I finished up on the circuit design and prototyped the first 16 channels to read current from 0 to 25 milliamps. I had one channel that was off by about 12 pounds at 7500 PSI and this was disturbing when all other channels were spot on. Zero and 15000 were spot on as well for the offending channel but it wasn't tracking the way it should.
Thinking and rethinking, replacing parts and the processor and of course it was the last thing I tried that fixed the problem. It was a leaking zener diode. It's been up and running for about 48 hours and still tracking. I brought up the fact that it was the last thing I did that fixed the problem. Once I was on a job with Van and he was searching through many storage containers for a particular part. When he found it, I ask him to look in one more container and he did. I said, “See there, it wasn't in the last place you looked." Just so I not always like someone else, occasionally I put my pants on both legs at the same time and once in a great while I get off of bed from the foot of the bed.
Back to the testing, and design of the 4-20 milliamp input module, 4 ma being zero PSI in this example and 20 milliamps being 15000 PSI and this was my problem channel before I found the leaking zener. Now it is sharp and spot on.
4 ma reads 0 5 ma reads 938 6 ma reads 1875 7 ma reads 2812 8 ma reads 3750 9 ma reads 4687 10 ma reads 5625 11 ma reads 6563 12 ma reads 7500 13 ma reads 8439 14 ma reads 9376 15 ma reads 10313 16 ma reads 11250 17 ma reads 12189 18 ma reads 13126 19 ma reads 13063 20 ma reads 15000
I know that these number are so close that it looks like I used a spreadsheet to come up with the numbers but using 16 bit integers this is wat the program is reporting. Nice and tight and I like it!!!
When we put this into production, the parts will be of a higher tolerance I would hope.