Automating Resistance Measurements

[jdmcdon2]

Hey guys I'm a Mechanical Engineering Senior working on a Senior Design project. Our project is to redesign a testing apparatus for a positive pulse mud telemetry tool. How the tool operates isn't really important to our project (if you're curious I'll explain it), but what is important is the measurements they want our test apparatus to be able to take. The project is a redesign of a current testing apparatus, the apparatus is operated manually by a technician and they want to automate it so that the tech just pushes a button on a computer and the device reads several measurements on it's own without prompting. So far we can automate, the RPM, Temperature, and stroke length measurements. We just have a question regarding the resistance measurements. The tool in use is below surface and is controlled above ground by an operator, the signals our sent to the tool through a wire, which sends signals to open/close a valve. The wire is composed of 6 different segments all bundled together they want to check the resistances between the wire while it's hot and again while it's cold. Righ now the tech performs this using a multimeter checking a-b a-c...etc. Again they want to automate this so the computer sends reads all these a-b signals without requiring the lab tech to touch wire to wire manually. My question is if we design a cap of sorts that allows a multimeter to be plugged into all 5 wires at once is there a way to filter the voltage sent by the multimeter to just two of the leads and then to another two so that it is essentially doing the A-B, A-C measurements automatically? If you guys could help I would appreciate it, my circuitry knowledge is limited at best.

-Josh

 

[Militoy]

The computer makes the problem fairly straightforward. I use several different pieces of automated test equipment (ATE) every day to perform the same task (along with other more complex tests). We bridge our ohmmeter leads in Kelvin configuration – as we are typically measuring down to a few milliohms, with 4-digit accuracy. It doesn’t sound from your description like you will need to go to quite this extreme to get your data. The first step will be to get the computer to switch leads for you. There are many low-cost relay-driver I/O boards available to perform this job. Even if you decide to build the circuit yourself, look on eBay first to get an idea of what’s available. I’ve seen 8-relay boards for around $25 – so it might not even make sense to build your own – unless that’s part of your assignment. Once you figure out the lead switching – you will need to time and datalog the measurements. Again – you can use the relay board to trigger the measurement operations. There are plenty of low-cost “canned” solutions to the datalogging – but we chose to build our own circuits. I use a matrix of switchable shunts, and select a shunt and reference voltage level appropriate to the DC resistance to be measured. We measure and log a voltage drop across the known shunt resistance to determine the unknown resistance. Our program is set up to switch to a higher or lower value shunt, and adjust the reference test voltage, in the event the initial measurement is outside of a specific voltage range. You may not need this function, if all of your downhole resistances are within a fairly narrow range.