I will tell you an interesting case. One integrator turned to me with a request to make a control system for kilns. So that the parts on the ramp function are warmed up, kept at a certain temperature and cooled down smoothly. For all this, it is necessary to develop a program that will be executed on the Owen PLC110 controller and a visualization program that will control the technological process. The terms of reference for the implementation looked more like the release of the Murzilka magazine. And I answered him that since I had rented out brick factories, I would definitely cope with some stoves.
The advance was paid and the work began. I quickly made a controller configuration, since it was easy to inherit my own, previously made, developments associated with a smooth increase in temperature. With the visualization program it was a little more difficult, but also bearable. Invited my companion who knows C # well to create order entry forms. We made a simulation program and debugged this process on simulators as we understood. It seems that the program is ready, everything works, take it and use it. But, as it turned out, the local personnel of the military enterprise had a different view of this program.
Thus, it turned out that the work that could have been completed in 2-3 days has been completed for almost 6 months. In the course of the work, it became clear that the temperature should be regulated at a minimum and maximum (by delta).
This medieval approach is very difficult to accept as how the PID regulation acts on the calculated setpoint. The reporting system should contain both graphical and textual information about the firing process carried out, should capture both the period of time before the start of the process and after. Each product should be able to select its own firing program, and different products with different temperature requirements can be inserted into the same furnace.
This scheme, if you think about it, is extremely conflicting. Titanium alloy is not equal to copper. And if one product improves, then the other will obviously be spoiled and sent to waste. There were nagging about the width of the fields in the order entry forms. Every time the operator insisted on his field width, the next time he entered it, it flew off, which caused a negative. The shim-regulation by the heating element also left a lot of better, since the process took place in a very wave-like manner.
The acceptance process was delayed. The fourth version of the comments is already underway, it is very difficult to imagine when the finish line will be. There is a feeling that the Customer, in the acceptance process, is himself busy thinking out how it should work for him, and complicates the conditions, thereby increasing the number of iterations for carrying out work. It turns out to be a real temporary trap.
#furnace, #DBMS, #heattreatment, #ramp, #Master, #OWEN, #PLC110Russian version