SAMF2000 said:
I never thought Replacing my Ribbon cables would cause my entire
system to stop booting. Seemed like an easy task...heck I just replaced
my CPU With no headaches ... Well After Replacing the Ribbon cables
with IDE Rounded cables And getting some "No 80 wire connecter "
error, I Decided to replace the old ribbon cables. Well to my
amazement same error. I then stated to try the jumpers on the MB. Then
I went into the Configuration and set the defaults... still same error.
I had to completely format the drive, reinstall windows and everything
to get my computer back. So my question: How can just replacing a cable
cause this much trouble and what could I have done short of
reinstalling and starting over? Any help would be greatly appreciated.
Sam
The problem with rounded cables is called "crosstalk".
If you look at the cable design, the 80 wire cable looks like
this.
. x . x . x . x .
The X character is a grounded wire, while the "." wires carry a
signal. The ground wire provides a reference for the signal,
and makes the cable look closer to a controlled impedance. It
improves the signal quality. The extra spacing between the
"." wires reduces interference from one "." wire on the other
"." wires on either side.
Now, roll up the cable, so that one layer of wire is
adjacent to the other. A small portion of the cross section
would look like this. Depending on how tightly the cable
is rolled, the displacement between layers, could take
on just about any relationship.
. x . x . x . x .
. x . x . x . x .
Now, some of the "." wires are getting closer to one
another, than they would be if the cable was kept flat.
The close signals interfere with one another, and sometimes
a "1" signal gets switched to a "0" signal. If the disk is
doing write operations, then corrupted data can get written
to the disk. (Note - you can reduce the interference, by
dropping down to ATA-33 or lower. Not that anyone would want
to do that, when perfectly good flat cables are available.)
A round cable must be fabricated, so that the interference
is no worse than it would be in a perfectly flat cable. I've
never heard of a standard for round cables, so there is nothing
to spell out good practice for the companies who make them.
(Although, if they have any engineers working for them,
they should know about this problem.)
You can see a product here, where they chop the cable into
sections, then lay the sections on top of one another. The
X's are offset, so each layer is shifted a precise amount.
The X's form a "box" around the dots. If I had to make a
cable, this is the way I would do it (from a theory
perspective). Simulations and testing should still be
carried out, to verify that the cable works. For example,
the impedance of the signals is changed by this structure,
which may or may not be important, depending on the Southbridge
IDE driver output impedance. At least I can see some "science"
in this product.
http://www.hardwarecooling.com/product_info.php/products_id/289
Not every claim in that advertisement, is for real. But the
cables could be a bit better than a randomly rolled up alternative.
None of my computers have rounded cables - for a reason
A better alternative, is to use SATA disks and SATA cables.
A lot less air resistance with SATA cables.
Paul
