Determining load on suspension links???

atomicjoe23

Well-Known Member
Joined
Feb 2, 2008
Posts
4,058
RDC Crypto
73
Location
Silverdale, WA
How would you determine the load placed on suspension links (in this case let's say a 4-link for a solid rear axle), worst case scenario?

Here is the info I think that I will need:

-sprung weight
-unsprung weight
-total vehicle weight
-corner weights
-link lengths
-wheelbase (?)
-trackwidth (?). . .maybe distance between the pivot points would be better, but I'm not sure that this would even matter. . .

plus the properties of the material being used (ultimate tensile strenght, yield strength, % elongation, % reduction in area, and modulus of elasticity, area or volume of the link. . .OD & wall thickness). . .the link info would only tell you if your selection of link material and size will be adequate for your vehicle

Am I missing anything?

Does anyone know an equation, or series of equations, I can use to calculate the worst case scenario load a suspension link will see?

Thanks for the help!!!
 

rickf

Well-Known Member
Joined
Oct 27, 2007
Posts
810
RDC Crypto
60
Location
Temecula, Ca
The g-force is the tricky one, that's what breaks things.
Getting accurate g-force calculations are really important (same force over shorter time = greater g-force).
There are engineering programs that will calculate those values, above my pay grade.
Many of the high-end fabricators have them.
 

Kritter

Krittro Campbell
Joined
Jun 18, 2001
Posts
10,914
RDC Crypto
299
Location
Chino, CA
Assuming simplest form possible a 4 link of just "links" only, with shocks on axle only sees braking and acceleration forces in the form of tension and compression since they are a 2 force member. Assuming no binding.

Start there.

What are you trying to accomplish by these calcs?

Once you figure those out, then start with shocks mounted on trailing arms.
 

atomicjoe23

Well-Known Member
Joined
Feb 2, 2008
Posts
4,058
RDC Crypto
73
Location
Silverdale, WA
Since I started my new job as a toolmaker apprentice I have been a LOT about materials, design, fabrication, etc., etc., etc. . .and I have really started to question some of the things going on in the off-road community (300M and aluminum links for rock crawlers/KOH rigs are the two that I'm currently rolling around in my head). . .since the aluminum links that I've seen are on rock crawlers/KOH buggies then assuming that the shocks are directly over the axle(s) is a pretty safe assumption.

I have found some formulas and I have the data. . .so now I'm going to compare the yield strengths, each links resistance to bending (as a simple beam supported on two ends with a load in the middle. . .greatly simplified, but a good way to directly compare solid aluminum links to tubular steel links), and do a cost vs. weight savings and cost vs. resistance to bending analysis and see if they are as good as some people are claiming to be. . .

. . .I've pretty much written 300M off as being way over-priced and the trade-offs in increased brittleness and a smaller margin between yield and failure (plus seeing three $1K 300M axleshafts snapping in a small buggy) have made me decide that 4340 is better in a torsioinal application (axleshafts) than 300M is. . .although 300M may still be a worthwhile option for gears. . .haven't done all the math on that yet.

. . .I've never been one to blindly follow the crowd. . .they "why" and "how" are very interesting to me.
 

Drayke

Member
Joined
Mar 9, 2007
Posts
20
RDC Crypto
1
Location
Torrance, CA
There is no real simple equation to determine loads in structural members for custom designs. You need to start with a FBD at the earliest and use your known geometry to locate the applicable forces on your links. From there you can apply whatever "worst-case" force you want, for example: 10g frontal impact on the rear tire. I've never done structural analysis on a link but my intuition tells me that bending forces are going to be the driving factor. A very simple analysis would probably be to analyze each link individually with a shear and bending moment diagram as a starting point.
 

scottm

Well-Known Member
Joined
Feb 7, 2005
Posts
1,902
RDC Crypto
601
Location
scottsdale AZ
Website
www.cemaz.com
As rick said, g-loads, or dynamic forces are the problem, and there are an infinite possabilities to consider. A rain rut may produce a 50 g shock but only compress the suspension 3 inches. Or you could come down on a funky flat landing and get a sustained 50 g shock that bottoms your bumpstops and deflects your chassis until the shocks go metal-to-metal, causing the trailing arms to bend. To do this right you would have to fit everything with accelerometers and drive it around, collecting tons of data on a million different variables. This will let you determine the minimum amount of material required to do the job with a certain factor of safety. Buildings and bridges have a fos of 5 to 10. Production cars and trucks are around 3. Military aircraft are only 1.5. The lower the fos, the lighter the vehicle and therefore more payload it can carry for any gross weight. Military pilots are considered relatively disposable, so they make do with a minimum fos. This whole process is called 'optimum design'. No one in offroad can quote you specifics simply because only companies like NASA and Lockheed need to and are capable of this level of engineeering. You could do it the hard way - start out real light and beef up as needed, or just go with with is known to work and live with greater weight.
 

atomicjoe23

Well-Known Member
Joined
Feb 2, 2008
Posts
4,058
RDC Crypto
73
Location
Silverdale, WA
There is no real simple equation to determine loads in structural members for custom designs. You need to start with a FBD at the earliest and use your known geometry to locate the applicable forces on your links. From there you can apply whatever "worst-case" force you want, for example: 10g frontal impact on the rear tire. I've never done structural analysis on a link but my intuition tells me that bending forces are going to be the driving factor. A very simple analysis would probably be to analyze each link individually with a shear and bending moment diagram as a starting point.

Thanks. . .this is the direction I had ended up going. . .determining section modulus, and then starting to plug and chug in the formulas I have found in the Machinery's Handbook and in a couple of statics books that I have. . .so far what I have done has definitely been simple analysis (I haven't reached ~dynamic analysis at all yet). . .

. . .what does FBD stand for?

You seem to know what I'm talking about so maybe one of you two can help answer a question for me. . .how does a suspension link fall into beam analysis? It's supported at both ends, but it's supporting weight (to an extent) on both ends. . .one end supporting the chassis and the other end supporting/supported by the tire/wheel. . .not sure where to go with loads to apply here. I was thinking of applying ~1/4 of the vehicle weight to the chassis side of the link (which is probably more than it really supports, but I don't know what ratio of weight the upper and lower links carry respectively. . .is it equal, or does one carry more of the load than the other??? I don't know and I'm not sure how to find out. . .other than asking because I don't have the resource to get hard data on it. . .I don't think. I guess I could start hooking spring scales up to the different links with no load and then slowly letting the chassis load the links. . .I'd need at least 4 spring scales capable of at least 1000 lbs. apiece I would think). . .anyway, there are a ton of variables and since it's just an approximate analysis I will have to make a lot of "educated assumptions". . .the data out will only be as good as the data in, but you gotta start somewhere, right.

As rick said, g-loads, or dynamic forces are the problem, and there are an infinite possabilities to consider. A rain rut may produce a 50 g shock but only compress the suspension 3 inches. Or you could come down on a funky flat landing and get a sustained 50 g shock that bottoms your bumpstops and deflects your chassis until the shocks go metal-to-metal, causing the trailing arms to bend. To do this right you would have to fit everything with accelerometers and drive it around, collecting tons of data on a million different variables. This will let you determine the minimum amount of material required to do the job with a certain factor of safety. Buildings and bridges have a fos of 5 to 10. Production cars and trucks are around 3. Military aircraft are only 1.5. The lower the fos, the lighter the vehicle and therefore more payload it can carry for any gross weight. Military pilots are considered relatively disposable, so they make do with a minimum fos. This whole process is called 'optimum design'. No one in offroad can quote you specifics simply because only companies like NASA and Lockheed need to and are capable of this level of engineeering. You could do it the hard way - start out real light and beef up as needed, or just go with with is known to work and live with greater weight.

Thanks for the FOS data. . .I would have thought an FOS of 4 for a vehicle and less for a race car (around 2, maybe less. . .maybe 1.5 like a plane as long as the vehicle survived).

I figured that most custom off-road fabrication shops don't do a ton of engineering on their products sheerly out of lack of resources to do so (money and equipment), but I would think that some of the larger manufacturers would have to do some engineering on their mass-produced and distributed parts if for no other reason than to cover themselves from a liability point of view.
 

Drayke

Member
Joined
Mar 9, 2007
Posts
20
RDC Crypto
1
Location
Torrance, CA
Sorry, FBD stands for Free Body Diagram and for any mechanical or dynamic analysis, an FBD is always the first step. If you screw up the FBD then the rest of your analysis will be incorrect. It is basically a diagram for a body which shows ALL of the applied forces and their directions. For a link, the forces I can think of are the reaction forces at the chassis mount, reaction forces at the axle, shock loads, link weight, an applied loading case, and lastly the chassis weight. Unfortunately the weight of the chassis as applied to a suspension member is never as straightforward as 1/4 of the total weight. There is an equation, which I have since forgotten and don't want to look for, that exists which will allow you to calculate this figure based on: total vehicle weight, length of the suspension member, and the unsprung weight and possibly something else, I just can't recall off hand. For a purely static analysis you can take the FBD and turn it into a beam analysis (google Shear and bending moment diagram) and find out how big your link needs to be. UNFORTUNATELY, the result you're going to get from a beam analysis isn't going to be even close to the results you'll get from a dynamic analysis. Aint that a b**ch. Hopefully that gets you started at least.
 

atomicjoe23

Well-Known Member
Joined
Feb 2, 2008
Posts
4,058
RDC Crypto
73
Location
Silverdale, WA
Sorry, FBD stands for Free Body Diagram and for any mechanical or dynamic analysis, an FBD is always the first step. If you screw up the FBD then the rest of your analysis will be incorrect. It is basically a diagram for a body which shows ALL of the applied forces and their directions.

Thanks Drayke. . .I know what a FBD is, I just haven't seen the term abbreviated before and it didn't cross my mind as such. . .

For a link, the forces I can think of are the reaction forces at the chassis mount, reaction forces at the axle, shock loads, link weight, an applied loading case, and lastly the chassis weight.

Thanks! That's what I thought the loads would be.

Unfortunately the weight of the chassis as applied to a suspension member is never as straightforward as 1/4 of the total weight.

That's what I thought.

There is an equation, which I have since forgotten and don't want to look for, that exists which will allow you to calculate this figure based on: total vehicle weight, length of the suspension member, and the unsprung weight and possibly something else, I just can't recall off hand.

Would you happen to remember what book (or type of book) the formula was in? I'm sure I can find it if I know where to start looking for it.

For a purely static analysis you can take the FBD and turn it into a beam analysis and find out how big your link needs to be. UNFORTUNATELY, the result you're going to get from a beam analysis isn't going to be even close to the results you'll get from a dynamic analysis. Aint that a b**ch. Hopefully that gets you started at least.

I hear you. . .the beam analysis (for a beam in both bending and shear) is more suitable for axle shafts than it is for links. . .I have already started a beam analysis for axle shaftsso I'm familiar with what your talking about.

Thanks again for the help! Greatly appreciated and lets me know that I'm on the right track.
 

scottm

Well-Known Member
Joined
Feb 7, 2005
Posts
1,902
RDC Crypto
601
Location
scottsdale AZ
Website
www.cemaz.com
Actually a beam analysis is exactly what you need for the lower link. There is a force at each end with an opposing force in the middle, like an inverted teeter-totter. The middle force is the coil-over when static, and co plus bypass when dynamic. And the static numbers are easy, but entirely useless except for an academic exercise, since the dynamic forces will be 30-40-50 times greater than static. A fairly easy way to determine dynamic loads would be to apply strain paint to a light trailing arm, something sure to bend or at least flex. I am curious about this too, as some arms hold up just fine but are far lighter than my monsters. Maybe trailing arms by conventional design wisdom are much heavier than they need to be.
 

atomicjoe23

Well-Known Member
Joined
Feb 2, 2008
Posts
4,058
RDC Crypto
73
Location
Silverdale, WA
Actually a beam analysis is exactly what you need for the lower link. There is a force at each end with an opposing force in the middle, like an inverted teeter-totter. The middle force is the coil-over when static, and co plus bypass when dynamic.

Assuming there is a C/O or bypass mounted in the middle, for this forum a safe assumption. . .for the vehicle I am currently working on (a '49 Jeep Willy's P/U truck) the shock will be located over the axle rock crawler style (my next two vehicles will be conventional desert suspension set-ups). . .

And the static numbers are easy, but entirely useless except for an academic exercise, since the dynamic forces will be 30-40-50 times greater than static. A fairly easy way to determine dynamic loads would be to apply strain paint to a light trailing arm, something sure to bend or at least flex. I am curious about this too, as some arms hold up just fine but are far lighter than my monsters. Maybe trailing arms by conventional design wisdom are much heavier than they need to be.

I am unfamiliar with strain paint. . .links???

Thanks for pointing this out ScottM!
 

partybarge_pilot

Well-Known Member
Joined
Feb 19, 2002
Posts
6,794
RDC Crypto
1,628
Location
Easton, KS
Hey Scott, you forgot to list when the driver blows a corner and uses the trailing arm to grade the berm on the side of the road calc.......
 

rich1

Well-Known Member
Joined
Jun 6, 2007
Posts
785
RDC Crypto
83
Location
Soldotna, AK
^^^that is called the screed calc. HAHA.
 

scottm

Well-Known Member
Joined
Feb 7, 2005
Posts
1,902
RDC Crypto
601
Location
scottsdale AZ
Website
www.cemaz.com
Ha yeah try designing for crashing conditions! Heck mine drag just getting off the trailer. The proper name for the stress indicating stuff is Stresscoat. It cracks at certain levels of tension and is a lot easier than fitting everything with strain gauges..

http://stresscoat.com/SCManual.pdf
 

Drayke

Member
Joined
Mar 9, 2007
Posts
20
RDC Crypto
1
Location
Torrance, CA
A fairly easy way to determine dynamic loads would be to apply strain paint to a light trailing arm, something sure to bend or at least flex.

In order to obtain the most accurate data possible (in terms of strain readings), you do not want the trailing arm to visibly yield. When loaded if the material "lets go" and sees plastic deformation your strain readings will either be way higher than what actually caused the yield, if not maxed out by your instrument.
 

atomicjoe23

Well-Known Member
Joined
Feb 2, 2008
Posts
4,058
RDC Crypto
73
Location
Silverdale, WA
In order to obtain the most accurate data possible (in terms of strain readings), you do not want the trailing arm to visibly yield. When loaded if the material "lets go" and sees plastic deformation your strain readings will either be way higher than what actually caused the yield, if not maxed out by your instrument.

I haven't read the StressCoat .pdf that ScottM posted up yet so I don't know what ranges this stuff is offered in, but it seems that it would be available in different stress ranges. . .so you would pick one with a stress range below the yield strength of the material and work your way down (hopefully) until you find the correct stress range that the link is subjected to. . .once you know the actual stress that the link is being subjected to then you would know what your factor of safety was as well. . .you could decide if they were overkill or not. . .

. . .but I would agree, plastic deformation is not what we want and would be bad, keep it in the elastic range. . .
 
Top