Screw Pull-Out Force in Polypropylene - Calculation

[506818]

I'm designing a system housed in an enclosure (Peli case style) that will hold 1 psi (0.0069 N/mm^2) pressure with a top panel held in by a number of screws.

The top panel has a surface area of 83607mm^2, therefore, a total force of = 0.0069*83607 = 577N. At present, I have 10 screws holding the part in place (57.7N per screw).

The case is made from polypropylene with M4x0.7 stainless steel screws used to hold the top panel in place. There are mounting holes on the case that I am planning on drilling larger and tapping (5 mm of thread used by screw) to fit the M4 screws. However, I need to ensure that the screws hold the top panel in place without stripping the plastic thread on the case. I'm not worried by the screw, more about the case.

I've found this calculation for internal threads shear force area (imperial-based):

https://www.engineersedge.com/thread_strength/thread_bolt_stress.htm

Associated calculator:

https://www.engineersedge.com/calculators/internal-thread-shear-area.htm

The values I have come up (metric):
n = 25.4/0.7 = 36.29
Le = 5 mm
Dsmin = 3.838 mm
Enmax = 3.523 mm
As = 0.07102 in^2 = 45.819 mm^2 (not sure if calculator is correct on the website as I get 30.71 mm^2)

Above thread dimensions calculated using:

http://www.amesweb.info/Screws/IsoMetricScrewThread.aspx

Then with F = shear strength * shear area
(shear for polypropylene = 0.5* tensile strength = 0.5*31 = 15.5 N/mm^2, data from here:

http://www.sdplastics.com/pdf/pp.pdf)

Therefore, the max force each screw can theoretically handle is:
F = 15.5*45.819 = 710 N

This seems to indicate that I should be fine, but it's all theoretical and would like some idea if I'm on the right path. Am I?

 

[desertfox]

Hi 1muffin

I have only skimmed over the calculation and it seems reasonable to me, is there any temperature involved? as that might change things.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[506818]

I'm just a bit sceptical as 72kg (~710N) hanging off one M4 screw in plastic seems a lot, but may be right!

There is temperature involved, but not that much that it'll affect the plastic greatly I hope. Also, if the calcs are correct I have a 710/57.7 = 12 safety factor.

 

[3DDave]

One thing to worry over is that this is the shear strength, but that depends on the material staying around the screw and not simply elastically expanding around the screw and reducing the shear area. The effect will be very part-geometry dependent; if it's a block vs in the edge of a narrow wall. Since the elastic modulus of plastics is so low compared to metal it probably merits a look.

 

[506818]

Hi 3DDave - great point, especially in this instance when I do have a very thin wall on one side. Any idea how to test this cost-effectively? I see there exists equipment for this purpose like the one here.

Also, any thoughts on if the calcs I've estimated are correct?

 

[drawoh]

1muffin,

You calculations are based on the force needed to hold your lid on. You want to grossly exceed this, and torque the screw down to its proof load.

I am checking your numbers using my own calculator. An M4X0.7 screw in A2-70 stainless should hold 3KN (670lb). That assumes you torque it to 75% of yield, and a friction factor of 0.2. I have no idea of what friction factor you assign when you are screwing into plastic. You need to figure out a strategy for torquing your screws down properly.

I am getting a thread depth of 20mm, to equal the strength of the screw. Compare the cost of your screw to the cost of your plastic part, and apply safety factors.[ ]

--
JHG

 

[3DDave]

If the material was infinitely rigid then the shear calcs given would be fine; it's only that plastics have such low rigidity compared to metals that it a thing to check. Plenty of snap-lock plastic items are designed to take advantage of this. I did not duplicate the effort, but the steps you present are in the order expected.

I'd get a sample of the plastic and drill and tap it. I would probably use a 10:1 lever and a fish-scale to get a cheap ball-park so that I don't have the full load spring-loaded and flinging parts around if the screw pulls loose.

If there is room, there are ultrasonincally welded inserts that avoid this problem.

 

[IRstuff]

How thick is your PP? I'd be curious how sure you are that the pressure will even hold, given that thin PP will likely bow outwards, making the seal problematic.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Pud]

For maximum pull-out strength use screws designed for plastics. e.g. "PT Screws". Google will give more info.

www.tynevalleyplastics.co.uk

Politicians like to panic, they need activity. It is their substitute for achievement.

 

[506818]

Thanks all - I'll try to cover the responses. Also added an image to show the problem in question.

[You calculations are based on the force needed to hold your lid on. You want to grossly exceed this, and torque the screw down to its proof load. I am checking your numbers using my own calculator. An M4X0.7 screw in A2-70 stainless should hold 3KN (670lb). That assumes you torque it to 75% of yield, and a friction factor of 0.2. I have no idea of what friction factor you assign when you are screwing into plastic. You need to figure out a strategy for torquing your screws down properly. I am getting a thread depth of 20mm, to equal the strength of the screw. Compare the cost of your screw to the cost of your plastic part, and apply safety factors. smile][/quote]
Due to the hole size/wall thickness/hole depth it will definitely be the case that fails first. Would you be able to provide the calculator you're using?

The reason I'm going down the tapped plastic route is that there is not enough space for inserts. Good thinking with the fish scale. Looking at it all, the screw has to hold 57.7N, so about 6kg which isn't all that much when I think about it in practical terms! I think I'll try and get some test going...

I will have a wall thickness of ~1.9mm either side of the screw - not a lot. The gasket between the top panel and the case is fairly large and takes up a lot of surface area, so the panel or case will have to bow a lot for the seal to fail.

This would be the ideal route, but I need it to seal and have opted for using sealing screws. Unfortunately, I can't find any self-tapping and sealing screws on the market, other than ones for roofing - not ideal for the look of the product.

 

[IRstuff]

Assuming a square, your cover has a perimeter of 1175 mm; ten screws seems a bit light.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[drawoh]

1muffin,

I have a JavaScript screw torque calculator up on my website. I am still debugging it. I was getting ready to post a question about plastic and friction factors up on forum725, when you came along.

You have a serious assembly problem here, which will be a lot worse when someone in the field tries to maintain this thing. You need a strategy for installing fasteners that will not strip your tapped holes. I suggest either nylon screws, or a different fastening system.

--
JHG

 

[Compositepro]

The design of this case is just silly and will not work. In a rectangular case the only thing holding the walls together against pressure is the lid, and the screws will be in shear. The screws will shear out of the thin plastic walls. Plastics are just that, plastic. Plastics creep at loads much lower than their tensile strength. The walls will also bow and inflate like a balloon.

 

[IRstuff]

My question was really about whether you could even hold pressure. Your cover plate looks to be on the order of 5-mm thickness.

I'm guessing that the bow in the cover plate will enough that you'll not have a pressure problem, rather, you'll have a lack of pressure problem.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[506818]

There is space to add more, just wanted the limit cost and assembly (also disassembly) time if not required.

If I can ensure the plastic doesn't shear/strip, then I will be happy. I can only find sealing screws in the form of stainless steel.We have plenty of torque screwdrivers to assemble without over-tightening.

Any improvement suggestions? I can do 2 things at present to reduce the load on the screws: increase the number of screws, and reduce the pressure seen to 0.5 psi (this will lead to other effects). The pressure rating is set by a pressure relief valve, but it would be favourable to have this at a higher pressure rating as the intent is to prevent air passing through into the box (condensation).

Cover plate is 3.2mm around the screw and 8mm for the main bulk. Could you explain further what you mean about the 'lack of pressure problem'?

 

[IRstuff]

On a normal pressure vessel, with steel covers and whatnot, a 100-mm fastener separation is plausible, since the steel is sufficiently stiff to resist bowing. So, the question is whether your PP is sufficiently stiff to resist bowing between fasteners and allowing the pressure to bleed away (lack of pressure problem)? If you doubled or tripled the number of fasteners, the load would be corresponding decreased, and the possibility of bleed-off substantially decreases.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[drawoh]

1muffin,

Rethink your design.

I am looking at my camera Pelican cases here, and I am pretty certain the plastic is glass reinforced. It certainly is not polypropylene. If you need something sealed, you should minimize the number of seals. If your flanges are outside the sealed area, your fasteners are outside the sealed area, and not a point of failure. Since the fasteners do not have to be sealed, you can use anything you want, like screws and nuts, with large washers.

You have torque wrenches in your plant. Will they have them out in the field? Do people out in the field read and follow your instructions?

The key to keeping screws done up is to tighten them down hard. A 5mm polypro wall does not permit proper screw tension, or torque.

Have you worked out how rigid your enclosure is? Under 1psi pressure, your lid and possibly your enclosure will bow, and let the air out. This is why people are recommending more screws, but they may be applying rules of thumb for metal enclosures.

--
JHG

 

[Tmoose]

"Could you explain further what you mean about the 'lack of pressure problem'?"


I believe a previous poster referred to event similar to what happens around 8 seconds here -

https://www.youtube.com/watch?v=JxJ05TdjQEY

 

[IRstuff]

Hopefully, it won't be that violent. Nevertheless I once spent 4 hrs in a meeting that worried about pressure buildup inside an enclosure, and how we would deal with it, bladders, relief valves, etc.

Spoiler

When it was finally built, we pressurized it, and about 8 seconds later, the pressure differential went to 0.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[drawoh]

IRstuff,

I have designed enclosures that were dust sealed, and then installed in aircraft. I worked out the forces on the covers at 10,000ft. I added breather desiccators to them to make sure there was no pressure differential. Pelican cases have pressure relief valves.

--
JHG