Improving Surface Roughness of Polypropylene and Measuring Surface Rou 3

[KatRawlings]

Hi everyone,

I was just wondering if anybody can offer me any tips or advice for a project I am working on at the moment. We have been carrying out shot blasting using various abrasives to improve the surface roughness of some polypropylene parts that we are making for the medical industry (1m to 2.5m diameters). Various abrasives have been tested such as garnet, glass bead AB 177-297 and 801 AC, brown aluminium oxide 80/100 and 180/220, white aluminium oxide 70, 120 and 180/220, kiln dried sand, blasting grit (came free with the Sealey shot blasting gun), Honite 13 glass bead, Type 2 plastic media (Urea Formaldehyde) and Type 5 plastic media (20/30 and 30/40 acrylic).

White aluminium oxide 180/220 and acrylic blast media gave us the best results. However, when an electron microscope was used to find the surface roughness of some polypropylene samples that had been blasted with aluminium oxide 180/220, we found that the surface roughness was around 2.5 microns Ra, when it needs to be 0.8 microns Ra.

We therefore still need to improve the surface finish considerably. Does anybody have any suggestions of how we can do this? We are currently carrying out more shot blasting testing with aluminium oxide and plastic media. We are also looking into dry ice blasting.

Also, we would like to be able to measure the surface roughness of our samples on-site while we are carrying out trials. I am therefore looking into hiring/buying a surface roughness testing machine. Can anyone recommend UK companies who might be able to help with this?

Thanks

Kat

 

[KatRawlings]

A previous thread was started a while ago regarding this project if anybody wants to see suggestions that have already been made:

http://www.eng-tips.com/viewthread.cfm?qid=289644&page=1

Also, please remember that these parts are 1.5-2m diameter so processes suggested need to work on large parts.

People have suggested several times that parts could be molded in the first place to achieve the desired surface finish, rather than being machined etc. However, this is not possible because the cost of molding would be far too high for such large parts when the quantities will be quite low.

 

[patprimmer]

You could try dipping the parts in liquid nitrogen before blasting.

PP is quite tough and ductile. It is reasonably soft and has a pronounced thin skin layer when moulded due to orientation of molecules at the surface and the quick chill of the surface layer interfering with crystallization.

The liquid nitrogen sends it hard and brittle.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[HDS]

There is a huge variety of surface roughness measurement equipment using everything from laser interferometry to touch probes and handheld to desktop. Mitutoyo is one of the big players.

2.5 microns Ra is a tight spec. It is more of a polished finish than a blasted finish.

There are lots of nuances to surface roughness, however, the instrument manufactures all have good literature to explain it.

 

[KatRawlings]

@patprimmer: I'm thinking it will be difficult to dip the 2.5m diameter discs in liquid nitrogen. Do you think it will improve the surface finish of the polypropylene? We are looking into dry ice blasting but nobody seems to know the affects of this on plastics.

@HDS: the spec is actually 0.8 microns Ra, we achieved 2.5 by blasting with aluminium oxide 180/220. Thanks, I'll check out Mitutoyo.

 

[patprimmer]

If the surface is acting in a ductile manner and this is upsetting the blasting, cryo treatment will help.

Liquid nitrogen is ot that expensive.

I have no experience in handling it so I don't know what suitable containers are made off, but I suspect a sheetmetal tank the correct size will do.

I know cryo is used to grind ductile plastics to fine powders with hammer mills. I know PP has a distinct surface layer and I expect that will have a negative influence on blasting.

Apart from that I cannot add anything.

I have zero knowledge of dry ice blasting. I think embrittlement temps of PP are about -10 to -20 deg C depending on the grade. I don't know if this is cold enough to give a surface hard enough to polish. I don't know how cold the dry ice will make it.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[unclesyd]

For the detection and elimination of the Al2O3 transfer tapes like products. Something like a Scotch type Tape, like the 3M brand packaging tape might detect the presence of Al2O3. A wider tape might be able to clean the surface if Al2O3 is present in a peel and forget approach.

The first two links are to a type of surface finishing media that is used on plastics. I don't know if it will work on your material as it is relatively soft compared to Acrylic, Polycarbonate, and nylon that I've worked with.

You may want to look at a prefinished sheet. One such type is the the third link. I have two cutting boards with different surfaces on each.

The fourth link is to some good information on Plasma treatments for plastic surfaces. This may help with the surface finish you need or improve the finish from the blasting.

The fifth link is to micromachining techniques that may be applicable to your process.

https://micro-surface.com/index.php?main_page=page&id=1&chapter=0

http://www.frasersaerospace.com/index-1.html

http://www.thecuttingboardfactory.com/

http://www.rovski.com.my/plasma_materials.htm

http://www.microsystems.uk.com/english/micro_molding_nano-structuring_medical_biotech_devices.html

 

[MikeHalloran]

We are still at a disadvantage here because you haven't revealed how you make the part now, or what it looks like, beyond a rough diameter and resemblance to a disk.

So, yet another shot in the dark: If you're starting with sheet and quantities are low, you could improve its surface finish by pressing the sheet between two polished metal platens, assisted by a little heat. Said platens don't have to be strong enough to contain injection pressure. They could be engraved to produce raised figures on the workpiece if desired, but the primary requirement for the platens is they have one highly polished surface, and some means of heating and then cooling the surface. Spray the platens with some mold release, clamp the sheet between them, turn on the heat (electric | steam | circulating oil | whatever) until the plastic surface temperature reaches its melting point at the applied clamp pressure, then turn off the heat and either let the assembly cool, or chill it so the plastic resolidifies.

You'd still need to spend some tooling money, but nowhere near what injection molding would cost.





Mike Halloran
Pembroke Pines, FL, USA

 

[unclesyd]

I agree with MikeHalloran that we only know the end results required and nothing on how you are getting there. Irregularities in machine large diameter materials is severely compromised by lack of rigidity in the part being machine.

Read the advantages of wet blasting. This site also has has a wealth of information on blast media.

http://wetblasting.com/wet-blasting.aspx

I can't recall seeing a repeatable surface finish for Nylon or most plastics. I would get in contact with Keyence about their surface measuring systems. We used this approach for surface finish on diamond lapped parts.

http://www.keyence.com/rss/vision/ljg/kzn_ap1.php#form

 

[btrueblood]

I think you are skipping past a lot of good suggestions because of a "not invented here" outlook, perhaps imposed on you by others.

Good surface finish and absence of burrs/sharps for plastics does not come from machining, the stuff just doesn't respond like metals do (at least, not polypropylene). Molding, or calendaring which I think is the process Mike described, are where I would be going, and it is certainly possible to do such processes "in house" for simple shapes like disks.

If I absolutely HAD to machine and deburr a polyprop sheet, and cutting tools weren't getting me good enough finish, I would definitely look at flame polishing, as others have suggested. You say it's uncontrolled, but there are ways to control any process, you just need to experiment.

I don't think any abrasive process is going to work for you, as you say you can't tolerate any imbedded remnants of the abrasives in your parts - abrasives can even embed into hard tool steels, so this is a non-starter. Unless you can make a reactable or water-dilutable salt work as the blast media. But I would not expect these processes to yield a 4 rms finish, either.

The only other process I could think of that might give good finishes on plastics would be a burnishing process, i.e. moving a rotating polished metal surface across the part, simultaneuously melting and smearing the surface to leave a polished face behind the tool. Not sure how well it would work in a micro-channeled part.

 

[prdave00]

What about vibratory finishing?

http://www.rodeco.com/vibsup-med-plastic.cfm

I have no basis for suggesting this - another shot in the dark.

I have to agree with some of the other posts. Sounds like you aren't able (or unwilling) to try and tackle this issue upstream and are instead trying to rely on "treatments". I think there comes a point where you need to express to your customer that achieving 0.8 microns Ra is starting to look unrealistic without resorting to injection molding or exploring other exotic plastic finishing processes/treatments. Maybe your customer needs to revisit the costs/benefits of having this near immaculate finish. Maybe they'll realize what they're asking for is not necessary to the function of their part.

As for surface texture metrology I'd reach out to Taylor-Hobson in the UK. They offer both stylus & non-contacting instruments.

 

[unclesyd]

I agree with MikeHalloran that we only know the end
results required and nothing on how you are getting there.
Irregularities in machine large diameter materials is severely
compromised by lack of rigidity in the part being machine.

Read the advantages of wet blasting. This site also has has a
wealth of information on blast media.

http://wetblasting.com/wet-blasting.aspx

I can't recall seeing a repeatable surface finish for Nylon or most plastics. I would get in contact with Keyence about their surface measuring systems. We used this approach for surface finish on diamond lapped parts.

http://www.keyence.com/rss/vision/ljg/kzn_ap1.php#form

Btrueblood,
I'm working on a project that is encumbered by by the NIH, Not Invented Here, syndrome. Every time we have a meeting and come up with a possible solution to the problems it is quickly overruled by someone in management. I mention many times that my philosophy is AIA , Anything Invented Anywhere.

 

[btrueblood]

" I mention many times that my philosophy is AIA , Anything Invented Anywhere. "

And it's why I keep coming back here, just to read some of your posts, Unclesyd . I think any of your responses to most OP's on this forum ought to come with automatic lps's. Even if it ends up being unhelpful in the end, just being able to say to one's boss, "well, we looked at a, b, c, d, e ... and nothing works as well as f" makes a lot of the endless what-if discussions cut and dried.

 

[btrueblood]

Realizing that I do have one more thing to add, regarding polyprop and polyethyl -enes, both materials form long chain polymers, and that fact is evident when you machine the material. Whenever I have looked at an abrasively finished surface on PP and PE, I am amazed at how many fine, hairlike fibers seem to have been magically grown from the surface. It's apparent that the cutting edges of the grains are grabbing and extruding material from the parent surface. A grit blast might avoid some of that effect, but (seems to me) there will always be a tendency to extrude, vs. cut, the material, resulting in a poorer finish than one would expect from experience with (e.g.) more metallic materials. My $.02

 

[patprimmer]

btrue

It won't extrude if it's cold enough.

Both PP and PE excellent fibres by a process of very high stretch ratios at a temperature above their softening point but below their melting point. The correct temperature is required to optimise the draw down or extrusion process.

Flame polishing can work although I don't think it will give a good enough surface for this.

Calendaring can definitely work if done right but should be done before the final shaping takes place as it will alter dimensions.

Possibly laminating with a biaxially oriented PP film might work.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[btrueblood]

Pat,

Yup, I think you are right. FWIW, have done that trick to machine a rubber part. I don't know how I'd go about keeping the surface of that big of a part that cold during a finishing operation (at least one with high removal rates), but then I wouldn't go there in the first place.

 

[HDS]

Pat has hit on the best solution. Shrink wrap or vacuum form a PP skin over the part.

 

[KatRawlings]

Thank you everyone for your suggestions. Sorry for not replying, I have had a busy few days. I have found a local company who will measure surface roughness for us, so we do not need to purchase or hire roughness testing equipment. So now I just have the problem of achieving the correct surface finish.

First I think it might be best to give you some more details of the part in question. We will be making around 20 a year of this particular part and they will vary in size from 1m diameter to 2.5m diameter. These “discs” have various features such as holes of different dimensions on the top and bottom. On the top surface there are ribs and grooves that lead from the edge of the circle to the centre. The ribs are 3mm in diameter and the grooves are wider at the edge of the circle and as small as 3mm in the centre. They are being fully machined on a CNC router using standard slot drills and the tolerances are all 0.1mm so it is important to ensure that this is the case.

I really do not mean to have a “not invented here” attitude and love the idea of “anything invented anywhere”, which is why I have decided to post on some forums to ask others for suggestions as part of my research, instead of just trying to come up with my own solution. However, due to the complex nature of the part, and due to me not being able to give away to many details, it is difficult for you to appreciate that some ideas won’t work for this part. I really have tried to give you all some more information in this post, feel free to ask more questions about the nature of the product and I will try to answer them as best as I can.

As far as injection molding is concerned, I have spoken to management to try to get more details about why this technique has been ruled out. Very few numbers will be produced and they vary in size so several expensive (remember it is complex and is not just a flat disc) molds would be needed. Also, it is a special type of polypropylene for the medical industry and is not easily moldable. Our customer did consider this process route at some point in the initial stages of the project and it is just not feasible.

Also, flame polishing has been suggested and I get the feeling people think I have rejected this idea with not much thought. However, we do use flame polishing within our company so I have spoken to experienced people who have seen the part and this is not going to work with the thin ribs and grooves and to experience the surface finish required.

Mike Halloran, I can’t press between two platens because I need to improve the surface finish in the grooves as well as on the top face.

I think it is important to add at this point as well that the reason the smooth surface finish is needed is not simply for aesthetic reasons, as the product is for the medical industry, so it needs to be so smooth due to hygiene reasons (a rougher surface will mean that there are more inclusions for bacteria to breed).

Thanks for trying to help me everyone, it is appreciated. I am still going to look into wet blasting and laminating with a film, I haven’t had chance yet. I am also getting some dry ice blasting trials done at the moment. Hope this post has cleared things up a bit anyway, please let me know if you have any more questions.

 

[KatRawlings]

Oops, sorry that was so long :s

 

[unclesyd]

Long doesn't matter it the thought of feedback to the Forum is what counts.