INJECTION MOLDING: Processing Nylon…and Other Problems

Over the years I have been asked frequently about molding moisture-sensitive resins. Recently, I received a question from a molder about a problem he was having with nylon. His moisture content was varying within the accepted limits of 0.20% to 0.005%. This range of moisture content influences the viscosity of the nylon significantly, which in turn will influence the process.

This issue applies not just to moisture-sensitive resins such as nylon, polycarbonate, PET, etc.—because the principles used to solve the problem can also be used to make good parts as lots or color changes occur. Normal viscosity variations in your process can be expected and can be dealt with by applying Scientific Molding methods.

Processing starts with understanding the resin and providing correct preparation. If you are dealing with nylon or other moisture-sensitive resins that have the possibility of undergoing hydrolysis during processing, you must dry the resin. Hydrolysis means the moisture will react with and degrade the polymer chain, making it shorter. This in turn degrades the properties of the resin and subsequent part.

It may not necessarily produce visible splay, as the water is consumed in the hydrolysis reaction. Those who look for splay to check for moisture will be fooled. Bottom line: You are not drying the resin just to get the water out; you are drying the resin to ensure it performs properly in the application. And no, you cannot see this low level of moisture.

The level of moisture content allowed for proper processing is resin dependent, so check with your supplier to find what is allowable. For nylon it is usually quoted as below 0.20%; most other moisture-sensitive resins need to be below 0.02%. Further, there are subtleties that the molder must consider. 

In the case of nylon, for example, if you are processing 30% glass-filled compound with a moisture content of 0.17%, (established with a moisture-specific method—weight loss is not acceptable), you may feel you are ready to process. But you’re not. The 0.20% moisture specification is for the nylon only, not the glass. If you have 30% glass and 70% nylon in your compound, with 0.17% moisture you have actually have 0.17 kg of water in 70 kg of nylon. So your nylon moisture content is actually 0.24%—too wet to process.

So you dry this nylon to 0.13% (or below) and maintain it in a dry environment so that it cannot pick up moisture before it is molded into parts. You have less than 30 minutes in a normal processing environment to use most dried resins before they will pick up enough moisture to necessitate re-drying.

Now that you understand the resin and have dried it properly, you can set up your processing conditions. For the purpose of discussion, let’s assume that the material varies in moisture content from 0.13% to 0.005% during production. Note that you cannot “overdry” nylon. Bone-dry nylon will actually make good parts. Typical for nylon, give the parts time to absorb moisture from the air and toughen up a bit before you subject them to any stress.

How do you set up a robust process with incoming material that varies in viscosity? Basically, you run the machine on cruise control. That is, keep fill time consistent to ±0.04 sec by allowing the machine to use whatever pressure it needs to fill the part 90% to 99.9% by volume. Essentially this is cruise control on your machine. For your machine to have this capability you must determine the appropriate Delta P required for it to compensate for viscosity changes (and it is not 10% more than pressure at transfer). In addition, ensure the machine is “load compensated.” These topics have been explained in previous columns (click here and here).

If you keep fill time consistent and are filling the cavity to the same level on each shot—that is, the check valve is working properly—filling or first-stage injection will be automatically compensated by your molding machine. Again, this assumes Delta P is maintained and the machine is load compensated. With a consistent fill time, the pressure at transfer correlates with viscosity. If you plotted the pressure at transfer for filling or first-stage injection for each shot, it should look like the accompanying graph.

The graph shows actual data for five lots of the same grade of material from the same supplier. It shows the first lot was stiffer, took more pressure, than lot two. Lot three was stiffer than both lots two and one, and so forth. Note the wide variation between lots. Within the lots there is some variation shot to shot. This shows the machine is compensating for all the variables in the process.

If you are getting the same pressure on every shot, something is wrong. Looking at the data, we can see variations between resin lots. Parts have to be inspected or measured to make sure they are to specifications. As you measure parts shot to shot and resin lot to lot, you’ll find the range of pressures that make acceptable parts. When the moisture, lot, color, or additive variation is out of the pressure range that has been proven to make good parts, the process has to be adjusted to bring the part back within specification.

Let’s say that lot one produced parts to specifications. Lot two did not. Because the second lot is easier flowing, the parts can be expected to be larger and heavier. To return the part to specification, the only machine variable that has to change is the hold pressure—not temperature, profile, etc. How much lower hold pressure? You will have to experiment, but the difference is usually not more than 400 psi plastic pressure. If the lot is stiffer, more hold pressure will be required. Believe it or not, all of the five lots in the graph, minus the lot-change shots, made good parts to a relatively high tolerance.

In short, set up filling or first-stage injection to adjust for the viscosity changes; take data over time and find what pressure range makes good parts. As long as first-stage fill is working properly the only change needed should be adjusting hold pressure. Do not play with temperatures. This should take care or 80–90% of your molds. There will be a few molds that our nemesis “Murphy” owns and they will require the next level of processing.

ABOUT THE AUTHOR

John Bozzelli is the founder of Injection Molding Solutions (Scientific Molding) in Midland, Mich., a provider of training and consulting services to injection molders, including LIMS, and other specialties. E-mail john@scientificmolding.com or visit scientificmolding.com.