This is the roll center, suspension pickup, and steering angle modification thread.

[95KA-Turbo]

They're from speedway motorsports and are ~20 bucks a piece. There's a link to them on the first page.


You have to buy their taper bit too though because they are not the same taper as the OEM tie rods.

[SoSideways]

See, if you have to get a drill bit to drill the stock taper to another taper, you might as well use a bolt there instead

Anyway, parts going to machine shop this week, so I should have everything on the car by the end of the week.

I will report back how the bolt will hold up.

Probably going to miss the event this month, which is on Saturday, so I will have to wait till next month I guess, to report back on how the grade 8 5/8 bolts will hold up for tie rod end duties.

[Prez]

subscribing post...

[Xplat]

Looking at these makes me shudder to think what kind of bumpsteer they will introduce into the suspension geometry. I'm sure though that many of us will never actually get to the point where we will want to correct this aspect of our suspension geometry by moving the steering rack forward any way possible. I am almost certain this setup adds bumpsteer because my Tein inner tie-rods came with spacers that supposedly add steering angle... Well it does, but at the price of additional bumpsteer and those spacers were only about 1/4" of an inch thick each.

This thing looks like it's at least half an inch wide PER side. The use of something like this to move the rack forward unfortunately corrects one thing but manages to fuck another aspect of suspension geometry since the inner tie rods are now pivoting about half an inch past the steering rack on each side.

However, this is speculation on my part as I'm not sure if a longer tie rod shank eliminates the additional bumpsteer from the inner tie rod being spaced from the rack in that manner. Not to mention the amount of bumpsteer the tie rod is already correcting simply by having the car lowered. Maybe their knuckles had to be redesigned as a result of the new inner tie rod pivot point. Gah. All this suspension geometry corrections/changes are blowing my mind right now. One can correct a problem, but reintroduce an old problem they had previously fixed as a result of fixing another problem. No wonder suspension geometry is sometimes referred to as a black art. This shit is evil and difficult to measure/model accurately for the most part.

Of course Driftworks may just deal with the added bumpsteer in their suspension geometry as drift events concentrate around one section of track and if theres little to no bumps then theres no BUMP steer problem right?

I think I'll stop now I'm just rambling.

[PoorMans180SX]

Those do not affect bumpsteer. I think you are confused.

Longer tie rods will not increase your bumpsteer.

Not having your LCA and tie rod parallel is what makes the car bumpsteer. This has to do with vertical spacing not horizontal.

[racepar1]

Originally Posted by ballinnmiami240sx

We tried the bolt method and you can tell that just by rolling off the trailor into the garage it bent. The taper GM stud Im sure can stand the beatings he is going to put on it this year. We are doing alot of R&D on all the suspension mods we did to 95ka-turbo's car. Hopefully in a month or 2 we can report back as to how the car performs.

Danny

The bolt bent because:

1: It was a shitty bolt

2: The spacers that were used were also shitty

3: It looks as though you installed the bolt upside down. The head side of the bolt SHOULD be basically a press fit into the spindle and the thread side should be hanging off the bottom of the spindle. It looks like you had it the other way.

4: it seems that the pivot was spaced down too much. You can only get away with an extra inch or so of spacing using a bolt like that.






It's funny how many people are under the impression that you need a 3" spacer to get your geometry good. When speaking about roll center you need to be referencing the roll center heighth to the CG heighth, NOT necessarily the ground. If you dropped your CG a few inches through lowering the car and managing weight you don't need to correct the roll center nearly as much.

[PoorMans180SX]

Originally Posted by racepar1

It's funny how many people are under the impression that you need a 3" spacer to get your geometry good. When speaking about roll center you need to be referencing the roll center heighth to the CG heighth, NOT necessarily the ground. If you dropped your CG a few inches through lowering the car and managing weight you don't need to correct the roll center nearly as much.

True, but about 2.25 inches is perfect on a car that tucks tire on 18's

[racepar1]

Originally Posted by PoorMans180SX

True, but about 2.25 inches is perfect on a car that tucks tire on 18's

Exactly how did you come to that conclusion?

[SoSideways]

Going to the machine shop tomorrow to drop off the parts to get machined, but here are some of the hardware.



62 degrees of misalignment without having to use high misalignment spacers.

That is also a 5/8" grade 8 bolt, which is substantially bigger than the stock ball joint stud, even at the thickest part of the tapered part.

[ballinnmiami240sx]

This subject always turns into he said she said.....lol. Until you actually try all these mods out on track we will never find out.

[PoorMans180SX]

Originally Posted by racepar1

Exactly how did you come to that conclusion?

Look at the driftworks car. Take crank centerline as the CG.

Driftworks front knuckles correct the geometry by 50mm and the arms correct by 15.

This places the roll center even with the crossmember, which is pretty much perfect.

The rear is obviously different.

The problem when people go to correct their roll centers is that they don't take into account the instant centers. When you do, you'll realize that your roll center is quite a bit lower than it would seem. Just taking where the lower arms angles intersect is not accurate, the roll center is at least 2-3 inches lower.

[racepar1]

Originally Posted by PoorMans180SX

Look at the driftworks car. Take crank centerline as the CG.

Driftworks front knuckles correct the geometry by 50mm and the arms correct by 15.

This places the roll center even with the crossmember, which is pretty much perfect.

The rear is obviously different.

The problem when people go to correct their roll centers is that they don't take into account the instant centers. When you do, you'll realize that your roll center is quite a bit lower than it would seem. Just taking where the lower arms angles intersect is not accurate, the roll center is at least 2-3 inches lower.

I would love to know why everyone seems to think that driftworks knows so well how/how much to correct roll center. I found a clear design flaw in their front knuckles just by looking at a side by side shot. MMDB found a flaw when he tried to bolt his e-brake up to them too! If they didn't think of that I seriously question what else they either didn't think of or did wrong. This ignorance is exactly why I hesitated to post in this thread in the first place.

I honestly do not know the correct answer to my own question. It takes a LOT of measurements and calculations to even get a clue of the length of your roll couple. I simply don't have that capability right now. That being said, from my own research the driftworks front knuckles are spaced more then is needed both in the front and ESPECIALLY in the rear. I would absolutely LOVE to hear from them how they came up with those numbers as I am not convinced that they truly understand what they did.

[SoSideways]

racepar1 - dude, he just told you why he said what he said.


In any case, I doubt the correction on either the DW's knuckles or what people are thinking about doing are worse than having stock hardware and slamming the car as low as possible to achieve a low CG.

If that were the case, people who have those DW knuckles, like mmdb, would be saying their car handles funny.

As for the rear knuckles not working with the Z32 e-brakes bit, did you see the reason why DW said the Z32 rear e-brake assembly didn't work?

[SicBastard]

I think that's his point. This is all just "I doubt" and "probably better" until some one measures their actual CG and posts up the pics that they took of their car at a measured ride height with overlaid angles, and then did at least some rudimentary modeling or bumpsteer/camber gain measurements. Dirt circletrack guys wouldn't dream of stepping out onto the track without a proper set-up measureing session. Why do we demand so much less from our parts suppliers?

[PoorMans180SX]

Originally Posted by racepar1

I would love to know why everyone seems to think that driftworks knows so well how/how much to correct roll center. I found a clear design flaw in their front knuckles just by looking at a side by side shot. MMDB found a flaw when he tried to bolt his e-brake up to them too! If they didn't think of that I seriously question what else they either didn't think of or did wrong. This ignorance is exactly why I hesitated to post in this thread in the first place.

I honestly do not know the correct answer to my own question. It takes a LOT of measurements and calculations to even get a clue of the length of your roll couple. I simply don't have that capability right now. That being said, from my own research the driftworks front knuckles are spaced more then is needed both in the front and ESPECIALLY in the rear. I would absolutely LOVE to hear from them how they came up with those numbers as I am not convinced that they truly understand what they did.

Sheesh man, I'm the one that started this thread. Calling me ignorant is a little over the top, don't you think? I've corrected myself several times in this thread, and am more than willing to learn. That's the reason I made this thread.

I'm just saying that if you want your roll center below your CG, and your CG is about crank centerline, then having your roll center even with the crossmember is about ideal. The Driftworks knuckles and arms put it there on their S15 drift car, which is pretty slammed.

All I did was this:



Now I know that's not the most accurate thing in the world, but allowing for inaccuracies and suspension compression when the car is actually on the ground and I think it's fair to say that the front roll center is just below the crossmember (or somewhere very close).

[racepar1]

Originally Posted by SoSideways

racepar1 - dude, he just told you why he said what he said.


In any case, I doubt the correction on either the DW's knuckles or what people are thinking about doing are worse than having stock hardware and slamming the car as low as possible to achieve a low CG.

If that were the case, people who have those DW knuckles, like mmdb, would be saying their car handles funny.

As for the rear knuckles not working with the Z32 e-brakes bit, did you see the reason why DW said the Z32 rear e-brake assembly didn't work?

I have little confidence in the ability of zilvia members to correctly assess suspension geometry. This thread has solidified my lack of confidence. I have even LESS confidens in zilvia members having the driving talent to be able to tell reliably if a suspension change is better or worse. And yes I did see why the e-brake didn't work. It didn't work because Driftworks didn't DESIGN the knuckles so that it could be bolted on because they didn't care. That goes right back to my previous statement here "just because something works great for someone else doesn't mean it's great for you".

Originally Posted by SicBastard

I think that's his point. This is all just "I doubt" and "probably better" until some one measures their actual CG and posts up the pics that they took of their car at a measured ride height with overlaid angles, and then did at least some rudimentary modeling or bumpsteer/camber gain measurements. Dirt circletrack guys wouldn't dream of stepping out onto the track without a proper set-up measureing session. Why do we demand so much less from our parts suppliers?

This is EXACTLY my point. People are just jocking the crap out of a part that they really know very little about.

Originally Posted by PoorMans180SX

Sheesh man, I'm the one that started this thread. Calling me ignorant is a little over the top, don't you think? I've corrected myself several times in this thread, and am more than willing to learn. That's the reason I made this thread.

I'm just saying that if you want your roll center below your CG, and your CG is about crank centerline, then having your roll center even with the crossmember is about ideal. The Driftworks knuckles and arms put it there on their S15 drift car, which is pretty slammed.

All I did was this:



Now I know that's not the most accurate thing in the world, but allowing for inaccuracies and suspension compression when the car is actually on the ground and I think it's fair to say that the front roll center is just below the crossmember (or somewhere very close).

Ignorant is not the same as stupid. Ignorant means that you don't know, stupid means that you can't learn. Calling someone ignorant isn't necessarily an insult. You have proved your ignorance before in this same thread, so no I don't think that it is "over the top" to say that. Trying to make the rough calcualtions that you are from a picture is EXTREMELY unreliable. You need rough measurements off a car at the least. I have spoken with people, an engineer actually, that had these rough measurements at his disposal and the conclusion that was reached was MUCH different then your conclusion. Now, I am not saying that I am 100% right, but I have more confidence in the people that I have referenced then a rough drawing on a picture. You are taking what driftworks did as the word of god himself (so to speak) instead of actually doing your own research and calculations. I will not be convinced untill I get concrete numbers. In any regard the DW knuckles are certainly not worth the money. MMDB was better off with what he already had IMO.

[95KA-Turbo]

I am fully planning - and have been from the beginning - to do a proper measurement/calculation session with my car once I get the motor in and running.

I am going to get the car up on a drive-on lift so I can get the car in whatever position I need it in to get the proper measurements. I'll definitely keep everyone updated with what happens with that.

I'm not expecting JGTC results, but I feel confident I can improve the suspension for the better while nearly tucking wheel on 17s.

[PoorMans180SX]

Originally Posted by racepar1

I have little confidence in the ability of zilvia, etc, etc, etc...

And I think you're making it more complicated than it really is.

We're measuring angles here. It's not like you have to make a bunch of complex calculations and plug them into some crazy formula and do a bunch of math to figure it out.

And CG isn't even that hard, but no one has corner weight scales to be able to take measurements and weights to find out exactly where the front and rear CG's are.

Yes, I know measurements off a car would be much better. I am all for getting the very best parts and suspension geometry, but no one is stepping up to the plate to offer anything else in the way of drop knuckles. Sometimes compromise is all that you can do. I know that I have no way of fabricating drop knuckles.

[racepar1]

Since I am regulated as to how much I can correct things like roll center, by the RTA "street tire" class rulebook, I am no longer very concerned with measuring out roll center. I'll just correct it the full inch that I am allowed and leave it at that. My main concern is whether or not I should really do anything to correct rear roll center at all with only an inch of correction in the front. I've decided to just raise the subframe up, like the new SPL bushings, and leave it at that in the rear. I figure that should help the roll axis issue. I may also make spacers so that I can adjust the rear subframe position. I am definitely going to replace the rear balljoints with sphericals so that small adjustments can be made as well.

What I AM really anxious to screw with is the front bumpsteer and rear camber/toe curves. If you have adjustable rear arms you can actually adjust the lengths of the ruca's and toe arms WITHOUT affecting the static camber/toe settings. This can be accomplished by adjusting the cam bolts and re-setting the camber/toe to wherever you want it. I think that by playing with those arm lengths and the length of the trac arm some amount of gains in rear traction can be made. I plan to rig up a plumb bob setup and spend a day on the rear alone.

The issue that I am the most curious about though is lowering the toe arm pivot on the rear knuckles. I wanna find a way to mock that up so that I can figure it out. All in due time though as my chasis is basically ready to go on a rotisserie at this point. Not much I can measure like that.

[racepar1]

Originally Posted by PoorMans180SX

And I think you're making it more complicated than it really is.

We're measuring angles here. It's not like you have to make a bunch of complex calculations and plug them into some crazy formula and do a bunch of math to figure it out.

And CG isn't even that hard, but no one has corner weight scales to be able to take measurements and weights to find out exactly where the front and rear CG's are.

Yes, I know measurements off a car would be much better. I am all for getting the very best parts and suspension geometry, but no one is stepping up to the plate to offer anything else in the way of drop knuckles. Sometimes compromise is all that you can do. I know that I have no way of fabricating drop knuckles.

You're right, CG isn't that complicated. BUT nobody that I know of has actually calculated it out. Also NOBODY that I know of has actually calculated out their roll center heighths either. I know a couple guys that have made rough measurements and educated guesses, but nothing concrete. Still I trust the rough estimates of people that I know and respect far above some british company that specializes in drifting and that I really know very little about.

The problem with trying to guestimate roll center from a picture is that a photographic image is not necessarily the correct to scale dimensions compared to the car. Images get distorted slightly depending on what angle the picture is taken from, lighting, camera lens, etc... I'm sorry, but that just doesn't do it for me, even as an estimate.

[DALAZ_68]

i need to ask this...because...frankly, i cant come up with an answer...

tie rod spacers...

they supposidly give you a bit more angle right? but i dont understand how...if say...ur steering rack/tierods were fully aligned...then u add spacers to the rie rods...

my initial thought is, that ur wheel would be front of car "<" rear of car to each other (not to such great degree but u get what im saying) obviously after alignment theyll be "=" to one another... so would that mean, that u cancelled out the increased length the spacers gave you...or am i missing something here?

[Xplat]

Originally Posted by PoorMans180SX

Those do not affect bumpsteer. I think you are confused.

Longer tie rods will not increase your bumpsteer.

Not having your LCA and tie rod parallel is what makes the car bumpsteer. This has to do with vertical spacing not horizontal.

What I mean was the balljoint from the inner tie rod is now pivoting further away from the rack. Thus, changing something unnacounted for in suspension geometry. The spacers may not even make a noticeable difference so long as you have LOOONG tie rod ends like SPL's for example which probably compensate for the added 'instability'.

I bring this up because I was experiencing some noticeable bumpsteer without being too far lowered. After some reading I decided to Remove the tie rod spacers that came with my Tein inner/outter tie rods and doing so helped remove 50% of the bumpsteer I was feeling the rest I know I know is from the angle of the outer tie rod. For test purposes I would always drive down the same stretch of shitty road testing the before and after. So, it's not like I was driving on a different smoother road after I removed the spacers to give a placebo effect of having 'removed' bumpsteer. It helped, I'm happy so I just thought I'd bring that up.

SPL has this to say about using tie rod spacers (which the DW product I'm referring to inevitably does)

"Note: Steering angle spacers are not recommended for road racing use; installation of steering angle spacers will create a small amount of bump steer due to modified steering rack geometry."

Found from the link below.

http://www.splparts.com/main4/parts/...tmultilink.htm

Hope that clears up what I was talking about.

EDIT:

Oh and something similar to that effect was also written by Tein in the instructions that came with my tie rods. I just threw them out a little while ago so I can't really quote them word for word. Something along the lines of not using the spacers for street use.

[Otto347]

Originally Posted by DALAZ_68

they supposidly give you a bit more angle right? but i dont understand how...if say...ur steering rack/tierods were fully aligned...then u add spacers to the rie rods...

They allow the main shaft of the rack to move further in on one side resulting in an extra Xmm on the other side. I thought the same thing when I first saw them but after tearing a rack apart I could see how they actually worked.

[CrimsonRockett]

Does explain why I felt more bump steer after adding the spacers to my Tein inner/outer rods.

Got a set of SPL inner/outer rods I'll be using next. Won't be driving the car for another year.

So much for that. Haha.

[SoSideways]

I can sort of see the reasoning behind that.

Since the tie rods' overall length will have to be slightly shorter to compensate for the spacers (I didn't think the actual tie rods were spaced further outward w/ the spacers, but instead allows for the tie rods to move further before hitting the stops?), the tie rods now have different arcs than the FLCAs.

I'm sure if you were able to space the tie rod ends down in such a way that if you can bring both the tie rod and the FLCA to travel at around the same arc (as close as possible), then your bumpsteer issues should be mostly eliminated.

[ManoNegra]

The new(er) Drift Tengoku 46 has a segment on angle adjusters parts and techniques
as well as roll center (shock!!)

[PoorMans180SX]

Any updates on the Driftworks rear knuckle mods? Sphericals?

[mmdb]

Wonderful folks over at Skullworks are hard at work on them. They've modified the knuckles to fit the z32 e-brake, and are awaiting parts to fit the sphericals in. I'll post up pix when that happens.

[mmdb]

Originally Posted by racepar1

I would love to know why everyone seems to think that driftworks knows so well how/how much to correct roll center. I found a clear design flaw in their front knuckles just by looking at a side by side shot. MMDB found a flaw when he tried to bolt his e-brake up to them too! If they didn't think of that I seriously question what else they either didn't think of or did wrong. This ignorance is exactly why I hesitated to post in this thread in the first place.

Yeap I agree. I was a bit disappointed when I didn't receive solid answers from DW when and asked where the roll center was placed, and if any calculations were made to logically place toe, camber, etc angles. It is apparent though that they've used SP-TEC's knuckles as a template to develop their knuckles. They're not going to openly admit it, but that doesn't matter. I'm glad they did though, because I'm assuming SP-TEC in Japan put a butt load of time developing it. Also the S15 that holds the ~55 second lap time in Tskuba uses those as well. Does that mean it'll make your car unstoppable on the track? No. But I'm sure it's far from hindering.

Originally Posted by SicBastard

I think that's his point. This is all just "I doubt" and "probably better" until some one measures their actual CG and posts up the pics that they took of their car at a measured ride height with overlaid angles, and then did at least some rudimentary modeling or bumpsteer/camber gain measurements. Dirt circletrack guys wouldn't dream of stepping out onto the track without a proper set-up measureing session. Why do we demand so much less from our parts suppliers?

Yeah very good point! But along those lines, I think I'm fortunate in that I've had several experienced drivers (one of them being a professional) comment on my car's handling. To me, that speaks more than the rough calculations and guesses here on the net. I'll find another track day and have some of the instructors give me feedback on my car without telling them what I did to it. What would be awesome though, is to have Tsuchiya drive it and give me some feedback. Maybe someday ^_~

In my experience, the 240 realm has been more of a try and trust type of affair to me. I've tried and spent a lot of money on new things. But it always comes down to buying parts that racers use, working with professional race shops, and working with the local race fabrication shop. In my opinion, the products from Japan tend to be overhype and over priced. But that's another thread. Sorry bout the rant.

[Silverbullet]

Originally Posted by PoorMans180SX

And I think you're making it more complicated than it really is.

We're measuring angles here. It's not like you have to make a bunch of complex calculations and plug them into some crazy formula and do a bunch of math to figure it out.

And CG isn't even that hard, but no one has corner weight scales to be able to take measurements and weights to find out exactly where the front and rear CG's are.

It may sound easy to figure these calculations, but without the proper tools, such as a Farro arm, it is quite a task. By just old school measuring, being to what one may refer to as "close enough" points will yield not so accurate results since we are working with angles that depend on other angles. There is lots of room for error.

Additionally, the method in finding the vertical CG height will be dependent on how accurate your scale is. Here is a real world example: We did a little experiment on finding the vertical CG height on a Formula Mazda race car with two different methods.
1- the tilt test, and 2- the classic lifting of the rear method. The tilt test is known to be more accurate, and it showed since we already knew the specs of the car. It was dead on. The tilt method though however, resulted in about an inch off (keep in mind this is also a smaller car than a 240 lol). The source of the error was most likely the accuracy of the scales. The ones we used were accurate upto a 1 lb. There was only 2 lb total of weight transfered to the front scales when lifted 10 inches from the ground. 2 lb. Just for kicks, we plugged in worst case + and -, and CG resulted in a couple inch of variation. With the combination of an inaccurate roll center location used with an inaccurate CG location, the final result can be significantly off. A couple inches may not sound alot, but on a smaller scale formula car, it can be dramatic depending what your trying to do with that information.