Mazda3 Forums banner

Angle of Control arms (Suspension theory thread)

83K views 511 replies 18 participants last post by  XCNuse  
#1 ·
If anyone can (and this would be fantastic if you can), get a picture of the underside of their car (with it sitting on the ground of course), so I can see the angle of the control arms with the prokit?

A measurement from the ground to the weld seam would be fantastic as well.

Pretty please!! Thanks.
 
#4 ·
If you need stock for comparison, I'd happily shoot a pic of the bottom of my 07. What've you got cooking Nuse?


Sent from my iPhone with Tapatalk.
 
#6 ·
A measure in inches to the weld seam and a picture of the angle would be the perfect way to do it.

It's beyond clear to me how much grip I'm losing by having a loose front end that likes to roll around without me knowing, but not even that sharp turns on normal driving roads and I run out of grip SO FAST it's pathetic.

It was particularly noticeable at two very specific times on this trip to maryland last weekend.
Once on the way up I hit a turnpike in South Carolina and it was a little damp, only going 40-50 (oily?) I almost plowed off the road which was scary as shit. I actually had to brake mid turn (no-no) but it held its line and I made the turn safely.

The other time was going up to the house at epicmeet and a buddy in a speed6 and I were sprinting up a hill, and the first turn into the neighborhood/area is a flat right and a flat/offcamber left, and once again not going that fast I lost all grip through my front end.

On-camber turns my car handles like a beast, off-camber I just get screwed over.
It was hard to notice this on the track because of how much more sticky the NT05s are, but it's very clear to me on street tires, my grip is non-existant.
Especially when a near stock car can take turns as fast / faster than me when that shouldn't at all be the case.
 
#8 ·
I've got the sheet around here somewhere.. maybe; but really it doesn't matter because it all needs to be raised and re-done.

It was close to a stock setup (not track), only difference was camber, which I'm running less of now than I was.
Majority was zeroed or near zeroed out, and camber stayed where it was. Toe I had more set to negative than positive but hardly any at all anyways.
 
#9 ·
Lower doesn't mean more grip. It can mean better handling due to a lower center of gravity, but that's dramatically less important than suspension geometry. That lower cg may not even matter if your roll centers drop so much that the roll moment actually increases.

Stiffness, which most people get only as a side effect of lowering springs, is generally good for grip. This is mostly because of A) controlling load transfer, and B) reduced body roll. For a-arm cars A is much more important, since they gain camber throughout their suspension travel. For strut cars like ours, But matters about equally as much as A.

Unfortunately, in our cars, the suspension geometry, which is the number one priority for making grip, doesn't work well with more than moderate lowering. Once you go beyond that, the lowered cg does nothing to help you. Ideally, there would be a way to adjust our roll centers, but nobody has come up with a good solution for that short of custom LCAs.

My point is that when you say a stock height car shouldn't be able to take turns quicker than you, you're basing that on the idea that low is fast, which is a myth. It's actually quite expected that a stock height car would turn faster than a significantly lower one.
 
#10 ·
Trust me this is known ;)

It's trying to find where that sweet spot of arm motion on our McStrut setup is.

Someone mentioned talking to Tripoint but I don't want to pay $2000 for whatever solution they have.
Guys on MSF are looking into it but no dice, I've looked into it and it could be done, but it won't happen (I don't have access to right people / shop stuff).

(btw I wasn't basing my point on lower is faster; it's the fact he's on mazdaspeed accessory springs and stock everything else. For the n00bs wondering why, yes you give an excellent explanation, but for me that isn't the case [but still good info to add in])


Hence why I've been bugging Corksport about it lol.
Only companies I can see helping is Corksport, AWR, or Tripoint, but chances are they'll never give us a solution (not worth the R&D for the handful of us that want it).
 
#11 ·
Nuse, can you post up a pic of your front suspension shot at ground level (if your LCA's are even visible based on your ride height?) With an angle finder and tape measure, I could probably help to find out if your camber is moving in a positive direction at some point in your suspension travel. That's something I dealt with in my 96 Miata track car when it was lowered WAY too much, and like you my front grip would all of a sudden just go completely away. This is why I always tune suspension on low grip street tires as track tires can mask a lot of these bad behaviors.

I'll have to dust off my suspension theory books (haven't done any number crunching for suspension mod/design since I quit racing when family and finances pushed it to the back burner) but if you'd like the help I'm more than willing to get involved and see if we can figure it out together. This, of course, would be a Hell of a lot easier if I didn't live several hundred miles from you but I'm sure we can figure it out.

My gut instinct based on your symptoms (and not doing the math or taking measurements, just an educated guess) is that your front suspension geometry is compromised due to ride height, and that you are correct in wanting to raise the suspension back up to correct. Not knowing more about your setup, that's all guesswork but at least it's not an uneducated guess.
 
#13 ·
My suspension setup right now is set up to be solid, but positive gain and awkward roll in front is ruining it as far as I can look into it and research I've done.

BC coilovers (dampening is set at/near critical; hard to really know without having them dynoed though).
10k front springs
8k rear springs (will likely be pushing these to 12k later this year)
hotchkiss front and rear sway bars (soft front, stiff rear).

When I had my front bar at full stiff I semi had more grip (and it makes sense due to body roll); BUT thanks to open diff the inside would lift and spin freely (no good of course).

I might be able to get pictures but it's rather useless lol; they're absolutely pointed down, not severely but going the wrong way nonetheless.
 
#14 · (Edited)
Phil: tuning suspension on street tires for use with race tires isn't ideal. You're right that race tires can cover a lot of mistakes, but they are also very different. A setup that gets 100% out of street tires will always get less than peak performance out of slicks.

Just as one example, consider camber gain versus body roll. Ideally, you adjust things so that in a maximum lateral acceleration turn, the body roll and camber gain cancel each other, leaving the tire square to the ground for maximum grip. Then you swap race tires on, and suddenly you have more grip, therefore more body roll, and you go beyond the sweet spot of the camber curve. You could be getting more out of the race tire with more spring rate, so you keep the body roll the same, and the camber happy.

Btw, you mentioned finding your suspension theory books. I think I happen to know what you're looking for so I can save you the headache. Mcstruts begin to gain +camber when the lca is perpendicular to the line from the strut top to the lower ball joint.
 
#16 ·
Phil: tuning suspension on street tires for use with race tires isn't ideal. You're right that race tires can cover a lot of mistakes, but they are also very different. A setup that gets 100% out of street tires will always get less than peak performance out of slicks.

Just as one example, consider camber gain versus body roll. Ideally, you adjust things so that in a maximum lateral acceleration turn, the body roll and camber gain cancel each other, leaving the tire square to the ground for maximum grip. Then you swap race tires on, and suddenly you have more grip, therefore more body roll, and you go beyond the sweet spot of the camber curve. You could be getting more out of the race tire with more spring rate, so you keep the body roll the same, and the camber happy.

Btw, you mentioned finding your suspension theory books. I think I happen to know what you're looking for so I can save you the headache. Mcstruts begin to gain +camber when the lca is perpendicular to the line from the strut top to the lower ball joint.
I wasn't advocating tuning suspension for race tires using street tires, only that I prefer to baseline on street tires before progressing to stickier tires. That helps you figure out if you have a geometry issue (as seams to be the issue here) that race tires will tend to cover up. As grip levels rise, so must spring rates and by extension damping rates/roll stiffness, but at least you're working from a good foundation.

I need to dust off my books, and my own memory. Coming from a double wishbone car (Miata) to a Mcstrut car the theories still apply, but some of the measuring points are different. I used to build spreadsheets to quickly figure wheel travel taking into account spring rate and roll stiffness to help us figure out if we were exceeding the amount of travel we had between static ride height and the point at which we compromised out geometry by the control arms moving beyond their normal range. That, and a bit of stopwatch trial and error, was how I figured out that I was killing my front end grip with by overlowering. I raised my car up 1.5" and increased the spring rate about 40% over what I was already running(already plenty stiffer than stock, but Miatas have pretty mild spring rates in stock form) and found a whole world of grip on street tires I hadn't found before.

Sometimes a set of fresh eyes helpe point things out that are being missed, but it sounds like XCNuse has things well under control. I was initially offering a stock pic/angles/measurements if he needed a control to compare his care and a Pro Kit lowered car.
 
#15 ·
For the sake of simplification and the fact none of us here are turning their cars into full time racecars.. street tires for reference are the way to go.

Besides, NT05s are hardly slicks lol (but when those go I'll definitely be going with a stickier compound most likely)
But.. I do want to keep this conversation alive as it's a very important topic that NEVER happens on this forum. Rarely on the MSF forum; but still majority of those guys just want to go fast in a straight line.

My question though is; IS there a negative camber gain somewhere in the arc of the control arm motion?
If not; then for a mcstrut setup.. wouldn't theoretically on a flat surface having a solid rock hard suspension and a shitload of negative camber be the best thing for it?
 
#17 ·
Oh I see what you mean. That makes sense. Both about street tires and your suspension setup.

If you guys figure out some of the suspension geometry, I'd definitely be interested. I'd love to know the actual position of the cg, and the instant roll centers at stock height.

If it will be any use I can get you pictures and measurements off my car. It's on Koni yellows and stock springs.
 
#18 ·
I mean I guess what I'm curious about (because I dont' know), is like I said there a point where the arc decreases camber and if so, where (within reason of course).

Otherwise, is the height of stock going to allow the best movement (not that I have much to begin with), vs a slight lowering as in say Prokit drop area.

Obviously (well.. I think to us in this conversation at least) decreasing the amount of arc will only be helpful (obviously soft springs out; reason I say prokit is for height, not spring rates).

That said when I drove a 2.0 sedan on prokits in the mountains (Nitto Neogens.. awful loud things) I guess it had to of been how weak the tires were because even that felt like I had no grip.

But I KNOW my Toyo Proxes 4 should have plenty more grip than they seem to produce (more-so in the rain).

I'm happy and comfortable with my setup as is, but I know I'm losing a lot in some spots. But I say I'm happy and comfortable otherwise because if I do find grip; I've put my car into 4 wheel slides before controllably with no issue. But I run in to understeer more often than I feel like I should be.
 
#23 ·
I mean I guess what I'm curious about (because I dont' know), is like I said there a point where the arc decreases camber and if so, where (within reason of course).
This is going to be when the LCA is perpendicular to the line from the strut top to the lower ball joint. That's guaranteed from basic geometry. What you'd need to measure then is the angle and length of this line at stock ride height.


You're right, everybody would love to know where the CG really is! Only way to be absolutely certain is with a crane that you can dangle the car from, but there are ways of doing a decent job of estimation. The fore/aft position should be deducible from the weight balance on the front/rear wheels. The height should be deducible in one of two ways:

1) tilt the car nose up/down while it's on scales. The height of the CG will dictate how quickly the front/rear balance shifts.

2) figure out the roll stiffness and roll center height, then collect data on lateral acceleration and body roll while driving. That should let you find the rolling moment, and thus the CG height.

The second method is nearly impossible to get correct, since you need to know the instant roll center during the turn, which is hard to be sure of. You also need to know the other numbers very accurately. I think method 1 is more reasonable, since it only requires some ramps and scales.

I hadn't even thought about scrub radius. As long as it's within reasonable limits, what does it impact, other than steering feel?
 
#21 ·
Nuse, as luck would have it, I have a stock 3 (2007 hatch) to get measurements from.

It'll take me some time to reacquaint myself with all the measurements, but I could probably lend a hand in this regard. Once I get some good baselines, I can tell you what angles/measurements I need from your car to do the math and figure it out. As you've mentioned, your higher spring rates will limit the suspension travel, but there's still a reasonable amount of travel you need to keep in the suspension to account for roll before your control arms start increasing positive camber which is washing out your grip. My guess is you're experiencing this in the front versus the rear because Macpherson strut suspensions (front) do not gain much camber in travel versus multilink/double wishbone (our rears are multilink) which gain a fair amount. This is more than likely the reason you're not running into a limitation on the rear.

I'll get engaged in this. It may be till next weekend before I can get too much done (recovering from a nasty stomach flu this past weekend, but at least I GOT MY A/C FIXED!!!) but I'll lend what knowledge I can. I just have to translate all this Miata specific knowledge to a new platform, and get rid of some of my old preset suggestions as they don't translate well to a 500lb heavier FWD car.
 
#22 ·
(about damn time; it's getting hot for sure here in GA!)

I can get very accurate measurements as I have access to every part imaginable thanks to Mazmart.
That said, that won't allow me to get angles of said parts (junk yard with half the cars taken apart) but I do have access to knuckles, control arms, etc.

It would be really freaking nice to get a very accurate reading on everything for the sake of the community.

I was reading through a section of VWVortex and they had done some math with their setup (slow day at work; can't complain). And oddly enough they began to see quite a large negative camber gain between a 1-2" drop (theoretically of course; dimensions were incorrect BUT... it goes to prove what I was asking about), and less negative gain at 0" drop, and of course decreasing after 2".. so the MacStrut can probably have a good negative camber gain, question is... where.

What would also be interesting is to see if I should go to a stiffer rear spring setup; otherwise I can keep my rears just as stiff and go to a softer spring up front no problem. (I just don't know right now).
 
#25 ·
You won't be able to find the vertical position of the CG without tilting the car, physics simply doesn't allow for it. You should only need F/R weight balance at a few different pitch angles though, so its not impossible to get the data. Just measure it level, measure it with the front on ramps, then with the rear on ramps.

For the other stuff, I agree -- simple measuring tools are quite good enough to measure the the lengths/angles of the suspension links.
 
#28 ·
Problem with transferring weight by lifting the car is that the suspension will compress/unload and skew your results.

You'll have to lock the suspension solid to find the CG.


Oh and: http://farnorthracing.com/autocross_secrets.html
We could also see how much suspension flex will affect the outcome by measuring fendergap/ride height during the corner weighting. If the change is neglible, it could be ignored. It will skew the results, but if we're talking a very small percentage then I vote it's not worth locking down the suspension. We're modding a production car, not racing F1 here.

Nuse's spring rates are already so high it shouldn't move much. On a stock car, that difference may be quite a bit. It's a valid point though.
 
#30 ·
Here's the silly angle of my LCAs.

I believe those that buy BCs and lower them all the way (stock setup) this is about where it will put the LCA angle at.

With that assumed; the Tein S-techs and Eibach Sportlines are probably near this angle as well (a bit less, but not much)

And I definitely need to sweep out my garage floor I know lol.

Image


Image
 
#37 ·
Jesus Nuse. I figured your geometry was off, but wow.

The only semi feasible way I can see correcting this is replacing a ballpoint with a spherical bearing, using a grade 8 bolt as the shank of the balljoint(drill out the taper in the hub), and use spacers between the bearing and the hub to space the control arm down.

That'll correct the angle of the control arm, but there are issues with durability to be concerned with.


Sent from my iPhone with Tapatalk.