Check valves are potentially troublesome devices. Here’s how to find out if yours is behaving properly.

Click Image to Enlarge

Check-valve function should not be taken for granted. Here are ways to test its performance.

Keeping your ducks in a row in molding is no easy chore; there are just too many ducks. The duck up for discussion in this column is melt delivery, particularly delivering the right amount of plastic to the mold. The screw, heater bands, and barrel have the job of melting and injecting hot plastic into the part cavities, but everything depends on the non-return valve or check ring to function properly.

There are various types of check rings: three-piece, four -piece, full-flow, ball check, etc. Which one is best? None—they all are problematic. It would make my day to see a group of molding company executives, quality directors, and six-sigma black belts witness one of these non-return valves functioning … not for a million shots but just for 10,000. They’d be forced to acknowledge that this mechanical device fouls up far more than three shots in a million—it’s more like at least three in 10,000.

My bet is that if the all the companies involved in molding spent their six-sigma budgets on improving the design of the non-return valve, the result would be a far more reliable device. But since that’s unlikely to happen, you’re on your own. Well, not entirely alone; here’s one way to test whether your check ring is working properly.

Let’s say you have a job running and you have been watching the cushion. You decide that the cushion is not as stable as you would like. That alone raises an interesting question: How much is the cushion allowed to vary? My view would be ± 1 mm. (Cushion for this article is defined as the amount of plastic in front of the screw at the end of the second stage or when the hold timer times out, not the minimum screw position as reported on some machines. There are two definitions.)

What’s causing the problem—the non-return or check valve or something else? Assuming that your machine is at safe operating conditions with all safeties functioning properly, take off pack and hold pressures, leave time on the hold or second-stage timer, and make about 20 shots. Line them all up; they should all be short and look nearly identical in size and weight. If not, you have an issue and need a better experiment.

Try the following, but before starting, ensure that you are switching from first to second stage via stroke position, and that you are using the minimum Delta P (the difference between the pressure used and the pressure limit you set on the machine). Perhaps 90% of the time, this should not be the maximum pressure the machine can reach. If your machine has first-stage pressure set at maximum, or if you don’t understand Delta P, set the first-stage pressure limit to the same pressure you use for pack-and-hold or second stage. Using second-stage pressure to test the non-return valve should insure against overpacking or damage to your runner or mold. If you feel this pressure will cause a problem, use a pressure that you feel is right. Safety first.

Then go back to making full parts by returning the hold pressure to the appropriate level for making good, well-packed parts. This is with the knowledge that without hold or second-stage pressure, parts are indeed short. Allow the process to stabilize for a few shots, then manually stop the press at the end of pack-and-hold or second stage. Do not open the mold and/or remove the part(s) until the experiment is complete (see below).

But use some common sense here; this experiment will take a few minutes, and if you are using a hot-runner system and are running a resin that can degrade quickly (like PVC), you may want to purge with fresh melt periodically.

Now, set the second-stage or pack-and-hold timer to about 12 sec, or whatever gives you enough time to watch screw movement. After this, adjust the shot size to about 80% of the capacity of the barrel. Manually rotate the screw to get it into this new shot-size position.

Important: Make sure your machine is cutting off first-stage filling on stroke or screw position and bring the cutoff or switchover position to within 0.25 in. (or 6 mm) of this new “80%” shot size. For example, if the shot size is 220 mm, stroke cutoff or transfer position should be 214 mm. Double-check everything and don’t stop until you’re confident you have things right. The goal here is that the first stage of injection will travel a very short distance and then go into a safe second-stage or hold pressure that is typical for this part.

Now manually inject and apply about 12 sec of hold. Watch for screw movement. Try to inject three times to check repeatability. Repeat the above at shot sizes of 60%, 40%, 20%, 10% and 5% of the barrel capacity. If the machine will not go to second stage in manual mode, set first-stage pressure the same as the second stage, or do the experiment in semi-automatic mode. Do the experiment with and without screw decompression after screw rotation. Decompression can make a big difference in check-valve seating.

If the screw creeps forward at any of these positions, you have a problem. It could mean a worn barrel or non-return valve, or both. You will have to pull the screw and check all components for wear. If the screw does not creep forward at all positions, you have a working non-return valve and barrel. Your cushion variation could be due to unmelt, contamination, or something else. If the screw does not move at the higher shot sizes but does move at the lower shot sizes, you have a good non-return valve, but your barrel is worn in the front section. The check valve is good because it held at the back positions—it moved forward at the lower shot sizes because plastic slips over the check valve, not through it.

Don’t think that all you have to do to check the function of the non-return valve is to watch the screw during injection to see when it rotates “backwards”—in the opposite direction from plasticating. Those days are gone, as nearly all machine builders now lock the screw from rotating during injection. Sometimes you can see it rotate backwards for a fraction of turn, but then it stops as they have locked the screw from rotating.

If you find that there is a problem with the check valve or the barrel, there is no satisfactory work-around. You have to bite the bullet and get the root cause fixed.
There is another telltale sign of a leaking non-return valve. Pull the hopper, empty the screw, and slowly decompress it. As the screw is being decompressed or sucked back, have someone (wearing the proper eye and head protection) look at the flights in the feed section as they go by. If you see any melted plastic on the feed flights, you have a problem. There are a number of things that can cause this, one of which is that the non-return valve is leaking so badly that the melt is being pushed all the way back to the feed throat. This will often cause bridging and be mistaken for too warm a feed-throat setting.