Anyone know about fiberglass springs?

[John in CR]

I've decided to go the flat spring route for compression of cells from Nissan Leaf packs, Chevy Volt packs, and A123 amp20's. My plan is to warp some plywood as my form, and then use fiberglass and epoxy to form the end plates that will also be flat springs. I'll just keep adding fiberglass layers until the "springs" have enough strength.

My biggest concerns are:
- Will flat springs made this way maintain their compressive strength over time, or will they deform...kinda like putting a significant weight at the end of a diving board for a very long time?
- Can I simply use cheap fiberglass cloth and layup epoxy.

All input is welcome.

 

[Punx0r]

Printed circuit board is ready-made fibreglass board, if that's any help?

I don't know whether fibreglass takes a set under sustained load, but I suspect not, as is the case with glass.

 

[John in CR]

Printed circuit board is ready-made fibreglass board, if that's any help?

I think circuit boards are too low a percentage of fiberglass, so it would be expensive and heavy. If I found some curved ones large enough at salvage prices it could be interesting though. Laying up glass is easy and cheap enough, and I can add layers until I get the strength I want. A123's call for 4lbs/sqin, so the springs will be significant, which is why I'd like to do it with glass instead of the heavy steel plates like used by Nissan and Chevy.

My thought is that gentle curve forming a spring will help prevent the end caps from simply bulging in the middle and coming up short on compression in the center like would happen if I simply made flat plates. I wonder if I should add extra uni-directional glass fibers perpendicular to the spring to ensure that once flat it doesn't bow the other way. I just don't know anything useful about flat springs, much less the forces while compressed under a static load, so I'm just winging it on gut feel and experience with spring clips and stuff.

I don't know whether fibreglass takes a set under sustained load, but I suspect not, as is the case with glass.

That makes sense. I was relying on the fact that fiberglass/epoxy leaf springs exist, and are claimed to have a longer life than steel leaf springs.

 

[Dauntless]

I'm not sure why a composite wouldn't take a "Set" under a "Sustained load," that's one of the things an Autoclave does is provide pressure on the part as it sets. If you form your fiberglas spring in a specific shape, it wants to maintain that shape. The question becomes strength and durability, not exactly the same things as each other. If your matrix doesn't set you can expect that means you didn't have a good catalyst and it didn't heat up enough. It's thermoset.

Dang, I'm a bad boy. I should have learned better by now how to tell you about rating the strength of your spring. Let's say you made a strip that's a foot long at the bottom, but each layer got an inch shorter until you had an inch each to compare how flexy each number of layers made it. Or you simply made a 3 layer layup, a 4 layer, etc. (I just guess that with 9.6 ounce bidirectional cloth you might like it anywhere from 3 to 6 layers, but that's a crude guess) you have ways of testing this out. An 8 pound fiberglas leaf spring (High pressure resin transfer molding) replaced a 41 pound steel spring on the Chevy Corvette, I doubt you need anything much for yours. Careful to not use too much of that epoxy, it actually increases the chances of cracking and crazing when you're resin rich.

As for "Cheap cloth," I'm not sure what that means but I'd say go with something like that 9.6 and it won't cost much. I only think of it as getting more expensive either because you have heavier cloth or there's something else in it, such as strands of kevlar, etc. Epoxy of course is better than polyester, hopefully you don't think of JB Weld, etc. at a time like this but the statement is vague. http://www.aircraftspruce.com/catalog/cmpages/ezpoxy.php This is cheap stuff that is good enough for airplanes.

Now you just need to find a representative part someone else is already making and adapt their process.

http://www.instructables.com/community/PowerisersJumping-Stilts/

http://krbuilder.org/MakingTailWheelSpring/index.html

 

[Punx0r]

To clarify, I meant a mechanical set i.e. plastic deformation under load, not set as in cure of the resin, although I understand the confusion...

If GRP is used for vehicle leaf springs then I say you have nothing to worry about - it sounds well suited

 

[John in CR]

Dauntless,

The change I was concerned about under long term load comes up as "creep" and other descriptions in searches, ie whether or not a simple fiberglass composite will lose some of it's spring. Of course I need to make sure of proper mixing and curing.

As far as cheap cloth I just mean can I make a durable spring with cloth, or do I need to invest time and money wrt fiber orientation? That may be necessary for optimum engineering in terms of strength and weight for something like a leaf spring, but I'm looking for cheap, easy, durable, and flexible in terms of being able to simply add more layers if the spring strength comes up short. That's nice because I don't have to make multiple versions and test...just add layers to the inside of the curve. The outside will need to be quite smooth and consistent, which I will accomplish with the initial form and probably a bit of sanding plus thin finish coat, since layup epoxy remains tacky.

Epoxy of course. Poly will simply crack and couldn't make a good spring. Vinyl ester resin might work and would be cheaper, but I've never been able to source some here in CR. I've worked with epoxy and glass since I was a kid, so it's in my comfort zone.

 

[Dauntless]

Yeah, well, you're still basically an inventor with this, there's going to be things you do that make this unique and the real answer will only come over time. Epoxy will do much to improve the long term over the esthers.

I grew up using such things as old diapers and old tshirts instead of fiberglas as the reinforcement in polyester resin, I don't recommend it. Depending on the reinforcement and matrix you use, your textile can provide up to 99% of the strength. And fiberglas is going away many times stronger than whatever material you have laying around to use instead. I don't remember if by weight or gauged to size, but fiberglas was once mentioned as 30 times as strong as cotton. Epoxy won't mean that much more to strength than an ester would. But it is more compatible than the esters, when I was using a part cotton/part polyester rag, the textile ester would melt in the resin ester, the result being rather frayed. I assume epoxy would leave the material less vulnerable, but I don't know for a fact. I wasn't making anything structural, just customizing bikes usually, but I was better off using paperboard.

In fact, if you want to act like I did when I was 12, you get an old cereal box and cut it out and bent it into the shape you need, then laminate it. You could use beezwax as a release to keep it from sticking to whatever you sue to hold the shape. But I never made a spring. Dying to hear how well it does. I can share my scant knowledge, but you ultimately have to experiment. (Which is fun until you set your garage on fire.)

Oh, but they make the Corvette leaf springs with polyurethane resin. Proprietary, I guess you can't get any of that. But there are urethanes available to do fiberglas with. The RC crowd has taken to using, would you believe, Varathane Diamond Finish with fiberglas. Water based urethane. I've never had something to experiment with this on, but they say it's fuel proof, so maybe good around the caustics forming on a battery?

My thought is that the urethane is an elastomer, so the flexibility might be valued over strength, but if it works then maybe the strength isn't as important afterall. Oh, Gorilla glue is urethane. Use just a little water so it doesn't foam with some of your cloth alternative and see how that does. If it's not good enough and you turn to fiberglas, you could try some of this. http://bjbenterprises.com/index.php/polyurethanes/castable/semi-rigid/all-purpose/fd-70-a-b-3168/

So a traditional approach to making a surfboard it 2 ply of 10 ounce cloth on top and 1 underneath. But they also get cocky and use 6 ounce cloth to make it lighter. If you're willing to experiment there's all sorts of things to try, but I think the cheap alternatives will fail, otherwise the expensive stuff wouldn't be popular, right?

http://www.faa.gov/regulations_policies/handbooks_manuals/aircraft/amt_airframe_handbook/media/ama_ch07.pdf

 

[amberwolf]

Wouldn't basing the spring-making method on something like skis work? (I don't know exactly how they are made, but they certainly are *usable* as leaf-springs, in some applications)

 

[John in CR]

Wouldn't basing the spring-making method on something like skis work? (I don't know exactly how they are made, but they certainly are *usable* as leaf-springs, in some applications)

I wish there were skis readily available down here...not for this project as I need to start with a curve, but some winter sports would be a nice change. This isn't Hawaii, so no hitting the beach and the slopes the same day here.

 

[Harold in CR]

I did a little research on Fiberglas springs. Corvette springs are/were made from Plastics mixed with resin. Didn't find any formula ratios, but, what I did find, was, the springs eventually went bad.

Wood is usually what is used with resins, for instance, water skis, I imagine snow skis, recurve bows, etc.

From my experiences, boat building, etc., layers of bi-axial would result in stiff parts. Flexing would eventually weaken the part and the strands would fray.

An example of multiple flexing would be recurve bow construction. Thin layers of wood, usually Maple would be glued to 2 thin layers of a plastic type strip, on both sides of the wood.

Maybe research that type of construction, to learn what materials are used ?

 

[The fingers]

We used to make surfboard skegs by laying up squares of fiberglass cloth and resin to the desired thickness over a square piece of plywood covered with wax paper. When it cured you would cut out and shape it into a foil that could even be machined and drilled for an attachment screw to fit a fin box. The length and thickness determined the flexibility. Some trial and error with the size and shape you want and the number of layers of # ounce cloth might give you an idea of how many laminations to use.

 

[e-beach]

My surf boards never lost shape, but they did scratch, fray, chip and break depending on pressure and impact.

I think fiberglass would be too brittle to use as a spring. Over time it would probably chafe and at some point crack.

Do you have any old beds with "bed springs" you could use?
Or maybe doubled or tripled up picture frame spring clips?



http://www.ebay.com/itm/Spring-Clip...hash=item565ca26b43:m:mprMWlcAnme98x7kKEmCnxQ

 

[speedmd]

All springs sag and get soft over time, no matter what they are made out of. Just a fact of life. Light weight glass /carbon epoxy x-country racing Ski will get soft if compressed /flattened out in quite short order is stored that way. Most likely a good item to study a bit as they measure /publish pressure distribution across the bottom every few CM on high end ones.

Beryllium copper makes one of the longest lasting springs and used in clocks and watches often if that helps on possible material options.

 

[Dauntless]

There's worn out for carrying more than a ton of car at speed and there's worn out for holding your batteries. FRP is still the Corvette spring. Fiberglas springs are very much in use, as are carbon fiber springs. I don't know that this would see use long enough for their to be an issue.

Skis are a good reference, as are the powerrisers. DYI skis always seem to have a wood core.

 

[e-beach]

............ DYI skis always seem to have a wood core.

A shaped laminate could work. Thin layers of veneer wood soaked in water and then dried to shape. Then layered with resins and cut to shape.....D may be on to something.

 

[John in CR]

These "springs" will just be an arc pressed flat against the front face of the 1st cell and the rear face of the last cell to maintain a compressive force on the cells. They won't be used to suspend the battery pack or act as moving springs.

I'll use precurved wood with mold release to create the springs, but I don't think it's the right thing to include in the composite as I believe the wood fibers will change under constant compression. I may be wrong though and that it's value is to make the spring structure stronger by creating space between composite layers. The composite leaf springs I've seen though had no filler though.

As a separate topic, I wish I could come up with a rectangular inner tube to use instead. It would still need some kind of support structure to hold it, but I see inflatable plates at each end of the pack as the surest way to achieve a measurable even pressure across the face of the cells with a valve giving an easy means to measure the pressure and vary it.

 

[speedmd]

A rectangular tube with end caps with a offset tensile rod (screw on one end) would create a relatively smooth arc that could be adjusted just by turning the nut. Fixed tension rods done commonly in pre stressed concrete.

 

[Dauntless]

You could use that urethane elastomer, or even silicone, to make a tight fit for the batteries inside. Wearing out doesn't become an issue unless you keep taking it apart.

 

[John in CR]

You could use that urethane elastomer, or even silicone, to make a tight fit for the batteries inside. Wearing out doesn't become an issue unless you keep taking it apart.

More than just a tight fit is required, and I believe uneven compression is common cause premature failures of prismatic cells. With cells I reasonably expect to last 5 or 10 years I want to do it right. We have to keep in mind these cells are made of many thin layers of very thin metal separated by thin layers of pasty substances, and deformations cause issue. With A123's recommended 4psi it seems like an insignificant pressure, that is until you consider it's over about a 6"x8" area, so it's hundreds of pounds of total force.

The car makers use inflexible metal end plates to ensure even pressure, but I want to do something much lighter, especially since it's only for 20-40 cells. I've used small plastic or metal plates to maintain a square surface plus electrical tape to create the compressive force for RC lipo prismatics, but my main concern was avoiding deformation of the corners. With large cells my worry is even pressure at the center, and a flat spring that in it's relaxed state is curved up away from the cell face at the end seems like the lighest space saving way to go. Otherwise the compression plates must be heavy, or include a space using structure to make it rigid.

 

[Harold in CR]

On my Volt pack, I squeezed all the cells together with bar clamps, used for gluing boards together for tables, etc. I had cut all the cooling tubes off, so, there are no rubber gaskets. It is a solid connection of plastic cell holders. I then fabricated straps from extra cell blocks and other heavy strapping I had, that I had to stretch fit over the entire 20 cell pack. I did use the factory end plates.

Then, I made up 2 end plates of a hard wood that will NOT compress. Those end plates are one side flat and 1 side convex, so the pressure is slightly tighter in the center. Looks like this I) . I had to work that down between the factory end plates and the straps. First strap separated at the welded joint. I re did the weld and the straps are VERY tight. When I loosened the clamps, there was NO noticeable slack. There is no way I could gain any more pressure, as the plastic holders are squeezed tightly together.

This as much pressure as I could come up with.

If you want, I could make a couple end plates and get them to you.