I suspect the answer was not the one that was hoped for but the rules are clear and we are not allowed to modify extension cables.
I wanted to comment on this as I doubt the use of the ferrite on the I2C extension cable makes much difference. They are used on the motor cables for EMI suppression. I’m pretty sure that if the use on IME cables was significant this would have been suggested by VEX when we had all the issues a couple of years ago.
It would be interesting to make a new comparison, use an IME cable with and without a ferrite on the same robot in the same location and see what the relative difference in performance is. If the results can show improvement then perhaps this could be a new product or a rule change to allow the modification in the future.
I tried this 2 years ago without much success. I used a toroid about an inch long but not sure of the real strength looping the 4 wire signal wire through it,
The data packets of the IME’s would come in without too much issue. No reported packet loss from jpearman’s sample IME packet interrogation code. So that went back to the static discharge theory.
I think clamping diodes either at the terminating point at the Cortex or as a dongle at each IME would help significantly more as it should stop any spikes passing through the signal wire at the Cortex. But will it clamp sufficiently fast enough as well as not interrupt normal signals?
I never got to making a sufficient circuit to try that though.
In the good news department, quad encoders have not failed us since.
Perhaps VEX should make a revision to the IMEs so they have diodes in them already. I would be nice to be able to use IMEs that are accurate because they are typically more convenient to use than the quad encoders.
