hi all,

again a weird phenomena ....
... last friday I was testing USB recovery software
... so I needed something to upset a good working USB connection
... so I hacked an electronic fly-flapper to get a an sustainable arc of 
about 0.5 mm.
... I held the arc (with about 10 cm wire) a few centimeters from PIC/FTDI

The DUT (Device Under Test) consisted mainly of a FTDI232BM + external 
eeprom + 2*16F877.
With the first DUT I was able to disturb the USB connection 8 times,
when I did the 9-the test, the DUT failed completeley.

So I took the next DUT, which already after 3 times failed compeletely.

It looked like the ftdi chip still could establish a correct USB connection,
and with one of the systems, the PIC even did his first words.
So most likely the PIC was damaged.

On monday I wanted to demonstrate the failing DUTs to a collegae , ...
... and you'll never guess, but the second DUT behaved compeletely normal !!
The first DUT was still performing the same problems.

Until today I hadn't time to investigate the problem with the second 
device, ...
... but you can already guess what happened when I wanted to investigate 
the first DUT ;-)

Has anyone an plausible explanation for this kind of behaviour ?
Or has anyone seen such behaviour before ?

I've seen this self healing effect before on a 16F84 or 16F628,
and mentioned in this list a few years ago.
The cause in that case was a bad programmer in combination
with very frequently programming (20 times an hour, for many hours).

thanks,
Stef Mientki

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist