Pirate 4x4 banner

New Version of my 4-link Analyzer & Request for Help

471K views 442 replies 184 participants last post by  Will.The.Thrill 
#1 · (Edited)
New Version of my 4-link Analyzer / Calculator

Edit 28 December, 2008
Here is the link to the latest 4-link calculator w/ Vetteboy79's travel mod.
http://mysite.verizon.net/triaged/files/4BarLinkV3.1d.zip

Edit 13 April, 2006
Link to the 3-link + Panhard Bar calculator.
http://mysite.verizon.net/triaged/files/3LinkV1.0bBETA.zip
Please note that this is a BETA version. I have not checked everything and it is possible that there is a mistake in there somewhere. I haven't taken the time to do a more up to date version.

Edit 12 November, 2005

I got the V1.5 HTML from Benjamin Porter (aka MNBen here).

I posted it up on my server. I am going to try and update it sometime to make sure I know what I am doing...and will then attempt to get the travel stuff in there as well as some more features for the next version.

http://mysite.verizon.net/triaged/4linkcalcv15html/index.html

Edit 12/4/2004
Version 3.0 is posted.

Top new features:
-Travel Worksheet that finds the geometry at bump/droop
-Calculates sprung mass CG and "anti-squat CG" (vehicle CG - rear axle CG)
-Fixed errors when bars are parallel (all that is required now is one set of bars to be non-parallel in top view)
-Made the material spec's customizable (you can add to the list if you want)
-Added a text file that explains some of the spreadsheet.
-More color coding
-Moved all the calc's to another page to make room for the drawing on the main page.
-Saved it is an Excel 5.0/97 file so people with old software can use it (doing so made the file HUGE so you might want to save it as the newer version if you have it).
-Name changed to be more "proper"

http://mysite.verizon.net/triaged/files/4BarLinkV3.0.zip

Edit 1/10/2004
I just finished V 2.0. This one draws a picture of your suspension for you. As always if you want to screw with it there is no password to the protection...it is just there to keep you from screwing it up if you don't know what you are doing. I added some more input cells but haven't done anything with them yet...mostly I wanted to draw tires in the top view and position the CG but it is starting to look cluttered anyway.
http://mysite.verizon.net/triaged/files/4LinkCalculatorV2.0a.zip
\Edit

Just uploaded V1.5. Greg72 from CK5 has helped me get it looking a lot better with everything layed out in a more fluent manner. He also added some pull down menus for the link materials. If you click on the cell a little down arrow will pop up next to it. Just click on the down arrow and you will be able to pick from a list of choices. I also added an extra sheet with the coordinate system on it for reference.

http://mysite.verizon.net/triaged/files/4LinkCalculatorV1.5.xls

V2.0 is being worked on (by Greg) that will draw a picture of your suspension geometry automatically. He has already drawn up one for the 2D anti-squat excel program he made and is trying to update it for the new 3D spreadsheet (which doesn't sound easy to me!).

I am working on making it cycle your suspension so you can see how some of these values change as your suspension compresses or droops (articulation would be nice as well but I have no idea even where to start on that).

I am also looking for some other people to help out.

I need someone who knows all this math stuff (most likely another mechanical engineer) to check my work and see if my assumptions are valid.

I would also like to have someone actually use it to analyze their suspension with it and give any notes on how it works in real life.


Take a look at the entire discussion and some other important info in the original CK5 thread
http://coloradok5.com/forums/showfl...6301/page/0/view/collapsed/sb/5/o/all/fpart/1

Also jeepboyben has put my version 1.0 into html/java for those who don't have excel. It still has some bugs to it so I would still recommend the excel version for now. You can check it out here
http://www.isd623.org/ben/jp/fourlink/
 
See less See more
#2 ·
Eskimo said:
I've been playing extensively with v1.1, and REALLY appreciate it... It's AMAZING to see how much higher the forces are on the links of a portal axle if mounted in a similar fashion to a normal axle.. No WONDER why Sunray is mounting wngrog's lower links to the portal boxes!
Nice to see that someone likes and is using it!

Eskimo said:
A few things: it would appear that changing tire size has no effect on things... But, since the link heights are really what change the forces acting upon the links, it's just a reference field, and doesn't really need to be filled in? Just checking my logic on this.
[/B]
At the moment the tire size does nothing. I would love to make it so that when you changed that value it would move all the 'z' values half of the change...I just haven't figured out how to do that. When I originally made V1.0 I calculated the torque at the axle (which needed tire diameter) and then the link forces. For V1.1 I got rid of that step. Also that is one of the values Greg needs to draw the picture of the suspension. An original version he made in 2D that would just calculate AS and a simplified force calculation drew in some 8 sided tires (must have been bias ply swampers:flipoff2: ) to help you visualize it.

Eskimo said:
Also, in the "material selection" field, where the calculated "link force" is displayed... this is the force put on the links at what point? a certain amount of the vehicle's weight on that end of the rig? More a curiosity...
[/B]
This is where some nasty assumptions come in. I assumed that 100% of the vehicle weight was on the rear tires (doing a wheel stand on flat ground!), that the tires had a coefficient of friction of 1 (this would be right if you had racing slicks on asphalt...it should be more around .8 but what the hell) and that you had enough engine power to break loose the tires. I would like to include an input value for the friction but nobody would know it anyway and '1' in these equations simply means you input nothing. If anything these assumptions are conservative and help account for some of the dynamic loads that were not considered.

Eskimo said:
Lastly, you mentioned in your thread on the colorado forum that a FS of 1 or 2 isn't deisred (except for lower link FS bending), but what would YOU feel comfortable with? I'm trying to match them up evenly, but get a 3 or higher across the board.. that seem "safe"? (no, i will not hold you accountable :D )

Thanks a ton, this newer version looks great! [/B]
I had a long email discussion with Greg about this. Factors of Safety are kinda like fudge factors. Engineers use the term to take care of things they didn't take into account (like any of the dynamic loads which could be HUGE!, fatigue, etc). This is why I want someone who has a 4-link to run these numbers and give some real world input. Maybe their Bending F.S. is only .5 and they never bend their links and they drive like Camo?

Anyway it came down to I think 3 might be ok for a trail only or comp rig, but I think I would want to see more like 6 on a street driven rig. I think you will see that if you have a F.S. of 3 (on anything but bending) that the links look almost stupid weak. This is the area I would like the most input on as I have never built my own 4-link. Also for the lower links I think if you have the bending over 1 all the rest will take care of themselves. In the example in the uploaded file if you change the lower links to 2" .375" wall the bending is 1.02 while the yield is 11.38 and the buckling is 8.92. From what I have read on this site that link material doesn't sound unreasonable for a full size truck (the suspension geometry that is in there is what Greg is designing for his truck but the materials I have screwed with).
 
#5 ·
82FB said:
I can't quite read which is x, y and z though? and if you could spell out for a dumbie what they are measuring?
quoted from Triaged:
"X" lengths are from the rear axle centerline forward
"Y" lengths are half (1/2) of the horisontal sepperation.
"Z" lengths are all measured from the ground.


Triaged, thanks for answering the questions I had.. Wish I could be of more help in developing....
 
#6 ·
It is hard to do something like this without setting up a coordinate system. It seems like that is the biggest thing most people want to try and stay away from. I figure that if you are going to go through all the work of putting in a link suspension taking a tape measure and making 12 measurements isn't too much to ask.

Eskimo, thanks for responding to 82FB's question...I was out wheeling my gay little 2wd:D.
 
#8 ·
I've used V1.1 to calculate lower rear links for our Volvo portaled Range Rover (www.hotchallenge.com)
It seems I get almost stupidly heavy tube for the link when designing for bending because of the truck standing on them.

Also I think as long as I enter the right Z-coordinates it doesn't matter it the axles are portals or not?

Will try again tonight with V1.5. Probably welding within a week...

Tobias
 
#11 ·
tobbjo said:
I've used V1.1 to calculate lower rear links for our Volvo portaled Range Rover (www.hotchallenge.com)
It seems I get almost stupidly heavy tube for the link when designing for bending because of the truck standing on them.

Also I think as long as I enter the right Z-coordinates it doesn't matter it the axles are portals or not?

Will try again tonight with V1.5. Probably welding within a week...

Tobias
Bending is the weakist way to load a link. That is why they will have to be so big. The longer the link is the weaker it will be in bending. The way the stress is calculated is with half of the weight of your truck resting on the middle of the link. If you don't think you could ever get that much weight on the link then you could aim for a FS of .75 or something in that area. Your other options are using 4130 and maybe even think about having it
heat treated. I think the asumption of half vehicle weight is valid and have heard of people on this site bending almost stupid thickness of tube. I guess you could also just try and not bounce them off every rock you see but what is the fun in that.

As for portals mattering with this spreadsheet they don't. It is because I based all of the stresses on available traction. There is another way to calculate it all though that Stephen Watson just posted up in the CK5 thread (that I hadn't thought of before). He used breaking strength of the axles to determine how strong the links should be. I am going to go back through my spreadsheet and change it so you could do it either way. The only time this will happen is when you have the tire stuck in a crack and bound up. Then you just want to make sure that the axle will brake before any of the links will (with some factor of safety). the only other area where portals will make a difference is like you said the "z" location of the links. If you put the lower links at the centerline of the axle tube it will be alot higher than if it was a non-portal axle.
 
#12 ·
The spreadsheet is calculating the link forces based on the tire centerline I assume, which is probably 1/2 of the tire height you enter. This could be a major error point also, my 42's usually sit at about 19" from the ground to the hub center which is more in line with a 38" dia. tire as far as the spreadsheet is concerned. That should be noted in there somewhere I think. This would mess up your link forces too, the formula could think my torque centerline is at 21" which is about where my lower link starts when really the centerline is at 19" and the lower link starts 3" farther up.
 
#13 ·
Actually at this point the spreadsheet doesn't use tire diameter for anything at all. It isn't in a single calculation. As long as you take the measurements with the tires at trail pressure it should turn out right. It uses the tire contact patch friction force and computes the reaction forces in all the links statically (not exactly a good assumption but you have to start somewhere). After your comments in the CK5 thread I was thinking of going back through and doing everything based on axle torque (which it doesn't use or calculate at the moment). To do that I might have to include something for the rolling radius of the tire (which sure would make drawing the picture of the tire hard!...some cos function along with the equation for a circle:confused: ). This is the same reason portals don't make any difference...because at the moment I'm not using axle torque...just tire contact friction force.
 
#15 ·
portals *do* make a difference.. you just have to adjust your Z axis numbers to allow for it, unless you plan on mounting your lower link 5" below the axle tube... :p

Accoridng to www.killeraxles.com , a Mog 404 is 4 3/8" offset, so if you take your standard axle Z measurements and move them both 4.3 inches up, you'll see that the forces have increased.... dramatically!
 
#16 ·
Perhaps I was a bit unclear, but that's just what I meant.
I move the Z-coordinates up to where the link is on the axle tube+portaldrop+tire radius, instead of position on tube+wheel radius.
But there are no hidden calculations where the portal axle would produce errors? I assume no, since wheel size is not needed.

Also COG is lifted because of the portal drop...
Because of wheel radius too!

T
 
#17 ·
Also COG is lifted because of the portal drop...
You know I forgot to put that note in...the CG height is the sprung mass CG (which is what is always used in anti-squat calculations). It doesn't include any of the unsprung mass. The weight however is the total vehicle weight. I'll make sure to add those destinctions. The rule of thumb for sprung mass CG that I have heard is either camshaft centerline or the top bellhousing bolt. If you were to measure it the way these guys do
http://www.jeepaholics.com/tech/cog/#_Toc535118705
You would have to subtract the CG of the axles/tires to get a value usable for AS calculations. If anyone actually does measure their rig I could help you turn the overall CG into a sprung CG. As no one can say that AS should have any one exact value (and you should make it adjustable as well!) I don't see this as a major issue. I didn't do this spreadsheet so that you could get it exactly how you like the first time...but hopefully it will get you close enough that only minor tweeking is needed.
 
#21 ·
I followed along with gillespie's "fundamentals of vehicle dynamics" to get link forces, it might be interesting to compare techniques to see what happens with the portal/non portal numbers. He used some assumptions we should probably watch out for.
 
#23 ·
Now I've played along a bit more while at the same time drawing the basic geometry on graph paper.

I' redesigning the triangulated four link that is standard on a RR into something for the portals on the rangie.
Lower links are 720 mm 28" stock and I will try to extend to between 1000 and 1200mm (40-48") . Upper link (triangulated with ball joint at axle) is 650mm (25") long stock.
Vertical separation at frame is 8" stock and at axle 4". X-separation at axle is 4" (lower link mounted 4" in front of axle centre lina and boll joint at axle centreline (in X))

Having the axle, ground and link attachement points in a graph in Excel would help greatly!

A few questions:
COG position in X?
We give Z, assume Y is in the centre, but X?
I change antisquat by changing wheel base so it is calculated there I think?

Pinion rotation. I have played with stock Rover rear setup, which makes the pinion rotate up at downtravel.
Then I made portals with 38,5 and longer lower links and moved uppe link axle mounting ahead of axle centreline instead of at axle centreline (X) Then I got downrotation of pinion.
When moving axle mount rear of centreline I get neutral pinion rotation. I think i might start to understand this, but not really.

If I build three different frame mountings for lower links I can adjust AS from 55, by 92 to 128% Roll centre angle moves from 9,3 by 11,9 to 14,4 (same as stock rover)
It seems I can choos from low AS (55%) low(ish) RCangle or high AS and high RC angle.

Will post pics!

Tobias
 
#24 ·
Anti-Squat depends only on the "z" coordinate (bummer because that is the hardist one to measure!). If you wanted to know how much lift you will get in the rear you would need to know the CG in the "x" direction (along with a host of other variables!).

As far as pinion angle goes I would like to put something in to account for it but it starts to get ugly quick. It depends on IC placement as well as relative link length. For now I would sugest making a simple 2D modle of what your suspension would look like from the side and cycling it. All it would have to be is some poster paper and some thumb tacks unless you have a CAD program to do it on.
 
#25 ·
I reread what i splanned for V2.0
It says 2D is already finished!

I foolde a while with Excel and got it to draw 2D, but the other way around from what we are used to...

T
 

Attachments

#26 ·
Real life....

COOL!

I just ran my rear 4 link into the spreadsheet.
The results.

Anti-Squat 16.97 %
RC slope 0.16 in/in ("-" = roll understeer; "+" = roll oversteer)
RC Height 21.76 in
RC Angle 9.25 degrees ("-" = roll understeer; "+" = roll oversteer)

Safety factors for the original material used (1.5 x .100" 4130) [only lower link shown]

Link Force -5,205
F.S. Yield 5.33
F.S. Buckling 6.77
F.S. Bending 0.81
F.S. Rod End 2.22

I've wheeled this setup HARD for several years now. Some comments:
I went for neutral Anti-Squat to prevent wheel hop. That worked great. Uncontrolled wheel hop is non existent. However rear squad from weight transfer is very noticeable on road. You'll also notice the roll overseer numbers, and again, on road that could be an issue. I do have to keep the rear shocks turned up to run at highway speeds or things get 'scary'. I also increase the rear spring rate to help reduce the rear squat on verticals. For a mall wheeler these would be a problem, but hard core stuff it works great!

As for FS numbers. After about a year, I had a nice bow to my lower links. A testament to 4130, they never collapsed or buckled! Early this year I replaced them with 1-3/8" 1018 bar. No more worries. I've never broken a heim (knock on wood), but the Teflon liners have been totally crushed out. This year I replaced two heims with spring bushing on the top links to add some cushion. I've been happy with that.

To do it over again, I'd work on reducing the roll oversteer. and probably add a little bit of antisquat back in. Just to make it a little more forgiving onroad.....

I wish I had this when I did mine. It would have saved many hours of hand calculations.... :)

Keep up the great work guys!
 
Top