Can you solve this plastic puzzle?

[Farnworth]

I have a part made of two mirrored pieces that are bolted together with two screws. The part is injected molded with 30% glass reinforced PA6.

History: I switched from a European manufacturer to a Taiwanese based one last year and have been having problems ever since. We use the same exact mold as before, however I believe the process has been changed somehow.

Problem: The new Taiwanese parts are weaker than the older European parts. If you over tighten the screws, you can strip the plastic out now, whereas it wasn't possible before. So you end up with a nut spinning freely inside the plastic part. Or it will simply compress the plastic until the nut goes through the plastic.

I tried switching the material supplier, the manufacturer was using a Taiwanese based supplier that I suspected didn't have the proper specifications for the material, so I switched to Dupont Zytel 73G30L NC010 (a much better material, equal to what we had with the old European parts) but it has only made the parts slightly better.

Here is a comparison of the two parts.
Bad part (Taiwanese part): Even after new material supplier, it still has these characteristics compared to the older part (good part). The surface is shinier and there are strange marks all over the surface, it's the only way I manage to find the different between the two when I have them in my hands.
Good part (European part): It has a smooth surface with a dull texture. It's a much stronger material.



What could cause these characteristics on the Bad part? Could this be related to it being weaker in strength to the Good part?

Thank you! I've been having many headaches over this one.

Alex

[rickbatey]

I think that the glass filler is too low on the current resins compared to the original resin. Glass resin levels can vary slightly and for some structural parts, a 3-5% drop in glass can cause bad parts to come out of a formerly good process. Also your melt index may be too high on the replacement resins versus the OE resin. Remember that although a resin should be a 1 to 1 replacement; they RARELY are without some serious process adjustments. There may also be some other fillers or processing aids in the OE resin that make your parts exactly as you need them to be.
Rick.

[brentb]

What is the crystallinity of the good vs bad part. Have them tested. Was mold run too cold on bad?

KOM
brent

[Farnworth]

What is the crystallinity of the good vs bad part. Have them tested. Was mold run too cold on bad?

KOM
brent

Thanks for the quick responses.
I suspect that the material is not the problem, as both materials (the taiwanese brand and the new dupont brand) came out with the same qualities in the plastic part. I could be wrong though.

Brent I think that may be a culprit, I'd like to have them tested. What should I ask to be tested exactly?
Our Taiwanese manufacturer clearly does not do any testing, they send me the parts without any additional information on whether they're as good as the originals, even though that's what I've been asking for.

[Suhas]

Hi,
1. Check to see if there is any significant part weight difference. If the Taiwanese parts are lighter, they are not packed out as much as they need and may suffer from failure.
2. Do a quick 4 corner study with Hold pressure and Mold temp and see if that helps. I am assuming that the gate seal study is done and that the time is optimized.
Let us know.
Suhas

[Farnworth]

Hi,
1. Check to see if there is any significant part weight difference. If the Taiwanese parts are lighter, they are not packed out as much as they need and may suffer from failure.
2. Do a quick 4 corner study with Hold pressure and Mold temp and see if that helps. I am assuming that the gate seal study is done and that the time is optimized.
Let us know.
Suhas

Hi Suhas,

I never assume stuff about Taiwanese suppliers, they will lie to save face hahah, so it's a game of cat and mouse trying to figure out problems.
I had a coworker in Holland test the parts with our old manufacturer. Here are his results:

"1. it's made out of the same material (nylon with 30% glass fibre).
2. we're using the good stuff (both Dupont and Akulon are very good).
3. it looks the same
4. it has the same weight
5. it doesn't have the same size! actually the diameter is 0.3mm smaller, which is very significant for our clamp!

The most likely conclusion is the following:
the raw material: granulate with glass fiber, is 4mm long, with 4mm glass fiber. When it's processed and melted to be inserted in the mold, the fiber breaks in half; 2mm length. Waste from production gets recycled. then the raw material is 2mm long. when it is used again for production it gets halved again; 1mm. With 1mm short glass fiber it can still be 30%, however, it does not have the same strength. it can also cause shrinkage.

So it could be that our suppliers are saying they're using dupont or akulon, but it might well be that they're lying. It could also be that they're using dupont or akulon, but they're supplying us waste, so they're cheating.

all other stuff is less likely. think of conditions (cooling, temperature of oil and water (40 and 50 degrees) or production time (40s)
"

I think this could also be a likely reason (broken fibers), as I have read about this type of problem as well.
I will be doing another set of tests with a different company today, I will update with results for those who are interested.

[brentb]

DSC is good from my experience. This is Differential Scanning Calorimetry, a thermal analysis. Suhas can expound more on this!
Most resin companies will do this for a molder for free! It can predict how a part may function in the field, depending on end use and requirements.

KOM

brent

[Suhas]

Hi,
1. Brent, you are correct with the DSC and it may give you some info. If the original parts have more crystallinity, then yes, the mold temps are probably too low. I do see some some flash on the parting lines and they may be reducing the mold temp to control the flash - not good anytime.
2. I think Alex may have something with the glass fibers. Have the glass fiber lengths checked on the parts. It may be a simple solution of reducing the back pressure and the screw speeds!
Keep us informed.
Suhas

[Farnworth]

Hi Suhas,

I just had a talk with a materials engineer from Miniwiz, they suggested the same thing about the flash on the parts, so I will definitely look into that.
They also thing the the marks are directly from the mold, meaning it might be damaged in some way (possibly corrosion). I'll be checking all these details once I visit the factory and fill you in on the little problems.

Thanks!

Alexandre

[Moldrite]

Alexandre,
Please post the original setup sheet if you can get it.

[Farnworth]

I visited the factory today and have some more information. I was not able to find any indication of problems.
They have a very nice factory, I believe it is up to spec and they take care into making their products. At this point, I think it may be a problem with injection parameters and machine.
They use an old injection machine (1986) from a Taiwanese based manufacturer. Their equipment seems to be well maintained.

-Cycle time (total production time per piece): 60 seconds (65 seconds for Dupont test)

-Machine size: 180 tons

-Raw PA6 + 30%gf drying time: 4 hours@100-150 celsius

-Colouring agent: Mass produced model is precoloured from supplier. Dupont test is not precoloured (colouring agent added at factory)

-After cooling: No water cooling is done after parts are complete (ambient air cooled)

Injection temperatures:​
Step 1 – 260 C
Step 2 – 250 C
Step 3 – 240 C
Step 4 – 230 C
Cooling – 23 C

Steps 1 to 4:
Injection pressure – 58kg/cm²
Speed – 35mm/sec
Time – 5 sec

-No material is being recycled
-No material is being mixed
-The mold seems to be in good shape. However, I noticed there is an indication of overheating on some of the ribbed sections. It is directly around the area where we have problems, where the nut goes. It's the only place where I saw any indication of overheating on the mold. I'm not sure if this is normal or not.

[rickbatey]

You could have an issue with the way the mold is watered: when correct parts are good and when wrong, bad parts. You need to ask for water diagram to confirm number of circuits. They could even have an issue with the injection unit; namely the NRV is leaking causing excessive melt temp and damage to the polymer. It might be from colorant package in the resin. Oftentimes they will use PE based colorant in PA parts. DSC should show the presence of PE. Have a burn test performed to confirm actual glass content.
Rick.

[Ten Fingers]

23C on the mold cooling seems really low - shouldn't it be closer to 80C or higher?

[Farnworth]

I thought that as well, but I'm still waiting for results from our original manufacturer.
The temperature is actually around 18C to 23C, at least that's what he told me. Unless he thinks I'm talking about something else. Isn't that very cold?

[rickbatey]

I agree that if that's the water temp, it's much too cold! The 80* C sounds much closer to the correct range and may even prove to be too cold based on the part geometry and Crystallinity required. I bet they dropped the water temp trying to reduce the cycle time.
Rick.

[nemanja]

Hi,

Is also posible, if you tell them that you will come to wisit company, that they make "scene" for you and you see that is all in best order, and whey you live again, they start again to do in same way.
We had same problem with parts from China (material also PA6 GF 30). Solution for us was sudden wisit to company in China with people from Shulman (our material supplyer).
In short, Chinise people worked with to low temperatures, speeds, holding preassure, just to save some money on energy. Instead heating tools for few hours they heat up for around 20min.
After wisit when changes are made, parts from China are same as parts form Tyco (before that wisit, Tyco parts was more expensive and from quality was way much better).
Here is maybie not case, but is also posible that Taiwan company do some things to save and diliver with lower quality.

Best regards,
Nemanja.

[Jendalf]

Hi I would agree with nemanja,

The state as you describe in the first thread points on some step of the nonhomogenity of the material. Also you can see the damaged tool (as somebody tried to take out (by sharp tool, screwdriwer maybe) the rest of the plastics from cavity after unsuccesfull ejecting). I woul definitelly focus on the process at your supplier. Go there set them parameters. Produce ok parts. Validate them and then stand at the keeping of the quality togather with wisits and audits.
Do you have evaluate your costs and time spending on this problem versus the old (ok) supplier???

The temperature that you write (the 23°C) is probably tempreature below the hopper.

J

[nemanja]

And if you make audit, don't tell them abouth. Do as surprise.
In that way you will be 100% shure what is happening there during normal production.

BR,
Nemanja.

[Farnworth]

Just an update to close this thread.

I ended up ordering a batch of the original material from Holland to Taiwan. I went directly to the factory and had them make the whole batch.
The problem seemed to be the material they were using, the Dupont material that I also tested wasn't as good.

Now the parts are back to their old self, the factory has a new supplier for this new material so should be all good. Thanks all for the tips!

[yansi88]

I have a part made of two mirrored pieces that are bolted together with two screws. The part is injected molded with 30% glass reinforced PA6.

History: I switched from a European manufacturer to a Taiwanese based one last year and have been having problems ever since. We use the same exact mold as before, however I believe the process has been changed somehow.

Problem: The new Taiwanese parts are weaker than the older European parts. If you over tighten the screws, you can strip the plastic out now, whereas it wasn't possible before. So you end up with a nut spinning freely inside the plastic part. Or it will simply compress the plastic until the nut goes through the plastic.

I tried switching the material supplier, the manufacturer was using a Taiwanese based supplier that I suspected didn't have the proper specifications for the material, so I switched to Dupont Zytel 73G30L NC010 (a much better material, equal to what we had with the old European parts) but it has only made the parts slightly better.

Here is a comparison of the two parts.
Bad part (Taiwanese part): Even after new material supplier, it still has these characteristics compared to the older part (good part). The surface is shinier and there are strange marks all over the surface, it's the only way I manage to find the different between the two when I have them in my hands.
Good part (European part): It has a smooth surface with a dull texture. It's a much stronger material.



What could cause these characteristics on the Bad part? Could this be related to it being weaker in strength to the Good part?

Thank you! I've been having many headaches over this one.

Alex

Hello, alex
we are a especial manufacturer for plastic injection mold. what can i do for you ?