Xilinx USB flatform cable length mistery ?

I've just ordered couple of these cables. To make them works with my JTAG connector, I built a little 10 inches thong adapter, one end connects to the existing Xilinx 2.5 mm ribbon connector, the other end fits to my JTAG connector. The total legnth of the Xilinx ribbon and the thong is about 16 inches.

Now comes the mistery, whenever I try to program the devices (proms) iMPACT 9.1i reported bla.. bla.. bla.. the device ID doesn't match bsld, It's really a headache, trying upgrade drivers, service packs, bsld,...nothing helped at all, I also ran all available speeds from slowest to max, result were the same

As my last trial I made the thong a half legnth as before, total legnth now about 11 inches, as you guess, it worked

Is there any doc/manual that specify the max legnth for this cable?

Thanks all

Reply to
Marlboro
Loading thread data ...

Yes. The official specification is "short"

Reply to
cs_posting

Signal Integrity is the big problem with specifying a maximum length for the ribbon cable. We have built a ribbon cable 30 ft long (using the same materials as listed in the PC4 Users Guide) and it worked just fine. We have also built a 12 inch cable and it didn't work. For this reason a maximum ribbon cable length is not specified. The ribbon cable that is shipped with the USB/PC4 cable has been tested and was designed with signal intergity in mind and will properly isolate the configuration signals.

I don't really follow the 'thong' implementation, but would recommend two alternatives. Build a single ribbon cable the length you need using the materials specified in the Users Guide (do not connect two individual cables together trying to increase the length) or use a longer USB cable (from the PC/Linux box to the cable pod).

-David

Reply to
davide

Hi Davide,

Sorry for my English, I meant it a "dongle" not a thong :))) Looks like I'm missing you something here. You have a 30ft ribbon cable working just fine and couldn't make it works at 1ft ? I know signal integrity is length depedency, but those numbers (30, 1) just building up my fuzziness

Regard,

Reply to
Marlboro

I can certainly understand the fuzziness as the varying ribbon lengths seem very contradictory. This was one of the factors that lead to the decision to not specify a maximum length. Just from personal observation, I have seen many groups create varying lengths (usually 3 ft or less) of ribbon cable for their specific needs without any problem. It is important to note though that there have been a small number of occasions where a custom length cable did not work for one reason or another.

-David

Reply to
davide

"Marlboro,"

The mystery is actually caused by reflections on the cable. If the reflection returns at exactly the wrong moment, the bits are confused.

The reflection is the incident wave hitting the receiver, bouncing back to the transmitter, and bouncing back (again) to the receiver.

If you has a signal integrity simulation program (eg Mentor's Hyperlynx), you could actually simulate this effect, and see it happening.

Another consideration is that ribbon cable will couple (cross-talk) to itself if it is rolled up on itself. We have see the cable not working, unrolled it, spread it out over the floor, and it then works just fine.

Try reading:

formatting link
or
formatting link

or one of the many resources on the web for transmission line theory.

Austin

Reply to
Austin Lesea

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.