Recurrent missing pulses
Posted by
jcc3inc
on 2008-03-17 13:17:19 UTC
Good Afternoon,
A couple of weeks ago I mentioned a problem I was having with excess
count from a circuit I have to generate 1 to 31 pulses for X, Y, Z
using the parallel port. Basically I use a 1 MHz clock into the
parallel port which requests data from the computer. After the
computer receives the 1 KHz pulse, it responds with the data it has
generated for each axis in sequence.
I was having extra data out which I could not explain. After some
further investigation, I found that for 15 times in 10 second
interval, one of my 1 KHz Out pulses was lost! And it is Very
consistent. (The lost Out pulse resulted in a repeat of the previous
data in the external circuit, that's why there were extra pulses
out.)
Now the old computer that I normally program on is a 133 MHz unit
that once had WIN98 on it; I removed enough WIN98 so that it runs OK
in DOS. The second computer I tried it on was a 500 MHz WIN98
switched over to DOS via the start-stop options. Both show the same
problem.
Have I overlooked something?? Does DOS have some housekeeping
function that I'm not aware of? Does WIN98 in the DOS mode still
interrupt things at 1.5 times per second??
I am absolutely sure that the 1KHz data In is continuous, but
carefully watching the Out data on the scope shows an
occasional "missing tooth" in the out pulse train. I would
appreciate any comments/suggestions any of you may have about this
problem.
Maybe one should get another hard drive and just put DOS and the
QBasic program on it?
Thanks,
Jack C.
A couple of weeks ago I mentioned a problem I was having with excess
count from a circuit I have to generate 1 to 31 pulses for X, Y, Z
using the parallel port. Basically I use a 1 MHz clock into the
parallel port which requests data from the computer. After the
computer receives the 1 KHz pulse, it responds with the data it has
generated for each axis in sequence.
I was having extra data out which I could not explain. After some
further investigation, I found that for 15 times in 10 second
interval, one of my 1 KHz Out pulses was lost! And it is Very
consistent. (The lost Out pulse resulted in a repeat of the previous
data in the external circuit, that's why there were extra pulses
out.)
Now the old computer that I normally program on is a 133 MHz unit
that once had WIN98 on it; I removed enough WIN98 so that it runs OK
in DOS. The second computer I tried it on was a 500 MHz WIN98
switched over to DOS via the start-stop options. Both show the same
problem.
Have I overlooked something?? Does DOS have some housekeeping
function that I'm not aware of? Does WIN98 in the DOS mode still
interrupt things at 1.5 times per second??
I am absolutely sure that the 1KHz data In is continuous, but
carefully watching the Out data on the scope shows an
occasional "missing tooth" in the out pulse train. I would
appreciate any comments/suggestions any of you may have about this
problem.
Maybe one should get another hard drive and just put DOS and the
QBasic program on it?
Thanks,
Jack C.