Typical Force on front LBJ Stud...

[entx]

Hey...I'm just trying to find out what some of you guys see as a range for the static forces on the lower ball joint stud...

Here's what I have... the static loaded weight if each front wheel is 1320 lbf. Using a 3/4/5 rule for y/x/z. I have 3960 lbf(x-axis), 5280 lbf (y-axis), and 6600 lbf (z-axis) at the tire contact patch.

Now that I have this I set up some FBD to calculate the reaction forces at the lower ball joint to determine the loads each of the lower arms carry... This is also done with various steering angle to determine the max loads each of the arm would see whether in tension or compression...

So based on the geometry of the suspension, I have calculated the following loads carried by each of the links that make up the lower control arm for the max loaded condition... 8512 lbf(Tension) and 13519 lbf(compression), these loads don't raise any red flags, but when I do analysis on the ball joint studs...I am getting ridiculous amount of strain in the material...

For example...the outer lower ball joint I am receiving 0.0229 in/in of strain which based on typical steel MOE of 30E6 this comes to 687 ksi...that's out of this world...and this is based on the factory ball joint design...So either my loads are assumed to be way out there, or there is something that I am overlooking...

So that is why I'm wanting to see what you guys have seen as typical design loads that the lower ball joints sees, which based on the above loads of 8512/13519 is a resultant load of 15976 lbf...

 

[MikeHalloran]

>> Using a 3/4/5 rule for y/x/z. <<

That's where you lost me. What rule?





Mike Halloran
Pembroke Pines, FL, USA

 

[entx]

3G's in the lateral direction
4G's in the longitudinal direction
5G's in the normal direction...

so for the a corner weight of 1320lbs...this give the forces of the x,y,z axis as a multiple of each...

 

[MikeHalloran]

Doesn't the tire's friction coefficient limit two of those forces?



Mike Halloran
Pembroke Pines, FL, USA

 

[entx]

Yes it would limit the longitudinal and lateral forces, but I'm using this to help build into the design some added SF that would otherwise account for impact loads, fatigue...etc. Besides, steel is allot cheaper the lawyers...lol Not all of us can afford a dynamic simulation software. So I have to do it old school... which means I need to build in a bit of uncertainty in the designed loads.. Think of it this way...after all said and done it comes out to a SF of about 4 give or take....so there is quite a bit of margin in there...

My next option is to attach some strain gages to the ball joint stud...and measure the max strain....but I'm hoping before I do all that...that there would be someone that has done enough analysis on suspensions to remember some of the typical loads the LBJ saw for a typical car with a GVW of 4500lbs produced...

 

[desertfox]

Hi entx

Why don't you upload your FBD then we can see better what your problems are.

desertfox

 

[GregLocock]

I know exactly where the 3/4/5 comes from, and you are right, it accounts for real world impact loads. Non aero racing cars are usually designed to 1/2/3.

The exact loads are very dependent on your suspension architecture and geometry.

Sounds to me like you are probably on the right track, albeit a tad on the high side. What size ball joint stud are you looking at?



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[BobM3]

I also think you're in the ballpark but a bit high. It's been years, but I used to strain gage ball joints, calibrate them in the lab and then collect road data. Pot hole and curb strikes were the most severe events.

I'd gage the joints to pick up bending from lateral and longitudinal loads. The calibrations were done on multiaxial load simulators where I could put known (measured by calibrated load cells) loads into the vehicle suspension and measure the output of the strain gaged joints. Lateral calibration loads were input to the suspension at the static loaded tire radius. Longitudinal calibration loads were applied at the spindle centerline.

As I recall, peak loads were between 2000 lbf and 3000 lbf for the pot holes and curb strikes. Of course, those are highly dynamic events so the static measurements used for the calibration process aren't really valid. All I could really say was that a during a highly dynamic event a strain was produced on the joint that equalled the strain produced by a static known load applied to the wheel adapter in the lab.

 

[patprimmer]

And here was me wondering what Pythagoras's theorem had to do with ball joint loads.

I am trying to envisage a strain gauge attached to a working ball joint stud.

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

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

for site rules

 

[entx]

Thanks for the responses guys....

First thing...after looking over my calcs over and over again...I may have not carried the correct sign convention on one of my results, so this hopefully will reduce my estimated loads...also I didn't really account for the z-axis loads...and after some thoughts on this and a rough calculations with this...it appears that the z-axis will further help reduce the load on the lower ball joint by counteracting some of the loads induced from the x and y axis, but I won't know this for sure until I start over with some fresh paper and a clean slate....so I need to clear my desk of all the clutter so I can start over...haha

Once I do this I'll post them on here to get y'alls input...we can post pdf's here can't we? I see the link to post on engineering.com, but not sure what formats we can post...


But you know it's funny...BobM3 mentioned he was measuring the dynamic loads to be in the range of 2000-3000lbf as when I was calculating the stresses based on assumed loads and got the way out there loads...I looked at the ball joint itself and worked from there and it wasn't until I was working in the 3500lbf range that the stresses started to become a bit more reasonable with about 180ksi if I recall....so I know I have to be way off...I just need to re-think my methodology...

Also this is in the future....but I will be making some new studs for the ball joints...and for me to get the yields strength needed, or at least I'm assuming I will need once I do the load calculations...I may be using 4340 QT to get the 200ksi range...but my question is can you easily localize heat treat a part...I assume there is some compound that can be applied to insulate area of the part to not receive heat treat....I don't want to heat treat the whole part, just the area that is highly stressed so I can retain ductility in the reaming part...I'm not sure if this would be done by induction heat treat or just in an oven... I may just need to ask this question in the materials group section...

GregLocock

The ball joint's diameter where the highest stresses are is Ø.654", which is the point right below the tapered fit of the steering knuckle....and the distance to this point from the load center is about .8"

Patprimmer

It's not hard to attach strain gages to a ball joint... you just won't get much life out of it....but its very doable...

 

[BobM3]

There's a good factor of safety on the ball joints. Before gauging them we'd machine flats on them (the studs) to help locate the gages, drill several radial holes and drill a hole along the axis of the stud. The radial holes not only allowed us to route the wires, they served as stress concentrations for the gages. In my time at the company there were dozens of ball joints modified like this and I never saw one break. We were hard on the cars during commissioning of the rigs. The suspensions saw hundreds of pot hole strikes on the simulators.

 

[GregLocock]

Bill did you mean to type 2000 and 3000 lbf? That seems very low to me.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[BobM3]

Greg -

You mean Bob? It's been over 10 years and it's quite possible I'm not recalling the numbers too well. The ball joints did make good load transducers though (much better than control arms) so they must of had a decent amount of strain.

Entx - I believe that the large lateral loads that occur during pot hole strikes are the result of the moment developed due to the the offset from the vertical input at the tire and the jounce bumper hit. Calculate what lateral loads you would develop at the ball joints when you force the suspension up into the jounce bumper and create 30-40 g's of vertical acceleration.

 

[GregLocock]

Sorry Bob, yes I meant you.

A 1.125" ball joint with a 5/8" thread is designed for a typical service load of 7364 lbf, I'd have no qualms about exceeding that if I could guarantee that it was sensibly loaded, for occasional impacts.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GregLocock]

Just thinking about it I suggest you reverse engineer some other designs, a 5/8 thread sounds marginal for a heavy car like you are proposing. Big hint.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[entx]

This morning...I redid all the calculations...and as it turns out I did in fact drop a negative sign somewhere in the mixed which skewed my final answer...and also this time I accounted for the side load that the z axis introduces, even though is relatively small compared to the x and y forces..it still accounts for a small bit...

This time I am coming up with about 6242 lbf... just want to check and see if this is in the range from what you guys see for your calculations...now remember, this figure accounts for the SF that I mentioned in the first post, which if I just used the loaded wheel weight for all three loads, about 1320 lbf...then the 6242 lbf would then workout to 1560 lbf...based on an average 4x SF that's built in already...

This is at least a bit more reasonable...though when I put this force, 6242 lbf, into the stock ball joint stud...the max strain is about .009 in/in which again based on typical steel MoE 30e6 this comes to about 270ksi for max bending stress...which is a bit more reasonable as we can have materials to this level...

Does anyone out there know what material is typically speced for ball joints...not sure if there is a common material speced for this or if it is speced for the designed loads...I may get PMI and tensile testing to see what the original material is so I can feel a bit more confident in my hand calculations and my assumptions on my loads...

 

[entx]

Bob,

when you mentioned the 30-40 G's I assume you would mean that don't account for any longitudinal or lateral loads...right? Only account for the vertical load...right? Also is this what you guys saw for the momentary strain, when you guys measured them?? That's alot higher than I would have ever looked at...

 

[GregLocock]

I suggest you look at the Aurora rod-end catalog, it gives you thread sizes ball sizes and max radial loads. And I really suggest you check what size rod ends people use in similar applications. Big hint repeated.

I don't understand your comment about safety factors, 3/4/5 is based on measured loads, it has NO factor of safety (apart from a bit of rounding up) in it at all.

I'm a little dubious that a lower ball joint would see a max force less than that applied at the CP.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[entx]

Greg,

Is there some resource from ball joint manufacturers which has the intended loads for given ball joints??? I tried looking at some ball joint manufacturers that I can think of...Moog, TRW and lemforder are the only one that come to mind...

Also this may be a bit of a dumb question...cause I have never been able to find info on ball joints...but the size of the ball joint...is that typically referred to as the diameter of the ball??? If so these are then 1-1/16" that I am reversing...

 

[GregLocock]

http://aurora.thomasnet.com/viewite...precision?&bc=100|1050|1053|3001038&forward=1

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.