Flex joints instead of Rodends

[Vector006]

I was wondering which teams use flex joints in place of rodends where the control arms mount to the chassis. By flex, i mean flat plates of steel or composite material. Im not so concerned that they will create a virtual pivot point or that they will add a slight bit of stiffness, but i am slightly worried about them buckling or fatiguing, especially on the LCA. I wanted to know if anyone had any pictures or horror stories to steer me in the right direction. Also, what materials have people used?

[Vector006]

I was wondering which teams use flex joints in place of rodends where the control arms mount to the chassis. By flex, i mean flat plates of steel or composite material. Im not so concerned that they will create a virtual pivot point or that they will add a slight bit of stiffness, but i am slightly worried about them buckling or fatiguing, especially on the LCA. I wanted to know if anyone had any pictures or horror stories to steer me in the right direction. Also, what materials have people used?

[Igor]

Did you think about how you are going to repair one after you hit something with your wheel?
I know they're cool, but if you can't repair it what's the point?

Igor

[ben]

I don't see why they would be any harder to repair than a conventional wishbone?

If you made the flexure out of steel like Lehigh tested a couple of years ago (and like Jordan did in F1 in 1997) the wishbone is going to be a lot cheaper than if you spec an expensive aerospace grade rod end/speherical.

Something I would have liked to try given another year.

Ben

[murpia]

I agree with Ben, I would like to see a good flexure FSAE car. Nearly every non-steering joint in an F1 suspension is a flexure these days.
Flexures have cost and weight advantage over sphericals if done right, but they are less tolerant of design changes as they have a much smaller 'operating window' than a spherical. So don't change your mind about your suspension geometry!
As to his repair comment, I assume Igor is comparing a threaded rod-end with an integrated flexure, rather than an integrated spherical with an integrated flexure...
Ian

[DJHache]

Not every wishbone mounting point on F1 cars is a flexure. In the rear of the Ferrari for instance there are still rodends because they allow more movement. For a lightweight fsae car there are issues with making the tab thin enough to allow movement but yet be tough enough to avoid buckling. 1/4" thick steel flexures will not allow for much movement, but 0.035" thick steel might have buckling or strain hardening problems. Titanium is a good alternative I guess.

[MikeWaggoner at UW]

F1 can use them on wishbones because their A-arms travel through small angles. SAE travels through larger angles, so designing flexures that don't fatigue may be more difficult.

[Igor]

I was indeed thinking of something else. We once had a discussion on integrating kevlar flex joints into the monocoque, which is kinda hard to repair.
Well, you can always bolt a door hinge there :-)

Igor

[Nate Notta]

So does anyone remember which teams have done this? I can scarcely remember what the joints look like, and I want to try to dig up some photos.


Nart

[Jon]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by MikeWaggoner at UW:
F1 can use them on wishbones because their A-arms travel through small angles. SAE travels through larger angles, so designing flexures that don't fatigue may be more difficult. <HR></BLOCKQUOTE>

My other concern, with a much smaller scale car, and larger angles of travel, designing flex joints large enough not to fatigue would have a noticeable effect on the spring rate at the wheel of the car. You can design around this to a point, but someone would need to do a little development work to prove its feasible.

[Ben Beacock]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Jon:

My other concern, with a much smaller scale car, and larger angles of travel, designing flex joints large enough not to fatigue would have a noticeable effect on the spring rate at the wheel of the car. <HR></BLOCKQUOTE>

Also, the addition of bending loads into the wishbones can't be very good either.

[Nate Notta]

So does anybody remember which teams have done this?

[Ben Steele]

I could be wrong, I'm fairly new to FSAE, but everything that I've heard and read indicates that there has never been an FSAE car that used flexures. Of course, that doesn't necessarily mean that research has not been done. I would be willing to bet that someone has looked into it.

Ben Steele

[Charlie]

At the SAE Motorsports Conference in 2002 (Indianapolis) there was a FSAE team with a booth that had flexures on display. Not on the car, but they had built some samples and from what I understood were intending to run them in 2003.

I believe it was the Colorado School of mines. But I am not 100% on that.

[TA]

Hey Jeremy

This might be a bit off the wall but I've seen polycarbonate (lexan) used in hinge applications. It has excellent tear strength and is fatigue resistant.

Mounted between two plates in double shear it would be pretty cheap and easy to repair as well.

Cheers, Ted

[DJHache]

the problem with carbon is gluing it to metal. There are some pretty cool epoxies out there and I know that there is a thread somewhere in here on them, but why would you want to go through the hassle of making sure you have no stiction in the rodends when most teams have basic compliance issues in their a-arms and tie rods?

[Travis Garrison]

I have heard stories of a drexel car from a ways back that had some flex joints...From what I gather the car had composite flex joints (glass?)...maybe one of their memebers will chime in here eventually...

IMO if you ar going to do it, save some weight while you are at it. Make composite a-arms with composite flex joints. End the joint in a plate that you bolt to your chassis, and you're done. Should be less weight, fatigue shouldn't be as much of an issue, and you shouldn't have to worry quite so much about peel as you would if you tried to intigrate a composite flexure with a steel arm.

Edit: Just one other thing, on the bending loads, I would be willing to be that the bending loads are already in most of our setups since the sphericals do tend to stick, at least a little, even when new.


Travis Garrison
UW FSAE

[Ben Steele]

Again, I'm fairly new so if I'm wrong you guys should feel free to tell me to shove it, but I have to disagree with you Travis. The bending loads you have to consider when using rodends are not even close to those you have to design for when considering flexures, even though there is some stiction, unless you somehow exceed the range of motion that your particular mounting system allows. I very much doubt that anybody would overlook something like that.
As far as carbon A-arms are concerned, one of our guys did a little bit of experimenting with making carbon tubes (for a clutch handle) and it turned out to be a pain in the butt. He never could figure out how to separate the part from the mold. That's not to say that someone who could devote a little more time couldn't figure it out and it would be awesome if they did.
I'm really in favor researching the flexure idea, though I doubt Auburn will be trying it this year. It'll be cool to see what you guys come up with.

Ben Steele

[Lehigh FSAE alum]

Lehigh, my former team was to my best knowledge the first team to have flexures on thier FSAE cars. I did the study as a senior project and the flexures were fitted to our 1999 test mule and successfully tested over an endurance simulation. Our 2003 car went to design competition with them however they werent run in the events; we just saw it as being too risky. Last competition (2004) we had full composite a arms with integral flex joints. They were half the weight of steel a-arms (5/8 x .035). These flexures worked however the FOS was very low they werent run. The high angle of our upper A arm was a contributing factor as well as our design being a pullrod system; the forces on the lower arms are simply too high to get the neccessary displacement and buckling resistance. My senior study was thorough using mathematical optimization to attain the geometry;
The design can be done, especially on the top; however the car should be designed with them in mind: 13 inch wheels helps, pushrod, low angle with horizontal. direct bolted & bonded connection to square tubing on chassis (not neccesary, double shear plates in bracket would be fine).
BTW: the main aussie judge (little fellow, really nice guy, not sure of the name) absolutely loved them: no rod ends in bending; no inboard rod ends, no "sticktion"...
I would love to share the report/pictures, just email me at tac5@lehigh.edu

[James Waltman]

I'm not certain but I think that Drexel beat you to the flex joints. The only picture that I can find of the car is in the FSAE history pdf.
http://www.sae.org/students/fsaehistory.pdf
Bottom right of page 5 (45) - the yellow car. Flex joints in the drive shafts as well if I am not mistaken
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Lehigh FSAE alum:
BTW: the main aussie judge (little fellow, really nice guy, not sure of the name) absolutely loved them: no rod ends in bending; no inboard rod ends, no "sticktion"... <HR></BLOCKQUOTE>
Could that be Pat Clarke?

I would love to see your report. I will send you an email soon.

Ted,
I would be concerned with the chemical resistance of polycarbonate in structural applications. It's not resistant to many chemicals we use in/on/around our cars. It is not resistant (rapid attack or attack over short time period will occur) to acetone, brake fluid, lacquers & thinners, and gasoline to name a few. I have seen what acetone does to polycarbonate (thanks to Mike Waggoner's demo) it makes it very brittle very fast.

A chart of chemical resistance for polycarbonate sheets (http://www.palram.com/CachedImages/0000001/t009_r00016_v0.PDF)

[Denny Trimble]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Lehigh FSAE alum:
...Our 2003 car went to design competition with them however they werent run in the events; we just saw it as being too risky. <HR></BLOCKQUOTE>

I believe that's against competition rules, to show up in Design with something you're not running in dynamic events.

[Amos]

OK.... don't know much about flex joints, but isn't one of the main points of this competition to have a maximum mechanical grip? Therefore if you have a component that adds to stiction/friction of the movement of your "optimised" suspension system, wouldn't it be a disadvantage. Why would we want to risk the tyre compliance with the road?

ie. We have a rodend - that is placed correctly so it doesnt deal with bending and allows excellent motion, or we have a joint that relies on stiffness for its integrity.

I think in this case we cant really compare to F1. An F1 car has what sort of crazy stiffness springs? Add that little bit extra from flex joints cos they try to get as stiff as possible - makes no difference.
Now we have a 200kg car with damn soft springs, if we add just a little bit of stiffness to this through a joint aren't we just adding another uncertainty to an already "hard enough to work out" system????

Wont these joints add to the damping and stiffness of the car?

[ben]

F1 use flexures to reduce stiction, because rod-ends bind up under load, they are plain bearings after all.

They also allow higher installation stiffness.

I think all those advantages would also be realised in an FSAE application of flexures but as has been mentioned the larger suspension movements with shorter wishbones presents a great challenge. Hence my original post that I almost choose it as a major area to explore in my final year.

Ben

[RBbugBITme]

oops... didn't realize this was from 2004!

[exFSAE]

Doesn't matter if its old or not.. did you have a specific question?

Since '04 there's been at least one team (Lehigh) that ran flexures, carbon ones at that, on their suspension.

Given the very high amount of friction / mechanical hysteresis in some of these cars suspensions it might not be a terrible idea to look into. You'd be surprised how little even a ChampCar steel flexure, such as the one I have handy, adds to the system. Feels stiff as hell at the joint, but if you clamp that in a vice and try moving the wheel-side point around it moves A LOT with little force.

[RBbugBITme]

I was just gonna say I could take a pic of our 89 car, its hanging on our wall.

[smb96]

TU Graz and TU Darmstadt are using flex joints this year. Graz is using steel and Darmstadt carbon fiber in one piece with the a-arms.

[Anvit Garg]

RBbugBITme: can you please post that picture?

[Dan G]

Lehigh's arms from 2006:

http://evilengineering.com/gallery/d/2660-1/IMG_0147.jpg

http://evilengineering.com/gallery/v/SAE/FSAE/2006/

[RiNaZ]

anvit, check out the picture gallery section of this forum and you'll find a topic called ERAU formula hybrid 2008. It should have a close up on its suspension.

[Composites Guy]

Hi All,

I am the designer of the above Lehigh Univ. composite suspension parts. Below are some better pictures. The a-arms represented a significant amount of design and testing back in 2005-2006. The car you have pictured was completed the night before competition back in 2006, so it didn't fair very well. However, our 2007 car featured a similar suspension. Both cars have had extensive run-time now and I feel confident saying that the flexture a-arms are a good viable option for FSAE. They can take a good hit with a rock, wrench or cone, which is suprising to most people. These arms are vacuum infused unidirectional carbon fiber with an H80 Divinycell foam core. They are designed to flex only for a region of about half an inch on the inboard end of the arm. In this flexing region of the arm there is no foam core. Please note that these cars also feature a novel glass-fiber antiroll which is light as a feather.

I hope to write a paper (when I finally get some time) on these arms, giving the design process, and cross-sections cut-aways of the flexture portion of the arm.

Feel free to ask questions.

Many thanks to Tal (Lehigh FSAE Alum from the beginning of this thread) for his early work showing that flexing carbon was a possibility.
http://i110.photobucket.com/albums/n114/lehigh_jack/DSCN3987-1.jpg
http://i110.photobucket.com/albums/n114/lehigh_jack/DSCN5992-1.jpg http://i110.photobucket.com/albums/n114/lehigh_jack/FSAE06005-1.jpg

[Composites Guy]

All right ERAU... nice job with the carbon flextures! Can we see some better pictures of perhaps an a-arm off of the car?

[vreihen]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Composites Guy:

Feel free to ask questions.
</div></BLOCKQUOTE>

Since you offered, http://fsae.com/groupee_common/emoticons/icon_smile.gif I'm curious what your testing revealed was a safe range of motion (in degrees) from horizontal with this type of flexure, so as to not fatigue the joint?????

[RiNaZ]

composites guy, what type of adhesive do you use for your metal to carbon bonding?
are they all unis? how many layers of uni?
how come i dont see the foam in the hole where the pushrod goes thru?
was it cure in an autoclave?
do you have pictures of your feather light ARB? http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[DART-CG]

Here are some pictures of our flexijoints made of carbon fibre from Tu Darmstadt

http://farm4.static.flickr.com/3148/2694923351_32eec27427_o.jpg

http://farm4.static.flickr.com/3138/2695741776_d4f43e2158_o.jpg

They were tested statically and dynamically and our know used in our gamma2008

[Composites Guy]

Dart: Thanks for posting the pictures... the carbon is beautiful. We do so many structural tests here that carbon's asthetics are often overlooked (but I know asthetics will count in the commercial world.) I'm curious as to how the inboard metal joint is designed. Those little metal tabs look quite small. What is hidden under the paint??? Have you embeded some trick piece in the body?

vreihen: I considered 1" bump and 1" droop measured from ride height at the wheel(to comply with FSAE minimum supsension travel.) The range of motion was tested accordingly. I made up a sample a-arm which was put through this travel 10^6 times.

RiNaZ: I used Henkel (formerly soverign chemical) epoxy E2119, which is some lovely stuff to work with. It has the look and consitency of tar, and it stays where you put it, and looks good with carbon. You can't see the foam in the hole because I inserted a small G10 phenolic block into the core prior to infusion. This makes for a finished and durable piece once the hole is cut. I have never seen an autoclave in person:-) Lehigh's specialty is vacuum infusion without needing an autoclave. Sometimes post cure is done in a light-bulb oven, though a black trashbag in the sunlight would do the trick too. The a-arms are completely uni, but I won't tell you the layering scheme (perhaps in a future paper.)

Below are two pictures pertaining to the lighweight anti-roll device that I mentioned. The basic idea is that bellcranks, shocks, push-rod, etc. are in the same plane. Two seperate wavy fiberglass springs are attached in an x-shaped pattern to the bellcranks. As the car rolls, one spring goes into tension, the other goes slack (and does nothing). As the car bumps (without roll) both strips stay (nearly) the same length (and do nothing). Thus these wavy strips provide the typical characteristics of an anti-roll device whilst being VERY lightweight. The strip's layup scheme and wavy shape determines the spring rate.

http://i110.photobucket.com/albums/n114/lehigh_jack/untitled.jpg

http://i110.photobucket.com/albums/n114/lehigh_jack/DSCN6004.jpg

[RiNaZ]

impressive, i dont think ive seen the ARB back in 06'. Are those seatbelt parts? I cant really see if you're just bonding the glass to one side of the ends or it's clamped with the metal (im asking about the ARB).

do you know how much the henkel epoxy is?
And do you still have the old A-arms from the first time you did it? Im curious to know how long can u use the A-arms before it starts to delam or sag just from the weight of car.

[Dan G]

What are the design benefits of this type of joint, over say a plain spherical? Lower cost/weight? How about compared to a simple poly bushing? I'd think the flex section would make a Z shape pretty easily, changing the effective arm length and goofing with all of the suspension geometry values.

For the guys that have used them, what do you feel are the weaknesses of your design?

[Parker]

I remember doing a double take at the Lehigh paddock back in 2006 at those ARB's, followed by a fairly big "doh" moment wondering how I had never seen something like that. They were insanely light as well if I remember correctly.

[cmeissen]

Did you guys measure how linear or not the effect of the fiberglass ARB's were on your roll stiffness? Also were they used in the front and rear? Thanks

[RiNaZ]

what kind of glass do you use for the ARB? it looks like 6oz surfboard glass and it looks really thin, how thick is it?

DART-CG, is the arm a one piece? i dont see the seam line in the close up. What do you guys use for core?

[J.R.]

If you look closely on the left wishbone, in the bottom picture, towards the tyre, you will notice a seam.

I am also curious about the effect on suspension geometry. Won't the pivot point move slightly during movement? Has anyone done flex joints successfully out of steel (mentioned earlier)? As a team, we don't have much experience with carbon, but weight savings is always something that we are going for.

Also, did anyone ever answer the question of reducing tyre compliance and changing spring damping ratios? I can see a change in spring ratio from the material, but how would damping be changed?

I'm new to suspension so fell free to let me know if I'm making some really dumb assumptions.

[DART-CG]

@Composites Guy: There's no tricky piece inside the monocoque, only the front roll-hoop and the steering bulkhead to which the metal plates are mounted to. CFRP inserts in the shell of the monocoque connect them to those parts.

@rinaz: i didn't design the wishbones and our design would slap me if I tell too much but the A-arms are a hollow two-piece design. A specially designed silicon bag for the autoclave process was used to achieve the hollow design. Due to the huge geometrical moment of inertia a wall thickness of 0,8mm was achieved at a total weight of 140g for the largest wishbone.

[Composites Guy]

Rinaz: Those aren't seat belt parts... we have a CNC abrasive waterjet cutter... the metal just sandwiches the glass and gives an abrasion resistant place to mount a bolt. I measured lap shear strength for the glue joints at about 2000 to 2500 lbs per one square inch (lap shear strength)... which is plenty.

To answer your question about the a-arms, we are still running the originals. When designed right composites basically don't fatigue. The only real problem we have is that some of the spherical housings (outboard end) weren't made by the most competent machinists, and a little slop around the spherical has become a lot of slop. To correct this we can (and have) machined out the spherical housing and bonded a new one in. Mostly aluminum was used for the metal bits, but ideally (for both bonding and strength) I'd use 6AL4V Ti. (The inboard tabs on the picture above are Ti).

[VinceL]

I am also interested in how teams running flex joints are predicting their suspension's kinematics. I'm assuming you're not using off the shelf kinematics packages. Are you purely using FEA to analyze your suspension's geometry, or are you inputting FEA results into lumped parameter models? It seems to me that developing the model would be as difficult as designing the joint.

And once the car is built are you K&C testing your suspension to see if your model matches reality?

BTW: DART-CG, those are some very nice looking A-Arms. I can't wait to see your car and all of the other European cars at FSG.