The Clean Machine
Recent adventures in the realm of processing problems led me to consider and approach to processing problems that seemed to make sense. Start with a "brand new" machine. This means that the machine must be completely cleaned out. In fact if there's a runnerless mold involved it may behoove you to clean it out as well.
How is this best done? At the risk of being irritating I say, "It depends". Each operation pretty much stands on its own, in spite of the fact that there are obvious differences between locations, materials, personnel and management practices - at least. Below you will find a couple of my examples:
A. A runnerless mold that kept on leaking and burning polycarbonate, causing "hot shots". If you want more details, please get in touch with me. Suffice it to say the designer/manufacturer of the runnerless system demanded tolerances that might be considered too close for the average moldmaker. We wound up converting all of these molds to conventional runners which allowed us to maintain production. We wound up cleaning the runnerless feed system in a fluidized bed cleaner. This machine got a lot of business from the molding shop. Ours was bought from Procedyne and had a large enough basket to hold screws and tips. The way they work is really neat. The "bed" is really comprising hard alumina particles which act as a fluid once the heat is applied and the air flow is established. We set it at 800 degrees Fahrenheit for polycarbonate. After leaving the dirty equipment in there for a predetermined time the basket would be hoisted up and the components checked. Typically there was some surface residue that had to be cleaned. The runnerless systems seemed to come out ready to go- at least that's the way it was told to me. The samples were nice and clean! Of course if you want to operate a piece of equipment like this it should be vented to remove the heat and resultant plastic fumes.
B. Another runnerless mold was running flame retardant glass-filled PET that gave us fits when starting. After consulting with the material supplier they sent an expert in who told us that there's no one in the world successfully running their material in runnerless systems. By the time he told us, we had figured out a way to do it. When we shut the mold down we ran polypropylene through the injection unit AND through the mold until we made some parts out of polypropylene. Then we shut it down with the polypropylene in the manifold, allowing us to get the PET material flowing at a much lower temperature then if we started with the PET. Once we established a workable cycle the mold just ran. We never told the consultant how we did it.
C. A third runnerless system gave us periodic burn marks that were extremely difficult to get rid of. One key finding from experiences such as this where we had to get into production was that in order to quickly clean out the degradation we INCREASED the runnerless system temperatures by about 100 degrees. The usual result was that the suspect cavity would show a lot more burn for about 15 minutes and then the parts would start to come out cleaner, allowing us to dial back the temperatures to normal levels.
Now some of these troubleshooting techniques may sound suspicious to you; however, they all worked under our circumstances. These approaches were derived from determining exactly what the problem was and then using our heads to overcome it. A common thread that wove through all of these situations was how a pattern was established and then responded to. Once we knew how to counteract the effect, we made a record of it in what we called our manufacturing layout. These tips were available to all technicians. Some technicians even developed their own techniques for overcoming typical shop floor problems. Other key factors are that:
1. One must understand how the material flows
2. The injection molding cycle can be UNBELIEVABLY repeatable!
We did a study once to determine how long startups took in each of our molding areas (Each one had a sort of specialty). In one area we started up in less than 15 minutes - all machines. In one other area it might take a full shift or more to bring the last machines into their cycles. The key differences were in hhow the process was shut down. The first case was processes that were exclusively polycarbonate, and the machines were purged using polycarbonate. The second area used polyethylene to purge and produced tons of splayed parts until the polyethylene was completely run out. The splay was essentially in the same place leading us to believe the problem was in the front of the barrel. That's were the polypropylene residue stayed after the purging was done. Our material supplier suggested using polystyrene but our technicians complained that it stunk. Another consideration was that the people who shut down did not start up, so they weren't as concerned about the ramifications, nor how long it took to start. It seemed that the answer all along was to purge using the material that was running, especially if that same material would be used for start up.
Another option we tried was chemical purging compounds. My experience is not extensive; however, it was generally good - even for PVC. One must use caution to use the proper purging compound, depending on what was running. Also, the stuff we tried released an aroma of ammonia which was objectionable. We could never agree that this was the best way.
If all else failed it would be great to start with a "clean machine" (Where have we heard that before?). Let's call the machinist over and have him pull the screw and put the affected components into the fluidized bed cleaner, then we start from scratch - less work for the technician, but more for the machinist. ONE LAST AND VERY IMPORTANT NOTE: None of these scenarios turned out as simply as they're presented. There are other factors to be considered in all cases, but there's not enough room here to include the boring details. Please let me know if you want more. Golubki@cox.net
Jerry Golmanavich |
