Material considerations for a sturdy cantilevered flat spring

[futureguy]

Hello all, long time reader first time poster.

I am prototyping a sporting goods product that requires a flat cantilever spring approximately 6 inches long. Mounted rigidly, and supporting a rigid bolt at the free end, the "flex zone" between these elements is actually only 2.25" long. I chose extruded polycarbonate plate for beta testing purposes due to it's high flexural strength to stiffness ratio (Fb/E), and the proper size seems to be around 3/8" thick and 2-3" wide (I=0.011 in^4, E*I= 4,125 lb*in^2). The cantilever loads are typically 0-75 pounds, and may be up to 200 pounds, with thousands of cycles. To give you an idea: under a 80lb load, the spring deflects about 12 degrees.

The PC is performing beautifully but eventually breaking, assumably from fatigue and high impact loads (although the impact is not directly to the PC part itself).

For purposes of continuing beta testing, is there a more reliable plastic that will remain elastic under these high deformations but not fracture catastrophically? ABS, Delrin, HDPE...all seem to be less favorable "on paper" but maybe are less prone to this brittle fracture failure?

I have begun to look at composites and spring steel as well. A leaf spring in a vehicle suspension is the best real world example I can find of a very sturdy flat spring, but I'm not sure if such a material could provide enough flex over only 2.25" unsupported length.

Thanks in advance for any advice

 

[patprimmer]

Traditionally the best thermoplastic for a spring at normal room temperature and humidity is acetal homopolymer, but it does not perform as well when moulded in very thick sections, like from memory, over about 6mm.

If it is over crystallising due to the thick section, it may be possible to design a ribbed or corrugated profile.

If acetal won't do it, 45 to 50% glass fibre reinforced nylon 6.6 might do it so long as extreme care is taken to maintain fibre orientation along the desired axis.

If normal glass fibre does not do it, then long glass fibre would be better, with a fibre length about equal or longer than section thickness. 60% long glass reinforced nylon 6.6 would be the optimum choice I would think.

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

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

for site rules

 

[futureguy]

Very helpful, thank you. Although the final design may be a ribbed or corrugated section, investing in the mold at this early stage of development is not in the budget...unless there is an easier way to prototype such a section?

In the immediate interest of continuing beta testing without the risk of brittle fracture failure: if I need to use <6mm acetal (is commercially available Delrin acceptable?), I can try to arrange 2 or 3 plates into a leafspring-like configuration. Viable?

The other low-tooling-cost option I can come up with is getting a pre-hardened spring steel sheet cut to size....

 

[futureguy]

as for glass-filled plastics, the data I've seen indicates that glass increases stiffness but does not significantly increase flexural strength, which work against me as I need a lot of elastic deflection in the piece.

 

[patprimmer]

Because the glass filled is stiff you can make it thin. The geometry may help.

A part machined from an acetal block will give a fair indication, but will not perform exactly like an injection moulded part for a number of reasons, being:-
1) The part made from block will be made from an extrusion grade of material that is a high molecular weight grade and has minimal if any nucleating agents and lubricants in the formulation.
2) The resin used to make the block might be a different colour. YES it can make a significant difference.
3) The machined part will not have the same molecular orientation and crystal structure as a part that is moulded.
4) The mill cutters tend to leave little notches in the part.

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

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

for site rules

 

[patprimmer]

I have seen prototype moulds for very simple shapes made by bolting a mould set with two flat plates and a spru bush, but no ejection.

A third plate is cut and filed or ground with a carbide bur, then that is roughly held between the two flat plates and the mould closed on it nd a shot taken.

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

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

for site rules

 

[Compositepro]

Generally thermoplastics make poor springs compared to steel or composites due to creep and low modulus. A composite spring would be much thinner and more durable. You can get thin sheets from McMaster-Carr. Steel would be cheaper but prone to corrosion.

 

[futureguy]

The main problem for this design is finding high strength coupled with high flexibility (low stiffness). So I want flexural strength and section modulus to be high, and young's modulus and moment of inertia to be low. With Fb/E being material properties and S/I being section properties.

It seems a flat "strap" shape is the most advantageous....for example a common C-channel laid sideways (C3x4.1), has a section modulus about equal to it's moment of inertia. While a wide rectangular section of roughly the same section modulus (.5"x5") has a moment of inertia about 4 times less, providing much more flex. This is why a leaf spring is not shaped like an I-beam....they have opposite purposes in terms of stiffness. Am I right?

Could anyone be a bit more specific about what kind of fiberglass/composite/or spring steel I'm looking for? I'm finding the flexural strength and modulus of elasticity much more difficult to research compared to plastics.

 

[patprimmer]

With composites a lot depends on the lay up or orientation of the fibre and fibre to resin ratio.

A lot of composites data list properties of individual fibres and resins rather than of a real world layed up composite.

Having outlined the problem, I can't really help with an answer other than to say be aware of what the data represents.

Thermoset resin based composites will certainly have a better performance as a spring than will a termoplastic.

Creep is the most common problem with thermoplastic springs under high loads.

By far the biggest application of plastic springs is keyboard key return springs and ball point pen point retracting mechanisms.

All are that I know are acetal but all are at very light permanent load as the at rest position is relaxed. Also, these are complex mouldings and high production numbers but under low load so thermoplastic has two major advantage and the dissadvantage is mitigated. This may well not be the case with your part.

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

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

for site rules