That's well explained - that's exactly how we test our systems before they go to the string test where the real end devices and possibly the gas turbines are connected.From when I worked in an alarm company's dev lab: these are resistance values seen by the controller to cause certain actions. Testing involved switching in resistors (of higher than normal precision -- often carefully set pots) to simulate each side of each boundary of the tolerance range specified for the controller. So that could be four tests for each controller action. Given that in production alarm and end of line resistors were the then standard 5% or 10% tolerance, the test resistance had to be set to better than 1% accuracy. That will be tighter now, probably looking for a test set accurate to 0.1%. Using digital potentiometers sounds like a good approach.Are all of these stuff resistors? Because I bet most of these parts can be emulated with a resistor bypassed by a transistor, and driven by a "simple" port expander .Digital flame detectors (end of line resistance: 10kohm; alarm resistance: 1kohm)
Digital smoke detectors (end of line resistance: 10kohm; alarm resistance: 1kohm)
Various push-buttons (end of line resistance: 10kohm; alarm resistance: 1kohm)
does one need to explore each and any value of the remaining resistors (perhaps, instead of adressable potentiometers, tw or for values would be enough (if number of truliy "continuoous restsiros is weak, maybe it is doable)
Statistics: Posted by Sph3x — Fri Feb 23, 2024 6:57 am