offset 3-link

[jerome]

I'm planning an 3-link rear for my S10, and would like some opinions on the numbers I'm getting just to make sure it will work out alright. Here's the theory of the offset third link:

http://home.earthlink.net/~whshope/id5.html (http://home.earthlink.net/%7Ewhshope/id5.html)

I am also using the calculator on this page.

All links will be 4" from the axle centerline vertically, directly above or below the axle centerline.

I will be using a panhard bar, as long as possible (about 40"), adjustable for height, and parallel to the ground and rearend.

Will be using coilovers, very close to vertically mounted

The lower links will be parallel to each other in plan view, and basically parallel to the ground in side view. I am assuming that only the lowers define roll axis, so I would have a neutral roll axis at ride height. I guess optimal would to have the roll axis slope downwards towards front so there is roll understeer, even in bump. The lower links can be very very long, but I am assuming that the gains are negligible north of 36".

The upper link will be offset to the passenger side of the car. It can be long, but it must be angled downwards (towards front) to provide anti-squat so it must stop around 20" in order to not reduce ground clearance.

The results from the calculator are attached.

I have 3 concerns with this suspension setup.

13" of offset to the side for the third link is very significant, will this cause problems with uneven braking because of squatting force on the passenger side?

Pinion angle change. From what I can see, the shorter upper link will make the pinion rotate downward in bump, which is good. I would assume that I should try to match the degree change of the pinion to the degree change of the driveshaft in both bump and droop. Is there a theory on what works best here?

I currently have the links sloping very slightly down towards front of car to create roll understeer. I havent ever seen a suspension that does this, is there something wrong with doing that?



Thanks,
Jerome

[jerome]

nobody knows?

[99 BUCKET]

I wish I did, I am currently going through the same " what do I need to do" process for my s-10. I am leaning toward a tri 4 link. or possibly a truck arm set up (not really sure bout this).

[jerome]

Ive seen your thread on s10forum, I really like it. I ended up with the same wheels, but I have a 383 sbc instead of the ls1. I've been through all the suspension possibilities. I wouldn't do the tri4 link if youre talking about having the angled arms on top, that puts the roll center too high. What I really wanted to do was the satchell link which is basically the opposite: parallel arms on top and angled arms underneath. (from axle ends to a crossmember in center). I ended up nixing the satchell link idea because I couldnt make the angled arms long enough because of my offset driveshaft position. I think you can do better than the truckarm, really the only reason to do it is if it is mandated (nascar). Satchell link, 3-link and torque arm (maybe decoupled) would be the places I would look first.

I'll let you know how this works out for me, I'm making the lower bar mounts right now, hopefully she'll sit on the ground in a couple weeks.

Jerome

[parsonsj]

Jerome,

Did I read right that the housing bracket centers are 4" from the axle centerline?

That's pretty short, and likely why your LCAs are angled down from the housing to the chassis. What size wheels are you using? You can put the housing brackets far enough down so they don't drag in the event of a blown tire. That's around 8" for an 18" wheel.

jp

[99 BUCKET]

well the vette rims are gone, I am now sitting with a 8.5 rear in the truck. I have kept low key on posting pics cause the truck is going through some body mods right now.
the suspension is driving me nuts. Keep going back and forth on what to do. I will defenatly keep an eye on what works out for yours.
On a side note I rode to work today with the a/c blowing for the first time. and it was great!!

[jerome]

I didnt plan on the brackets being too far from centerline, just because this is a truck, so the frame will be relatively visible, and a mount jutting down from it will also be. Herb Adams says anywhere from 3" to 7" is good, with 5" being average. I have 18" rear wheels. The bottom of the differential is about 5 3/4" below axle centerline, so I would like to stay above that. They will probably end up being at about 5" from centerline. Thanks for the tip.

Setting the mounts further from the axles doesnt really have an effect on the LCA angle. When figuring out the math, I wanted to have roll understeer, and the only way to get it is to point the LCA's downwards towards front of car.

Jerome

[MonzaRacer]

As for me if your building a rear suspension why the 3 link I never was satisfied how 3 links handled.
I would prefer, if your set on using a panhard bar (rememberthis casues side to side movement so it does make the rearend shift) a parallel 4 link over the 3 link and as for the angle of the control arms you really need to figure the focal point of the link(s) , how they intersect and where. the higher the center point (closer to the roof ) makes it less likely to lift the front of the vehicle.
Honestly we set up an S10 a couple yrs back after I got my parrallel 4 link and air ride from Air Ride Technologies.
We used thier set up but cut the bars shorter on top nad they sat on the inside of the frame. We divorced the front arm mounts so the upper and lower could be tuned for better handling , kind of like a drag 4 link but he was driving it. He swapped in a modded 12 bolt and when his coil over blew out he bought Shockwave V3.0 double adjustable.
His best rode course intersection came with the point roughly even with the drivers waist.
Now for his forays to the drag strip we definatelly tune it different. and the truck gets a lot of attention when he ran an autocross course and whipped a Vette.
Anyway good luck and remember experimenting is the fun part of tuning and rgardless of bar angle plot your intersection point.
Take the imaginary centerline from the center /front of the camshaftand draw a line to the downward intersection point of the truck "perfectly ballanced on the frame lie a teeter totter. that is the reference point to play from then your intersection point for more "lift is to the front of that and for less is behind then your moment arm is the height of that lift point.
Theories are best left on paper till they are proven.
Also dont forget your pinion angle the more down angle in relation to the driveshaft adds more "lift or grip" think of it this way if your rear end dips on acceleration your loosing traction, if your suspension splits or lifts up like scissors then your on the right track. and pionion angle will give you more or less bite and intheory the driveline should at the moment of launch become zero pinion angleand then return to a down angle as forces subside.
Good luck
Lee Abel
AFTERMARKET PERFORMANCE

[parsonsj]

Setting the mounts further from the axles doesnt really have an effect on the LCA angle. When figuring out the math, I wanted to have roll understeer, and the only way to get it is to point the LCA's downwards towards front of car.I agree that your downward LCA will cause roll understeer. I'm wondering why you want that. If you are going to go to the trouble to design your own suspension, why not try for a neutral (or near neutral) response to roll? Just curious.

Also, why are you limiting your design to parallel LCAs in plan view? Having them converge to the frame can help you manage roll steer as well.

One other reason to make the housing brackets longer is to reduce the stress on them, the rod ends, and the control arms.

jp

[jerome]

Thanks for the advice, I appreciate it.

Assuming a 40" panhard that is level at ride height, 3" of change in either bump or droop gives .113" lateral movement, for a max change of .226" of lateral movement for any bump or droop condition. I can tolerate a quarter inch of imperfection in exchange for the difficulty of setting up a watts properly.

If I understand correctly, a parallel 4-link cannot provide both roll neutral/under steer and antisquat at the same time. The links need to point up for antisquat, and the links need to point down for roll understeer. I thought when you have the links not being parallel, you have a kinematic bind in roll.

Dont get me wrong here, I see that the system worked for you, but on paper, I dont know how it worked. The bit about the pinion angle helps alot, I'll look into calculating how much change I'm getting.

http://www.ridetech.com/productinfo/airbars10.asp
Air Ride has a 3-link application for the S10, mine will be similar, but will be laterally located with the panhard instead of the wishbone for an adjustable roll center.

[jerome]

John-
I had a brain fart, I was thinking that I would want understeer at ride height to compensate for squat under acceleration when I wouldnt want oversteer. But then I realized that I squat would increase the roll understeer. Also, I guess roll steer is more into play at high speed steady state than at low speed acceleration.
Anyways, I've decided that parallel to the ground and neutral roll steer is probably the way to go.

If I were to converge the lower control arms, that would help define a roll center height, but would not be enough strength to do it on its own, meaning I would still need the panhard bar. Do the roll center heights they define need to be matched to each other, or is there no problem with letting them be different for an "averaged" real roll center?

About the plan view convergence, I dont understand how it helps with roll steer. Could you explain that to me?
Looking over the numbers, I noticed that the SVSA is pretty short at 33 inches. Herb Adams recommends at least 42 inches to reduce possibility of brake hop. Increasing SVSA would either reduce anti-squat or increase roll oversteer or both. If plan view convergence of LCA could help manage roll steer, and allow me to angle LCA up a little that would be great.

I included the new numbers

Correct me if I'm wrong, but I think a longer housing bracket would put less stress on the bracket and the weld to the axle, but it would put more stress on the rod end and the control arm. It's all a balance really.

Thanks,
Jerome

[Norm Peterson]

A small amount of roll understeer makes the car slightly more predictable (read: easier to drive hard and less demanding to drive under normal driving conditions). For relaxed street driving, you probably want only enough to keep you out of a roll oversteer condition due to other effects (such as bushing compliance which may differ from side to side based on the different loading, or the suspension moving into droop due to encountering a gentle dip in the road while turning).

Plan view convergence of the LCAs moves their intersection to a finite position, and the axle's roll axis is actually drawn between the PHB at mid-chassis and the LCA intersection point. This changes the rate at which the roll steer varies with ride height (slows it down) and can permit the LCAs to run uphill toward the chassis while still providing roll understeer depending on PHB height.

For any given traction force and UCA vertical pickup coordinate, the forces in the LCAs drop as their bracket length increases (LCA vertical coordinate drops). But not by as much as the bracket length increases, so LCA bracket bending stress and weld stresses still increase.


Norm

[Mean 69]

Jerome, nice work so far. Just a few points in that I don't have a bunch of free time these days.

One, I'd look pretty hard at lift properties under braking, you'd like to "minimize" it on your application. Trucks typically have low rear weight ratio percentage, and pretty long "polar moments." This is the primary reason the OEM's put anti-lock on the REAR of the pickups a few years back: lift under braking would greatly unload the tires, causing them to lock well before the fronts. The SVSA number you noted is (in my professional opinion) FAR too short. Yes, Adam's suggests a bit over 40", other reliable folks in the literature and those who build fast cars suggest, and typically design for SVSA's equal to or greater than 60". All things being equal, (which they can't in this comparison) this would reduce the A/S value relative to your example.

I would recommend buying a copy of Milliken's Race Car Vehicle Dynamics. It's spendy, but it helps fill in some of the gaps that Adams, Puhn, Van Valkenberg, etc. books leave you hanging on. Based on the questions you're asking, you're ready for it, but it's more the graduate textbook style than it is "fun" to read.

Angling the LCA's downward can also induce harshness over bumps, the axle assembly would be forced to move towards the front of the car in bump, and nature just doesn't like that. I'd be a bit less worried with a few percentage points of axle steer than I would sacrificing some of the other trade off attributes. Make a few different pickup locations for the control arms so you can play with it all in practice.

I'd recommend exercising caution in large offset upper links, look closely at the behavior of the system in roll before you declare victory.

Nice work, very best of luck,
Mark

[parsonsj]

f I were to converge the lower control arms, that would help define a roll center height, but would not be enough strength to do it on its own, meaning I would still need the panhard bar. I'm not trying to remove the need for a panhard, I'm merely saying that some convergence (say 5 degrees or so) will allow you more flexibility in dealing with roll steer.

jp

[jerome]

http://corner-carvers.com/forums/showpost.php?p=348547&postcount=21

probably the most informative post i've read explaining plan view convergence with two things defining roll center height.

now that understand this a little more, I need to put some plan view convergence in the lowers. That way, I can get the LCA upwards, good antisquat, good SVSA and good roll steer numbers.

What is a good guideline for roll center height? LCA upwards forces the panhard bar up a bit. Is axle centerline a good height to shoot for, or do I want it lower?

[jerome]

I got a good balance going of svsa, antisquat, and roll steer.
intersection of control arms side view height: 12"
panhard mounting height: 13"
so there's the roll understeer
side view swing arm: 57.6"
there's the more reasonable SVSA
if brake hop becomes a problem, then I can tune it out at the cost of antisquat
antisquat: 94.6%
Pinion will be 0 at ride height and will lower when back of car squats, so that is in the right direction. From what I've been reading, pinion angle is not that critical for traction, but that seems to contradict what's said in drag racing literature. Anyways, the pinion is moving in the right direction, and I've heard 60% is a good starting point for ratio of upper to lower length (longitudinally).
I have 16% of convergence in lower links, (10 inches jutting in for 36 inches lengthwise)
There is only one thing that isnt optimised, and that is the offset of the upper link. According to Shope's calculator, I would need around 20 inches of offset, which is squarely underneath my frame rail and thus unachievable. I will put the offset at around 15" and just live with not quite complete cancellation of the driveshaft torque (which isnt really THAT big of a deal, since nobody even seems to have a problem with it a sub-500hp levels).
BTW, correction of earlier posts, panhard will be from frame on drivers to axle on passengers. Here's a thread telling why:
http://www.corner-carvers.com/forums/showthread.php?t=2806

the (hopefully) final numbers are attached...I know somebody will probably point out another shortcoming about 5 min after I post this. :)
Thanks for all your help I really appreciate it, and I have learned alot from all this planning. I hope my discoveries will help others setup their 3-links.

I will put build pics up here as I get them

Jerome

[99 BUCKET]

were do you live, so you can do mine. lol

I look forward and fear this stage. which happens to be the next big project....

[jerome]

haha, dont be fooled, I'm just a newb. I don't yet know whether all this will actually work. BTW I'm in Tacoma, WA. Here's some more info and pics on my truck: http://www.fquick.com/garages/Chevrolet/S10_Pickup/1982/6101/

I'm gonna buy the links premade. I'm using Johnny Joints

https://static1.pt-content.com/images/pt/2007/06/9114-1.jpg

1 1/4-12 thread with 1.75OD internally threaded tube.
This way I'll be able to adjust link length while on the truck.
I went and bought some 3x3" square tubing .25 wall and will cut my upper mount from that (fishmouth and slot/drill). I still have to decide how to make lowers. I can do them like the upper, but they will be straight, and let the joints make up for the 16 degree angle, or I will have to think up how to make angled mounts. Coilovers will mount off the back of these mounts, and the panhard will route behind the shocks.

I want to make an adjustable panhard like this one, but a little heavier duty:
https://static1.pt-content.com/images/noimg.gif

For the front, I will be running ATS spindles, SPC uppers, stock lowers with del-a-lum bushings, howe ball joints, speedway spring adjusters, 600lb 5.5x9.5" springs, stock steering stuff (maybe a quicker box), and C5 brakes. Still have to decide on shocks, but will probably go for the single adjustable varishocks.

zz4blazer also has a pro-touring S10, check it out for some inspiration. theres gettin to be quite a few of us on here now.

Jerome

[99 BUCKET]

I am familiar with zz4's. I have a similar set up in mind for the front. but possibly a tubular lower with some coilovers (non qa1)

I already have a full zq8 suspersion under mine. fully poly, with the koni adjustable shocks and I eliminated the rag joint.

I had planed for the c5 set up also with hydraboost. I have a set up with 17's on there now and do not want to change rims again.

[99 BUCKET]

just reviewed you pics, nice project. what rear did you end up with?

[MonzaRacer]

Ok on the point of the Road Grater setup from ART its not a racing or autocross suspension it is a ground scraping unit and if you look at it its actually what some would call a hybrid 4 link but they removed one of the upper links and tied the upper arms together, mostly for packaging.
From what experience I have gleaned over the years I think the term you are using(antisquat) is simply what I was talking about the motion of differential roll under acceleration is what makes a car connect and start moving. For a simple example look at your fore finger and mifddle finger on your right hand think of your tire being tied to your middle finger. Now if you open your fingers or get suspension seperation than the tire is driven into the tractive surface, if the scissor is closing your lifting the tire and braeaking loose.
with a parralel system I had always used spherical air craft bushings to aleviate bind as they move in multiple plains (very usefull in GM Abodies as none of the arms are parrallel ) but they also let a parallel setup stay unbound.
The big trick I learned in making any rear drive vehicle to hook is make it hook up in the straights and NOY unload a tire in a corner and not bind so as to stop the motion of the suspension. then properly shock and swaybar it and it goes very simply.
Most people overthing rear suspensions to the point of too much geometry and that gets complex.
But if the 3 link fits what you dreamed up go for it but watch the pinion angle as too much and your gonna break loose on take off regardless of your design. Most people forget that. I had a friend do that and when we wondered why it just lit the tires he had the pinion angle pointing up and it never had any bite.
I alos figured out the having my upper and lower bars at same length doesnt work for a handling car and that some degree of shortening on my front bars made me control things better, Regardless of all the books, tips and tricks or math i never could get overthat one.
I have also helped several bomber dirt cars work out problems in thier suspensions and one guy cleaned his car up and took it to an autocross somewhere and took 3 rd place! and then he went to the drag strip and it ran(with only a lowly little 307 SBC) low 7s and had 1.31 60 foot times. He took it out one night and ran a local crooked road and he held it to the floor (thisis without any tire stagger or wierds spring work or anything) and was clocked at 127 and he never broke a tire loose.
and all this from some old school Abody and some super secret controlarms and a little attention to the straight line performance.

[Norm Peterson]

I wouldn't worry too much about not being able to get the full 20" offset as long as your limited slip is in decent shape. Or 100.0% anti-squat for that matter (depending on what sort of driving or seriousness of competition we're talking about).

There is, at least theoretically, a downside under braking. But since you should probably be setting the rear braking percentage in a truck used as a toy somewhat lower than its OE level (which must account for a significant load in the bed and possibly trailer tongue loading), it should be minimal. Less offset than 20" would help this situation, though.

I would imagine that the Johnny joints will be rigid enough to prevent axle steer under power (courtesy of the now-asymmetric lower link loading). OE rubber-bushed control arms might not be quite as friendly in this respect, and this compliance effect would add algebraically to the geometric axle steer that you get under power (as the car rolls slightly to the right in reaction to drivetrain torque).

I haven't run any numbers - I did have the actual equations in a spreadsheet somewhere but can't find it. Anyway, I'm guessing that even a 10" offset would work OK. Sounds like 50% dynamic equalization . . .


Norm

[jerome]

I'm putting the fuel cell behind the rear axle. Bad for moment of inertia, but good for weight distribution. I have a Ford 8.8 from an explorer. 4.10 traction lok. I am putting carbon rebuild clutches in for the limited slip and already got the axles and rotors redrilled. They are stock explorer disc brakes, single piston. I am going to end up with a dual master cylinder manual brake setup so I will be able to tune it to have very little rear braking. Front braking is C5 so it will already dwarf the rear braking performance.

The Ford pinion is already offset a few inches to the passenger, and I will probably end up putting the third link just right of the pumpkin (about 13" if i remember correctly). Norm, that was my thinking exactly, As long as I am on the right side of the axle, I will be getting some dynamic cancellation, and since I cant get the full effect, I may as well just put the third link where it is convenient, and get partial cancellation, seeing as driveshaft torque shouldn't be a huge issue anyways. That way I'll get less negative braking effects too.

monza- I wasnt actually going to use the ART 3-link, just referring to it. Yes, the finger analogy is the anti squat I was talking about. I'm still not quite sure what pinion angle does for traction, but I think the reason why the shorter upper works for you is because it points pinion angle down when the car squats.
I found this quote in a google search:
"Pinion angle plays a big part in how hard the car launches. With a sturdy rear suspension (such as with boxed control arms and hard bushings), the rearend doesn’t rotate much, so pinion angle doesn’t change a lot. Typically, the more the pinion points down (negative pinion angle), the softer the hit to the tires."

It didnt go into detail, and I dont understand why it is, but thats probably what you mean, Monza.

99bucket- the C5 brakes are also considered to be better than the c6z's by alot of people, you will be saving money while youre at it too. I nixed the hydroboost idea, and I'll be going with a dual mc brake system w/ balance bar setup. Since all four corners will be disc brake, it shouldnt be that bad, even for street driving. Check with Savitske, they have the varishocks coilover conversion for the S10.

[99 BUCKET]

will do, going to check out the coilover now.

[99 BUCKET]

hows the rear of the s-10 comming?

[jerome]

been working on other areas., got new headers, put bellhousing on, and realized that the floor and part of the firewall need to be reshaped, as well as the seat area (seat sits too high)...so I think I will have to do a completely new floor, which sucks, but it means it'll be done right.

just bought 10 pieces of 3/8"x3"x5.5" cold rolled plate to make axle tabs with today. I'll be making some of them tomorrow. All I have is an angle grinder, drill, and bench grinder so it'll be slow work. Johnny Joints are backordered til the end of the month, and the electrician is coming end of next week to put a 220v line in the garage for my new welder. I'm itching to get this thing rolling. Front suspension is backordered, and I'm a slow worker so rear isn't coming fast either. I hope I can get it rolling by mid-july, but thats a stretch.

Jerome

[parsonsj]

Jerome,

3/8" ? Or did you mean 3/16"? 3/8 will take you a very long time, and isn't necessary. And will be hard to fully weld.

3/16 is more than adequate for suspension brackets.

jp

[jerome]

yeah, I meant 3/8". I realize not that 3/8" is overkill, but I already bought the metal, so I might as well use it. I need a drill bit for the bolt hole. The bolt is 5/8", do you usually go 1/32nd oversize for a hole? I attached a pic of what the tabs look like now. Took about an hour to do each one. Drilled overlapping holes to make general shape, angle grinder to smooth it out, and hand filed to fit.

John, I've been looking through your photos. Your rear suspension looks really nice. I was just wondering how you decided on a bar. i.e. how did you pick a rate, and where did you get it?