is fiberglass legal to use on a first robot?

[mechanicalbrain]

Originally Posted by ngreen

We used fiberglass unistrut on this years robot for the elevator lift and arm. Great material. Saved us a ton of weight and stong enough in testing to hold at least 4-5 tetras at 5-6 feet or at least one mentor trying to break it. Look at 1108 Teaser #1 and #2 to see our fiberglass or I'm sure you can find a competition picture somewhere.

Please keep in mind the different types of strengths. While it can take weight consider that if they make something like their entire frame out of this they might want a padding to keep from chipping and splintering.

[Gdeaver]

With composite structures the orientation of the fibers affect it's characteristics. One problem with pultrusions is the large majority of fibers are in the length direction. Because of this point impacts are a problem with Pultrusions. This can be partially addressed by filling the pultrusion with 2 lb, 2 part expanding foam. With pultrusions a good way to increase the point impact strength and increase the load bearing capacity is with tube in tube construction. We took a 2" 1/8 wall and a 1" 1/8 square tube pultrusion. The 1" was inserted in the 2". The space between the 2 on one end was filled with epoxy and chopped carbon fiber for 4" and bolts where set in the epoxy to mount the hinge hardware. The space between the 2 tubes was then filled with the 2 part foam with in 6" of the other end and then that end was filled with epoxy fiber mixture. The resulting structure Has point impact resistance and is stronger than going to heavy wall units and is still very light. With composite work I would strongly recommend using an amine laminating epoxy and not polyester resins. The surface hardness and strength is much better with epoxy.

[Veselin Kolev]

Fiberglass gets its strength from the glass fibers that make it. If you drill a hole in a fiberglass box beam, the glass fibers are no longer continuous, so you compensate the beam's strength. If you want to have a continuous box beam with cuts in it, you have to get it custom woven by a company, and that is pricey. I would stick to using fiberglass welding glue, or if you must, only drill screw holes in the very ends of the beam. That way you retain as much of the beam's strenght as possible.

[mechanicalbrain]

Originally Posted by Veselin Kolev

Fiberglass gets its strength from the glass fibers that make it. If you drill a hole in a fiberglass box beam, the glass fibers are no longer continuous, so you compensate the beam's strength. If you want to have a continuous box beam with cuts in it, you have to get it custom woven by a company, and that is pricey. I would stick to using fiberglass welding glue, or if you must, only drill screw holes in the very ends of the beam. That way you retain as much of the beam's strenght as possible.

One thing to keep in mind with fiberglass as well as carbon fiber is the epoxy. I know heavy carbon fiber weave has a much weight to strength ratio than stainless steel but it doesn't take sharp impact well (it shatters because its brittle) and while Kevlar weave can take impacts extremely well over a series of hits the epoxy will turn to dust and the Kevlar turns back to fabric. I'm willing to bet fiberglass has a similar limitation due to the epoxy or glass (one or the other will give). Fiberglass is actually brittle (I found this out from experimentation) It can break of in chips or even form finite cracks due to minuscule air bubbles that weaken it.

[EricH]

Originally Posted by mechanicalbrain

I know heavy carbon fiber weave has a much weight to strength ratio than stainless steel but it doesnt take sharp impact well (it shatters because its brittle)

Run a robot at full speed into a properly built sheet (or something similar) of carbon fiber, then come on here and say that. 330 ran an old robot into a piece many times and the only damage was a few scratches and scrapes.

Originally Posted by mechanicalbrain

Im willing to bet fiberglass has a similiar limitation due to the epoxy or glass (one or the other will give).

Not necessarily. If you drill a hole through fiberglass, you might have an easy time breaking it, but if you keep holes to the minimum, and only use it above the frame (or as the frame as previously described), you should be OK. I don't kno how brittle fiberglass is, but 330's arm in 2004 was mostly fiberglass, and it didn't break easily.

[mechanicalbrain]

Originally Posted by EricH

Run a robot at full speed into a properly built sheet (or something similar) of carbon fiber, then come on here and say that. 330 ran an old robot into a piece many times and the only damage was a few scratches and scrapes.

Actually thats me quoting the owner of www.sollercomposites.com in a talk we had two days ago. Since its his profession and he owns one of the leading companies in the field Im assuming he knew what he was talking about. I don't always randomly make up facts.

[EricH]

Originally Posted by mechanicalbrain

Actually thats me quoting the owner of www.sollercomposites.com in a talk we had two days ago. Since its his profession and he owns one of the leading companies in the field Im assuming he knew what he was talking about. I don't always randomly make up facts.

My dad directed the making of our test materials (as well as obtaining the carbon fiber and epoxy). He also has worked with the stuff for a living, particularly its uses in aircraft. I would presume that he knows the subject very well also.

The other thing that may affect the strength is the thickness of the epoxy. The thicker the better to a point, and we laid it on thick, then laid it over a mold and let it cure.

[mechanicalbrain]

Originally Posted by EricH

My dad directed the making of our test materials (as well as obtaining the carbon fiber and epoxy). He also has worked with the stuff for a living, particularly its uses in aircraft. I would presume that he knows the subject very well also.

The other thing that may affect the strength is the thickness of the epoxy. The thicker the better to a point, and we laid it on thick, then laid it over a mold and let it cure.

Well that or adding extra layers. Like i said kevlar has 20% the stength of cabon but takes a beating much better so it makes a good outer layer...... wait no were still talking about fiberglass right? yeah all i know is that the tube i posted a pic of is now chipped. The lesson: don't hit them with hammers.

[Norm M.]

Originally Posted by mechanicalbrain

Fiberglass is actually brittle (I found this out from experimentation) It can break of in chips or even form finite cracks due to minuscule air bubbles that weaken it.

The landing gear on my airplane is solid fiberglass. Not brittle. Also the main spar.

Fiberglass is a very generic term. A lot of people think of the polyester resin and chopped glass. In airplanes we use woven cloth and a two part epoxy. This epoxy has a much greater pot life, and gives you time to smooth everything out.

A couple sources for materials to learn with: Wicks Aircraft and Aircraft Spruce have a composite kit which includes a manual from Rutan's plans. The kit includes epoxy, cloth, different types of foam, and some flox and micro balloons. Another source for epoxy and cloth is at West Marine stores. The West System epoxy with slow hardner works quite well.

We have used fiberglass laid up over a large salad bowl (about 4 layers of 7 ounce cloth) to make 'hands' to grasp a large ball. 2000 game, maybe? We also used about 12 layers to reinforce some plywood to make a spring as part of a ball thrower mechanism.

[mechanicalbrain]

Originally Posted by Norm M.

The landing gear on my airplane is solid fiberglass. Not brittle. Also the main spar.

Fiberglass is a very generic term. A lot of people think of the polyester resin and chopped glass. In airplanes we use woven cloth and a two part epoxy. This epoxy has a much greater pot life, and gives you time to smooth everything out.

A couple sources for materials to learn with: Wicks Aircraft and Aircraft Spruce have a composite kit which includes a manual from Rutan's plans. The kit includes epoxy, cloth, different types of foam, and some flox and micro balloons. Another source for epoxy and cloth is at West Marine stores. The West System epoxy with slow hardner works quite well.

We have used fiberglass laid up over a large salad bowl (about 4 layers of 7 ounce cloth) to make 'hands' to grasp a large ball. 2000 game, maybe? We also used about 12 layers to reinforce some plywood to make a spring as part of a ball thrower mechanism.

Uhh i think your talking more about composites like on the site i posted. i was talking about the tube i posted a pic of but yeah thats the material i referred to. What we might do is a secret combination of layers of various sheathing to make our frame (assuming we can afford it)

[Gdeaver]

As this thread is showing composites is a deep subject. That is why I suggest that any teams that are interested in using composites should do the research and testing now. There is allot to learn and the build season is too short to perfect composite methods unless your team has a someone with experience. Every year I have mentor I have work a composite structure in some where on the bot to give the students an exposure to this important technology. I would like more teams to work this in to their robot too. A good start would be to take a 1' by 1' piece of 1/4" birch plywood and laminate 1 layer of 6 oz s glass on both sides and do some testing to see what this simple process does to the plywood. I can post some sites to help. If they like the results they can go from there. Get some fiber in your robot's diet.

[mechanicalbrain]

Originally Posted by Gdeaver

As this thread is showing composites is a deep subject. That is why I suggest that any teams that are interested in using composites should do the research and testing now. There is allot to learn and the build season is too short to perfect composite methods unless your team has a someone with experience. Every year I have mentor I have work a composite structure in some where on the bot to give the students an exposure to this important technology. I would like more teams to work this in to their robot too. A good start would be to take a 1' by 1' piece of 1/4" birch plywood and laminate 1 layer of 6 oz s glass on both sides and do some testing to see what this simple process does to the plywood. I can post some sites to help. If they like the results they can go from there. Get some fiber in your robot's diet.

Yeah now's the time to research. However I know that if you want to do a large portion of your robot in a composite be for warned that its expensive so: do your research before hand, talk to manufacturers for advice (even if you have people who have experience or have it yourself nobody knows more then the manufacturer), be safe (some of this stuff has the potential to be very nasty), and if possible get sponsorship (a cost estimate of a frame can easily exceed $1,000 and thats at wholesale!)

[ChrisH]

Originally Posted by Gdeaver

With composite structures the orientation of the fibers affect it's characteristics. One problem with pultrusions is the large majority of fibers are in the length direction. Because of this point impacts are a problem with Pultrusions. This can be partially addressed by filling the pultrusion with 2 lb, 2 part expanding foam. With pultrusions a good way to increase the point impact strength and increase the load bearing capacity is with tube in tube construction. We took a 2" 1/8 wall and a 1" 1/8 square tube pultrusion. The 1" was inserted in the 2". The space between the 2 on one end was filled with epoxy and chopped carbon fiber for 4" and bolts where set in the epoxy to mount the hinge hardware. The space between the 2 tubes was then filled with the 2 part foam with in 6" of the other end and then that end was filled with epoxy fiber mixture. The resulting structure Has point impact resistance and is stronger than going to heavy wall units and is still very light. With composite work I would strongly recommend using an amine laminating epoxy and not polyester resins. The surface hardness and strength is much better with epoxy.

For most commercially available pultrusions, the only transverse fibers are in a random mat finish ply on the outside. The finish ply has practically zero strength. So transverse properties come completely from the matrix resin. If you use a pultrusion the best way to transmit a large torque into it (like on one end of an arm) is to put it into a socket. Another method is to put a rod up the inside and wrap wet layup fiberglass around the outside. Be sure to use a matrix resin that is compatible with the resin in you pultrusion, epoxy with epoxy and polyester with polyester.

Drilling a couple of bolt holes in the side is a recipe for disaster. If you must do something like this, drill through the top and bottom rather than the side. Use big washers to spread the clamp loads and torque the heck out of it. The whole point is to minimize loads that tend to split the pultrusion open. Though if you do this you probably also want to fill the gaps between the rod and the pultrusion with epoxy or similar so that when you clamp down you don't break it in compression.

Composites can be great materials but the real trick is getting the load in and back out. Bond whenever you can because bolt is a four letter word. Bolts lead to stress concentrations and tend to cause failures in unpredictable ways. Unless you use the specialized fasteners we do here at the bird farm, but $50-$80 each is a little expensive for most teams.