Custom programming and a little creativity go a long way for this customer.

A few years ago we had a customer who wanted to blend seven different ingredients to make a single new product. Sounds pretty simple, right? Normally you would simply dose each product into a tank, agitate and the transfer the new solution to a storage tank. As an added requirement, this customer needed to blend these seven products into a single pipe on their way to the storage tank. Well that doesn’t sound so bad either, just set a master flow rate and let each product run off of a ratio to the master flow rate. Should be good, no?

Ok, so maybe it’s not that easy; we ran into a few snags. For example, what happens if one of the products cannot reach the desired flow rate during production? What if the run ends and the total amount of each product in the tank does not meet up with their tolerances? What if everything starts fine and then half way through production one of the individual product lines plugs?

The first way we approached these issues was to set a master flow rate and let each product run off of a ratio to the master flow rate. If one of the lines was unable to achieve the desired ratio we would step down the master flow rate. Off to a good start. We had it working better than just letting the lines do their own thing, but the issues weren’t completely solved. This solution still allowed for the incorrect amount of product in the finished product tank when finished.

After a few discussions we decided to approach the problem from a new angle. What we would do is still have the master flow rate, but this time there would be a master total as well. Each individual line would have a targeted total based off of the ratio set. A totalizer was created based off of time and the requested flow rate. Then you are able to give your control loops a set point of 0, and make your process variable the error of the total. If your first line needs to have a total of 10 gallons after the first minute and the actual total was 8 the control loop would drive the valve open more in order to make up the difference. We also created a ramp function. The system would start with a set point lower than the requested master rate. This would allow each individual product to initiate flow. After a preset amount of time the master rate would ramp up over a period of time to achieve the desired rate. If one of the product lines would get behind to the point that the control valve would be open over a specified amount, the master set point would ramp back down until all products would catch up.

After weeks of development and testing it was time to implement. After a day of trying to get the equipment to operate correctly, our theory worked out as predicted. In less than 10 minutes the entire line was up and running exactly as desired. Each product line was flowing at the required ratio with the total in the tank matching the same requested ratio.

How does your team tackle new problems? Have another idea how we could have addressed this particular issue? Let me know.

By Jeff De Boer