Rod end failures everywhere!

[osubeaver]

I have attented the formula SAE compeition last year and this year, and I don't remember seeing nearly as many rod end failures (if any) last year than this year. Maybe I just didn't notice last year, but this year I saw at least 5 failures of rod ends, with the most dramatic being the car who lost two front wheels! Does anyone think there is a good reason for this, or is it just poor use of rod ends?

[osubeaver]

I have attented the formula SAE compeition last year and this year, and I don't remember seeing nearly as many rod end failures (if any) last year than this year. Maybe I just didn't notice last year, but this year I saw at least 5 failures of rod ends, with the most dramatic being the car who lost two front wheels! Does anyone think there is a good reason for this, or is it just poor use of rod ends?

[drivetrainUW-Platt]

1/4" rod ends in shear....end of story http://fsae.com/groupee_common/emoticons/icon_frown.gif

[Wray]

Yeah, I think a lot of teams are down to 1/4" rod ends, and I still see a lot of rod ends in bending. I think some teams aren't really figuring out how big the loads in those rod ends are either. At some point, you have to either go with a larger rod end, or an alloy steel rod end that costs more.

[jack]

where is carroll smith when we need him...

[Denny Trimble]

My thoughts exactly, Jack.

It's a good idea to move away from rodends at the outer balljoints, at the very least. Our '05 car saved a lot of weight by going to sphericals on all of the control arms, and using rodends only for pushrods and tierods.

Don't skimp on rodend quality, either. They're expensive, but we've seen what happens when they fail. They can take down the whole corner of your suspension, put your car into a curb, or worse. Is that worth saving $5 each?

And, I almost forgot (assumed all teams would do this), please carefully calculate the loads and stresses on the parts you buy as well as the parts you design.

[ben]

We ran 1/4" high-misalignment Aurora sphericals on all wishbones and never had a problem. We did switch to the 1/4" rod end on the inboard ends for ease of manufacture and still had no problems.

Ben

[Jarrod]

anyone who had failures care to post what quality they were using? Load ratings on the different aurora series vary significantly, 2100lbs on MM-4, up to 5260lbs on an AM-4.

[Didier Beaudoin]

As far as I know, the team wholost two front wheels had spherical bearings, it's the welds on the a-arms that broke.

[MikeWaggoner at UW]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Didier Beaudoin:
As far as I know, the team wholost two front wheels had spherical bearings, it's the welds on the a-arms that broke. </div></BLOCKQUOTE>

When I was at WWU, we had a baja car (not on a team I was on) fail because of crap welding. I really think MIG welding should be banned from these competitions, because it's too easy to get poor penetration.

[BryanP]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by BryanP:
Consider the go kart. No go kart manufacturer would consider using 1/4 inch rod ends on their karts. This is for a vehicle that weighs 200 pounds and is powered by a 5,1/2 horsepower engine. Race cars live a very hard life. They are subjected to many bumps and loads that are not easy to simulate such as being dropped while being loaded onto the trailer, etc. etc. etc.
Remember, Before you finish first, first you have to finish.
Regards
Bryan<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by osubeaver:
I have attented the formula SAE compeition last year and this year, and I don't remember seeing nearly as many rod end failures (if any) last year than this year. Maybe I just didn't notice last year, but this year I saw at least 5 failures of rod ends, with the most dramatic being the car who lost two front wheels! Does anyone think there is a good reason for this, or is it just poor use of rod ends? </div></BLOCKQUOTE> </div></BLOCKQUOTE>

[1975BMW2002]

The most likely cause for a rod-end failing is loading in bending. Do not ever load a rod-end in bending. It is asking for trouble. The judges will not like it either. The earlier allusion to Carroll Smith for those of you who didn't get it is that he would harp on this. It is poor engineering. One easy way to wind up loading a rod-end in bending is to use a rod end at the outboard suspension point on say the lower arm, and using a push rod attached inboard of it on the lower arm. (Actually it would have to be inboard unless you could figure out how to attach it directly to the upright.) This will place the rod-end in bending, spelling disaster.

Use Aurora, not Mcmaster stuff.

We used 5/16 even though the numbers said we could use 1/4, but it is nicer to have that safety factor.

Still bumming over a fuel pump that died with less than 2 laps left and FTD at that point... Oh well.

Bill

[CMURacing - Prometheus]

wrt our testing failure, it was most likely due to the shitty parking lot we have and driving the car over a drainage gate. i suspect a&m's was the same at endurance, a shock load combined with some fatiguing.

so far, the mcmaster ones have worked fine, but they are two-piece, and i suspect the two-piece auroras would work too.

[Hoosier Daddy]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Didier Beaudoin:
As far as I know, the team wholost two front wheels had spherical bearings, it's the welds on the a-arms that broke. </div></BLOCKQUOTE>

That is not true. I pick up the right wheel - the one that didn't hit the lady. Looked like it snaped at the treads, both at the upper and the lower... Additionally, as I recall they sure didn't look like steel...

[Didier Beaudoin]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Hoosier Daddy:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Didier Beaudoin:
As far as I know, the team wholost two front wheels had spherical bearings, it's the welds on the a-arms that broke. </div></BLOCKQUOTE>

That is not true. I pick up the right wheel - the one that didn't hit the lady. Looked like it snaped at the treads, both at the upper and the lower... Additionally, as I recall they sure didn't look like steel... </div></BLOCKQUOTE>

Oh well, you must be right, I didn't see it, I was only told it was the welds that failed.

[BryanP]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by BryanP:
By the way, does anyone know if the lady is okay? I sure hope so.
Bryan<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Didier Beaudoin:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Hoosier Daddy:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Didier Beaudoin:
As far as I know, the team wholost two front wheels had spherical bearings, it's the welds on the a-arms that broke. </div></BLOCKQUOTE>

That is not true. I pick up the right wheel - the one that didn't hit the lady. Looked like it snaped at the treads, both at the upper and the lower... Additionally, as I recall they sure didn't look like steel... </div></BLOCKQUOTE>

Oh well, you must be right, I didn't see it, I was only told it was the welds that failed. </div></BLOCKQUOTE> </div></BLOCKQUOTE>

[Frank]

no-one would put 1/4" rod ends on a kart?

WTF are you talking about?

1/4" Heat treated rod ends are universally used in fsae

and don't try and talk about reliability, our 2004 car has some serious miles on it

(don't put rod ends in bending) ((not even a little bit of bending))

[Denny Trimble]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by 1975BMW2002:
...One easy way to wind up loading a rod-end in bending is to use a rod end at the outboard suspension point on say the lower arm, and using a push rod attached inboard of it on the lower arm.
</div></BLOCKQUOTE>

Don't forget about brake torque acting on the upright; it's one of the largest forces on the car, so even the lower balljoint in a pullrod car, or the upper balljoint in a pushrod car, will see forces that would put a rodend shank in bending. Don't do it!

[adrial]

You can put rod ends in bending...its not the most horrible thing in the world. You just have to design around that. Sticking a rod end in a threaded insert is a lot easier than pressing a spherical in.

[BryanP]

Name one brand of kart that uses them!
There is no problem with using 1/4 " rod ends. I guess I must be posting to the wrong topic.
Oh yeah, (don't drive on a rough parking lot) , ((not even one with little bumps.))
When a person has to swear they really don't have much to say.<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Frank:
no-one would put 1/4" rod ends on a kart?

WTF are you talking about?

1/4" Heat treated rod ends are universally used in fsae

and don't try and talk about reliability, our 2004 car has some serious miles on it

(don't put rod ends in bending) ((not even a little bit of bending)) </div></BLOCKQUOTE>

[osubeaver]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">(Actually it would have to be inboard unless you could figure out how to attach it directly to the upright.) This will place the rod-end in bending, spelling disaster. </div></BLOCKQUOTE>

So does anyone actually attach their pushrods to a node directly at the upright?

[Z]

Re: "don't put rod-ends in bending"

All rod-ends will always feel a bit of bending. The weight of the steering tie-rod puts its rod-ends in bending. I would be interested/amused to see the wording of any proposed rule banning "rod-ends in bending".

But more to the point:

How heavy are all the rod-ends on a typical FSAE car?

How much weight is saved by going to the next smaller size?

How bad do you feel when your wheels fall off because you figured you could save, err, 100gms on a 250kg car??

The above questions also apply to wheel bearings, which were apparently another common failure. These parts are amongst the most critical on the car. Why take shortcuts with them? I note on other threads that design loadings at each wheel are something like 1.5G X&Y, and maybe 2-5G Z (based on that wheel's static load). Groannn! The car should be designed to do a one armed handstand, PLUS shock load (aka safety, or stupidity factor, which we are all subject to), because one day that's what it will do.


The best way to save weight is by getting rid of unnecessary hardware, and there is a lot of that on most FSAE cars. Remember, if it ain't there, it can't break, weighs nothing, costs nothing, etc., etc...

Z

Bugger! 102 posts. I was going to stop at 100. (Gotta get some real work done...)

[sprintcar8]

The main reason that I see for these failures is that most team are running the rod ends in a horizontal position. This is bad because it limits the rod ends range of motion. If they simply turned them horizontally they would be less likely to bind up and fail. The racing surface was terrible this year, with huge pot holes in the track. Texas A&M hit one, that I know for sure of, and this is what caused their failure. I not sure if the teams took into account the over travel the suspension would go through in this type of situation. I also feel that the rod ends being used are too small for their application, of course to save weight. But look at what it costs in the long run.

[BryanP]

Frank,
My response to your post was far from complete. That is because in 23 years of teaching I have had students swear at me less than a handful of times. In each case the situation involved a person who was highly stressed and they were involved in something such as a fight. We place a very high value in my classes on treating everybody with dignity and respect. Needless to say I was rather upset at the time I responded.
Now to address your points.
First. " 1/4" heat treated rod ends are universally used in fsae"�. Just because something is universally used does not mean that it is the correct or optimum solution. In the 1960' s roll bars were universally about 4 or 5 inches below the drivers helmet. What was accepted practice by the best motor sport engineers in the world then is now regarded as horribly unsafe and illegal.
Your proof of your argument was"and don't try and talk about reliability, our 2004 car has some serious miles on it"�. Once again consider the race cars of the 60's. Many teams drove their race cars with the low roll bars for the racing life of the car with no problems. Does this prove that height of roll bar was adequate? A more recent situation was the reluctance of Nascar, particularly certain major teams to adopt the HANS device. We know all too well the outcome of that.
Designing a race car is all about making appropriate compromises. Initial choices in design will amplify throughout the design of the car. An example of this is the Gurney Eagle formula one car. In an effort to save weight many components of the car were built of magnesium, a highly flammable material. A result of this was the decision that the driver would be safer to be thrown from the car in the event of an accident rather than risk being caught in a fire so no seat belts were installed in the car. Is this good engineering? It is up to the individual designer to decide their design priorities.
Your design priorities will have far reaching consequences for many people particularly as you begin designing faster cars. As the designer you have complete responsibility for the car's safety. A poor design can cause death or permanent injury to you, A poor design can cause death or permanent injury to your driver, A poor design can cause death or permanent injury to other competitors, A poor design can cause death or permanent injury to spectators. If any of the above happen you, the designer will have to live with it for the rest of your life.
I have a strong personal bias in favor of light cars. However, In any area that I consider critical to safety and reliability I am willing to accept a weight penalty to ensure safety and reliability.
My reference to karts is because I have the highest respect for the professionals who make their living doing what is a hobby for most of us. If I am doing something radically different from what they are doing I will check and recheck my calculations very carefully.
Regards
Bryan<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Frank:
no-one would put 1/4" rod ends on a kart?

WTF are you talking about?

1/4" Heat treated rod ends are universally used in fsae

and don't try and talk about reliability, our 2004 car has some serious miles on it

(don't put rod ends in bending) ((not even a little bit of bending)) </div></BLOCKQUOTE>

[sprintcar8]

Bryan,
I totally agree with you. Weight is very important. But as engineers safety is or at least should be the utmost concern. I use lots of rod ends in sprint car racing. They are aluminum 1/2"-5/8" depending where its on the car. The designs have not changed over many years, and frankly Im not too worried about them because they are being used like they are supposed to be, in the vertical position where no bind can occur. The only time I have seen a rod end fail is in a crash that decommisions the car. In closing, let us not forget the fundamental canon of ASME that says, "safety is paramount".

[Wizard]

I saw more teams using spherical bearings in control arms than in years past, however I was disappointed to see that not everyone is staking bearings into the housings. With some practice it is fast and easy to do.

If anyone is interested in spherical staking dies email me.

ekm24@drexel.edu

[kozak]

rotating the rod end 90 deg,this would take care of the binding problem, does this cause any other problems, sprintcar8. also does any one have a pic of the spherical bearing setup that you are talking about.

[sprintcar8]

I have not found any problems as far as failure. However, they are still rod ends and can fail at any moment just like any other part. It does limit horizontal motion of the rod so mounting can somtimes be cumbersome. But in a suspension, you dont want horizontal motion, you need vertical, up and down motion. Just be sure not to force the rod into place as this will stress the threads right off the bat. Also be sure not to overtighten the jam nut, as this "squashes" the threads at the rod face creating a stress riser. Another problem that you may run into is how to make the mount for the rod end in the vertical position. The bars are only so thick so modifications may need to be made to supply sufficient mounts.
Hope this helps, let me know if you need any other info.

[kozak]

wait i think i missed something, what bars? and does any one have a pic of the spherical bearing set up.

[Marshall Grice]

sprintcar8, i kind dissagree with your view on the rodends. My thoughts are that you mount the rod end in the direction that requires the most angular displacement. For most fsae cars steering is the largest angle to accomodate so that means mounting the rodends with the hole perp to the ground. Also the tightening of the jam nut is a good thing. it creates a preload on the threads which eliminates cyclical loading on the threads thus reducing the fatigue on the threads. So you want the jam nut damn tight, obviously not stripped though. Typical preload is about 75% of the yield strength.

I do agree that rodends as ball joints is a horrible idea though. Far too heavy.

[Frank]

Bryan P

After:

A heaps of statics (maths)
B using strain gauges and a DA box built specifically for the job
C using a tensile machine to test a few different rod ends
D designing 4 SAE cars (the last two were ok, perhaps ask a few judges)

I have come to the following conclusions:

A 1/4" heat treated rod ends are in most cases the best choice for:
inboard a-arm joints, push (or pull) rods, and toe links

I would not use "non heat treated" 1/4" rod ends for these applications

Smaller heat treated rod ends (such as 10/32 or 6mm) might be OK for these applications, depending upon the surface encountered, the mass of the car, and the specific suspension geometry (which of course will effect the load, on each member). I'd be very careful for a while, and probably do some fluro magnetic particle testing on these units.

Rod ends used (deliberately) in bending such as is the practice on rustic racecars
((at the outboard end of an a-arm, to enable camber adjustment))
would certainly need a rod end larger than 1/4".

Personally, I don't like the outboard rod end design too much, but that's a long sordid argument....

As for your comments about roll hoops, HANS devices, and go-karts...

well, I ask again, what are you on about? We are talking about rod ends on FSAE cars, are we not?

This is what I call "going down in flames".

This is the last time I post on these forums. I've had enough.

Kind Regards to all the helpful people over the last few years..

Special Thanks to:

Charlie Ping
Denny Trimble
Micheal Jones
Scott Wordley
Big Bird
Handles
James Waltman
Vin Honda
Kevin Hayward
and of course Pat Clark http://fsae.com/groupee_common/emoticons/icon_smile.gif

(IMO search these people's posts, you'll get the best answers)

Frank
www.uq.edu.au/fsae (http://www.uq.edu.au/fsae)

[Denny Trimble]

Cheers Frank, it's been good talking with you, and I appreciate your contributions.

You'll have to make up another account so you can come back now and then to give us all hell...

That's 1000 posts for me, I guess I'll have a beer http://fsae.com/groupee_common/emoticons/icon_smile.gif

[cdkrishen]

hello all,

just to clarify any discussion about the team that lost both front wheels at the recent compeition, here are some pictures i snagged just as it happened. i did have to poke my camera between people's legs, which was rather awkward...

i am by no means a suspension guy...but i can see sheered rod ends and a ball joint that failed at the weld. also its amazing how their break lines poppped right off too.

http://home.cwru.edu/~cdk5/images/fsae/DSC01090.jpg

http://home.cwru.edu/~cdk5/images/fsae/DSC01091.jpg

http://home.cwru.edu/~cdk5/images/fsae/DSC01092.jpg

http://home.cwru.edu/~cdk5/images/fsae/DSC01094.jpg

[Denny Trimble]

How's this for a failure sequence:
1) Upper balljoint (aluminum rodend in bending from brake torque) failed
2) Brakes locked wheels to uprights, and the kinetic energy of the 60mph wheels was transferred to the upright assembly with no upper restraint. This spun the upright/wheel around the lower balljoint and tore the LBJ (spherical bearing) out of the a-arm
3) Tierod outer joint (aluminum rodend) and brake lines failed from the wind-up.

[Tony K]

Speaking of failures... did anyone catch the European Grand Prix and see the suspension failure on Kimi's car? Now that was impressive. I've never seen a flat-spotted tire cause that much vibration, yet, it was a rather large flat-spot. Add in a litte braking force, and you've got yourself some nice carbon shards where your suspension used to be.

[James Waltman]

Frank,
Thanks for hanging out here for the last few years

We have learned a lot from you too

I'll miss your choppy writing style


----------------
Calvin,
Check your PMs


.

[sprintcar8]

Marshall, I never said not to tighen the jam nut, I said not to overtighten. Also, be careful when talking about steering angle as compared to suspension travel and wheel geometry as I am refering to. Try an experiment as to what Im saying. Use a bar with a rod end and see which direction you have the most travel. Simply insert a bolt and then move the bar in the vertical and then horizontal motions with the rod end in each different position. Let me know what you find or if Ive confused you more. we need to make sure that we all learn from the mistakes made by others this year.

[syoung]

Thanks Calvin, they're... 'interesting' photos! http://fsae.com/groupee_common/emoticons/icon_redface.gif

Re: the Grand Prix, I'm interested to see if anybody thinks McLaren should have pitted Kimi? I know I wouldn't have changed tyres if I was the driver or race engineer.

[vinHonda]

Re: Kimi's failure.

As his race engineer, I would have yelled at him over the radio for flat spotting the tire in the first place trying to pass a BACKMARKER Villeneuve....

Then when he gets out of the car, whip a steering at his helmut.

I am a big Kimi fan and believe he's got talent..but if u want to win a championship, you have to put on your noggin and realize a tire with a flat spot on it isn't a fast tire.

Cheers,

[SD]

Has anyone heard of inboard 1/4-28 threaded 1/4 eye rod ends failing?

As others have mentioned and as I have found out the calculated loads indicate these should be more than sufficient.

Thanks,
Spencer

[Agent4573]

I'm sure someone out there has designed a system that will break a 1/4 inch rod end on the inboard side.

If you did the calculations and everything came out alright, as one of our alumni always says, "Trust the numbers"

[RacingManiac]

We DNF in 2005's enduro in detroit with a 10-32 rodend failure on the rear pullrod. One of our member sent it to a Bombardier's engineer and it was said to be a fatigue failure primarily. We upsized the rodend to 1/4-28 for England and have had no problem...

The thing is, the rodend on the competition car have had probably at least 2 month of testing on it. It was the smallest size rodend we've ever used and in hindsight we should've replaced them before the competition.....oh well...

[smokinstang65]

Does anyone have a picture of how they are locating press fitted bearings to the a-arms for ball joints? I'm curious to see the positioning and how they are linked to the tubular a-arms. Email me please at

rdl2108 at columbia dot edu

[Schumi_Jr]

We ran 10-32 rodends on our pullrods from 2001 to 2004 without an incident. Most of these cars weighed between 420-450 lbs. We used bearings from Spherical Bearing Corporation (sorry, don't remember the PN). We finally had a failure on our 2005 car the October after competing at FSAE and Formula Student. This car had Aurora Bearings and this particular bearing was a standard rodend- the ones from SBC were a higher grade piece. In 2006 we switched to a 1/4-1/4 rodend for our pushrods- primarily for stiffness reasons.

[kwancho]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by smokinstang65:
Does anyone have a picture of how they are locating press fitted bearings to the a-arms for ball joints? I'm curious to see the positioning and how they are linked to the tubular a-arms. Email me please at

rdl2108 at columbia dot edu </div></BLOCKQUOTE>

If all goes well, we'll be welding A-arms tomorrow night, and I'll email you some pictures of our setup afterwards. Hint: waterjet plate.

[Sathersc]

Aw http://fsae.com/groupee_common/emoticons/icon_frown.gif I want access to a water jet. Our A-arms are being done on a piece of wood with really big washers. Not my preferred method by any means.

I also just turned the pieces we're pressing the spherical bearings into tonight. I'll try to get a picture or two up tomorrow.

[kwancho]

We paid to have it outsourced. Only like $100. I'm going to try and convince our shop manager to get one. That'd be sweet.
Just to make sure, for those captures, you're welding them to the arms, then finishing the bore, then pressing, right? Welding it is going to get it out of round.

[flavorPacket]

are you gonna stake or use snap rings?

[Sathersc]

Snap rings. As far as the process goes, I couldn't think of a good way to do the snap ring groove after it's been welded (first year lack of experience). Instead I turned it down before hand. We'll see how it turns out. Our lathes are not the most precise either, so it's kind of a lose-lose do the best you can situation.

Here's how it's turned out so far:

http://www.miamiredhawkracing.com/gallery/generators/preview.php?imageid=182
http://www.miamiredhawkracing.com/gallery/generators/preview.php?imageid=172

[Andy K]

Whatever you do, don't weld with the spherical bearing in place. If you do, you'll notice that the bearing doesn't move anymore. http://fsae.com/groupee_common/emoticons/icon_wink.gif We made the mistake once, and I promptly machined a steel insert with the same dimensions as the spherical to be used while welding to properly locate it.

[BenoJ]

Sathersc - Why are you using a rod end for your outboard ball joint? Have we learned nothing from all the failures that have already occurred in competition, including those involving wheels falling off and hurting people? I believe if there is one lesson to be learned in FSAE, it is to not use threaded components in bending, especially one that sees any sort of cyclic loading, or you are just asking for trouble.

[kwancho]

Uh, so, basically, the A-arm fabrication tonight was a failure. Pictures tomorrow.

[kwancho]

Pictures!
This is the setup
http://img233.imageshack.us/img233/674/aarmfabrication029qi1.th.jpg (http://img233.imageshack.us/my.php?image=aarmfabrication029qi1.jpg)
http://img217.imageshack.us/img217/6405/aarmfabrication030wr4.th.jpg (http://img217.imageshack.us/my.php?image=aarmfabrication030wr4.jpg)
The steel block at the bottom serves to hold the capture in the right place. It works horribly. We're probably going to press a shaft in with a shoulder that holds the capture at the right height.

First uberjanky a-arm.
http://img245.imageshack.us/img245/2109/aarmfabrication031hi7.th.jpg (http://img245.imageshack.us/my.php?image=aarmfabrication031hi7.jpg)

Other things that went wrong:
Tubes cut too short leaving a whole lot of rodend thread sticking out
Crimping makes the tubes too wide leaving part of the tube sticking out the side
Spherical captures made out of some crappy steel, so we did this with our hands:
http://img201.imageshack.us/img201/6371/aarmfabrication033uq6.th.jpg (http://img201.imageshack.us/my.php?image=aarmfabrication033uq6.jpg)

But, in theory, the setup should work if we change the way we hold the capture.

[Schumi_Jr]

Alex,

I don't know what your plans are for the a-arm with tons of exposed threads. FYI the tech inspectors at FSAE may not let those through. We've had troubles at tech with 1/2 that amount of exposed threads.

We fixtured our spherical housings exactly as you described. I would recommend you undersize the locating "posts" by 0.002-0.004" so you can easily remove them after welding

[kwancho]

Yeah, we're going to buy more tube and make more inserts.
We'll definitely do the post. We were worried at first about not being able to take them off, but it's still a far better method than what we have now.
Thanks for the reply.

Out of curiosity, I assume you had different posts for upper and lower a-arms. Were the posts just slide fits into the plate?

[BeaverGuy]

Alex, why are you crushing the tube ends? I assume it is to give you more area for welding to your spherical capture. Have you considered using a setup with tabs on the spherical capture instead of it just being round? You would notch the tubes so the tabs slide into the tube then weld it up. Here is a picture from manufacture of our '05 A-Arms to show what I'm talking about.
http://home.comcast.net/~joshuagillett/Picture_022.JPG
I wish I had a better one but I cant seem to find the others at the moment.

[kwancho]

Yeah, for more welding surface. No longer a fan of the crimping. I'd actually done the tab setup you described for a project over the summer, though in the middle of the arm to join 3 tubes together. That seems pretty simple though. We might have an issue in fitting into the double-shear clevises in the uprights, though. Check the Y-shaped upright pics on our website.

Oh, no problem with water getting into the tubes and corroding? Do you think you lose any stiffness from not picking up the top and bottom half of the tube?

Here's another thought:
Waterjet a sorta triangular shaped capture, and weld a perpendicular plate that you weld to. Means no semicircle grinds in the tubes.
Here's a picture:
http://img408.imageshack.us/img408/5708/part1xl0.th.jpg (http://img408.imageshack.us/my.php?image=part1xl0.jpg)

[billywight]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Here's another thought: </div></BLOCKQUOTE>


Remember that welds in tension are bad, shear is good...

[BeaverGuy]

Didn't check for corrosion in any of those A-Arms but they have been rebuilt a couple times due to crashes and the ends are now welded shut because of concerns that water might get in. I didn't do the design but the only issue I would see with stiffness would be in a bending of the tab which shouldn't happen. However, in '06 the tab length was increased to have more weld area. As far as I know there have been no issues with them on either the '05 or '06 cars which are still both running strong.

[Sathersc]

We're using rod-ends simply because of inexperience and time constraints. I've been fighting an uphill battle even to get some ball joints on the car. Hopefully before the competition I'll have convinced my team and the shop advisor that we should switch to ball joints all around. I intentionally left enough room in the design for them, so it's not a hard retrofit.

It's hard to get a first year team, most of whom haven't seen the sophistication of these cars to really put the time and effort required into putting together a higher teir car. But I digress, that's another thread.