Measuring high voltage using basic stamp?

CuriousOne

Hello.

I need to measure with say 1% accuracy voltage from 200 to 330v. As basic stamp manual says, you can use capacitor charge/discharge time to determine the input voltage, but my problem is, that voltage source is charge pump, which delivers very little current, so, I will have to use very high impedance voltage divider resistor, which will cause measuring capacitor to charge slowly, which itself will cause very low integration speed. Any solutions?

Mike Green

Use a smaller capacitor.

You want 1% accuracy and I don't think you'll get that using RCTIME. Basic Stamp timing is somewhat temperature sensitive. It's certainly sensitive to supply voltage and that's not regulated to 1%. You will need an external analog to digital converter with an external reference source and you'll need to supply that reference and make sure it's accurate to at least 1%.

CuriousOne

Actually, I need to capture two moments, when voltage reaches 200v and when it reaches 310v and measure time, how long it had taken. So I think, maybe I use two neon bulbs feed via divider adjusted to specific voltage, and each bulb illuminates dedicated phototransistor, which generates logical level for stamp?

Mike Green

How much variation do you think is involved in neon bulbs lighting? How temperature sensitive do you think they are? What would be the threshold for the phototransistor? It would be one thing if you were looking at 10-20% accuracy and repeatability, but you're not.

CuriousOne

I did some preliminary tests. At room temperature, "on" voltage for particular neon lamp is quite stable (if feed with DC current, with AC, it really has large variation), accuracy better than 5% is possible I believe, however, I haven't done extended temperature range tests so far.