Hard drive cable question -

[bobgoodwin]

Gene Heskett wrote:

>On Wednesday 22 February 2006 18:48, bobgoodwin wrote:
>  
>
>>Does it matter which forty pin connector plugs into the master and
>>slave drives?
>>
>>I have come to the realization after struggling with some problems
>>that there are a lot more wires in the ribbon cable than there are
>>connector pins.
>>
>>    
>>
>Then you have an ata100 or faster cable.  And yes, in many cases, the 
>drives position on the cable is important.  As in scsi systems, the 
>last drive on the cable must assume that its job is to terminate the 
>cable and try, by the termination enabled when is either jumpered as 
>Master, or is in CS mode and the last drive, to absorb the echos from 
>the signal transitions.  By playing mix and match, possibly with the 
>drive thats functioning in master mode, on the middle connector, and a 
>drive set as slave on the end connector, you are just asking for data 
>integrity problems.  The signal edges under those conditions will ring 
>like a bell causeing data errors galore.
>
>Also, if the cable is a cable select cable, as evidenced by having a 
>black connector on what should be the motherboard and, and 2 other 
>colors are used for the other two connectors, then the drives should be 
>jumpered for CS, and the master drive must be on the far end of the 
>cable.
>
>Also as a reference, these 80 pin cables should NEVER be turned end for 
>end, as its only the black connector that is suppposed to be plugged 
>into the motherboard that has the proper grounding connected to the 
>alternate wires in that 80 conductor cable, thereby serving as 
>interconductor shielding by haveing a grounded wire between each active 
>wire, thereby reducing crosstalk by a considerable amount.
>
>  
>
>>I now see that Maxtor designates the connector at the end of the cable
>>as the "master" and the one just below it as the "slave."  I presently
>>have only one drive jumpered as master with FC4 running on it and it
>>doesn't seem to care which data cable connector is attached to it,
>>but when I put in the second drive, jumpered as "slave" the computer
>>won't boot.  I decided the drive was bad but now I'm wondering? It
>>may be the wrong connectors plugged into the drives.
>>
>>I thought "cable select" cables had wires obviously crossed in the
>>ribbon cable but that may not be true with this 80 wire ribbon?
>>    
>>
>
>Untrue, that is used only rarely even in the scsi world.
>
>  
>
>>Can someone clear this up for me.
>>
>>Thanks.
>>
>>Bob Goodwin   Zuni, Virginia
>>    
>>
Ok, I'm glad I asked that question.  It clears up some gaps in my 
knowledge,  I hadn't considered the effect of unterminated lines on fast 
pulses but that would be a reason for always connecting both ends of the 
cable, leaving one end free, using the center connection only would 
leave a bunch of unterminated stubs which might cause some ringing.

I did not realize that alternate wires were grounded either ...

Anyway thanks to all.

Bob Goodwin