[ List Subjects ][ Post Message ]
Subject: Re: fuel level detector
[ View Followups ][ Post Followup ]
Date: 11/03/02 at 9:05 AM
Posted by: Ray
In Reply to: fuel level detector posted by xwindy on 10/30/02 at 3:23 AM:
We are doing that where I work. It has problems unless you do it right. The last thing to do is to place a "guard" ring/enclosure around the sensing plate so that the capacitor only sees the region of interest.
This should get rid of sensitivity to the surrounding enviroment.
Then you need to estimate the amount of background capacitance and the capacitance change due to the fluid level changing.
The way they are doing it at work is using a phase lock loop of some sort and measuring the frequency shift. Since it has given them problems I am not sure this is the best method. Perhaps measuring the time it takes to reach a certain voltage level with a certain current charging the plates is better.
I can not give you definite answer. I am inclined to switching the measuring circuit between a referance capcitance (that reflects the physical situation with respect to humidity and temperature) and the real measurement capacitance.
This would allow you to monitor just the change in capcitance; remember to make the referance so that the only difference between it and measurement is the fluid.
You should try to detail all of the sources of error quantitatively before you start and then design the circuit. After that compare and check that your original error estimates and controls actually did what you intend.
A designer should always make he knows what his design is doing.
Post a Follow-up: