Regrind Use ?

[TC]

How about liability? I remember the sad story told by an automotive molder at a seminar I attended who said he spent 3 days on the witness stand being hounded about regrind vis-a-vis part failure:(Though I think he was absolved,he told the seminar that he would not have regrind in his plants because testifying about it was the worst week of his life.

[Bader]

Louis


This is a dynamite topic!


Mitesh, i talked about the cascading method of using re-grind in contrast to RAP (Regrind as produced).

Will discuss it later this evening.

In haste,



Bader




[email protected]
http://www.mediplas.com

[RD]

Ok, My thoughts on regind. I don't see any problems running it back as long as long if it is a percentage of 25% or less . Yes I have used up to and including 100% on certain jobs with and without problems. ( depending on the parts) We have a couple of customers that will only let us use 10% of fisrt generation regrind only, stated on there part print. We also run some reprocessed material that is a pain in the a** because of the lot to lot differences
. We do sell some of our runners , sprews, rejects, it is usally up to the buyer if they want it ground or unground. Here is my complaint if you will being a processer, The engineers build the tools with the shortest sprew possible to save on waste and I end up using a 10 inch nozzle. I try to tell them that the shorter the nozzle I can use the better so that the material doesn't degrade. It is kind of A catch 22. They are doing what they are told but it sometimes does not work. I try to tell them that the thermocouple is at the end of the nozzle and to keep it from freezing it has to run hotter at the tip witch causes the degrading. Maybe we need to covert our presses to read the nozzle tip also.


Rick

[Clive A Maier]

Rick,

I agree. I don't care for very long nozzles either. Much better to use a hot bush in the mold once the nozzle extension gets beyond maybe 100mm.

Regards,
Clive A Maier
Econology Ltd

[M&M]

What the differance between the long nozzle and the hot spure? Seems as thought they do the same thing as far as how the resin is treated right. They are both hot and an added area post screwtip right? Hot spures produce less regrind but isn't that about it?

regards

M&M

[Clive A Maier]

The principle is the same but most long extension nozzles are home-built one-offs whereas hot bushes are produced in volume. There is a lot more development and engineering invested in the latter and it shows. Long nozzles flap about in the breeze too, and it is hard to check that they are seating properly at all times. If they don't, there is a hell of a mess and probably a need for another new set of heater bands and a thermocouple. So I prefer a hot bush in most cases.

Anyone remember nozzles with undercuts for pulling out reverse taper sprues?


Regards,
Clive A Maier
Econology Ltd

[j]

long nozzles also use more area around the C/l in the mold design because of their diameters but in the right situations (many different molds with similar injection strategies) save money over using a hot sprue in each of the molds.
I have an a friend whose father was designing a multiple nozzle system years ago for Impco as an injection strategy for large or multiple parts at the same time Hot Manifolds were first being evolved in mold constuction. The evolution resulted in mold hot sides.

Another solution was also P/L injection, which i still find useful when confronted with a large part that needs to be gated on a side in a one cavity situation where putting it off center in a tool requires a much larger molding machine.
I've even used a hot manifold sideways in a mold to accomplish this when a P/L machine was not available.

j

[RD]

The biggest differance would be that you have better control of the temp in the hot sprew.
Long nozzles on some jobs are ok like with a PP. The one's I don't like are with unfilled nylons & PC's

Rick

[Bader]

Hello All

Using regrind in any ratio affects two things, Cost and Quality.


I agree to Tylers approach that if use of 100% regrind produces parts between your AQL's (Acceptable Quality Levels), feel free to use any percentage of regrind.

Obviously, we cannot consume more regrind than is being produced but on many occasions there are two or more parts being made in the same material. This gives use the opportunity of making Part A in 100 virgin while part B is made of 100% regrind!

CASCADING METHOD:

Generally, RAP (Regrind As Produced) method is used for regrinding. The sprue is collected for the shift and handed over to the granulator where it is crushed and mixed with virgin material in a pre-set ratio and brought back to the machine for processing.

On some occasions, the process is automated. Conveyers are used to crush the waste throgh on-line slow speed granulators. Prorportional valve or auger feeds are used to supply virgin and regrind in a pre-fixed ration to the hopper.

If you are using RAP on continous basis, there will ALWAYS be material from the first generation, no matter how small in percentage but.....it will be there!

In contrast to this methodoligy of regind (RAP), i suggest using CASCADE method of regrind and it is best explained by an example:

Shot weight : 100 gms
Net wt of part : 75 gms
Batch size: 500,000 parts

Using RAP, the parts from the first lot will be made from virgin material. Parts from secon lot will be made of 75% virgin and 25% regrind. Third lot : 75% virgin - 18% from 1st gen and 7% from second gen (approx) ....and so on!

Using Cascading method, the first lot of (say) 300,000 parts is made from 100% virgin mateiral. This will produce a regrind of 7,500 kg.
Run this 7,500 kg (100% regrind) to produce 100,000 parts and 2,500 Kg of second regrind.
Run this 2,500 Kg (100% second regrind) to produce 33,333 parts and 833 kg of third regrind.
Run 833 kg to produce 11,106 parts and 277 kg of fourth regrind.
Sell this 277 kg of regrind and start all over again!

The benefit of this method is that you always have the regrind history of the product and you exactly know the what percentage of regrind is there since there is 100% of only ONE generation.
It is cost effective and easy to manage.

Ofcourse there are certain do's and dont's with each method and becuase the variables are so many, we cannot safely say that one method is the best one!


Bader
[email protected]
http://www.mediplas.com

[Aahz]

Bader,

Are you saying to run 100% regrind and 100% virgin in the same product? Doesnt this reek havoc on the process? In most cases buddy if I have very much variation in the process it kicks off a whole string of costly events to assure product quality. Whereas if I stay at a given regrind level the process is set accordingly and should not vary run to run.


Testing 100% regrind parts for validation purposes is typically a long and expensive process the customer would rather not absorb. I as a custom molder cannot absorb the cost either. Therefore we have determined (Automotive through appliance and comsumer products) that a stable process of a given regrind precentage is a more cost effective and a higher quality method.

I would have to say that RAP has proven to be very effective. Nothing is the "Best" method for all situations there is no panacea for regrind handling methods.

Whatever works best for you is the best thing to do..


GBYA

Aahz

<i> "May ye live as long as Ye want and never want as long as ye live" Old Irish toast </i>

[Louis]

Originally posted by: Aahz
Bader,

Are you saying to run 100% regrind and 100% virgin in the same product? Doesnt this reek havoc on the process? In most cases buddy if I have very much variation in the process it kicks off a whole string of costly events to assure product quality. Whereas if I stay at a given regrind level the process is set accordingly and should not vary run to run.


Testing 100% regrind parts for validation purposes is typically a long and expensive process the customer would rather not absorb. I as a custom molder cannot absorb the cost either. Therefore we have determined (Automotive through appliance and comsumer products) that a stable process of a given regrind precentage is a more cost effective and a higher quality method.

I would have to say that RAP has proven to be very effective. Nothing is the "Best" method for all situations there is no panacea for regrind handling methods.

Whatever works best for you is the best thing to do..

GBYA

Aahz

Aahz

The cost to qualify a part made with 25% or 100% should be the same - no?

As for process changes, I suppose there may be some, but then the qualification process would include the determination that there was indeed a process - and hopefully someone recorded it. So, a slight change in parameters may need to take place for each level or pass that Bader spoke about.

I kind of like the idea.

Of course beside the press granulators with a closed loop system would be the way to go, but that only works well when the volumes are high.

I'm still learning - there has been much food for thought.

<b> <i> Louis </i> </b>
Silly Poly Questions <i> Moderator</i>
<a href="mailto:[email protected]" target="_new">[email protected]</a>

To see a world in a grain of sand
And heaven in a wild flower,
Hold infinity in the palm of your hand
And eternity in an hour.
-- William Blake

[Aahz]

Louis no actually you would have to test the product twice using the aforementioned process whereas if you use 25% all the time then you only test it once..

If you run Virgin then regrind both must be tested or does one just assume that if regrind works virgin will also? Though a logical assumption not one a product engineer is likely to make. Some regrind is always generated in the sampling/debugging phase of a tool thus when parts are finally submitted to the customer for testing they represent production intent conditions. Many times adjustments will be made after prototypes are made and these adjustments often must endure the same rigorous testing the proto pcs did.

SO the cost is twice what it should be. Small adjustments to the machine parameters are exactly what we as custom molders are trying to eliminate. Small adjustments for regrind are NOT repeatable....just ask M&M he deals with it every day. The size and shape of the granulated pellet varies too much to provide a stable repeatable process.

With QS900 every process adjustment is has to be logged. Multiple processes for the same machine and same part would be very difficult for the auditor to understand. I can hear him asking now..." What happens if someone gets the wrong set-up sheet?" The way it is now there is only one per part number for each machine it can run in. SO all they need to do is match up the machine number to the part number. Add in three or four more varations and trouble lurks on the horizon. Not to mention the Doc/data control issues with having up to a dozen different process sheets for the same part number.

No sir for injection molding I do believe the 25% RAp is about the best we have come up with so far....to each his own however.. Just want to make sure you are getting a clear picture ..


GBYA

Aahz

<i> "May ye live as long as Ye want and never want as long as ye live" Old Irish toast </i>

[This message has been edited by Aahz (Edited: 08/20/01).]

[Bader]

Aahz (Sir)

Testing 100% regrind is not at all a costly process. It is as expensive a boiling two eggs instead of one in the same pan!

Just like you would test a part processed at two different heats or in two different materials...you will test a part produced from regrind.....it is simple!

I ask you why 25% ....why not 27% or 22% ?
It have been hearing of numbers associated with regrind and it does not make any sense or logic to me.

Regarding the mistake of machine setting, well Aaahz...this mistake can be made anywhere! Nowadays the machines come where you can save the machine settings and recall again.

I have tried both methods and i can safely say that cascading method of using regrind is best if it does not affect your product.

You have tried only method (RAP) and this explains you inclination towards RAP. I am sure if you give a serious thought to this, you might would like to make the change...!!!


Bader
[email protected]
http://www.mediplas.com

[Aahz]

Sorry Bader ol buddy we will just have to agree to disagree on this one..

Gave it plenty of thought last time and still came to the same conclusion... The way we do it now is the most cost effective repeatable method discovered yet.. For us anyway.

No Bader it is not as simple as boiling two eggs instead of one. When an automaker tests a part for evnirnmental/dirability testing they do it in acceleration tanks usually. These tanks run through cycles of hot/cold/wet/dry and combinations of all of the above. While the part is under load or functioning. Now imagine the jigs and fixtures that are required to do all this testing. I have had to sit on jobs for months while they were in these tests. When the customer asks for 30 samples he intends them to go to specific places for specific tests. They will only test X number of parts in each "tank". If you send him regrind and virgin parts now he has to track them both seperately $$$ he has to lengthen the testing time by double $$$$ and he has to do twice the paper work when finished $$$$$$. This is one of probably a dozen reasons Bader why the cascading method of regrind usage would not be accepted by my customer base..


As I said before buddy....if it works for you great....personally I think you are throwing away good money and material. That is just my opinion though..


BTW 25% only because someone somewhere decided that it was perfectly acceptable for most resins and almost all resin manufacturers agree...

It is also easy to calculate for the material handlers....1 bucket regrind three buckets virgin...LOL


As for machine settings ..Bader My point is that you cannot consistantly reproduce regrind. It will vary a good deal from lot to lot. Therefore even the "saved" info will be useless because this is a different lot of regrind then when it was saved before. Also the paper trail alone for all those different processes would be difficult to control. No sir I think we are better off with one process per machine and a consistant batch everytime.

The natiral inconsistancy of the regrind is deminished when using 25%. It is generally pretty easy to lock down a process under these conditions....change the precentage and things begin to go awry sometimes.

Keep in mind Bader we go through a couple million pounds of material a year. It is definately in our best interest to find the most cost effective method of use..


Welcome back buddy we missed ya..



GBYA

Aahz

<i> "May ye live as long as Ye want and never want as long as ye live" Old Irish toast </i>

[Bader]

Hi Aahz...

I knew one day we would agree to disagree on somehting...and here is comes!

At the end of the day it boils down to the $'s....
How much you save through RAP and how much you save through Cascade method of regrind (keeping the quality aspect within tolerances)

I have always carried out accelerated stabilty test on parts produced under more than one variable. This is a normal practice...you save a lot of man hours.
Imagine the man hours to check 50 parts from lot A and 50 of lot B at different times and the cost to check them both at the same time.
I cannot understand how can the second cost be double the first one!!!

To conclude:

The regrind is either sold or re-used.
Either way, it incurs a loss or an expense.

There is not doubt that the expense to regrind is much less than the loss to sell at throw-away price.
Question is, what method to adopt?

I used to apply RAP method. I still use it with some of my parts but with a closed loop recycling system. where the waste is recycled within 15 minutes after it is produced. It works well!
For certain products (that are large in volume) i find Cascade method to be very cost affective.


Hope to hear others views on this


Bader
[email protected]
http://www.mediplas.com