Crossover function in the standard.

Crossover function in the standard.

NewsGroups | Search | Tools
 comp.dcom.lans.ethernet  Post an article  get this group's latest topics as an RSS feed add this group's latest topics to your My MSN content add this group's latest topics to your My Yahoo content  add this group's latest topics to your Google content  YahooMyWeb Yahoo!  Google Google  Windows Live Favorites Windows Live  del.icio.us del.icio.us  digg digg  Add to Netscape Netscape
Subject Author Date
Crossover function in the standard. Trebor Mushroom 06-23-2008
Posted by Trebor Mushroom on June 23, 2008, 4:34 pm
If you were  Registered and logged in, you could reply and use other advanced thread options
Hi!

I'm looking for an authoritative answer to the following
question (preferably by referencing specific pages from
the IEEE standard):

Is the direct connection between two stations (i.e. use of the
so called crossover cable) explicitly supported by the standard?

In other words is it guaranteed to work if both adapters are
compatible with the standard?

Background:

We sell a sophisticated piece of equipment built around an IPC
board running quite old Linux kernel (the 10/100MBit Ethernet
adapter is a part of the chip set, for that reason I cannot
give you the exact make/model). Nevertheless it works flawlessly
with all kind of hubs/switches all over the world.
The manufacturer of the PC board claims that adapter is compatible
with the 802.3 standard).

One of our customers is complaining that direct connection
with some PCs does not work, specifically the "LINK" LED
doesn't come on. The customer believes that the network
adapter in our product is not compliant with the standard.

From experience I know all to well that a connection like
that sometimes just don't work. Not only with this particular
product but with *any* two Ethernet adapters. In some cases
fixing the speed and/or duplex settings does the trick, but
still some combinations of adapters won't work no mater what.

Is the Link Integrity Protocol mechanism guaranteed to
work when connecting two adapters directly using the
crossover cable?

I read good book about Ethernet some time ago, and I have
vague recollection that using short crossover cables
could cause problems (the ones I tried are >= 1m).

We test this extensively on loads of properly configured
PCs, using correct (and tested) cables. On average 3 out
of 10 doesn't work in this scenario.
Through the switch (or hub) it works like a charm.

TiA,
Trebor.

P.S. Please CC me on replies, I can't read the newsgroup
at work.

Posted by glen herrmannsfeldt on June 23, 2008, 9:56 pm
Trebor Mushroom wrote:

> I'm looking for an authoritative answer to the following
> question (preferably by referencing specific pages from
> the IEEE standard):

> Is the direct connection between two stations (i.e. use of the
> so called crossover cable) explicitly supported by the standard?

I don't know if the wording is in the standard, but as
I understand it, it is required to work. The only difference
between MDI and MDI-X ports is the crossover of the wires.

> In other words is it guaranteed to work if both adapters are
> compatible with the standard?

> Background:

> We sell a sophisticated piece of equipment built around an IPC
> board running quite old Linux kernel (the 10/100MBit Ethernet
> adapter is a part of the chip set, for that reason I cannot
> give you the exact make/model). Nevertheless it works flawlessly
> with all kind of hubs/switches all over the world.
> The manufacturer of the PC board claims that adapter is compatible
> with the 802.3 standard).

In the early, and not so early, days of 100baseTX I knew of
some 10baseT devices that would not connect properly
to 10/100 ports.

Also, I once had a cable modem that wouldn't connect properly
to a 100baseTX NIC at 100. Connecting through a 10baseT
repeater it worked fine. (At the time cable wasn't that
fast, anyway, though it may be now.)

> One of our customers is complaining that direct connection
> with some PCs does not work, specifically the "LINK" LED
> doesn't come on. The customer believes that the network
> adapter in our product is not compliant with the standard.

> From experience I know all to well that a connection like
> that sometimes just don't work. Not only with this particular
> product but with *any* two Ethernet adapters. In some cases
> fixing the speed and/or duplex settings does the trick, but
> still some combinations of adapters won't work no mater what.

It seems that there are devices that won't work together.
As far as I know, that is unrelated to crossover,
NIC-NIC, NIC-repeater, NIC-switch, repeater-repeater,
or any other combination.

> Is the Link Integrity Protocol mechanism guaranteed to
> work when connecting two adapters directly using the
> crossover cable?

I believe so, yes. Though there were some pre-standard
devices that didn't do link pulses at all.

> I read good book about Ethernet some time ago, and I have
> vague recollection that using short crossover cables
> could cause problems (the ones I tried are >= 1m).

As I understand it, there is no lower limit for UTP ethernet.

> We test this extensively on loads of properly configured
> PCs, using correct (and tested) cables. On average 3 out
> of 10 doesn't work in this scenario.
> Through the switch (or hub) it works like a charm.

-- glen


Posted by on June 26, 2008, 10:04 am
> Trebor Mushroom wrote:
> > I'm looking for an authoritative answer to the following
> > question (preferably by referencing specific pages from
> > the IEEE standard):
> > Is the direct connection between two stations (i.e. use of the
> > so called crossover cable) explicitly supported by the standard?
>
> I don't know if the wording is in the standard, but as
> I understand it, it is required to work. =A0The only difference
> between MDI and MDI-X ports is the crossover of the wires.
>
> > In other words is it guaranteed to work if both adapters are
> > compatible with the standard?
> > Background:
> > We sell a sophisticated piece of equipment built around an IPC
> > board running quite old Linux kernel (the 10/100MBit Ethernet
> > adapter is a part of the chip set, for that reason I cannot
> > give you the exact make/model). Nevertheless it works flawlessly
> > with all kind of hubs/switches all over the world.
> > The manufacturer of the PC board claims that adapter is compatible
> > with the 802.3 standard).
>
> In the early, and not so early, days of 100baseTX I knew of
> some 10baseT devices that would not connect properly
> to 10/100 ports.
>
> Also, I once had a cable modem that wouldn't connect properly
> to a 100baseTX NIC at 100. =A0Connecting through a 10baseT
> repeater it worked fine. =A0(At the time cable wasn't that
> fast, anyway, though it may be now.)
>
> > One of our customers is complaining that direct connection
> > with some PCs does not work, specifically the "LINK" LED
> > doesn't come on. The customer believes that the network
> > adapter in our product is not compliant with the standard.
> > =A0From experience I know all to well that a connection like
> > that sometimes just don't work. Not only with this particular
> > product but with *any* two Ethernet adapters. In some cases
> > fixing the speed and/or duplex settings does the trick, but
> > still some combinations of adapters won't work no mater what.
>
> It seems that there are devices that won't work together.
> As far as I know, that is unrelated to crossover,
> NIC-NIC, NIC-repeater, NIC-switch, repeater-repeater,
> or any other combination.
>
> > Is the Link Integrity Protocol mechanism guaranteed to
> > work when connecting two adapters directly using the
> > crossover cable?
>
> I believe so, yes. =A0Though there were some pre-standard
> devices that didn't do link pulses at all.
>
> > I read good book about Ethernet some time ago, and I have
> > vague recollection that using short crossover cables
> > could cause problems (the ones I tried are >=3D 1m).
>
> As I understand it, there is no lower limit for UTP ethernet.
>
> > We test this extensively on loads of properly configured
> > PCs, using correct (and tested) cables. On average 3 out
> > of 10 doesn't work in this scenario.
> > Through the switch (or hub) it works like a charm.
>
> -- glen

To the OP:

Your experience is different from mine:

> We test this extensively on loads of properly configured
> PCs, using correct (and tested) cables. On average 3 out
> of 10 doesn't work in this scenario.
> Through the switch (or hub) it works like a charm.

Well when I do this it ALWAYS works.

I have done it hundreds of times over more than 10 years and
it just works. PC to PC, Router to Router, Switch to Switch.

Of course sometimes a miss-configuration results in
terrible performance due to duplex missmatch.

I would worry first of all about your cables. Are the pairs
wired correctly?
How do you test them?


Posted by Trebor Mushroom on June 26, 2008, 4:53 pm
Bod43@hotmail.co.uk wrote:

> I would worry first of all about your cables. Are the pairs
> wired correctly?
> How do you test them?

Ready-made, certified cables tested using stock standard cable tester.

With regard to my original question, I dig deeper and have found
some problems with MB layout. Basically the network adapter
(Intel btw) is completely integrated in the chipset, with just four
lines on the outside (+ 3 lines to control LEDs). On the motherboard
there is only integrated RJ45 module jack (with transformer & LEDs).

The design violates the rules regarding the length and spacing
of the routes. I still have to find what exactly is wrong with
it but probably too long tracks increase loses to unacceptable
level and it doesn't work with some adapters.

Best regards,
Trebor.




Posted by glen herrmannsfeldt on June 27, 2008, 4:04 pm
Trebor Mushroom wrote:
(snip)

> With regard to my original question, I dig deeper and have found
> some problems with MB layout. Basically the network adapter
> (Intel btw) is completely integrated in the chipset, with just four
> lines on the outside (+ 3 lines to control LEDs). On the motherboard
> there is only integrated RJ45 module jack (with transformer & LEDs).

It shouldn't be that hard to get right before the transformer.
It is a little harder to keep the signals balanced after the
transformer through long PCB traces.

> The design violates the rules regarding the length and spacing
> of the routes. I still have to find what exactly is wrong with
> it but probably too long tracks increase loses to unacceptable
> level and it doesn't work with some adapters.

It has to last through 100m of cable, so a few inches of PCB
trace normally won't have that much loss. It gets somewhat
harder for gigabit, but you didn't say that. It is nice to
impedance match, though, in any case.

-- glen


Similar ThreadsPosted
Has anyone ever seen a crossover AUI cable? May 20, 2005, 3:39 am
Problem with Crossover Cable August 28, 2005, 1:18 pm
Crossover Vs. Patch Cables March 9, 2006, 11:15 am
LF: Crossover Ethernet Cable 25 feet August 25, 2006, 3:27 pm
standard for auto-negotiation May 24, 2007, 12:57 pm
Can a standard NIC be triggered to send an electrical signal? August 30, 2005, 8:28 am
More Than One Fiber Optic 100 Megabit Ethernet Standard? June 30, 2007, 12:28 am
Is there anybody here having IEEE's DSRC/WAVE 1609 standard documents? February 20, 2008, 8:55 am

other useful resources:
The Federal Communications Commission (FCC)
Telecommunications Industry Association
Electronic and Software Security Products and Services
International Telecommunication Union

Custom CGI Perl and PHP programming by 1-Script.com

Contact Us | Privacy Policy
The site map in XML format XML site map