??? 05/30/07 07:49 Read: times |
#140031 - Avoid using interrupts where possible Responding to: ???'s previous message |
Ali, I gather the RUN signal is very low speed (if it is connected to a lamp it most certainly is), so connecting this to an interrupt is probably not advisable. Why so? Your incoming signal may pick up interference from various sources -mobile phones, two way radios etc so unless you filter the signal extensively, you have the possibility of your system crawling to a halt if too many interrupts happen. Whereas, if you poll, you don't care if the input signal is changing faster than expected, your load on the processor is constant (or near enough) and you end up with a more robust system.
If this is for an industrial system, why not use a PLC? This sounds like exactly what you want. These are cheap, fully certified and built for purpose. You could have your solution in days rather than weeks. To build your own controller will take time, money and you'd want to have a bit of experience to better or equal even the cheapest PLCs. |
Topic | Author | Date |
12 parallel pulses counting | 01/01/70 00:00 | |
How fast are these pulses? | 01/01/70 00:00 | |
pulses speed | 01/01/70 00:00 | |
More info | 01/01/70 00:00 | |
Pulse rate | 01/01/70 00:00 | |
Isn't it slightly more complicated? | 01/01/70 00:00 | |
Stretching pulses | 01/01/70 00:00 | |
Another Quick Way | 01/01/70 00:00 | |
Sampling timer with a buffer | 01/01/70 00:00 | |
Showing Vertical Couners | 01/01/70 00:00 | |
Yes other interrupt are also there | 01/01/70 00:00 | |
Avoid using interrupts where possible | 01/01/70 00:00 | |
Run signals frequency... | 01/01/70 00:00 | |
PLC can be dirt cheap | 01/01/70 00:00 | |
PLCs | 01/01/70 00:00 | |
an xcellent idea![]() | 01/01/70 00:00 | |
Sample with 2 74LS165 parallel loadShiftRegisiters | 01/01/70 00:00 | |
would not the output pins drop... | 01/01/70 00:00 | |
No Extra Hardware Needed | 01/01/70 00:00 | |
No Extra Hardware Needed - I do not agree | 01/01/70 00:00 |