[Ti] FireWire Port Problem

[b]

Tarik paused, thought it over, and spoke thusly:

>On Tuesday, September 30, 2003, at 05:41  pm, b wrote:
>
>>External devices 'power-up' on the FireWire bus, but aren't 
>>detected, so, who knows? Thanks for your input.
>
>Yes I believe these are the same symptoms.
>
>I have seen 3 other Ti's fail in the same way, all 400/500 models. 
>There is nothing at fault with your power supply, I think has to do 
>with some kind of "arcing" phenomenon within the IEEE 1394 
>implementation Apple used back then...but that's my hazard at a 
>guess.
>
>--
>Tarik

After some research, I find it is, indeed, arcing. Power in the 
cable, or device, that gets accidentally shunted over to the data 
line (s) in the cable or device. This fries the PHY box that is on 
the host firewire port. (It is responsible for sending power out of 
the 'extra' wires in the ports that also supply power in addition to 
data.

Usually a cable (that either twists, or is mounted backwards (hard 
but possible, apparently), is the culprit. The advice, in terms of 
prevention, is to use high quality cables, change them often 
(monthly?), be careful on plugging-in, don't let the cable 'twist' 
anywhere near the connector ends, and if possible, use 3-way ( 
External Power/Off/Bus Power) devices.

A short, high-quality cable, that might have developed an arc due to 
subtle 'twists', over time, probably fried the port on my Powerbook. 
If Apple had implemented the anti-Static/anti-Power over/under 
circuit on the port, earlier, it all would have been avoided.

As for me, given current dire straits, I may just wait ( quite a 
while ) for an UltraWide SCSI PC card, and go that route (skipping 
FireWire/USB and the slow drives, altogether). I hope such a thing 
exists. In the meantime, I am all the way out of business.

~flipper