Instrumentation Troubleshooting Part 4
By Mark Williamson
Quality, quantity, resolution, accuracy and design are all very important aspects in any system and it doesn’t matter if we are talking about signal levels, power supplies, hydraulics, air supply, fuel, chemical additives, media transport or whatever else composes a system that performs a series of tasks that eventually translates into us receiving a paycheck in a timely manner.
I’m going to start off talking about air. Ample clean dry air is essential for the proper operation of most all processes, from system controls, to breathing apparatuses, to bulk distribution systems.
Control valves and actuators are not cheap and neither is human life. Moisture can be a killer. Any time you have a pressure drop in an air-line, the temperature drops and if moisture is present, when dew point is reached you have water in your air-line. That is not a good thing. I have seen lots of damage caused by water in control valves, actuators, current to pressure transducers, brake systems and if you can imagine the horror to someone in a contained environmental suit getting a slug of water.
Condensation at a pressure drop can sometimes freeze causing an ice block, the reason alcohol is injected at meter runs in the oilfield.
Driers sometime fail and even when a failsafe is in place they may malfunction. When the outside humidity increases, more moisture in introduced into the system where ambient air is being compressed. There is not a lot we can do about the weather except to have proper safeguards in place. Regular checks on air tanks to drain any water buildup even if automatic dumps are in place is important.
In the design stage, hopefully the engineers provided for an adequate air supply for all the instrumentation needs but as systems get some age leaks can be a problem. Enough small leaks make for large leaks and the air supply may be strained in supplying adequate air for the operation.
You may think of yourself as an electronics person but when working on systems, keep an eye out for the mechanical things as well. It may be your own life you save.
There have been several plant explosions recently and in the past, some of which we may never know the real cause. Many years ago, working on a cryogenic plant I noticed a stainless steel tubing was run alongside a pressure vessel and from vibration had worn a groove into the pressure vessel. I immediately contacted the plant foreman and we put the plant into emergency shutdown. Countless lives may have been saved that day.
Vibration or repetitive movement can cause wear and wear is damage. I have seen hydraulic lines that flex, rub through wiring bundles and wear through the hydraulic lines when the encounter a metal edge.
At another new cryogenic plant build that we were near completion on and performing commissioning startup, the plant was going into emergency shutdown when the dehydration beds switched out. It took some time to figure it out but it ended up being an inadequate air supply to the actuator that switched the bypass valves. The valves were switching, it just took too long and sensing low pressure in the switchover was the culprit.
Even though you have the proper signal going to a device, doesn’t always mean you will get the proper results. Restricted flow in the air or hydraulics or over pressure can widely vary your results. I have seen hydraulic pumps damaged from debris left behind in construction of the hydraulic tanks. A costly mistake and even more costly if it happens at a critical stage of an operation.
There are many types of compressors and pumps, centrifugal, piston, turbine, gear, impeller and rotary just to name a few. Just because your controls go from zero to 100 percent doesn’t mean that the pumps perform at zero to 100 percent. Centrifugal pumps can cavitate when run at too high a speed. This leads to decreased performance, entrained gasses in the fluid and will affect fluid measurements as in flow rate and bulk density.
Gear type pumps are used sometimes in putting additives into a mixture and they need to be sized appropriately for the volume to be pumped if proper loading is to be expected. Running them from hydraulic motors is quite common but at very low speeds they may be jerky or quit. In this situation, a gear reduction to the pump might be warranted.
On yet another plant startup, one of the product pumps was vapor locking. Even though the plant was built to the engineer’s specifications, the run of the product line had an unnecessarily long run of pipe going to the inlet of the pump. A visit with the pipe fitters and welders, we fabricated a different run, putting a slight drop in the piping so that the gas bubbles could flow back to the tower and the problem was solved. A week later the engineer came out to address the problem. My answer was to him, “Draw it up and correct your As-Builts.” He was happy and so was the customer.
Back to instrumentation. We have a desire and a need for some level of control and monitoring. A lot of this is hidden in the mumbo jumbo deep inside the software someone wrote. A proper human machine interface is nice to have and if designed properly, makes for a safer and more effective work environment. Being pretty, with nice looking graphs and charts is good but it is only as good as what information it received.
The higher the bit resolution the finer we can granulate the measurements but at some point the ability to measure will exceed the precision of the device being measured. The accuracy of the device and the repeatability are both factors in the equation.
Wear and buildup on all sensors is another factor. Especially when working with slurry, it can be abrasive and the pipe will wear out. In calibrating, most all density devices, the inside diameter needs to be known for the formulas to work. As the pipe wears, the inside diameter increases and causes the readings to be skewed. If you are working with a range of zero to 30 pounds per gallon and the instrument only has precision of .1 pounds, an ADC with 10 bits of resolution would do the job. They are most often calibrated with a carrier fluid of water which is not always the case.
In fracturing wells, the water may be processed or reclaimed water containing a level of salt to varying degrees. This water has been standing in ponds where stratification can occur meaning that the varying levels will have a different salt content. The fresher water will be on top and the saltier, heavier water will be on the bottom. As the water is used over a period of time the weight of the carrier fluid will change affecting your density measurements.
In pressure pumping activities, piping has to be recertified on an ongoing basis. Portable gauges sometimes have their pipe replaced. In one of these cases a technician called me in the early hours of the morning. He had replaced the pipe on a density gauge and now it was not reading properly. The “new” pipe even though it was certified was more worn that the pipe he removed and the gain was adjusted too high. This was on a Thermo SGD-O gauge. Simple fix, power down, crank R-35 clockwise, power it up and adjust for 9.6 volts. The gain being too high for the thinner pipe had over saturated the circuit making adjustment impossible.
Moisture in circuits is not a good thing. Arcing in high voltage circuits, changing values of components, electrolysis and corrosion in connections all resulting in flawed readings and possible damaged circuits.
You can have the best systems with the best sensors on the market but a bad wire or faulty connection can still be your downfall.
Keep it neat, clean, tight and dry. Don’t make it a nightmare for the next technician that comes in to troubleshoot. Do a good job, fix the problem and make it look like you were never there.