Welcome to the NEW Infrared Training Forum. To view the old message board, please click here.World Class Infrared Training for Thermography Professionals


Post Reply 
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Transmittance in molded plastic parts
02-02-2014, 05:48 PM
Post: #1
Transmittance in molded plastic parts
Hello everyone,

I have been working on bringing thermal imaging into our injection molding facility. It's been on my project list for a long time but has either been cost prohibitive or I was not in an appropriate role for developing the system. All things have aligned and I have begun the process of getting key team members certified and collecting baselines for comparison later on. It's become a truly large task because everywhere we look we are seeing different potential applications. We have narrowed our focus at this time to PdM and in process monitoring of the molded parts for defects which will eventually be used towards the goal of parametric release of final goods.

We currently use a large amount of HDPE in our mix of parts that we mold so it seemed like a good place to start in identifying potential areas for improvement with infrared imaging. It didn't take long to notice that the HDPE was transmitting a lot more IR than we had expected. This problem was compounded by its level of transmittance varying during each stage of cooling as the crystallinity of the polymer increased. It has been suggested that a filter specifically designed for HDPE would be the best route and is something that I am looking into but I wanted to check here for advice on alternative ways of combating transmittance. For instance, our parts are all removed from the mold using robots and could be placed anywhere for the IR camera to image. This could be on a surface with a known temp that could be accounted for in the software. This is just one idea I have come up with but I'm sure there must be others.

Purchasing a filter for HDPE could be justified due to our large volume of parts using this polymer but on smaller, low volume jobs running a variety of different engineering grade polymers it would not be practical to purchase a lens for each one. What would the best approach be when inspecting such a wide assortment of polymers?

We had been initially using a handheld model from Testo but recently purchased a Flir A315 for process improvement R&D. I was at the plant working on setting up the external camera trigger on one of our machines and grabbed a screen shot of a few of our parts that show the problem. On the long vertical portion the wall thickness is so low that the plastic is almost invisible compared to the steel temp inside of it and only on the really thick handle portion are we able to get a "good" reading.

   

Any advice would be appreciated. Thanks.
Find all posts by this user
Quote this message in a reply
02-03-2014, 05:22 PM
Post: #2
RE: Transmittance in molded plastic parts
Can you provide a visible image so we can identify what we are looking at? It would also be helpful if you could point out the areas you mentioned, the steel and the almost transparent areas, and the thick handle. Thanks
Find all posts by this user
Quote this message in a reply
02-04-2014, 11:14 AM
Post: #3
RE: Transmittance in molded plastic parts
Yeah sure, sorry.

   

In the image the red arrow is pointing towards the thin wall section. Its actually the entire vertical length of that section and the blue arrow is pointing to a "thumb tab" type area that has a very thick wall and transmittance isn't as much of an issue.

I had a good thermal image of the thin wall section with my fingers behind it and you could see the outline of my fingers through the part to show the issue but I am unable to locate it at this moment.

Eventually, when I have enough data to start the build I will not be imaging inside of the machine with all of it's IR unfriendly elements of low emissivity metals and reflections from hundreds of different heat sources. I will have the part removal robot place the parts in an area where i can control the environment. Inside of this environment I can design for issues like emissivity and reflections but the transmittance is a harder issue. I had also originally envisioned a system where the accuracy of the temp reading was important but not necessarily critical. The planned system was looking for temp differences and overall trends that could identify to a technician that a problem was forming. The end users have asked for a program that could be used on several different machines. Being able to match actual temps becomes much more important when this functionality is included.

I've reached out to Flir to get a quote on a filter for HDPE but as I move on I expect to encounter many different polymers with varying infrared properties and having a game plan in place for how to deal with this during the R&D phase would be helpful.
Find all posts by this user
Quote this message in a reply
02-04-2014, 12:48 PM (This post was last modified: 02-04-2014 12:49 PM by Bobberry.)
Post: #4
RE: Transmittance in molded plastic parts
I don’t see any other option other than using a camera with the correct filter fitted, particularly if you need measurement. This may mean, that for different polymers, you would need different filters. There may be some common properties that would allow one filter to be used on a number of polymers.

You have a semi transparent material, and a component in a transient state, this is almost impossible to deal with as a combination. I assume from your post that you are looking at components that are cooling, this means that the inside will be warmer than the surface, your camera is seeing radiation emitted from the surface of the object, and some energy transmitted from the hotter inside, and some energy transmitted through the component. There is no way to separate these out, and to measure the surface temperature you would have to be able to separate them.

Also on the thicker part, you still have emitted energy from the surface and energy transmitted through from the hot inside. I think that true accurate temperature measurement is not possible without a filter. However without the measurement, you might be still able to figure out what is happening, but it may take a bit of playing around. You have a real challenge, I love stuff like this.

BTW, what camera are you using?
Find all posts by this user
Quote this message in a reply
02-04-2014, 02:14 PM
Post: #5
RE: Transmittance in molded plastic parts
I am using a Flir A315 for the R&D of potential applications. Once I overcome some of these hurdles I may be able to spec a lower cost camera that has just the functionality that is required for the application.

Bobberry, I definitely love challenges like this. Its been great finally getting the opportunity to work with thermal imaging. Until recently my exposure had been limited to a few handheld units that were so expensive that the Plant Managers would keep them in their offices under lock and key. They never got broken but unfortunately never got used either.

Yesterday we had MoviTherm onsite to conduct training on their IRControl software. I set up the software to record still frame images on each cycle of the molding machine with AOI's placed across the same four cavities as in my pictures above. I then ran the report and when the results were graphed it looked like this:

   

Notice the cyclic wave pattern that spans 30 shots. It's a gradual rise in temp followed by a sudden drop. The range was only 3 or 4 degrees but it was repeating and consistent so it stuck out. After investigating it today I determined that it was our closed loop water control unit cycling to maintain it's temp set point. The long thin walled sections of the parts are cooled by metal cores that each have tiny water pipes in them called bubblers. Since the pipes are so small they easily get clogged from scale buildup and cause downtime to disassemble and remove the blockage. If you notice the cavity represented in purple on the graph does not follow the wave pattern of the others. I suspect that this cavity has restricted flow on its water circuit and plan to set up and investigate later this afternoon.

I'm sharing this because the amount of things that can be monitored using infrared imaging in my industry is huge and even when you are least expecting it, something like this shows itself. The ability to see the flow problems before they affect part quality is a huge benefit and it was discovered while training on the software. (I still need more data to confirm this but the frequency matches up perfectly with the timing of the water control units venting cycle)

Great stuff!
Find all posts by this user
Quote this message in a reply
02-05-2014, 02:09 PM
Post: #6
RE: Transmittance in molded plastic parts
Hi
A315 is a LW model

HDPE presents a strong absorption in MW between 2800-3000 cm-1, or around 3.5 µm. Your camera is not sensitive in that range
There is another absorption band in LW before 14 µm, but the response of the A315 at this wavelength is very limited.


I would suggest to switch from LW to MW.

Raphael.
FLIR ATS.
Find all posts by this user
Quote this message in a reply
02-16-2014, 04:56 PM
Post: #7
RE: Transmittance in molded plastic parts
Great info RDanjoux. Thanks.

I was trying to make some sense of the HDPE graphs detailing the absorption rates but unfortunately another major project at work has shown up and I've had to put this one on hold for a little while. Our PM scheduling program at work was found to vulnerable to several other depts. reporting inputs and as a result I've had to focus on completing baseline inspections on all our machines. This presents another great IR application but that's for another thread.

What effect would a filter specific to HDPE have with the A315? I've been told of an application with plastic film manufacturing where the plastic was almost invisible to the IR camera due to it's transmissivity but when a filter was used the camera could image the film. I'm asking this question because I'm not entirely sure I am grasping what a filter would be filtering out of an image to correct for transmissive materials.
Find all posts by this user
Quote this message in a reply
02-17-2014, 01:51 PM
Post: #8
RE: Transmittance in molded plastic parts
(02-16-2014 04:56 PM)ArikB Wrote:  Great info RDanjoux. Thanks.

I was trying to make some sense of the HDPE graphs detailing the absorption rates but unfortunately another major project at work has shown up and I've had to put this one on hold for a little while. Our PM scheduling program at work was found to vulnerable to several other depts. reporting inputs and as a result I've had to focus on completing baseline inspections on all our machines. This presents another great IR application but that's for another thread.

What effect would a filter specific to HDPE have with the A315? I've been told of an application with plastic film manufacturing where the plastic was almost invisible to the IR camera due to it's transmissivity but when a filter was used the camera could image the film. I'm asking this question because I'm not entirely sure I am grasping what a filter would be filtering out of an image to correct for transmissive materials.


Hi ArikB,

Raphael's post really says that the A315 is not the most appropriate camera to use. The A315 is a LW camera, you will get better results with a MW.

When you use the correct filter, you a would be filtering out the part of the spectrum where the material is transmissive, and the camera will only see the part of the spectrum where the absorption is high, therefore the transmission will be low and the emissivity will be high.

I have to say, that what you are trying to do is difficult, and I would recommend that you do a level 2 course if you have not already done so.
Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)
Copyright © 1999-2012 FLIR Systems, Inc.