Sometimes the problems might seem bizarre or weird and working in the field, sometimes you will find yourself alone or with no technical assistance due to whatever circumstances. It makes it a bit more difficult when the problem is critical to an operation and you have the big wigs breathing down your neck and asking when it’s going to be fixed.
Be analytical in your thinking. I tend to take a systems approach, ask a few questions if the information I received was incomplete or unclear. Was it working when you started the process? What was happening when it failed? Is the reading just not scaling properly or is there no reading at all?
Broken wires, corrosion of connections, loose connections, power surges, improper grounding, overheating, water or moisture, dead batteries, power failure and don’t forget the one that the brass never wants to hear, operator failure.
A lot of time there are multiple ways that the same type of reading is determined or something to reference the reading in order to know that the reading is wrong to begin with if it wasn’t a total failure.
I’ve seen enough operator errors that I think I could write a book on stupid user tricks.
I will start with current devices. Most common current devices work in the 4 to 20 milliamp range and either source or sink the current. Sourcing, the devices provides the current using its own power source and sinking, the devices is supplied with an excitation voltage. A current reading of more than 20 mA or less than 4 mA is generally and error condition.
Never check voltage when your meter is in current mode. You will blow the fuse in your meter. I don’t know how many times I traveled many miles to find that the technician at the location had a blown fuse in his multimeter.
If the device is a sinking device, check the excitation voltage. This varies among devices but it is very common to have 24 vdc supplied. You can usually find a way to put your current meter in series with the circuit to see if there is current. If it is a sourcing device, you can disconnect the current loop from the device and measure the current at the device.
In some systems, new fuse blocks have a LED indication for when a fuse is blown and sometimes these fuse blocks may be mounted where it is not obvious. With no load on a power circuit from one of these types of fuse holders, a good voltage will be read through the resistor and LED but when a load is applied, there is no power.
In outside environments, I have seen many times that the pin in the connector going to a pressure transducer is gone. Eat up by electrolysis from moisture in the connection. I’ve seen this in butt splices as well and sometimes from solder connections that were not properly cleaned that led to corrosion.
On one installation in a mobile process, the crew had to switch channels on their radios. Previous to this everything had been working fine but after switching frequencies on their radios, every time they keyed the radio in the data acquisition van, their density reading spiked. The density gauge transmitter, radio and data acquisition system were all on separate isolated power supplies. The small length of the current loop to the acquisition system was the just right to make an antenna that was picking up from when the radio was keyed. A common ground between the instrument transmitter and the acquisition system solved the problem.
I’ve been to several sites where the current devices were not properly scaled or where an operator used a span function the zero the instrument or where cables were swapped and the sensor they thought they were reading was a different sensor.
A fairly popular nuclear density gauge from Thermo Fisher is the SGDO and when maintained is quiet dependable. When running slurry that is abrasive, the pipe will wear and the inside diameter changes thus changing the readings. Also there is a rubber grommet than holds the ion chamber in place.
I’ve seen it all with these. Mounted in a high vibration situation the techs had installed it upside down to help keep the moisture out of the connection. Vibration caused the ion chamber to slip in its enclosure causing a higher reading. Have also seen them with the grommet broken and they were dancing loose in the enclosure.
In transit and when not in use, nuclear gauges that are equipped with shutters are required to have the shutters closed and locked and the gauge itself securely locked to prevent exposure to the radiation and prevent theft. Several times after a tech had been working on a transmitter problem, I discovered the shutter still closed.
Stick level sensors from various makers work well in maintaining levels in mixing tubs or monitoring levels in storage vessels. Sometimes a problem can be with buildup on the stick depending on the solution being measured. I saw one that came in from the field and couldn’t find a problem with the signal. The control valve settings on the acquisition and control system was scaled improperly. Instead of 4 to 20 mA, it was scaled 40 to 20 mA for the output.
In controlling valves or actuators with current, a lot of the time it will involve an I2P or Current to Pressure transducer that uses air pressure to actuate the valve or actuator. It is very important to have filtered dry air in these applications. Water or condensation in these will make for a bad day.
In my next installment I plan to cover some pulsed signals and flow meters.
Comments