“Can we use this?”, asked the client when pointing to the two grease kettles during a recent commissioning visit, “without using this?”, when pointing to the two control panels.
One morning, during the next commissioning visit, we turned the heating on at the Control Panel and nothing happened and then one of the staff told us that all the manual valves were closed and they only open them when the automatic valves that they feed open, instead of leaving them permanently open and relying on the automatic valves to do their job.
Going back to 1990, our first foray into process automation was started by a somewhat fiery boss. “I want automation!” he demanded, “I don’t care what it does, I just want the plant to look like the most technically advanced that there it to make us stand out from the competition!” Sometimes management or salespeople demand added value but they don’t necessarily understand how to do it.
It was so easy to apply the knee jerk reaction of: “Automation is easy, all you do is put in a Process Logic Controller, connect sensors to its input and connect its outputs to valve actuators and then program it to do the process for you.” We went on to learn that it was not so simple.
Our first learning experience with a Process Logic Controller, in response to our boss’s demand, was to create a sequencing operation to automate a grease batch in a pressure kettle. Although the sequencing worked, we did not automate the loading the solids, resorting instead to the PLC sounding a factory horn to tell the operators when to put in the Calcium Hydroxide and Distilled Tallow Fatty Acid. The latter had to be added with shovels, making it particularly hard work.
With hindsight, we realised it would have been better to mechanise and automate the solids additions, for example, keeping the Distilled Tallow Fatty Acid in a hot tank as a liquid to be added by mass flowmeter, but perhaps, left the sequencing manually controlled to give the operators flexibility to adjust for batch to batch variations.
Bad experiences like this act as a major deterrence. Recently, we were told by a company in the Midlands, “We buy in tanks for our plant and the first thing we do is remove the load cells”. Presumably, they had a bad experience with automation.
The problem is that we look at automation in an emotional way. On the one hand, we get excited and enthusiastic by it and can see its potential as a highly visible way to add value. Think of the number of pictures one sees of factory robots on car assembly lines. On the other hand, it frightens us, partly because an invisible fault can shut down a plant that is fine in every other respect. A loose wire in a control panel can stop a changeover valve working, leading to a cooling circuit pumping into a heating circuit. A missing minus sign or overflowing data variable can plunge a self-guided missile or rocket into the Atlantic Ocean.
Historically, just because automation sometimes goes wrong, doesn’t mean it should be avoided. We need to adapt and work with it so that it can help us in a beneficial way. We need to use automation in a smarter way and use it where we really need it.
An early success was the semi-automation of base oil additions for a grease plant. Traditionally, the valves were manual and scattered around the site and the flowmeter with mechanical totalizer was in a remote location. So, the simple task of adding base oil involved a lot more legwork than it should have. A simple system involving two rotary switches, an electronic counter with totalizer and trip and a series of valve actuators, turned this operation into a task that could be done in less than a minute from one location.
Now, we are prioritising safety, reduction of the level of physical work and better control to improve quality and consistency. One safety example is replacing the traditional drum and broomstick for disposing caustic solutions, such as Lithium Hydroxide, with a pre-mixer with a lid that seals. Physical work can be reduced by mechanisms such as vacuum powder handling. Quality and consistency can be improved by using process vessel load cells or mass flowmeters. A method that simultaneously addresses safety issues and degree of control is to automate heating and cooling circuits, both of which can be run from a single controller. Not only does this manage temperature better but it prevents errors such as the wrong valve combination causing heating oil to be pumped into the cooling circuit or vice versa.
In conclusion, our view is that before automating, the best thing to do is to look at the existing process first to identify the manual handling, the safety and control issues, then automate to resolve the most serious issues first.