Three-link, need some suggestions

[matty b]

Well Im starting to plan out my project and Im looking for some suggestions. Im building a 65 malibu and after much reading Im a bit stumped on the rear suspension plan. For starters Id prefer to retain the stock frame because of cost and if I came across a frame that fit my budget Id jump all over but like so many things in life that probably wont happen. So, through my research Ive seen two things, first is that a three link is great for handling and that a triangulated 4 link is pretty damn good too. Being that the chevelle has a triagulated 4 link stock I must ask, Is it worth the time and expense to change to a diferent supension? I know I have a wide array of off the shelf bolt in parts that dont look half bad at all and I wonder if those will get me a rear suspension set up that will be capable as a canyon carver which is what i plan for the car to be. I would like the car to be able to hang with a vette or bimmer if I want to and since I cant ride a sport bike anymore yet love mosquito ridge rd Id like to be able to drive it instead. Ok, so if the stock design is not that great and a three link is prefable I have a few more questions. I prefer to make things whenever possible and from the pics Ive seen the dont look terribly complicated but I havent been able to find much if any info on choosing a preferable length for the links just that longer links are preferable and that the lower links should be longer than the top. Beyond that I dunno how I should set it up dimensionally. The setup on the IImuch car looks fairly straight forward and If I went that route Id probably copy it quite a bit. I figured I could setup a crossmember between the stock frame rails, install a crossmember for the coilovers and weld on brakets where needed and call it good. Way less invasive than a drag race 4 link and I think it would get me the performance I would like. If any of you guys can chime in and give me your recommendations and any hints on setting this up it would help me tremendously.

[Lowend]

I got here 1st !
http://www.lateral-dynamics.com/

[matty b]

Ive seen that but I figured it was only for camaros :hmm::hmm: if im wrong then this could be a route to take although Id prefer to build something out myself if I can figure out what dimensions it should be. Anyone have any links on figuring that out? all i come across is stuff related to 4 links and drag racing but nothing about a three link and the associated calculations. If it becomes to complicated Ill start looking for kits, with lateral dynamics and ame as the two places that ive seen that sell a kit. also am i overlooking the capabilties of an upgrades stock suspension? If it will work Id rather by bolt in parts and call it good. I already have a 12 bolt posi in it so I save a bunch if I dont go to a 9 in although I like them better for a few reasons, cost of upgrades being the biggest one.

[MrQuick]

with a few bolt ons and changes you can get that tri-four to work real good. Don't let the intrigue of a 3 link to get you side tracked.

[jerome]

You really really need to understand what you are doing before you start welding. I can't stress this enough. You need to do all the calculations. Here's what you are looking to optimize:

antisquat: depending on application, you want different things, but you could start at 100%. In side view, antisquat line is from center of contact patch to center of gravity height at the front wheel center. 100% antisquat is when imaginary intersection of upper and lower links falls on that line.

roll steer: looking for slight understeer. that means your roll axis needs to angle up from front to back in side view. for parallel lower links (in plan view), the roll axis is the angle of your bars, so you can't optimize everything with this. you want to choose converging lower links. This way, your roll axis is drawn from your panhard bar height to the imaginary intersection of your lower links. As long as your panhard bar height is slightly above this convergence point height, then your roll steer will be good.

side view swing arm length: shoot for over 50 inches. too short gives you brake hop. defined by distance in side view from rearend to intersection of lower and upper links

vertical distance of link attachment from rearend centerline: as long as possible without touching ground when you get a flat tire.

link distance from center of rearend: as close to wheel as possible

PHB length: as long as possible, attach to drivers side frame, pass side rearend

upper link length: make it so antisquat remains constant as rearend undergoes travel

upper link placement: on passenger side of rearend will somewhat lessen the effects of driveshaft torque

coilovers: as far out and as vertical as possible

Hope that helped, I know it's really hard to grasp, but if you're serious about getting it right, start drawing it out, and asking for help on here with specific numbers, alot of people will just do it for you, alot have already made spreadsheets for their own vehicles.

Jerome

[MrQuick]

++1 excellent post

watts linkage would work better with a stock configured set up.

[matty b]

that helps a ton, got a question, would the calculations for a three link correlate closely to that of a 4 link?? if thats the case I have more than enough information at my finger tips to come up with some beginning numbers. Also I came across a watts link setup from a place called totally polished. one of the truck mags did an install on it and am wondering if anyone has run one of these and what they thought about it. thanks alot

[Randy67]

I did a 3-link for my 67 el camino and I found out a few things along the way. My car isn't running yet so I can't say how it works yet, hopefully in 3 months I will know more.

1. The angle of the rear shocks limits the length of the panhard bar. I have a 36" bar but could have had it as long as 40-42" if the shock wasn't in the way. The bar isn't straight across as viewed from above, since the shocks required the frame mount to be about 2" farther back than I wanted. Or a longer mount is needed on the frame, I used a stock car mount to save fab time. A Watts link would probably be a better idea, since the shocks wouldn't affect it as much if the center is frame mounted and not on the rear axle.

2. I used the stock crossmember for the 3rd link, but did brace it quite a bit. Still, it would be better to make a stronger crossmember.

3. With the 3rd link center mounted (and not wanting to cut up the floors) the adjustment are limited. The pinion yoke gets close to the crossmember when the suspension is compressed and the floor is right above the crossmember. El caminos have less floor space above the axle as well (chevelles have more room, not sure how much more) so I had to put the link right above the diff housing. If you are willing to cut more out, you will have more adjustment options. Or you could just have the 3rd link offset to the pass side.

4. I was able to run tailpipes, but el caminos have the fuel tank 6-8" farther back than a chevelle, so you may not be able to run tailpipes. My tailpipes run over the axle and the panhard bar/brace and then drop down. Side exit exhaust would be an option.

Hope some of this helps.

[Norm Peterson]

would the calculations for a three link correlate closely to that of a 4 link?? if thats the case I have more than enough information at my finger tips to come up with some beginning numbers. Also I came across a watts link setup from a place called totally polished. one of the truck mags did an install on it and am wondering if anyone has run one of these and what they thought about it. thanks alot
The anti-squat will correlate well, since in side view there isn't much difference between the projected view of a pair of converging uppers vs a single fore/aft single link in true length, and the PHB or Watts link can normally be ignored. Only the vertical and fore/aft pivot coordinates of the upper and lower links matter.

The roll steer won't correlate, as the PHB or a chassis-mounted Watts link pivot move downward as the rear ride height drops, while the virtual intersection of the converging uppers rises with a dropping ride height. A diff-mounted Watts link pivot doesn't migrate at all. That affects how the amount of roll steer varies as the suspension moves.


Norm

[matty b]

well I think if i go this route Im gonna go with a diff mounted watts link, Ive seen a few designs that use it that I like. Griggs racing uses a diff mounted one and if they do it I dont think I would argue with it. So it appears that the biggest thing I have to do is figure out my pickup points in side view and go from there right? Ive got a ton of useful info from you guys, really helpful.

[BillyShope]

The last world class sports racing car with a beam rear axle was the early C-Type Jaguar. (Okay, there's probably been another one since, but it would be difficult to name a more significant car.) Anyway, it had a 3link with the "odd" link offset to the right to dynamically cancel the unloading of the right rear tire during acceleration.

At my site, I have a spreadsheet for the determination of link angles and placement for proper cancellation. There is also a spreadsheet for determination of the optimum ratio of upper link length to lower link length.

You can also achieve the cancellation, with a centrally located odd link, by angling the odd link and one or both of the symmetrical links. There is also a spreadsheet for this arrangement.

With the 3link, it often works out that the odd link should be below. This is almost always the case with a tubbed drag car. If you do place the odd link below, DO NOT use 1 inch spaghetti for that tube. Picture a tube positioned vertically and supporting the weight of 4 or 5 of your cars and you have an idea of how strong that tube has to be.

http://home.earthlink.net/~whshope

[796spdbu]

Jerome, thanks for the info, I will use this in the design of my set up for my truck project ive been planning.
Mr. quick, thanks for the input about stock suspension upgrades.
Matty B keep us posted and post lots of pics!!

[rocketman]

Get with Travisb, thats his login here, he is building a 65 chevelle with a 3 link it looks good. or do a search kenny davis hot rods

[matty b]

I came across this http://www.powerblocktv.com/special/4BarLinkV3.0.zip its a pretty cool and FREE calculator. Its for a 4 link but Ill get over it. Ok so I have this now and what Id like to know is there any magic numbers I should be shooting for? Anti squat, roll center and rch.. stuff like that. As far as the lower links are concerned Im planning on a simple cross member in the same location as the lower stock mounts. I think i wanna use the brackets from air ride for the lower link from their tribar setup http://www.ridetech.com/productinfo/productpopup.asp?prodname=Tri%2DLink+Axle+Bracket&prodID=A113&price=%2435%2E00
and probably weld some tabs on top of the center section for the upper third link. I want to hopefully use the stock frame to mount a bar for the upper cross member so that I dont have to cut in the car if possible. So thats my plan if you guys can steer me in the right direction on numbers and alternative hardware than what im lookin at im all ears. thanks

[jason@gmachine]

You really really need to understand what you are doing before you start welding. I can't stress this enough. You need to do all the calculations. Here's what you are looking to optimize:

antisquat: depending on application, you want different things, but you could start at 100%. In side view, antisquat line is from center of contact patch to center of gravity height at the front wheel center. 100% antisquat is when imaginary intersection of upper and lower links falls on that line.

roll steer: looking for slight understeer. that means your roll axis needs to angle up from front to back in side view. for parallel lower links (in plan view), the roll axis is the angle of your bars, so you can't optimize everything with this. you want to choose converging lower links. This way, your roll axis is drawn from your panhard bar height to the imaginary intersection of your lower links. As long as your panhard bar height is slightly above this convergence point height, then your roll steer will be good.

side view swing arm length: shoot for over 50 inches. too short gives you brake hop. defined by distance in side view from rearend to intersection of lower and upper links

vertical distance of link attachment from rearend centerline: as long as possible without touching ground when you get a flat tire.

link distance from center of rearend: as close to wheel as possible

PHB length: as long as possible, attach to drivers side frame, pass side rearend

upper link length: make it so antisquat remains constant as rearend undergoes travel

upper link placement: on passenger side of rearend will somewhat lessen the effects of driveshaft torque

coilovers: as far out and as vertical as possible

Hope that helped, I know it's really hard to grasp, but if you're serious about getting it right, start drawing it out, and asking for help on here with specific numbers, alot of people will just do it for you, alot have already made spreadsheets for their own vehicles.

Jerome

good post.


Jason

[jason@gmachine]

I came across this http://www.powerblocktv.com/special/4BarLinkV3.0.zip its a pretty cool and FREE calculator. Its for a 4 link but Ill get over it. Ok so I have this now and what Id like to know is there any magic numbers I should be shooting for? Anti squat, roll center and rch.. stuff like that. As far as the lower links are concerned Im planning on a simple cross member in the same location as the lower stock mounts. I think i wanna use the brackets from air ride for the lower link from their tribar setup http://www.ridetech.com/productinfo/productpopup.asp?prodname=Tri%2DLink+Axle+Bracket&prodID=A113&price=%2435%2E00
and probably weld some tabs on top of the center section for the upper third link. I want to hopefully use the stock frame to mount a bar for the upper cross member so that I dont have to cut in the car if possible. So thats my plan if you guys can steer me in the right direction on numbers and alternative hardware than what im lookin at im all ears. thanks

I like that tab, if placed properly then im sure it will work great. When you are ready to collect the materials shoot me a pm of your list. I can give you some advice on material usage and spec of each application. Even on the heims, bungs, spacers, and tubing.

Jason

[BillyShope]

I came across this http://www.powerblocktv.com/special/4BarLinkV3.0.zip its a pretty cool and FREE calculator.

The spreadsheet on Page 13 of my site also provides this information and, again, it's free. In addition, my spreadsheet also allows the rear pivot points to be ahead of, or behind, the rear axle.

As for roll steer characteristics, keep in mind that only the steering is changed by roll steer. Wheel loads are unaffected. What I'm getting at is that the dynamic gremlins associated with oversteer (critical speed, etc.) are not involved. So, if you can accustom yourself to the steering inputs required, a bit of roll oversteer is of no concern.

http://home.earthlink.net/~whshope

[matty b]

yeah your calculators are really helpful I just have a hard time visualizing things and bein able to see the anti squat line and such makes it easier for a total newbie like me to see if im goin in the right direction. This stuff is pretty complicated and a picture is definitley a million words in this case.

[Elusive R]

https://www.pro-touring.com/forum/showthread.php?t=12946

:drive1:

I took all of the pictures down when the car was stolen, but it's doable and it works very nice. If you're short on time/money, you'd probably be best off focusing your money elsewhere. Be careful if you do go for it!

Ryan

[matty b]

Im gonna bite the bullet and buy a rear end housing from Art morrison along with the watts link they have. The rest of the stuff should be fairly straight forward. Im goin this route cause I dont feel like having to jig all the crap up on the rear end when I can have all off it pre welded and ready to install. I plan on starting this in probably a month or so, gotta get the motor trans installed first :D Ill keep you guys posted

[Exhausted]

Matty, I am building my own 3 link, was working on the crossmeber to mount the upper link tonight before I signed up to this site. I can't give you all the technical ifo, you may be asking, but my friend that is explaining as I build, (lol) does understand, and I would be willing to ask and tell you anything you need to know to build your own

[Exhausted]

also, I am using my stock leaf spring front mounting points for my lower links, 9" ford rearend, DSE minitubs, DSE Sub Frame connectors, Not sure on my coilovers, just yet, I will buy Heim Joints and rods, but I will fabricate most of everything else. (1971 Nova)

[matty b]

id like to see some pics if you wanna put em up. You using the stock rails or did you go ahead and put in new rails and tub it? I know our procedures will be way different but luckily I think mine will be the easier way.

[BillyShope]

Matty, I am building my own 3 link, was working on the crossmeber to mount the upper link tonight before I signed up to this site. I can't give you all the technical ifo, you may be asking, but my friend that is explaining as I build, (lol) does understand, and I would be willing to ask and tell you anything you need to know to build your own

Are you offsetting the odd link? If you offset it correctly, you can dynamically cancel the effect of driveshaft torque on rear wheel loading. In other words, your rear tires can be equally loaded during launch with any value of driveshaft torque.

See Page 18 of my site for setup information. Check out Pages 4, 5, and 6 for a method to verify your setup.

http://home.earthlink.net/~whshope

[baggins]

what is more advantageous... parallel lower links or triangulated???

[matty b]

i believe that angled lower bars are better, why and how much I dunno but I read its better

[BillyShope]

what is more advantageous... parallel lower links or triangulated???

As usual, there's no correct answer. It depends on your application. Angulation of trailing links (either top or bottom) is usually done to avoid the need for a Panhard rod to take lateral loads while cornering. It's very difficult, however, to provide convenient adjustment with triangulated links. In other words, you're usually stuck with the percent anti-squat of the original setup. With links that are parallel to the long axis of the car, you need a Panhard, but, on the other hand, it's relatively easy to provide adjustment.

And, of course, there's the matter of packaging. With any given car, one setup is going to be easier to install than the other.

So, it comes down to what you want to do with the car. If it's only going to see street duty, there's really no reason for tinkering with the percent anti-squat and the triangulated setup will be quite adequate. But, if it's essentially a dragstrip car, you probably will appreciate the adjustment and you'll go with the parallel arrangement.

http://home.earthlink.net/~whshope

[Exhausted]

In reference to BillyShope's question, My upper link will be centered give or take an inch or two. I'm a drag racer at heart, I know a 4 link, would be better, but My car will very seldom see the track, It's just a car to cruise short distances, I plan on using mid 80's camaro rear bucket seats, mainly for looks not for function. Another friend of mine is building a 3 link, it's almost done, with the upper link, set about 4 inches toward the center from the right lower link. But it is mainly a drag car, maybe burgers on Sunday afternoon. I'll take a look at your sitght, thanks for the info.

[Exhausted]

Mattyb, I thought long and hard about mini tubs apposed to gutting the car, building new rails, and buying stamped tubs from jegs or summit, the mini's I think are over priced, but hey that guy needs to make a living to. They look great. I am not gonna go real wide for my rear wheels, thinking about 17x9.5 maybe 10's, I have 16x8 Torque Thrust 2's with a 5-1/4" back spacing, front and rear, on the car now. So, keeping the stock frame rails, going with mini tubs.

[Norm Peterson]

what is more advantageous... parallel lower links or triangulated???It's something of a compromise either way. A little plan view convergence of the lowers makes the axle roll steer vary less with changes in rear ride height (a good thing for handling 'predictability'). But converging lowers may force you to choose a slightly higher roll center than might be do-able with parallel LCAs (depending on their side view inclinations, among other things).


Norm

[matty b]

Mattyb, I thought long and hard about mini tubs apposed to gutting the car, building new rails, and buying stamped tubs from jegs or summit, the mini's I think are over priced, but hey that guy needs to make a living to. They look great. I am not gonna go real wide for my rear wheels, thinking about 17x9.5 maybe 10's, I have 16x8 Torque Thrust 2's with a 5-1/4" back spacing, front and rear, on the car now. So, keeping the stock frame rails, going with mini tubs.

im looking at alstons catalog at the moment and I think Im gonna call them about a rear clip. A clean sheet of paper is probably gonna work the best as I can fit the tire I want, and not have to really compromise on anything. Damn this is snowballing into something I didnt really expect to do but a deep dish 335 sounds awfully tasty.

[baggins]

so how would you go about finding what the best angle to put the lower arms at... packaging doesnt really come into play, other than cleanliness, because all i have done so far is bare rails and a mid crossmember

[Norm Peterson]

It's a process of iteration, where you're trying to get the roll center at some general height and axle roll steer and anti-squat to within some ranges, Best to work with some sort of software or at least a spreadsheet, because you'll perhaps start with a roll center height and either roll steer or anti-squat and see what the unspecified one ends up at. And you'll want to see how all of these vary as ride height is varied (at least).

Conventional wisdom seems to put anti-squat in the 50% range for a road course oriented car and 100% for a street/strip car. Roll steer numbers in the 3%-ish ballpark (as a vehicle roll understeer effect) for a cornering sort of car are reasonable, maybe a little less for serious competition.

Or you can start with a known setup (or a reasonable, perhaps scaled-off-a-picture guess at one) and iterate from there.


Norm

[baggins]

thanks Norm. man this whole suspension thing is pretty tough to grasp but slowly im adding a little more to my knowledge. thanks again

[35spline]

I am bringing this topic back to ask a few questions on what has already been stated. From a list things to optimize when desinging a 3-link


Originally Posted by jerome
vertical distance of link attachment from rearend centerline: as long as possible without touching ground when you get a flat tire.
What is the benifit of doing this verses doing just the opposite? I was thinking about maintaining ground clearance with the 3-link I am planning.

Here is a spreadsheet that captures what I have so far. I would really appreciate any suggestions to improve on what I have so far.
https://www.pro-touring.com/forum/images/attach/doc.gif

[jerome]

The longer the distance is from the rearend centerline to the link attachment, the less force the rearend will transfer into the links. When you accelerate, a rotational torque is created in the housing opposite of the rotation of the tires. The force created is the torque divided by the length of the lever arm (the distance we are talking about). If this distance is large, then the force created is small. This allows your link bars to not be ridiculously beefy. There is a limit to the benefit that this provides

I think a good ballpark for this distance is somewhere between 5 and 7 inches. Ground clearance is a good point, but the rearend attachments should probably be very close to the tires, so it is less of a problem. If you do 5 inches, your links will probably still be higher than the bottom of the rearend.

Hope that helps

Jerome

[35spline]

My lower links will be near the tires and out of the way but I was thinking more about the clearance at the frame end since the links converge at a 15 degree angle and are basically horizontal. As for the links I have 1.75" dia .120 DOM tube and the large Johnny Joints. I thought this would be strong enough for a 345/30/19 radial tire even on a high HP car.

Sorry I could not get my excel spreadsheet to load even when copied into a word doc. If someone could please send a pm on how to do this I will post the info. Here it is just pasted into the post (not good).


3 Link Calculator v1.0 - Vector-Based with Graphical Worksheet Add-In


Vehicle Specifications:






Wheelbase 108.0 in = Mess with these cells


Tire Diameter 27.00 in





Tire Rolling Radius 13.50 in Created 2004.06.17




Sprung Mass CG 'Z' 20.00 in by




Weight 3,400 lb Dan Barcroft




Suspension Geometry:


Geometry Summary:

Upper Link x y z
Anti-Squat 95.45 %
Frame End 36.00 -12.00 14.00 in Roll Axis Angle -0.66 degrees ( - roll understeer, + roll oversteer) Axle End 0.00 -12.00 21.00 in Roll Center Height 10.93 in
Lower Links x y

Instant Center X-Axis 56.57 in
Frame End 36.00 10.00 10.00 in Instant Center Z-Axis 10.00 in
Axle End 0.00 18.00 10.00 in



Pannhard Bar x y





Frame End -6.00 16.50 11.00 in



Axle End -6.00 -17.00 11.00 in



https://static1.pt-content.com/images/noimg.gif


13.00

[BillyShope]

35spline, if you're concerned about ground clearance, I would recommend you go to symmetrical links on the top and the odd link below. Page 18 of my site will give you the necessary geometry for equal rear tire loading and 100% anti-squat.

http://home.earthlink.net/~whshope

[BillyShope]

Or, if you want to keep the symmetrical links on the bottom, you can move the rear mounting points forward and up. There's no requirement that the link mounting points be directly above or below the axle centerline.

http://home.earthlink.net/~whshope

[35spline]

Or, if you want to keep the symmetrical links on the bottom, you can move the rear mounting points forward and up. There's no requirement that the link mounting points be directly above or below the axle centerline.

Thanks, that's a good point. I finally got the rear housing under the car and set at ride height to see how everything would fit. It looks like I can maintain about 6 1/2" to 7" of ground clearance to the lowest point of my link brackets using Jerome's vertical spacing suggestion.