I am standing in the picture above with two weighing machines where I designed and built the electrics from straight out of my head. At the time I was hired by Trans-Packers Services they were doing a lot of Vertical Form Fill Seal jobs. This company never had their own full time electrician. I was their first. They used to pay a lot of money to have their machines rebuilt by outside contractors at a very high price. Now they had their own in house guy and to see what I could do I was given these 2 projects to make a couple of weigh machines for them.
These machines originally were point contact scales. That is a counter weight under each weigh bucket was set to a desired weigh value. As the bucket filled with product an arm came up with a needle point electrical contact on it. When the bucket weight matched the target weight the point contact would touch another pad to close an electrical circuit to energize the opening of the weigh bucket doors to dump the product into a collection chute and then begin the cycle again. These scales worked fine at low speeds but then they get kind of erratic at higher speeds and in packaging profits are all about high speed and high accuracy which translates into lower cost.
To improve these machines I decided to buy digital strain gauges and load cell sensors and to interface them to a new machine control system that ran the machine through a PLC. You can see some step by step photos of how that was done in the slide show with the link on the left.
The machine behind me also did vertical form fill seal bagging while the machine in the front of me just did weighing and dispensing into a preformed container like a can, bottle, bag or pouch. The machine on the table on the left is a pad printer used to print a label on the roll of plastic film used to make the bags on the vertical form fill seal machine. You can view a slide show of how I rebuilt that printer under the date coder tab in this section of the site.
These machines both worked, but they worked slow. In retrospect the reason was because I used PLCs and PLC type software to create the programs that ran them. The problem with this type of software is that the CPU scans the program from top to bottom or left to right in serial fashion searching for changes in events to act upon. Depending on how big the program is these scans could take milliseconds to complete, which are like hours in terms of production time. If an event changed on an element after a scan it would not be detected until the rescan of the entire program. Special interrupts could be programmed and assigned to specific inputs to get a real time response but then the program would get very messy and the code would start to look like elements connected with spaghetti
What I should have done is bought high speed single board computers that can handle multiple threads simultaneously at once in real time. I would have to program this in something like C language targeting a Real Time OS (RTOS). This would have taken much more time to do and is harder to debug and I was now working at a company where the only way they believe work is occurring is if you are twisting a screwdriver in your hands and not tapping keys on a laptop. Of course little by little as their newer technology machines started to break down they came to realize that software is a necessary evil they cannot avoid, but if they could they would
Because of that I limit all new PLC Installations to what I call SPLICE POINT TECHNOLOGY, which I will describe further in the PLC section of this site.