Rear roll center mods and modified rear control arm with rod ends

[blueovalz]

Jon lit a fire under my back side (as he usually does when he starts one of his many notable projects) to modify the rear suspension of the BlueOvalZ (A.K.A. The Ongoing Project).

 

I finished up the control arm mods using Jon's idea, but with slight modifications to suite my personal taste. I chose to use a single 1/4 X 2" steel bar for the rod end support.

 

 

 

 

Prior to this change in the lower control arm, I wanted to raise the roll center of the rear. I had some spare struts, so I modified one to test. and if I decide it is worth more work, I'll repeat it for the other side. Below is the modified strut with the lower portion extended by two inches. I used 1/4" flat bar for all material, and then used 1/8" sheet on the outside to overlap previous welds.

 

Jig I built to ensure everything was correctly aligned when completed:

 

On the car to test fit and align the tube ends for the control arm modifications:

 

Before and after:

 

Another view:

[JMortensen]

That really gets the roll center up there. Are you going to run softer springs? I imagine that would be the reason you did this mod.

 

Looking at the pics looks like you can run some much longer end links on the swaybar now. You know, while you're at it...

 

I like the way you made the jig for the control arm. I'm going to need to see if mine will work. I think I may have some fitment issues. We'll see. Looks like you turned them down a bit? Or at least took the shoulder off the end of the tube end.

[v80z]

Forgive my ignorance. I am always interested in what you are doing.

 

What is roll center?

 

What changes to the geometry does your mod effect?

 

I need to study more.

 

thanks

 

Joe

[johnc]

Just curious, why are you raising the rear roll center?

[boodlefoof]

Roll center height is the imaginary point about which the body tries to roll. Raising the roll center height reduces the angle the body will roll in a corner. Sort of like adding a larger sway bar. Except that raising the roll center height too much can cause sudden jerking motions as the suspension handles a turn.

 

I'm not as up on strut suspensions as SLA suspensions, but I think you measure RCH in a strut suspension as follows: Draw out a line parallel to the lower control arm towards the center of the car. Then, draw a line at a 90* angle from the top of your strut towards the center of the car. These two lines will intersect somewhere... Then draw a line from that intersection point to the ground at the center of the tire. The point at which this line crosses the car's centerline is the RCH.

[blueovalz]

The main reason I wanted to do this was to get the control arm geometry back to what I would call "reasonable". It is angled up from horizontal (at the outboard end) on my existing set-up. I'd like to get a little more camber action as well as get the roll center back up a few inches above ground. Kinda experimenting right now.

[johnc]

I forgot how much you had lowered the car. Getting the rear roll center back above ground is a good idea.

 

A couple months ago I purchased Mark Ortiz's "Managing Your Anti" video. I and a couple of my customers watched it a number of times because the concepts are hard to grasp. In general he takes a force vector approach to determining roll centers, anti-roll, and anti-dive which is different then the more traditional geometric approach.

 

We made some changes to the roll centers on a SCCA Solo2 DSP BMW 325i and noticed some subtle, but very nice attitude changes. Under hard braking the entire chassis squats down instead of pitching forward. Back to back to back testing at a practice last month showed a 3 tenth improvement in course times. At the nationals the car is currently 4th overall in DSP with the final runs today. The car is first overall in DSPL and won the overall L2 championship in ProSolo.

 

I need to learn more about this force vector stuff. I feel like a miner in a cave without a flashlight.

[JMortensen]

A couple months ago I purchased Mark Ortiz's "Managing Your Anti" video. I and a couple of my customers watched it a number of times because the concepts are hard to grasp. In general he takes a force vector approach to determining roll centers, anti-roll, and anti-dive which is different then the more traditional geometric approach.

 

We made some changes to the roll centers on a SCCA Solo2 DSP BMW 325i and noticed some subtle, but very nice attitude changes. Under hard braking the entire chassis squats down instead of pitching forward. Back to back to back testing at a practice last month showed a 3 tenth improvement in course times. At the nationals the car is currently 4th overall in DSP with the final runs today. The car is first overall in DSPL and won the overall L2 championship in ProSolo.

Sounds like an interesting video. Did you raise the rear roll center to get it to squat on braking? That's the one (and probably only) thing I think semi-trailing arm rear suspension has over our chapman struts. Porsche 944's do that really nicely too; the whole cars just sets down when you stand on the brakes. Of course, if its a bumpy corner entrance the rear suspension gets a bit locked up by the braking action... then I think our Z's still have the advantage.

[johnc]

Initially the customer did the traditional modified E36 thing and adjusted the front to gain maximum static camber. That was when he was running 245 width tires. When he went with 285 width tires he didn't adjust anything. I had to talk with him for a few months and finally convinced him to reduce the static negative camber and only use what he could get through the camber plates. That freed up the offset rear LCA bushing so we could play with the roll centers and anti-dive.

 

We took out all anti-dive and were able to run 700 lb. front springs and reduce the front ARB size. The car pitched forward more but would really bite on turn in and just fly around corners. Most recently, after the video, we added in a little anti-dive in the front (asymetrically in fact - based on the corner weights), brought the front roll center up, and adjusted the spring rate. Now he's got the best of both worlds and is faster.

 

Sorry, forgot to mention that yes, we did raise the rear roll center, but not as much as the front. We probably have a little more work to do back there but the SP rules limit what can be done.

[tube80z]

A couple months ago I purchased Mark Ortiz's "Managing Your Anti" video. I and a couple of my customers watched it a number of times because the concepts are hard to grasp. In general he takes a force vector approach to determining roll centers, anti-roll, and anti-dive which is different then the more traditional geometric approach.

 

I have his video and brought it up when I took the Motec seminar with Claude Rouelle. He spends a lot of time discussing suspension kinematics. What I got out of that class was the classical approach to thinking about RCs in general isn't really very good and we don't really have decent tools to work with them. What I mean by that is most software doesn't really roll the car realistically and how the points move (RC, IC, etc.) isn't real and tire and component deflections can cause large changes.

 

You can analyze geometry and some decent cars show a lot of RC movement, which many books tell you is bad. But when you use the force based approach you will find that this isn't the case. Claude went on to say that movement isn't so bad as velocity. That's one item that really needs to be watched.

 

I don't know if you get racetech magazine or not but they had a really good three part article on force based RCs from Andy Thorby. Like anything the more you know the more complex it gets but his explanations were very good and he provided a few tools to help.

 

We made some changes to the roll centers on a SCCA Solo2 DSP BMW 325i and noticed some subtle, but very nice attitude changes. Under hard braking the entire chassis squats down instead of pitching forward.

 

In the motec seminar Claude goes over pitch centers and how they work, which is exactly what you're seeing here. If you get a chance his class is really a good one to take and will shine a bright light down the mineshaft.

 

Cary

[v80z]

Now I see the difference between the before and after. Thanks for the explanations. You have raised the wheel in relation to the car. This effectively lowers the car. Correct?. The first impression I get is that it becomes more difficult to raise and inside wheel in the cirner. Correct?

[JMortensen]

Not really. It doesn't change the height of the car, instead this lowers the outer pivot of the control arm. In simple terms if the control arm is level and the car gets a large lateral load, no jacking effect is produced. If the control arm angles up from the inboard to the outboard pivot, and you introduce the lateral load, the side load will compress the suspension. If it angles down, it will extend the suspension. That's the simple version. In reality there is weight transfer that compresses the outside suspension, so what you get by having the control arm point down is the effect of slightly reducing the body roll that you would have with the control arm level, or maybe more significantly reducing it from what it would be with the control arm pointing upwards.

 

The other ways to combat roll are with larger sway bars and stiffer springs. So you can get the same amount of roll in a car by raising the roll center, not raising the roll center and adding heavier sway bars, or by lowering the roll center and adding stiffer springs and/or sway bars.

 

If the roll center is REALLY high, like an old Baja Bug for instance, the suspension might extend drastically with a side load and tend to want to make the car roll over. But Zs, especially the ones with really big wheels like Terry has, generally are lowered to the point where the roll center is underground, and that, plus the fact that they're now running low in the travel, means that seriously stiff springs must be used, or you're driving around on the bumpstops.

[Jolane]

Terry,

 

I like the way you modified the LCA's. Looks like a nice clean setup with a lot of adjustability. Did you mount the heim's with spacers between them and the strut? Are you using bolts throught the heims, or the stock pin?

 

Thanks,

Joshua

[blueovalz]

Yes, I used 1/4" spacers between the pin boss (strut) and the rod ends. I don't feel these are necessary, but I wanted some means of being able to tune up the fit, and to allow for a small amount of placement error in my assembly (welding of the critical parts). If needed, these spacers could be changed (thinned or thickened) so that the strut could be moved forward or rearward a small amount if I happened to have the rod ends not perfectly placed, but as things turned out, they were perfect (only because I assembled everything on the car, and then tacked in the tube ends to insure everything bolted up in the correct position).

 

The OEM pin is used, but modified by cutting 5/8" threads on the ends to tighten the assembly in place. I could have used a 5/8" bolt as well, and this would simplify things a great deal, but I was a little paranoid about the .010" slack fit between the bolt and the 3/4"-5/8" reducers that fit into the 3/4" rod ends. As it turned out, the poor tolerance control in these machined reducers (bought on Ebay) gave them a snug fit on the OEM pin, so I chose to use the pin instead of the bolt. FWIW, a 5/8" bolt has a shank that is in reality .620" in diameter (I found all bolt shanks are about 1% smaller than there nominal size).

[Jolane]

Thanks for a detailed reply Blue. Just to make sure I understand, you threaded the stock pin to 5/8" thread (corse probably), then used a nut to tighten everything down. It looks like you also put the stock nuts on perhaps as a jam nut?

 

Again, those looks great.

 

Thank you,

Joshua

[blueovalz]

Yes, yes, and yes. The only tricky thing about all of this is the precision required in the placement of the rod ends. It all looked fairly simple and straight forward when I started, but midway through the design stage, I realized that with no rubber to allow some degree of misalignment, I had to get the rod ends in the "dead-on" correct position. Not only to align the strut correctly in the very important pitch axis(the yaw axis could be adjusted out with the rod ends), but also in the rod end spacing (so that the balls would not be preloaded in the longitudinal axis when the assembly is tightened together). The rod end spacing issue could be corrected by changing the exact width of the 1/4" spacers (adding or subtracting a few thousanths), which is why these were added in case a future strut change had slightly different pin boss dimensions from the strut I am currently using.

 

Any misalignment in either of these two areas would not bode well for the project. It makes a jig imperative, and gives me some insight into the difficulty in using multiple rod ends for suspension arms of this design. This set-up leaves little (if any) room for error.

 

[JMortensen]

I did not use a jig when I welded my control arms up, and that may come back to haunt me. Terry increased the need for precision by the way that he set his strut housing up though. Rather than use a smaller ~.618" (5/8") bolt, he used the larger .631" spindle pin. In order to fit the outer rod end to the .631" (~16mm) spindle pin, he used a 5/8" thread, 3/4" overside rod end with a spacer turned down to .631 for a nice tight fit on the pin.

 

In the thread that started both of our projects, the main issue for Terry was that the spindle pin was "loose" on the polyurethane bushing sleeve. I said at the time and I still think that the slop poses no issue, because once the spindle pin is tightened down the sleeves would be captured and would not move. I'm counting on that being true, because the ~.013" slop, I'm hoping, will allow my control arms to bolt up without putting a side load on the strut itself. If not, I have a few options. One is to get the same rod ends that Terry did and have offset bushings made. The other, and I think I'll try this first if I need to, is to take a 5/8" bolt, cut a section out, and drill a 12mm hole offset in it, then use this to space top of the strut within the monoball at the camber plate without any preload. The last is to cut the tube with the threaded tube ends off and start over. The first two seem like it would work, so hopefully the third will not be needed.