Validation of moulding maschine

how to validate a moulding maschine ( moulding process ) which parameters of moulding process are the most important for validation.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mar 21, 7:19 am, snipped-for-privacy@wp.pl wrote:

Before I answer your question, I need to make sure what you mean by "validate". "Validation" is a term that is often used in ISO qualification, and similarly in GMP/GLP for medical devices in the United States. You can validate a process, but only for a specific product.
Maybe you mean "validate" in a broader sense. (Don't worry about the language barrier. Your English is fine, certainly better than my Polish. "Jak sie macz" is all the more Polish my grandmother was ever able to get into my thick skull. It's just that "validate" is a word with lots meanings, some of which are very specific.) In general terms, it is my opinion that the process temperatures are probably the most important variables , specifically the melt temperature and mold temperature. Molding pressure and cycle time would be a close second. Certainly the consistancy of the resin is critical, as variations in it can easily overwhelm all the process conditions, which again means that it is best to focus on the properties of the product and understand how the process variables alter those properties.
John Aspen Research - www.aspenresearch.com "Turning Questions into Answers"
Opinions expressed herein are my own and may not represent those of my employer.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 21 Mar, 14:44, " snipped-for-privacy@aspenresearch.com"

Hallo Nice to meet You I don't even suspected that someone will answer me. You have asked me what I mean, so in plant that I'am working for ( Medical devices ) we have a moulding maschines made by Arburg. I have to do a validation for a Moulding process. We have many parameters that are important for moulding process ( tight pressure, Molding pressure , Cooling time , "heating cylinder.zone" , moulding speed. etc.). Bright now I need to translate Parameter Card from Germany to Englich . Tomorrow I'll send you more about this parameters and I'll explain You what I have done and what I want to do.
Best Regards Kris
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@wp.pl wrote:

Kris,
Validation of a process depends mostly on your own internal procedures, the parts you are making and their risk indices.
As a general example (at least, the process I was accustomed to, hopefully, your own internal procedures will not be so demanding):
First was a review the final product design FMEA. If it is done correctly, you should be able to find the component(s) in the document that you are responsible for to see if it has a high RPN value for any potential failure modes (RPN=probability of occurrence x level severity x probability of detect-ability). There are certain break points of RPN values that will dictate the risk index level. There are usually (depending on internal procedures) risk index levels of 0 (lowest), 1,2,3 and 4. The higher the number, the higher the risk to the patient (the highest equates to catastrophic failure, or death).
Some procedures dictate that a risk index of 4 (or the highest level) in the dFMEA is unacceptable, and would force a redesign of the product. As you would guess, the FMEA meetings can get quite argumentative when deciding on rankings for RPN values. (If your components carry the extremely values, be careful. Push back on the design team if you feel there would be no way to validate to this type of critical level. Be honest and firm.)
I would then need to establish a pFMEA (process FMEA) that would outline the vital portions of the process and the potential failure modes (making sure you cover any possible modes in the dFMEA as well) along with the associated RPN values and risk indices, same as the dFMEA.
Most of these failure modes would carry the same RI as the dFMEA, unless there was a specific reason to make some failure modes higher (not covered in the dFMEA). (This would automatically cause an up revision to the dFMEA, not something project teams like to do, so be sure that there is good representation at the design level for what ever you are responsible for. Plastics guys always get the blame...!)
For those component items that are risk index 0 in the design FMEA, you may chose not to validate, or you may validate based on business reasons (some procedures may automatically upgrade any "0" to a 1, making all processes go through some level of validation).
The risk indices will determine the level of validation needed, i.e., sampling plan and acceptability standards for OQ/PQ portions of the validation (assuming you already have in place the proper IQ for the equipment. Some may also OQ the equipment). Usual standards may be CpK or PpK minimum values. Your own internal procedures will dictate how many OQ and PQ lots you will need to make (usually 1 High OQ, 1 Low OQ, 3 PQ nominal lots).
The MVP (master validation plan) would spell out all of this out up front, and would list the PQ (nominal) conditions along with the OQ (challenge limits) conditions of those parameters you have stipulated that you will need to control (and rationalization for those that you won't control) and reference any IQ documents for the equipment.
Then, go back and update the pFMEA if there are any provisions for validation reducing potential risks, and setup a process specification document.
Anyway, that's kind of an abbreviated version of a process validation I used to go through, which could include a lot more documents than this (design drawings and specifications, software validations, etc).
I have gone through the process many times. As always, one needs to ensure they follow their own internal procedures. You want to cover all your bases, but as always, try to keep it realistic.....I actually teamed up with some former colleagues, and we decided to draw up a general extrusion pFMEA, purposely being as anal as possible about it. We finally gave up 2 weeks about 300 pages later. It would have taken months to validate a single part using that approach.
I have completed validations that have taken up to 2 - 6" binders to hold all of the paperwork for less than a dozen extruded parts for neurovascular access therapeutic catheters, about 30 pounds of paper (back in the days before electronic documentation systems) that took over 6 months to complete.
Just try walking that around for signatures........one arm gets longer than the other ;-)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Halo
In my job we don't use any FMEA tools , so we have to do it in another way. We have a parameter sheet for our moulding machines. We want to choose the most important moulding parameters of them, I mean parameters, which have the impact on moulding process. ( parameters we have : cylinder heating zone ( our moulding machine consists 4 heating zones ) , closing speed of the form, mold safeguard pressure, injection speed , screw speed , injection pressure , holding pressure , injection time , holding pressure time , cooling-off time , injection delay ) we think that the most important are : cylinder heating zone number 4 ,injection speed , injection pressure and injection time.
I will try to describe our problem. We want to check, how big impact this 4 parameters listed above have for moulding process. If you have any suggestions, that we suppose to test another parameters please give us advice. We have two ways to do it.
1.When we verify 4 parameters in the same time ( I mean looking for upper and lower tolerances ) the found tolerances are very narrow because all this 4 parameters together posses big influence for moulding process. The another not-validated parameters during this test have nominal value.
2.When we test this 4 parameters each separately the found tolerances are wider because one parameter have smaller influence. All of the rest ( of course not validated and 3 validated have nominal value ). And we must to do it with each from 4 validated parameters.
Please advice us, is this second way enough for validation process.
Thanks in advance and best regards
Kris
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Halo
In my job we don't use any FMEA tools , so we have to do it in another way. We have a parameter sheet for our moulding machines. We want to choose the most important moulding parameters of them, I mean parameters, which have the impact on moulding process. ( parameters we have : cylinder heating zone ( our moulding machine consists 4 heating zones ) , closing speed of the form, mold safeguard pressure, injection speed , screw speed , injection pressure , holding pressure , injection time , holding pressure time , cooling-off time , injection delay ) we think that the most important are : cylinder heating zone number 4 ,injection speed , injection pressure and injection time.
I will try to describe our problem. We want to check, how big impact this 4 parameters listed above have for moulding process. If you have any suggestions, that we suppose to test another parameters please give us advice. We have two ways to do it.
1.When we verify 4 parameters in the same time ( I mean looking for upper and lower tolerances ) the found tolerances are very narrow because all this 4 parameters together posses big influence for moulding process. The another not-validated parameters during this test have nominal value.
2.When we test this 4 parameters each separately the found tolerances are wider because one parameter have smaller influence. All of the rest ( of course not validated and 3 validated have nominal value ). And we must to do it with each from 4 validated parameters.
Please advice us, is this second way enough for validation process.
Thanks in advance and best regards
Kris ---------------------------------
Though I don't have a lot of experience with molding, I would say that with only 4 critical inputs, you could probably conduct a full factorial DOE (condition changes) to test all of the interactions. This should give you enough data to support your validation.
There are many excel templates available to set this up. Here is just one of them..
http://www.qimacros.com/free-lean-six-sigma-tips/design-of-experiments.html
You could also download a full working version of Minitab (for 30 days) which can set up your DOE and analyze the data for you.
http://minitab.com /
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Thanks a lot for advice Greetings from Poland Kriss
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

No problem, Kriss, and greetings from the US of A to you. Good luck.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.