KryoFlux - Incorrect Signal Levels

2009-11-16

With regards to the problem of incorrect signal levels, lets look at the facts:

  1. We hardly, if ever, get wrong logical levels from 3.5” drives.
  2. We frequently get them from 5.25” drives.

Issues:

  • Was this never considered a problem by different hardware manufacturers over the course of 30 years? It seems unlikely.
  • It’s only a problem if you can detect and use a signal that is too short by the electrical specification to be considered valid. A (working) glitch filter would solve it.
  • It’s only a problem if there are cases that are considered as 0 by TTL (5V) standards, but just signal an 1 by at 3.3V. Level shifters would take care of that.
  • It could be an issue with the chip.

It’s very possible to be a combination of all of the above, but in the hardware errata we found a section on PIO (Programmed Input/Output) Electrical Characteristics, where a certain condition causes leakage through VDDIO (power supply for Input/Output) and incorrect signal levels. It recommends what we had intended to do - use an external pull-ups to solve the problem. So hopefully, this is just down to the internal pull-ups not working very well with the 5.25” drives electrical specs. We plan to disable the internal pull-ups and add external circuitry (simply different/better resistors) in an attempt to solve the problem.