58 replies to this topic

[John Brundage]

I am working on modifying a vintage race car and installing a Watts link. My thought was to make the Watts link adjustable with different center pin hights (and corresponding mount options for the links) to tune the car at the track. After reading old forums and posts, I still am not sure which is more desireable---
1--mount the centering pin on the axle
2--mount the centering pin to the frame.
I have seen the question asked, but never truly answered.
PS--I did try to bump some of these old threads but they were closed.

[gruntguru]

Originally posted by John Brundage
I am working on modifying a vintage race car and installing a Watts link. My thought was to make the Watts link adjustable with different center pin hights (and corresponding mount options for the links) to tune the car at the track. After reading old forums and posts, I still am not sure which is more desireable---
1--mount the centering pin on the axle
2--mount the centering pin to the frame.
I have seen the question asked, but never truly answered.
PS--I did try to bump some of these old threads but they were closed.

Centre pin on the axle seems to be the more popular choice.

Although having the centre link attached to the axle probably adds more mass to the axle than option 2, the extra mass is at the axle centre and therefore only one quarter of its mass is added to the unsprung weight of each wheel in the most common suspension mode of single-wheel deflection, none of its mass is added to unsprung mass in pure roll (one wheel in bump, other wheel in equal rebound) and the full mass is added only in a pure pitch situation. In summary - for most modes the unsprung mass will be less with the centre pin on the axle.

For small adjustments to roll centre height it is usually OK to adjust only the centre pin because the Watts linkage provides linear location over a fairly large travel.

[LMP900]

The chassis-mounted rocker keeps the vertical distance between centre of gravity and roll centre constant IIRC. That's a good thing in my opinion.

[Bill Sherwood]

Here's my simple racer as it was during the rebuild a while back.

The Watts Link sits horizontally under the diff, and it seems to work well.



It works well enough because the suspension doesn't move much on such cars. On a road car you may well be better off putting the Watts Link on the chassis as it will maintain a better relationship to the front roll centre.

[cheapracer]

(waiting for Greg to chime in here)

Why a Watts anyway, Panhard is just as good if you can get the length and very easy to adjust the RC trackside. I kind of like the Satchell type 4 link setup.

Theres pretty pictures and small sentences at this site - http://www.carbibles...sion_bible.html

Pretty pictures here for ideas - (i like to look!)http://www.rodandcus...ing/index1.html

By the way, how's about your Lambda sensor Mate, need any help with that?

[John Brundage]

I originally considered a Panhard bar. But because the car is driven on road courses, my goal was to make the car feel the same on left and right turns. With the panhard bar the static roll center is at different hights on left and right turns. (assuming the the turns are exact mirror images of each other.) The rear axle is in a Alfa SZ. That originally uses an upper a-arm to one side of the diff housing to locate the axle. The arm will be removed and replaced with an offset link. The Watts link will locate the axle and lower the static roll centre from above the rear axle to a lower position.
I have seen the link located under the diff, but this would bind with the suspension travel that the Alfa has.
Would you rather have the rear roll center move with the body or be more fixed on the axle? With it mounted on the frame, would you have a better contact patch as the car moves through turns? The Alfa suspension is fairly soft in the rear and the car runs on narrow vintage Dunlop racing tires. The front static roll center is a hair below ground level with normally stiffer front springs. Currently the car cas 500 front and 175 rear springs before the modifications. These springs will change also. The front also has a 1" sway bar. The front roll centrer is original. Originally the springs were very soft and the car had alot of roll. We are trying to make the car work better.
Gruntguru--I am seeing Mustang people arguing that the frame mounted center link works better and this is "the hot set-up".
Is there any data that suggests that one has certain advantges over the other? Looking at the frame on the Alfa, I beleive it is easier to fab a strong adjustable mount from the frame. Whereas with a diff mount, the end pieces needed to locate the end links may be hard to locate securely to the frame (due to the length of the arms hanging from the frame)
I prefer to make it adjustable to give me some tuning ability at the track. These cars do not have much adjustability.

[cheapracer]

You do get lateral body movement with a Watts when it's mounted to the axle. Mounted to the body gives no lateral movement.

Mounted to the body will give you more weight on the fronts under braking/turn in and vice versa as the roll line tilts forward and back - this could be good or bad, who knows until you drive it.

You can actually have a setup to fit the Panhard rod both ways to suit different tracks but to be honest I don't think it's a great difference that slight difference between L and R turns. Offset the driver weight with the appropriate side of mounting? That would be axle mount on the opposite side to the Drivers side if i figure correctly, giving a lower RC for the drivers side front tyre (left hand drive car, right hand bend) and higher RC for the less weighted side to add some weight to that front tyre (left hand drive car, left hand bend).

[phantom II]

I built a round tube mild steel space frame chassis roadster that weighed 1800lbs with a cast iron block 330hp Chev V8 then changed the engine to a 430 hp all aluminum LSx. I first built it with a stick axle and later replaced it with an IRS. My new roadster has 480hp to the back wheels and an IRS.
On the live axle, I used 3 trailing links and a watts. I can pass on a few things that I discovered.
Why do you want to change your set up? What's it doing that you don't like? How much power do you have?
Is the chassis your own? What is the wheelbase?
What is an offset link? Is it trailing link but not in the center? It must be placed in the center above the diff and trailing.
I positioned the 3 links to give me some anti squat and some roll understeer. You may not need this geometry. The Watts pin must be as low as possible on the diff, this gives the lowest roll center. Don't put it on the chassis. If your front roll center is as low as you say, which I doubt, you may experience a roll couple that you may not be able cope with. You will probably find that your front roll center is higher than what you claim if the Italians had anything to do with it. They were the masters of stick axle location and the corresponding chassis dynamics.

The first thing I had to do with my IRS was to lower the front roll center when I lowered the rear with the IRS. The front must be lower than the rear however, but not as much as you describe and less with IRS. Rather than an adjustable Watts, try to lower the CG. Where is the gas tank? The battery? where is the CG? If the links are long enough on the Watts, there is going to be no lateral shift on the axle for that amount of travel. Just limit suspension travel especially droop. Does the diff have a cover? Fabricate or billet(Al) a new cover and bell crank with a pin as low as possible.
If you must, adjust on the chassis. Got pictures?
The Mustang has a Reese Bar and two trailing arms plus a watts. Thats a whole different ball game. Your drive shaft is probably short, so length and angle of the original A link is probably correct from the side view. The rear mount should be on top of the diff and the link trailing. A symmetrical upper trailing A link mounted in the center may be a better bet than the Watts link if you can, but the RC will be raised. It would be aesier to make, space permitting.
Leave Panhard Rods to NASCAR.

Originally posted by John Brundage
I originally considered a Panhard bar. But because the car is driven on road courses, my goal was to make the car feel the same on left and right turns. With the panhard bar the static roll center is at different hights on left and right turns. (assuming the the turns are exact mirror images of each other.) The rear axle is in a Alfa SZ. That originally uses an upper a-arm to one side of the diff housing to locate the axle. The arm will be removed and replaced with an offset link. The Watts link will locate the axle and lower the static roll centre from above the rear axle to a lower position.
I have seen the link located under the diff, but this would bind with the suspension travel that the Alfa has.
Would you rather have the rear roll center move with the body or be more fixed on the axle? With it mounted on the frame, would you have a better contact patch as the car moves through turns? The Alfa suspension is fairly soft in the rear and the car runs on narrow vintage Dunlop racing tires. The front static roll center is a hair below ground level with normally stiffer front springs. Currently the car cas 500 front and 175 rear springs before the modifications. These springs will change also. The front also has a 1" sway bar. The front roll centrer is original. Originally the springs were very soft and the car had alot of roll. We are trying to make the car work better.
Gruntguru--I am seeing Mustang people arguing that the frame mounted center link works better and this is "the hot set-up".
Is there any data that suggests that one has certain advantges over the other? Looking at the frame on the Alfa, I beleive it is easier to fab a strong adjustable mount from the frame. Whereas with a diff mount, the end pieces needed to locate the end links may be hard to locate securely to the frame (due to the length of the arms hanging from the frame)
I prefer to make it adjustable to give me some tuning ability at the track. These cars do not have much adjustability.

[gruntguru]

Originally posted by John Brundage
Gruntguru--I am seeing Mustang people arguing that the frame mounted center link works better and this is "the hot set-up".
Is there any data that suggests that one has certain advantges over the other? Looking at the frame on the Alfa, I beleive it is easier to fab a strong adjustable mount from the frame. Whereas with a diff mount, the end pieces needed to locate the end links may be hard to locate securely to the frame (due to the length of the arms hanging from the frame)
I prefer to make it adjustable to give me some tuning ability at the track. These cars do not have much adjustability.

As you know - my comments spoke to the unsprung weight issues. On the other hand the issue of variable RC height is important too - especially if you have a lot of travel and soft spring-rates. So - focus on unsprung weight if you race bumpy tracks, run low profile tyres, RC is set low and you can firm up the damping for the track. Fixed RC - CG distance is more important if the track is smoother, tyres are taller, RC set high, you are stuck with soft damping and the rear height varies significantly around the track.

Having said all that - I reckon either set-up will work fine for you so go with the one that's easier to install on the Alfa.

[gruntguru]

Originally posted by cheapracer
You do get lateral body movement with a Watts when it's mounted to the axle. Mounted to the body gives no lateral movement.

How come?

[phantom II]

How come?

Originally posted by gruntguru
run low profile tyres

[gruntguru]

Originally posted by phantom II
How come?

The lower the tyre profile the more you have to rely on the suspension to soak up the bumps.

[phantom II]

Why would you want to do that?

Originally posted by gruntguru
The lower the tyre profile the more you have to rely on the suspension to soak up the bumps.

[John Brundage]

The car was generally not good. The driver was not happy with the car. The driver has experience with other Alfas, Lotus 23, Cobra etc. I have in car footage of the driver struggling with the car. Originally Alfa used a very soft suspension with front springs in the 200lb range and rears about 100. The car is original and was made in 1959 or 1960. The wheelbase is relatively short. Horsepower is in the 150 range. Old videos of the car showed how much they would lean. A common update to these cars is to put stiffer front springs--at least 500lbs and the rear about 175. A 1" sway bar is a common addition to the front. This car had these things. Before I started with the project of improving the car's handling, I measured the suspension and plotted everything on Performance Trends software. I found that due to old accident damage the front static roll center was way off to one side and a hair below ground level. Shop notes from other 101 series Alfas shows the front static roll center at or below ground. The rear suspension has 2 lower arms mounted below the axle and proceed forward to the frame. A single a-frame is used to locate the rear axle at the top. The two legs of the A frame are mounted to the body and the single end is mounted to an ear on the diff housing. The ear is offset to the driver's side and is mounted approx 3" above the center line of the axle. (from memory--the exact numbers are in my other computer). This puts the static rear roll center high and to the side. Using either a Watts or Panhard would require the upper A-frame be changed to a single link to help control axle "roll" and pinion angle. The Watts or Panhard would be used to locate the axle and lower the rear static roll center. Using the Panhard in the program gave me a change of at least 1.5" of rear roll center hight between left and right turns. I experimented with different angles and lengths to optimize it. The upper link being offset to one side is not helping, however there is no room to move it to the center. The rear diff housing does not have a cover. It a cast aluminum housing that the axle tubes bolt to. A plate would be fabbed to fit the the back of the diff housing if that set-up were to be used.
As I re-assemble the suspension, everything is being located and centered to the vehicle centerline. All the hights and pick-up points to the front suspension have been corrected. The steering linkage is centered and The front will be bump-steered. I am putting in adjustable spring perches to make it easier to corner balance the car and make changes to ride hight. I prefer to use the Watts link to have a good static rear roll center. Being that the computer progam is not driving the car, I want to make the rear center pin adjustable to adjust the car based on the driver's feedback and preferences.
Weight cannot be easily moved--the fuel cell is in the original fuel tank location. The battery etc. are all in their original locations. The car runs the original drum brakes and narrow hard compound Dunlop racing tires. I do not want to make the car into a go-kart; I just want to make it handle better without ruining an original car. The Watts link, adjustable spring perches etc. is all removable.
With the Watts link center pin frame mounted I understand that the roll center will move more than if the pin was mounted to the diff. Is this bad? Will this help or hurt handling? My thought that if it is relative to the changes in the front, it would be a good thing. Again, making the hight adjustable, I could tune this (or so I hope). Due to stock rear quarter panels and original rims etc. I am trying to limit lateral movement of the axle. There is not much tire clearence to the fender lip. I am at home at the moment and do not have any pictures. I will ahve to dig some up.

[John Brundage]

Originally posted by phantom II
How come?

We can't run low profile tires. The tires need to be original size. (and hard compound)

[cheapracer]

Originally posted by gruntguru
How come?

Because with the center bolt on the axle as the body rolls one link can angle up effecttively shortening that side. When its mounted on the body it can't do that.

[gruntguru]

Originally posted by cheapracer
Because with the center bolt on the axle as the body rolls one link can angle up effecttively shortening that side. When its mounted on the body it can't do that.

Does that mean the body doesn't roll about the centre pin axis ie roll centre? (Referring to the centre pin on axle system)

[cheapracer]

I'm not quite sure if I'm being baited here or not.

Go find a watts linked car such as an early Mazda RX7, stick a jack under 1 side and see what happens - as the car goes up if the opposite side doesn't go down in proportion then the body will move across to that side as it goes up as in an arc, so the answer is no, it doesn't roll around the center bolt. When mounted on the body the car will simply go straight up, body roll or not, consistantly rolling around the center bolt.

If you have the patience to freeze frame this rubbish (I apologise in advance) you can see the effect on this RX7 http://www.youtube.c...feature=related

[gruntguru]

Originally posted by cheapracer
I'm not quite sure if I'm being baited here or not.

Go find a watts linked car such as an early Mazda RX7, stick a jack under 1 side and see what happens - as the car goes up if the opposite side doesn't go down in proportion then the body will move across to that side as it goes up as in an arc, so the answer is no, it doesn't roll around the center bolt. When mounted on the body the car will simply go straight up, body roll or not, consistantly rolling around the center bolt.

If you have the patience to freeze frame this rubbish (I apologise in advance) you can see the effect on this RX7 http://www.youtube.c...feature=related

No baiting here - I'm not into asking questions when I'm sure of the answer.

Shouldn't you stick the jack under one side of the axle? Jacking under the bodywork changes the rear ride height unlike pure roll on the track.

Couldn't see the effect on the video. What elapsed time should I be looking at?

[Joe Bosworth]

This discussion seems to baffle all conventional mechanism design that I have been subjected to.

The centre pin of a watts link as I learned it can only move in a straight line vertically to a line drawn between the two extremity pivot points. There can be no lateral movement if the pivot points are the same height from the road surface.

I have never seen a panhard rod installation that provided a different roll centre in right and left hand turns but i suppiose one can be designed. Save me the trouble and show us the congfiguration of one that does this.

In trying to rectify handling problems I find it it easier to design and install a system that allows quick roll centre changes and to use that as a tuning tool than it is to change spring rates. Once you know what roll centre you really want then worrying about unsprung weight might become worthwhile but for any solid rear axle car the added weight of the Watts pivot point is immaterial in the whole scheeme of things.

A smidgeon of reality is worth a ton of theory.

[John Brundage]

JB--
I didn't realize that the RC would be different left and right in roll until I ran it on the program. Because the upper axle locate link is not on the center line may influence this. There is no room to locate it on the centerline. I tried the panhard with different lenghts, angles, pick-up points etc. to try to get it as close to equal as possible without success. Again, I was just trying to make it theorietically even from side to side assuming the turns were the same (which they never are). If I am doing this exercise, I want to take the time to decide which may be best and get it as close to even as possible, instead of just mounting a panhard or Watts and seeing what happens. I would rather start with something that is balanced and work from there.
These cars do have some roll to them. It is not a go-kart or track car running with limited suspension movement. It uses narrow racing dunlop tires.
BTW--Didn't NASCAR boys used to move the panhard to the opposite side when running roadcourses?
I don't know about COT though.

[phantom II]

I was at the David Piper Races near Pretoria S Africa last month. There were bunches of Alphas. I always thought that they handled well with their customary lean. Your best bet would be to contact the many Alfa forums to seek a solution to your problem, what ever that may be. I'm not sure what Alfa you have or which one had 150hp.
Can you find a picture in one of you books of the chassis lay out or pictures of your car in action(video). I can design a chassis and suspension from scratch not quite to Newey's standard, mind you, but fixing an existing design can be tricky, especially not having driven it or seeing it. I know toe out bump steer is an integral part to Porsche front suspension design to interact and compliment the rear geometry. Is that what you have? I notice old Alfas inside front wheel toes out and gets seriously negative in turns.












The Trac Rod or Panhard Rod on NACARs not only changes the thrust line by moving it left to right, but also roll center.

Originally posted by John Brundage
JB--
I didn't realize that the RC would be different left and right in roll until I ran it on the program. Because the upper axle locate link is not on the center line may influence this. There is no room to locate it on the centerline. I tried the panhard with different lenghts, angles, pick-up points etc. to try to get it as close to equal as possible without success. Again, I was just trying to make it theoretically even from side to side assuming the turns were the same (which they never are). If I am doing this exercise, I want to take the time to decide which may be best and get it as close to even as possible, instead of just mounting a panhard or Watts and seeing what happens. I would rather start with something that is balanced and work from there.
These cars do have some roll to them. It is not a go-kart or track car running with limited suspension movement. It uses narrow racing dunlop tires.
BTW--Didn't NASCAR boys used to move the panhard to the opposite side when running roadcourses?
I don't know about COT though.

[Powersteer]

Thought that if the Watts crank were center of the axle it should influence a typical roll center that is center on a live axle,similar on one that is mounted to the chassis. If it was lower than axle center it would bring down roll center and load up the suspension more from side forces.

[John Brundage]

Originally posted by Powersteer
Thought that if the Watts crank were center of the axle it should influence a typical roll center that is center on a live axle,similar on one that is mounted to the chassis. If it was lower than axle center it would bring down roll center and load up the suspension more from side forces.

Still thinking about this, but thought I would take a look at a Ferrari 250 GTO. It is original with the original Watts link center pin on the diff housing. The center pin is below the axle center line.

[Joe Bosworth]

John

I have just come into tune with you on the panhard rod/roll centre thing.

If your panhard rod mounts from chassis to dif centre there is no RC change with body lean. If your rod mounts from chassis on one side to past dif housing centerline to fixing near wheel fully over on the opposite side of chassis mount there will be a roll centre change right turn and left turn. With a heavily leaning car as you are using the chassis mount rises in one kind of turn and falls with the other with the RC raising and lowering about half the distance of the rise/fall of the chassis mount.

in a close to nil leaning car this is of no significance. With high softly sprung things it might be significant. Shorten the rod to eliminate or minimise the affect but you might then say that the lateral displacement of shorter bar will be felt. I have driven plenty of short bar stuff and never felt it but also have driven much long rod stuff albeit not leaners. So if you are concerned about the real affects of the RC or lateral movement of rods you go to the Watts linkage that avoids both probleems. Please don't comp;ain that the Watts setup adds weight over and above a panhard rod because the world is filled with compromises. It is just a metter of which compromise you wish to live with.

My recommendation is to throw the computer programme away, do something, measure and test the results and run or alter from that baseline. The programmes are only a guide and none that I know of corelate to real life lap times. You don't get that correlation unless you run back to back cars or back to back tests on the same car. By that time you are past the need for a computer programme.

Regards

[John Brundage]

JB
Thanks for the reply, I wan't concerned about the weight of the Watts. 20 lbs or so won't make a difference to us. We are not going for a World Championship. The is my first time using the computer program to look at suspension. I have always done it old school and adjusted accordingly at the track. I always started with getting a car square at the shop. Due to the complaints of the Alfa, and my constant desire to improve, I plugged the existing info from the car into the computer. I learned alot by changing things in the computer and seeing the electronic outcome. But as I already noted earlier "Being that the computer progam is not driving the car, I want to make the rear center pin adjustable to adjust the car based on the driver's feedback and preferences."

Powersteer-- Nice pics. Alot do handle well with the lean, for some reason this car did not. I have found problems that I am rectifying be getting the suspension and steering square. I have seen dropped spindles and panhard bars added to 101 Alfas with mixed results. I don't have all the details regarding the differences between those installs, I am not sure why some were better than others. The 105 Alfas had different front suspension. An Autodelta GTA used a pin mounted on the diff cover, sliding in a track for rear roll center location. I will be adjusting bumpsteer once I have everything back together. I usually try to get as close to .001 as possible. It will be harder with the original steering arms and tierod ends. I will have to re-check Ackerman so that I will understand what the wheels will be doing in a turn.

I am sold on the Watts link with an adjustable center pin that is frame mounted. Time will tell.

Now, the lower rear roll center will most likely require a spring change.....Hmm any thoughts?

[mariner]

Just a thought. The car seems to have a tubular or space frame from your picture so I assume that bits can be fabricated onto the chassis without too much hassle.

If so you could consider a forward facing bottom "A" frame with the single ball joint on the chassis with a few alternative mounts on two vertical "L" section plates ( for rigidity).

Why put the roll centre on the chassis? Well if you have a high C of G and soft springs and then use an axle based RC point then the roll couple of the C of G versus the RC will change all the time.

IF you fix the RC actually ON the chassis then the roll couple from the C of G to the RC is always constant. This may not be perfect suspension theory but you talk about the driver not being happy so if you do as above you may get more roll couple consistency. That may be a better compromise than thinking about the RC to ground height when thinng are bouncing up and down.

[John Brundage]

Mariner--
The picture earlier in the thread is not mine. The car in question is a stock Alfa with stock frame rails. It uses a lower arm on either side and the single upper link.
I understand what you are saying, but the rear roll center changes on a formula car and they do rather well. My thought was to get front and rear changes to work well together. This is why I am opting for an adjustable rear center pin.

[mariner]

As it is a stock Alfa then modifying the chasis is trickier.

If you cannot get the roll centre low enough then you could take a leaf out of Arthur Mallock's U2 book and go for a Mumford link. that is symetrical and can even give below ground RC if you really want to. It was described in an old "racecar Enginering " I think.

Here is a diagram

http://www.ltu.edu/e..._mechanical.asp

[NRoshier]

perhaps this would be interesting? It is mostly a commercially available kit for an Escort.

[cheapracer]

Originally posted by gruntguru

Shouldn't you stick the jack under one side of the axle? Jacking under the bodywork changes the rear ride height unlike pure roll on the track.

Well I agree and that would be fine if there was such a thing, but if the track was that smooth then no suspension at all would be required - cars go up and down over bumps, undulations etc. as well as roll.

Mariner has just said what I should have said, the RC to CG doesn't change when the center pin is on the body but on the axle it does and most certainly has lateral movement as the body is rolling around the combined geometry of the center pin, the bellcrank and the 2 links.

[cheapracer]

......................

[gruntguru]

Originally posted by cheapracer
Well I agree and that would be fine if there was such a thing, but if the track was that smooth then no suspension at all would be required - cars go up and down over bumps, undulations etc. as well as roll.

Mariner has just said what I should have said, the RC to CG doesn't change when the center pin is on the body but on the axle it does and most certainly has lateral movement as the body is rolling around the combined geometry of the center pin, the bellcrank and the 2 links.

OK - thanks.

[cheapracer]

Originally posted by John Brundage
Mariner--
My thought was to get front and rear changes to work well together. This is why I am opting for an adjustable rear center pin.

FWIW, mount the center pin mounting as far back as possble to make the longest distance you can between front and rear RC's for a bit of a gain in overall stability - reducing the change of pitch of the roll line under braking etc.

[cheapracer]

Originally posted by gruntguru
OK - thanks.

Is that it? Just OK? Darn, darn I was looking forward to being baited, I think you did this on purpose to make me feel hollow, thats even worse than baiting me, bastard

[cheapracer]

Originally posted by cheapracer


FWIW, mount the center pin mounting as far back as possble to make the longest distance you can between front and rear RC's for a bit of a gain in overall stability - reducing the change of pitch of the roll line under braking etc.

Actually I need to retract this, I was generalising with theoretical RC's but when it's actually a fixed RC pivot on the body as in this case then the pitch will always be the same, sorry.

There could be some yaw effects depending on the mounting point though, Greg??

[phantom II]

You have to fix one thing at a time otherwise you won't know what is working. You really aught to read some books. I don't wan't to write one here. I think your roll couple is something you should address first. That seems to be the major problem from what you describe. It would be nice to know what the car is doing. Anyway, assuming this is the problem, starting with a Watts is a good thing to lower rear RC. Trust your instincts.
Roll couple percentage is the effective wheel rates, in roll, of each axle of the vehicle as a ratio of the vehicle's total roll rate. Roll Couple Percentage is critical in accurately balancing the handling of a vehicle.
A vehicle with a roll couple percentage of 70% will transfer 70% of its sprung weight transfer at the front of the vehicle during cornering.
Weight transfer during cornering, acceleration or braking is usually calculated per individual wheel and compared with the static weights for the same wheels. Cornering wheel weights requires knowing the static wheel weights and adding or subtracting the unsprung, sprung and jacking forces at each wheel. Very important.
Sprung Weight Transfer is the weight transferred by only the weight of the vehicle resting on the springs not the total vehicle weight. Calculating this requires knowing the vehicles sprung weight, the front and rear roll center heights and the sprung center of gravity height which is used to calculate the roll moment arm length. Calculating the front and rear sprung weight transfer will require knowing the roll couple percentage.*****
The roll axis is the line through the front and rear roll centers that the vehicle rolls around during cornering. The distance from this axis to the sprung center of gravity height is the roll moment arm length.
If you have corner weight scales and an hydraulic jack, you can determine the CG.
The total sprung weight transfer is equal to the G-force times the sprung weight times the roll moment arm length divided by the effective track width. Sorry, I don't know how to type math. The front sprung weight transfer is calculated by multiplying the roll couple percentage times the total sprung weight transfer. The rear is just the total minus the front transfer. Leave the front suspension as it is and have fun with the back.
Again, what is the car doing? The stock steering arm and bump is the last thing I'd fool with. How much surgery do the rules allow anyway?
I have a F250 Lusso which has the same chassis and running gear as the GTO. It has leaf springs at the rear and no Watts although the SWD did.
With your low front RC, the rear must be made as low as possible. This is where the Watts comes in.
The bushings on the axle and body must be on the same plane as the bellcrank pin. There will be no lateral movement even if you could move the axle 6" vertically. IE: Provided the locate hard points are about 48" apart. Check this movement on your computer.
Don't expect miracles with an adjustable pin, just make it low as possible and fixed.
How long are the trailing arms and where are the springs located on these links? Can you lengthen them into the body?
Mounted where they are already lowers the roll center statically and if they are long enough, dynamically also but there is such a thing as jacking forces.
These can be thought of as the centripetal force pushing diagonally upward from the tire contact patch into the suspension roll center. That is why I ask where are the rear springs are mounted? Alfa may have soft springs there for a reason. Can the springs be place on top or behind the axle at its extremities(wide apart) to reduce jacking? This way, you can have longer springs and shocks and a higher rate. Coil binding is a bad thing and this is something that you probably have already, especially at the back.
Jacking forces are calculated by taking the front sprung weight times the G-force times the front roll center height divided by the track width. The rear is calculated the same way. Check this at the rear each time you change something and catalog it. One thing at a time, please and remember, you are doing this for fun.

Originally posted by John Brundage
The 105 Alfas had different front suspension. An Autodelta GTA used a pin mounted on the diff cover, sliding in a track for rear roll center location. I will be adjusting bumpsteer once I have everything back together. I usually try to get as close to .001 as possible. It will be harder with the original steering arms and tierod ends. I will have to re-check Ackerman so that I will understand what the wheels will be doing in a turn.

I am sold on the Watts link with an adjustable center pin that is frame mounted. Time will tell.

Now, the lower rear roll center will most likely require a spring change.....Hmm any thoughts?

[Greg Locock]

Sorry I'm a bit pushed for time, my honest answer is that most drivers (including me) can't reliably pick which way a Panhard rod is mounted (ie left or right to body), so a competent Watts is going to be fine, whatever the installation details.

I'd rather have the roll centre at a height that is stable wrt to the contact patch, so that implies that pin on axle should be better, by that theory. However for racers pin on body seems to work fine.

Incidentally I strongly suspect that the RCH for a live axle, and an IRS, are not the same thing at all. Nobody would design an IRS with a RCH of 307 mm, but coil sprung live axles I've worked on (three and a half designs so far) has had an RCH of at least that much, and they handle just fine on smooth roads. A good watts setup is as good as a poor IRS (say a semi trailing arm. no prizes for guessing which) on smoothish roads, the only pickable differences in favour of the IRS were stability on unsealed roads, and traction out of corners, which is obviously vital for a race car.

[John Brundage]

Checking my notes, the roll couple percentage was about 70%
The lower rear axle rods are mounted below the axle and are parrallel to the ground at ride hight. The rear lower rods are 24" long. The single rear upper is about 8.5" long. These cannot be relocated. The coil springs are mounted on top of the axle, directly above the lower rod mount. The shocks are mounted in the middle of the spring perch on top of the axle. These springs are soft and will not be in bind--too few coils etc. The spring perches on the axle are 32" on center and the axle measures aprox. 53" from axle flange to axle flange. We were going to mount the Watts link axle supports at the rod mount location. I still need to input all the changes into the program from the repairs/centering of the front suspension.
Were you suggesting that the center pin, which will be chassis mounted, be at the same plane as the rods, below the axle center line?
The 250 GTO does have a Watts link. The car I am looking at is all original and never modified or restored.
However, I looked at a 1960 250 GT. That does not have a Watts link.

Mariner--I did look into the Mumford link, but settled on the Watts

NRoshier--I have found similar designs for the Mustang. This is what we will be designing for the Alfa.

[phantom II]

The bellcrank must be on the diff and not on the chassis. Everything looks good except for two things. You say that the springs are 32" apart? Can you get them further apart, even if you angle them? Use your head for this one.***** The trailing arms are good but the upper link is too short.
Here is a suggestion.
Remove the upper link. Fabricate a Reese bar about 40" long. Check out NTSOS homepage. You may be able to use the gearbox mount hard points with large flex rubber bushes. Make sure that the Reese Bar and the axle move as a unit. The pin can be mounted on the Reese Bar under the diff. The axle bracket is welded as far as possible away from the diff under the axle. Half of the bell crank will protrude behind the diff.
Drop a bracket down from the chassis for the other Watts link and use Heims every where.
Everything can be done in aluminum. All bolt on.
Fat high profile tires are a good thing. The lower the pressure, the more self damping properties they have and in some sidewall designs, can improve contact patch on this axle. You can experiment with tire pressures.

Originally posted by John Brundage
Checking my notes, the roll couple percentage was about 70%
The lower rear axle rods are mounted below the axle and are parallel to the ground at ride hight. The rear lower rods are 24" long. The single rear upper is about 8.5" long. These cannot be relocated. The coil springs are mounted on top of the axle, directly above the lower rod mount. The shocks are mounted in the middle of the spring perch on top of the axle. These springs are soft and will not be in bind--too few coils etc. The spring perches on the axle are 32" on center and the axle measures aprox. 53" from axle flange to axle flange. We were going to mount the Watts link axle supports at the rod mount location. I still need to input all the changes into the program from the repairs/centering of the front suspension.
Were you suggesting that the center pin, which will be chassis mounted, be at the same plane as the rods, below the axle center line?
The 250 GTO does have a Watts link. The car I am looking at is all original and never modified or restored.
However, I looked at a 1960 250 GT. That does not have a Watts link.

Mariner--I did look into the Mumford link, but settled on the Watts

NRoshier--I have found similar designs for the Mustang. This is what we will be designing for the Alfa.

[John Brundage]

Originally posted by phantom II
The bellcrank must be on the diff and not on the chassis. Everything looks good except for two things. You say that the springs are 32" apart? Can you get them further apart, even if you angle them? Use your head for this one.***** The trailing arms are good but the upper link is too short.
Here is a suggestion.
Remove the upper link. Fabricate a Reese bar about 40" long. Check out NTSOS homepage. You may be able to use the gearbox mount hard points with large flex rubber bushes. Make sure that the Reese Bar and the axle move as a unit. The pin can be mounted on the Reese Bar under the diff. The axle bracket is welded as far as possible away from the diff under the axle. Half of the bell crank will protrude behind the diff.
Drop a bracket down from the chassis for the other Watts link and use Heims every where.
Everything can be done in aluminum. All bolt on.
Fat high profile tires are a good thing. The lower the pressure, the more self damping properties they have and in some sidewall designs, can improve contact patch on this axle. You can experiment with tire pressures.

I checked out NTSOS homepage. The rear end on the blond is perfectly aligned. And the car is quite nice as well.
The center cection on the Alfa diff is a cast aluminum piece. It does not have a back cover or any cover. The axle tubes bolt to it. Modifying the housing by welding would most likely distort it. This one one of the reasons I have decided on the frame mount. I understand that the characteristics are different, but will still definately improve the car. The upper link is also at an angle and the pin is not on the centerline. There was no room to mount it on the center. The Reese bar is interesting. The car has a two piece driveshaft with a body mounted centerbearing. Running a link past the centerbearing to the gearbox seems like it would stress the centerbearing. But I think I will need to stick with the original upper link. The suspension needs to be original. Installing a panhard or watts in this case, is a stretch of the rules, but is done. Tire size is per the vehicle and we cannot alter the size or compound. We cannot alter spring location either. The tops of the springs are mounted in up-side down buckets (spring towers) inside the car at about the hight of the top of the rear seat. It makes everything an interesting challenge. To try to maintain originality and get the cars to run differently (and better, hopefully) than they did many years ago.

[cheapracer]

Originally posted by phantom II
The bellcrank must be on the diff and not on the chassis.

I think it would be reasonable etiquette to give a reason why when a sentence contains "must".....

Anyone had experience with a Jacobs link by the way?

[phantom II]

Foxtrot foxtrot sierra, what kind of foxtrot alpha hotel are you, you foxtrot mike? I hate wasting my foxtroting time, you foxtrot charlie, you.
Now there is more information. Is this a guessing game? It is grossly inconsiderate and unfair of you to ask a question on this forum in this frustrating manor and expect help from the honorable members who are otherwise generous with their help.
So its got a two peice drive shaft, has it? What the foxtrot else?

41 foxtrot posts and still no foxtrot pictures of the foxtrot chassis and we still don't know exactly what Alfa you have? You won't even state what the foxtrot is wrong with the foxtrot car.
Since when did I suggest that you must weld the foxtrot housing, you foxtrot lema?
If you were a decent human being, you would have furnished this information in your first post including the rules and regulations.
I'll tell you where you can place the foxtrot Watts. Somewhere, where the moon don't never shine.

Originally posted by John Brundage
Modifying the housing by welding would most likely distort it.
The Reese bar is interesting. The car has a two piece driveshaft with a body mounted centerbearing. Running a link past the centerbearing to the gearbox seems like it would stress the centerbearing. The suspension needs to be original. Installing a panhard or watts in this case, is a stretch of the rules, but is done. We cannot alter spring location either. It makes everything an interesting challenge. To try to maintain originality and get the cars to run differently (and better, hopefully) than they did many years ago.

[phantom II]

See above post. Out put is based on input.

Originally posted by cheapracer


I think it would be reasonable etiquette to give a reason why when a sentence contains "must".....

Anyone had experience with a Jacobs link by the way?

[John Brundage]

I sincerely appreciate all constructive answers to my initial question which was regarding the prefered mounting of a Watts link--frame or rearend mounted. When asked I have answered that the car is an Alfa SZ and also refered to it as a 101 series Alfa. I have said that it is being vintage raced. Most vintage race governing bodies dictate that the car must be period correct; though each has different regs. I have also said that the car is original with original suspension (except for springs) and frame rails. I have also said more than a couple times that we need to run the narrow racing Dunlops. The only information I did not supply was pictures. That is because the rear axle is out of the car and apart. I know that you did not mention welding the pin and that you said to mount the pin on the Reese, I wanted to mention that this too was not an option. I asked the question here because of the knowledge here. I am sure that some are familiar with an Alfa SZ or know what it is. I did not resort to putting some one down for not knowing that a GTO had a Watts link, nor was I ever nasty when I repeated a statement that had been made earlier in the thread.
I do appreciate the earlier responces, but I would like to know why, in your (Phantom II) opinion should the bellcrank be on the diff and not on the chassis? That is the original question--which is the preffered method and why.

[phantom II]

You can't just go and add things to a car that has already had many issues addressed in it's original design, such as rear axle location and the myriad of things that are entailed therein unless a thorough investigation of said design is executed. You must understand what you have before you can modify it. You can buy a nice suspension design program from Lotus for about $600.

"I am working on modifying a vintage race car and installing a Watts link."

If it were a general question, it would be an entirely different matter. When you said you had 150hp Alfa, I didn't think it could be a 1960s car. I had no issue with the required tires other than you were going to add other things which were not included on the original car. A contradiction I would think. You went all over the place with Ackerman, bump steer and other such and yet you wanted answers for a specific as yet non existent component. For the umpteenth time, I have yet to discover what the handling problem of the car is. I struggled to get sufficient information to get the full picture and the two part drive shaft was the last straw of my patience. Pictures are worth a thousand words.
You can't remove the upper link, yet you are considering attaching links either to both sides of the axle and fabricating a mount point for the bellcrank on the chassis or mounting it on the diff housing. Either way, some surgery would be required which tomy mind it is not in keeping with the letter of the law.
With this new information, I can safely say that unless you remove the upper locate link, you can't install a Watts either way. The two part drive shaft is there to address the poor location and dimensions of the upper link. If this link is too rigid, another thing I don't know, the Watts pin will sheer or brake off its mount point, be it on the frame or the axle and it won't do diddly for RC. The RC is already as low as it can go because of the design of the lower trailing arms.
If you mount the Watts bellcrank to the chassis, the violence would even be greater between it and the upper link.
Depending on how one locates the axle, the location of the watts bellcrank will be determined. Generally, the pin on the diff is the best for the least migration of the RC.

I have a very expensive die cast model of a 62 GTO. It doesnt have a Watts. I have a real 63 250 Lusso which has the same chassis and running gear as the GTO and it doesn't have a Watts. That is not to say that some of the 39 cars didn't.

Originally posted by John Brundage
I sincerely appreciate all constructive answers to my initial question which was regarding the prefered mounting of a Watts link--frame or rearend mounted. When asked I have answered that the car is an Alfa SZ and also refered to it as a 101 series Alfa. I have said that it is being vintage raced. Most vintage race governing bodies dictate that the car must be period correct; though each has different regs. I have also said that the car is original with original suspension (except for springs) and frame rails. I have also said more than a couple times that we need to run the narrow racing Dunlops. The only information I did not supply was pictures. That is because the rear axle is out of the car and apart. I know that you did not mention welding the pin and that you said to mount the pin on the Reese, I wanted to mention that this too was not an option. I asked the question here because of the knowledge here. I am sure that some are familiar with an Alfa SZ or know what it is. I did not resort to putting some one down for not knowing that a GTO had a Watts link, nor was I ever nasty when I repeated a statement that had been made earlier in the thread.
I do appreciate the earlier responces, but I would like to know why, in your (Phantom II) opinion should the bellcrank be on the diff and not on the chassis? That is the original question--which is the preffered method and why.

[John Brundage]

Actually,
I am using a program which pointed out some issues that I have corrected in the front geometry. The car would push, then snap oversteer among other things. It was generally ill mannered. The corrections I made to the front, which I mentioned earlier would make the car handle better and take away most of the issues. This just happens to be a good time, with everything apart to make the car work even beter. Alfas have been modified with panhard bars to lower the rear roll center. I was opting for a Watts link instead. Yes the Watts is a stretch of the rules (and so is the panhard). The single upper link was originally an A frame as I mentioned earlier that would locate the rear end. When the panhard is installed, and in this case the Watts, the upper A-link has one leg removed which allows the rear end to move side to side. The panhard or Watts will then locate the rear axle to eliminate this movement. The rear upper linkmount on the side of the diff is the rear static roll center (if I am not mistaken) This pin is not on the center line unfortunately, but it is the way the suspension was originally designed by Alfa. The Watts will lower the static roll center. With the program I found the roll center differences left and right in roll when using the panhard rod, probably in part due to the original offset upper link. As I stated earlier, I tried different panhard mount, lengths etc. in the program to reduce the differences before deciding to try a Watts. The program showed that a Watts eliminated this. I will try to get a picture of parts on the bench.

[John Brundage]

I did run a simulation with both the Watts and panhard to see if there was a bind issue. The closer I got to eliminating the roll center differences with the panhard, I was putting the suspension in bind. I had no bind issues with the Watts on the simulation.

[cheapracer]

Originally posted by phantom II
Foxtrot foxtrot sierra, what kind of foxtrot alpha hotel are you, you foxtrot mike? I hate wasting my foxtroting time, you foxtrot charlie, you.
Now there is more information. Is this a guessing game? It is grossly inconsiderate and unfair of you to ask a question on this forum in this frustrating manor and expect help from the honorable members who are otherwise generous with their help.
So its got a two peice drive shaft, has it? What the foxtrot else?

41 foxtrot posts and still no foxtrot pictures of the foxtrot chassis and we still don't know exactly what Alfa you have? You won't even state what the foxtrot is wrong with the foxtrot car.
Since when did I suggest that you must weld the foxtrot housing, you foxtrot lema?
If you were a decent human being, you would have furnished this information in your first post including the rules and regulations.
I'll tell you where you can place the foxtrot Watts. Somewhere, where the moon don't never shine.

I needed that, ahhh I'm having a bad day and that one just cracked me up!

[cheapracer]

Originally posted by John Brundage
I did run a simulation with both the Watts and panhard to see if there was a bind issue. The closer I got to eliminating the roll center differences with the panhard, I was putting the suspension in bind. I had no bind issues with the Watts on the simulation.

How can you bind a Panhard Rod? It goes up and down in a single plane, either theres something in the way or theres not, it's that simple ffs

If your using Performance Trends then often you can run into binding thats not actually there for real. I have real working suspension models the software either binds or won't accept in the first instance. Lucky I was building cars before there was such things as PC's