
all the loss mechanisms over your channel as well as the drive and the
receiver. The minimum limit of the rise/fall time is related to the noise
budget. The rise time plays as a big contributor for the following noise
sources.
. Crosstalk
. SSN
. Reflection
Let's say you have a compliance channel that has 10dB loss at your Nyquist
frequency, for example, then the compliance spec could just define the Tx
amplitude and the Rx amplitude such that 20*log10(Vrx/Vtx)=10dB. However, in
reality, that is not the case. The loss budget in the link is usually bigger
than just the channel loss because of various noise sources as shown above. For
example, a -40dB of aggressor crosstalk source contributes 1~2dB loss of signal
in a victim channel depending on your channel conditions, so your loss budget
should be at least 12dB(10dB+2dB) in the case of using 10dB loss channel as
mentioned above. As a result, the compliance spec could call for higher Tx
amplitude as one of the way of meeting the Rx sensitivity spec.
As explained, you have to put those noises under control for your loss budget
and, hence, the rise time should also be controlled. This is one of the reasons
why most of high-speed standards define Tr/Tf spec as the greater than a
certain number.
2) Eye mask test.
There are several items you could try.
. Look for your parasitic capacitance on your transmission line due to your
board/package design that may cause too much loss of high frequency components.
. Double check your termination resistance at your Tx.
. You said you had tried PRE-empahsis, but you don't like it because it boosts
a transition bit. Then, you may want to try DE-emphasis with increased output
swing, which utilizes the non-transition bits.
. Without seeing your waveform, I cannot say exactly. But, your periodic
jitter(or random jitter) could cause your test to be failed. If your scope has
a CDR function, you can measure the eye with CDR setup that represents your Rx
jitter transfer function, so your unwanted low frequency jitter components can
be filtered out.
. Better yet, many scopes have the equalizer function to mimic the Rx
equalization, you may want to use it if you have one.
Regards,
TK
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
Behalf Of ???
Sent: Thursday, December 15, 2011 11:21 PM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] The question for HDMI compliance spec & Debuging method
Dear everyone,
This is my first trial to ask question on SI-LIST. Wish everyone is
welcomimgwith my question.
1) when I see the HDMI compliance test report(H/W), the Rise/Fall time spec
is >=75.000ps .
why the spec is >=75ps, not <= (max time limit) ?
I think the rise/fall time should be shorter for better SI performance.
Is there any specific reason for restricting rise/fall time to be larger
than75ps?
2) Now my system has a problem with EYE test on 1080 60p with 148.5Mhz.
The envelope of Rising signal is touching the EYE MASK since the envelope is
too slow. (rise time is 290ps)
Would you please share some methods to make shorter rise time on the
system?
( I tried the pre-Amphasis option for boosting signal, but it causes
criticalovershooting which i can't accept)
* the trace lane impedance is about 100ohm in this system.
Best Regards,
Chang
[IMG][IMG]
------------------------------------------------------------------
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 technical documents are available at:
http://www.si-list.net
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 technical documents are available at:
http://www.si-list.net
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:
