Injection Molding: Pressure, Time, Temperature and Calibration

[Bubba]

Injection Molding

We use several injection molding presses at our facility to manufacture many different components out of varied thermoplastics. Up until now, we have calibrated the temperature controllers only. We rationalized that the pressures and times were reference numbers that were verified by the output of the machine. In other words, everytime a mold was installed in the press, the first shot would be inspected to verify that it meets specifications. If it did not, then the pressures and time could be adjusted until a good shot was produced. Then parts would be sampled throughout the run of product and at the end to make sure that the product did not drift out of spec. Temperature was considered to be a critical parameter due to the possible degradation of material that could not be inspected for if the temperature was inaccurate.

Now, we have leadership that wants to see calibrations on time and pressure. I am not very knowledgable about injection molding, but I notice that there are a lot of times and pressures available to calibrate on each press. I guess my question is simply what makes sense to calibrate on these things, and what should be left as reference only? We need to meet ISO 9000, FDA, and shortly ISO 13485 requirements. The presses that we have include Arburg, Kawaguchi, and Toshiba.

Does anyone out there have any guidance that I could use?

 

[Jerry Eldred]

Re: Injection Molding

There are a few directions you could go. I am not at all familiar with FDA requirements, so any replies do not take that into account.

With the above qualifier, I have seen similar situations in my industry (some nearly identical). I will give you one caution. Please read this detail carefully, as I want to be sure you fully understand my implications.

There is the potential that you could degrade your process control by calibrating every instrument on the system. The explanation is that your process specifications are typically based on stability rather than absolute accuracy of any given indicator on the production tool. I don't know individual specs on the meters and transducers on the tool used in your factory. However, it is possible that the calibrated tolerances of the individual process monitor/control instrumentation, when added together cumulatively may be much worse than the allowable stability limits of your process.

It is common to use a well developed system of Statistical Process Control (SPC) as a means to assure the process is maintained where it needs to be. Then the absolute accuracy aspect is achieved through some sort of test of the finished product at the output of the process step.

The trick is knowing which process monitors must be monitored, and what amount of drift to look for.

If you must calibrate time and pressure, I would recommend a process review by competent technical personnel. Determine what the needed spec is for the process. Then document that as the calibrated accuracy. Many digital timers are much more accurate than process specs. So it may be pretty painless, and cause minimal (if any discernable) shift to the process.

Regarding pressure, that is the aspect I wonder how the accuracy, resolution and stability compare with your process specs. If they are high quality digital pressure controller and transducers (be sure to calibrate transducer and display unit - together as a matched set if applicable). It is possible in that circumstance that this would also minimally impact process.

If process instrumentation is much better than process specs, it is possible that you'll improve confidence in the process. Alternatively, be sure to review the relationship between process control limits and instrument tolerances before adjusting. I have gotten a few engineers pretty irate when I adjusted instruments to their nominal, while at the same time throwing the process completely out of control.

Not sure if this is any help.

 

[howste]

Re: Injection Molding

My short answer would be this: you don't need to calibrate them if you are verifying the output of the process. The 9001 and 13485 standards say "Where necessary to ensure valid results, measuring equipment shall be calibrated..."

If you can't verify the output, it falls under 7.5.2, and you must validate in some way - probably including calibration.

I also am not familiar with FDA requirements, but for 9001 and 13485, the above answer applies.

 

[Howard Atkins]

I think that there are 2 issues here if I remember my technology.

1. Calibration of the temp, time and pressure just shows that the gauges etc give a correct reading and does not make any difference to the process.
2. Injection moulding machines despite the parameters that you enter have an interaction during work that effects the conditions e.g. after the screw has heated up the heater bands supply less than a half of the heat for the process and the shear produces the rest. Pressures and times are interelated and one works by pressure and the time is a result or the opposite.

In my opinion the calibration is unnecesasary as it does not provide any useful information.
SPC on the process is also a problem with the introduction of closed loop machines the parameters of pressure and time are altered constantly to ensure conformity.
The best method of control IMHO is monitoring of the overall cycle time and the plasticiation time (the time it takes to fill the screw, this shows the temperature, viscosity of the material and thus the uniformity of the plastic) as they should give the best picture of the process.

 

[satsoe]

I got a generic question on setting up tolerances for pressure transducer calibration. Right now we go with the rated manufacturer tolerances. I was told by my peers that for our process performance our gates could be opened up little more. What shall be the best way to accomplish this statistically?.

 

[gszekely]

Hello Bubba
See this thread as well:
http://elsmar.com/Forums/showthread.php?t=14619
and the last post, where I have added several references, sites you can find more about IM:
http://elsmar.com/Forums/showpost.php?p=198126&postcount=32
In my opinion we are talking about 2 different things:
1) Injection Molding Machine reliability, reapetability, stability, in case of temperatures, pressures, strokes, times etc
for this check out the attacment, and see these threads on IMM forum site:
http://www.immnet.com/cgi-bin/dcfor...w_thread&om=1107&forum=DCForumID1&archive=yes
http://www.immnet.com/cgi-bin/dcfor...w_thread&om=2770&forum=DCForumID1&archive=yes
http://www.immnet.com/cgi-bin/dcfor...w_thread&om=3246&forum=DCForumID1&archive=yes
2) all the above in case of a master set up for a specific product
See this attachment I have posted in another thread for start up qualifications, decupled or scientific molding:
http://elsmar.com/Forums/attachment.php?attachmentid=6491&d=1172333050
Hope this helps, if you have further questions let see if we can help.
Good luck
BR
György

 

[gordman]

Thank you for the links they are really usefull.

 

[Jeff Frost]

Gordman,

When calibrating any system like a ram press or injection molder you must remember that it is a system and not just individual components and the system must be calibrated per the recommendations of the manufacture.

First starts by contacting the manufacture of the injection molder and ask them what must be calibrated, how it must be done, and who they recommend in your area. Make sure that the calibration lab you use has the capability to calibrate the molder per the manufactures recommendations.