LocostUSA.com

Learning how to build Lotus Seven replicas...together!
It is currently November 22, 2019, 6:27 am

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 160 posts ]  Go to page Previous  1 ... 7, 8, 9, 10, 11  Next
Author Message
PostPosted: December 24, 2014, 5:24 pm 
Offline

Joined: October 19, 2009, 9:36 pm
Posts: 2139
Location: meadview arizona
i need some feedback here guys!

_________________
this story shall the good man teach his son,
and chrispin chrispian shall ne'er go by,
from this day to the end of the world.
but we in it shall be remembered.


Top
 Profile  
Reply with quote  
PostPosted: December 24, 2014, 11:30 pm 
Offline
User avatar

Joined: October 23, 2010, 2:40 am
Posts: 1123
john hennessy wrote:
is there a way to display the numbers i have come up with on the forum?
John, try the SnipIt tool (part of Windows) to take partial screen shots, save them as jpg's and then upload them to your posts like pictures. Alternatively, you can use ScreenPrint, but the process has a few more steps.

Quote:
i have a fluctuation of .053" of toe at rim in 2" of bump or 2" of droop, do you think this is within reason?
Can you run smaller increments of bump/droop and look at the bump steer toe-in curves? At smaller suspension movements, the toe might be quite small.

Quote:
i have limited rc lateral movement to within 1" and it is below ground and doesn't migrate above ground in 1" of droop with 4 degrees of roll and 3" of rack movement.
Some consider lateral rc movement a non-problem. Your number looks pretty good as is.

Hope this helps. Hopefully those with more expertise can chime in after the holidays.

_________________
Cheers, Tom

My Car9 build: viewtopic.php?f=35&t=14613
"It's the construction of the car-the sheer lunacy and joy of making diverse parts come together and work as one-that counts."

Ultima Spyder, Northstar 4.0, Porsche G50/52


Top
 Profile  
Reply with quote  
PostPosted: December 25, 2014, 12:07 am 
Offline
User avatar

Joined: December 1, 2014, 3:57 pm
Posts: 8
Location: W.Poland, ME
Another method to save an image of an active window, press "Alt"+"Print Scrn"... you can then "paste" that image into any image software. (for those with older versions of Windows)

I prefer to see the bump steer at .015" per inch or less... & which way depends on if it's front or rear steer. (rear steer = out / front steer = in).

Now, with some front ends based on stock parts that's not always possible... but .030"/inch should be doable.... unless the geometry is way out of whack.

Count me in the "lateral kinematic RC migration has little, if any, effect" camp... I don't care much about lateral migration of the kinematic RC anymore... the late Bill Mitchell's ("Racing By The Numbers" software, including "WinGeo") force-based calculations showed the effect of lateral kinematic RC migration to be minimal & not worth handicapping other areas to minimize it... keeping it low to minimize jacking means much more than how far it moves right or left.... of course, Wishbone predates Mitchell's force-based calculations by about a decade.

That's not to say something like Wishbone isn't still useful... for the price it has it's purpose... & kudos to kf2qd for taking the time to migrate it to Windows... but you need to keep in mind that some of the theory has evolved when using it.

If it takes letting the RC skitter off to the side in order to get a better camber change curve, so be it... you're better off than handicapping the camber curve for the sake of keeping the RC pinned... again, so long as it stays low.

I've always considered the term "Roll Center" to be something of a misnomer really... which is why I started consciously trying to specify it as the "Kinematic Roll Center" more... it's too easy to see the simple term "roll center", look at a drawing, & get an image of the RC being like the center of a wheel that the chassis rotates around... when that's not the way it works... if that were the case then, in theory, you could place the RC outside the inside tire of a turn and have both sides of the suspension go into compression in that turn... which obviously isn't likely to happen in the real world.

Your spring rates, anti-roll bar rate, caster change, CG height, & RC height all influence the exact location the chassis will appear to rotate about... not how far east or west the kinematic RC may have migrated to.

Here's a link to a paper Mitchell handed out during his presentation at an SCCA Competition Clinic in Cleveland, OH on March 17, 2007... titled: Roll Center Myths and Reality

http://f500.us/files/Roll%20Center%20Myths-Vehicle%20Dynamics%202007-Mitchell.pdf

_________________
Image
http://F500.us


Last edited by Dix on December 25, 2014, 10:21 am, edited 1 time in total.

Top
 Profile  
Reply with quote  
PostPosted: December 25, 2014, 4:10 am 
Offline

Joined: October 19, 2009, 9:36 pm
Posts: 2139
Location: meadview arizona
o.k. i have a static toe in of 0.125, this is front steer.

in 1/2" bump i have 0.127, thats 0.002

in 1" bump i have 0.132, thats 0.007

in 1.5" bump i have 0.141, thats 0.016

in 2" bump i have 0.152, thats 0.027

in 1/2" droop i have 0.126, thats 0.001

in 1"droop i have 0.132, thats 0.007

in 1.5" droop i have 0.141, thats 0.016

in 2" droop i have 0.153, thats 0.028

caster is 4.46 deg positive static

camber is 1deg negative static

lift is unloaded .33 neg and loaded is .041 pos at 3 deg roll and 3" of rack

ackermann is 74.7%

there is not much i can do about the scrub radius at 2.25" but i may install power steering, the track width is based on the hub flanges at this time, i have moved the wheel center inboard which raises the roll center a little but not much else changes, just to see the effect of reducing the scrub.

i will get to reading the mitchell stuff tomorrow.

_________________
this story shall the good man teach his son,
and chrispin chrispian shall ne'er go by,
from this day to the end of the world.
but we in it shall be remembered.


Top
 Profile  
Reply with quote  
PostPosted: December 25, 2014, 2:34 pm 
Offline
User avatar

Joined: April 26, 2008, 6:06 pm
Posts: 3017
Location: Under the weather. (Seattle)
Dix wrote:
Image
Dix wrote:
Image


Looking at these got me thinking a bit...Yes a very dangerous thing. :shock:

I think it's worth noting that both of these are typical illustrations showing the rear anti-squat line as drawn for a ring & pinion live rear axle, where the toque is reacted from the axle housing, through the trailing links, and into the chassis. This makes sense, as the applications where anti-squat is the most important most commonly run solid axles anyways. But that's not the only anti-squat being produced. It is typically the dominant one though, such that the effects of the forward thrust from the rear axle are generally ignored. For an IRS though, this torque is reacted directly through the differential into the chassis, which results in a direct weight transfer through the springs if I'm not mistaken. So an IRS necessarily experiences 0% geometric anti-squat from the same torque reaction. All of the resulting anti-squat in an IRS comes from the forward thrust. Thus the anti-squat line would be drawn from the center of the rear wheel through the instant center, rather than the contact patch through the instant center. This makes for dramatically different results.

I had not thought about this before, but much like an IRS, the F500/F600 style rear suspensions I'm most familiar with would not have any direct torque reaction through the trailing links either. I had never considered this before. So they would have 0% anti-squat from that as well. The thrust based anti-squat component remains the same for all of the suspension designs, and the torque reaction would be through the upper side of the chain/belt. I'm not yet sure how it would be graphically (or otherwise) determined, nor which of the two would generally be the dominant factor for typical geometries.

_________________
-Justin

Also follow my build on blogspot, tumblr, or instagram and twitter (GarageOdyssey)


Top
 Profile  
Reply with quote  
PostPosted: December 25, 2014, 3:20 pm 
Offline
User avatar

Joined: April 26, 2008, 6:06 pm
Posts: 3017
Location: Under the weather. (Seattle)
John,
Since you're not looking at more than two inches of travel from ride height, is it safe to say that this is a track oriented suspension you're working on?

_________________
-Justin

Also follow my build on blogspot, tumblr, or instagram and twitter (GarageOdyssey)


Top
 Profile  
Reply with quote  
PostPosted: December 25, 2014, 3:43 pm 
Offline

Joined: October 19, 2009, 9:36 pm
Posts: 2139
Location: meadview arizona
in Arizona the roads are so bad that the concept of suspension is a bit pointless, only kidding, yes this is for autocrossing.

having read the mitchell paper i think i understand what he is saying, in that the force application point would migrate upwards due to jacking effect the more the vehicle rolls and thus there are two forces applied to the chassis, one lateral and one vertical.

however, in all the sketches the roll center is depicted above ground, i believe this is to more clearly visualise his opinion, if the roll center is below ground then the kinematic roll center is also lower and thus the force application point is also low.

as far as a jacking effect, this to me is a self cancelling effect as the more jacking you have, the shorter the lever from the force application point to the c of g reducing the lateral effect and thus chassis roll which then reduces the jacking

the problem with the roll center moving from below ground to above ground is an indicator of the instant center hight movement from below to above ground and as this occurs the force application point moves up at a faster rate causing unpredicable handling due to the roll reduction caused.

this will reduce roll and any camber change in the suspension resulting in loss of grip when driven on the limit, suddenly you find you are over the limit of grip.

just my opinion.

_________________
this story shall the good man teach his son,
and chrispin chrispian shall ne'er go by,
from this day to the end of the world.
but we in it shall be remembered.


Top
 Profile  
Reply with quote  
PostPosted: December 25, 2014, 4:09 pm 
Offline

Joined: October 19, 2009, 9:36 pm
Posts: 2139
Location: meadview arizona
Justin,

the power applied to the wheels is causing an equal and oposite reaction in the chassis, in that if the wheel was held still the chassis would rotate.

however this would require the moment to be transmitted through the upper(in tension) part of the chain, pulling the chain rearwards.

as there is suspension, then the net effect would be to cause the axle to move forward until the suspension compresses or relaxes to give the shortest distance between the trans sprocket and the axle sprocket, wether it relaxes or compresses depends on the pivot point, real or imaginary, most motocross bikes are designed to relax the suspension to force the wheel down onto the ground, at first this would appear negative as it looks like weight is being lifted off the rear which would deminish grip but it is in fact the amount of grip and torque applied causing the rear to rise as when the tire losses grip the back of the frame drops.

if you look at the top part of the chain as the top link of a four bar system, then the trailing arm of the suspension could be the lower arm of the four bar, unfortunately, the front pivot of the chain is the output shaft of the trans and is probably low in the chassis relative to the axle, not exactly though, the chain itself is the pivot and its the links so at the front, the pivot is the last link going onto the sprocket and likewise the first link off the axle sprocket.

what i am saying is by changing the diameter of the sprockets, you can change the chassis and axle pivots just like moving the upper link in a four bar

_________________
this story shall the good man teach his son,
and chrispin chrispian shall ne'er go by,
from this day to the end of the world.
but we in it shall be remembered.


Top
 Profile  
Reply with quote  
PostPosted: December 27, 2014, 3:10 pm 
Offline
User avatar

Joined: December 1, 2014, 3:57 pm
Posts: 8
Location: W.Poland, ME
Driven5 wrote:
I think it's worth noting that both of these are typical illustrations showing the rear anti-squat line as drawn for a ring & pinion live rear axle, where the toque is reacted from the axle housing, through the trailing links, and into the chassis.

True, I probably should have noted that IRS is calculated differently... but regardless, if the program is actually calculating it the way I believe it is (as in the 1st diagram) then it's not going to be anywhere near correct unless your front brake bias percentage is reasonably close to your rear weight bias percentage. (for calculating anti-dive)

Also, the use of the term "squat" in the descriptions, if not just a misapplied substitution for "dive", would imply use for calculation of rear anti-squat... and would be dead wrong regardless if used for either case (solid axle/IRS) on a 2WD vehicle... although, I suppose you could "fool" it into coming up with the correct anti-squat by inputting the "CG Long" as equal to the wheelbase... as that would indeed force the program to place the 100% line at the correct angle.

In which case (I'm assuming here - based on nothing but observation, not calculations) the "%anti I/B" would be the IRS calculation.

EDIT: Come to think of it... that trick could also be used to make the program calculate the anti-dive correctly... simply input your front brake percentage (as a percentage of the wheelbase) for your "CG Long" input... all you're looking for in an anti dive calculation is the have the CGH at the same percentage of the WB toward the rear axle as your front brake percentage... so it's really a simple workaround... & the anti-dive/squat is the only calculation effected by the CG Long input... don't know why I didn't think of it before.

For example.... you have a 100" WB & use 70% front brake.... so you'd input the "CG Long" as "-70".

Driven5 wrote:
This makes sense, as the applications where anti-squat is the most important most commonly run solid axles anyways. But that's not the only anti-squat being produced. It is typically the dominant one though, such that the effects of the forward thrust from the rear axle are generally ignored.

True enough... & they can be ignored mostly due to the angles commonly used.

Most trailing arm type suspensions have rather shallow angles (with respect to the ground plane) & fairly long side-view swingarms (in my case infinate - parallel 4 bar)... so the vast majority of the thrust action from the axle is in the forward direction parallel (or very near parallel) to the ground... makes sense, that's the direction you want the power to acceleration the vehicle... power applied in any ther direction would be wasted... if maximum acceleration is the goal.

Now, there are some special cases where you might want to trade off a little acceleration efficiency for another purpose... John's MX example is a good one... a motocross bike needs to plant the rear tire hard otherwise it will just spin on the loose surface & not go anywhere... so you use some of your available power to plant the rear tire.... other off-road racing vehicles use the principle as well... a drag racer coming off the line is another example.

Driven5 wrote:
I had not thought about this before, but much like an IRS, the F500/F600 style rear suspensions I'm most familiar with would not have any direct torque reaction through the trailing links either. I had never considered this before. So they would have 0% anti-squat from that as well. The thrust based anti-squat component remains the same for all of the suspension designs, and the torque reaction would be through the upper side of the chain/belt. I'm not yet sure how it would be graphically (or otherwise) determined, nor which of the two would generally be the dominant factor for typical geometries.


Not sure what F500 chassis you've seen but the typical Invader/Late-KBS & the later Red Devils all use a typical 4-link (w/birdcage) & panhard bar or watts link.... I'm reasonably sure that the Maverick/Scorpion & NovaCars do as well... & they all must use a solid rear axle per the rules... IRS is not legal (nor are differentials BTW)... so there is definately an anti-squat componant to the suspension linkages & it's 100% line would intersect the rear tire contact patch.

Some of the earlier "F440"-based cars did use some odd rear "suspension" (if you can call them that) arrangements... but everything I've seen from the early-mid 1990s-up have all been 4-link based.

Now, the torque reaction through the belt & how much of an effect it has is an intersting question... closest analogy to F500 is a mini-sprint & they do get a noticable effect on handling depending on where the SVSA's IC is located vs the drive chain angle... then again, they have a better HP:weight ratio & race on dirt... so there is that difference to consider in how much effect the belt drive really has on an F500... on pavement, with near 20 less HP & 200lb more weight. (vs a 600cc mini-sprint)

The "line" formed by the upper side of the belt is going to vary with gearing... & the negative anti-squat I'm seeing in my car may indeed have been an original attempt to copy the angle of the belt with the trailing arms to minimize it's effect on suspension loading... or it may be out of necessity to maintain belt tension throughout the arc of axle travel... answer to that particular question is YTBD, & is getting way off-topic for this thread.

JOHN - regarding your results... I'd call the bump steer pretty decent & if you're happy with everything else... congrats. :mrgreen:

Only thing I'd question is your static toe... 1/8" toe in (1/4" total) is a lot & the wrong way for autocross (especially a front-steer)... unless you meant it to be -1/8" (or -1/4" total) toe (out).... that I would find resonable for a front-steer AXer.

Something I don't think has been mentioned in this topic... but if you want to "share" a project for others to view/comment on, besides taking a "screen capture" you could use the BB"Code" tags to copy/paste your Wishbone Perameters file into a post... unlike just copying it into a post, the "Code" tags preserve the formatting (such as the extra spaces between characters)... just open the ".wsh" file with notepad, then copy & paste that between the "code" tags... here's an example using the car I've been talking about...

Code:

19.5
46.75
80
11
-45.44
-1.75
0.271
0  2.375  6.875
17  2.125  4.75
-4.375  9.875  7.375
-0.875  9.875  7.00
10.75  9  5.875
0  5.75  20.875
-5.6875  12.75  18.375
-0.875  12.75  19.00


Now, anyone can simply copy what appears in the "code" box into Notepad & save it with the ".wsh" file extension. (NOTE: the 1st line needs to be a blank carriage return)

_________________
Image
http://F500.us


Top
 Profile  
Reply with quote  
PostPosted: December 27, 2014, 8:54 pm 
Offline
User avatar

Joined: April 26, 2008, 6:06 pm
Posts: 3017
Location: Under the weather. (Seattle)
John,
So that component of anti-squat would essentially be a matter of using the side view swing arm (instant center to axle center) to determine how the suspension would react to the tension side sprocket points of tangency trying to pull themselves closer together...That seems to correlate well with what I was thinking. Thanks!
:cheers:

Due to the limited wheel travel for the car you're considering, I would think that the bump steer is probably ok. I have a feeling you would end up wanting more static camber, but that can be adjusted on the final product. How are you coming up with your roll angle and steering estimates? Is the relatively low caster being driven by mechanical trail/steering effort concerns, or something else? What are your reasons for going with the below ground roll center? The illustrations are for above ground roll centers, in large part because even most race cars choose to do this. Yes there are arguments for cars having underground roll centers, but for constant quick transient maneuvers I would think that the greater responsiveness from an above ground roll center would generally be one of the dominant factors.

_________________
-Justin

Also follow my build on blogspot, tumblr, or instagram and twitter (GarageOdyssey)


Last edited by Driven5 on December 27, 2014, 9:33 pm, edited 5 times in total.

Top
 Profile  
Reply with quote  
PostPosted: December 27, 2014, 8:59 pm 
Offline
User avatar

Joined: April 26, 2008, 6:06 pm
Posts: 3017
Location: Under the weather. (Seattle)
Dix wrote:
...all use a typical 4-link (w/birdcage)...
Yep, that is exactly what I was referencing. A ring and pinion live axle reacts the axle torque into the trailing links directly through the single piece axle housing. This is what creates the anti-squat forces as graphically determined in the previous illustrations. A chain/belt driven shaft as used in F500/F600 does not react any of the axle torque into the trailing links through the birdcages, because there is nothing to actually transmit torque into the birdcages themselves. Thus this component of anti-squat, as traditionally derived for ring and pinion live axles, would necessarily be zero for such a birdcage located shaft.


Dix wrote:
I prefer to see the bump steer at .015" per inch or less... & which way depends on if it's front or rear steer. (rear steer = out / front steer = in).
Dix wrote:
1/8" toe in (1/4" total) is a lot & the wrong way for autocross (especially a front-steer)...
If I'm understanding these statements correctly, you're claiming that front steer and rear steer each are preferred to have the wheel toe in opposite directions from one another?...I would be interested in learning more about the reasoning behind this idea.

_________________
-Justin

Also follow my build on blogspot, tumblr, or instagram and twitter (GarageOdyssey)


Top
 Profile  
Reply with quote  
PostPosted: December 28, 2014, 9:36 pm 
Offline

Joined: October 19, 2009, 9:36 pm
Posts: 2139
Location: meadview arizona
it would appear that i made the assumption that the toe figure was overall, wheel to wheel, is this not the case, the degree figure is 1.1 degrees.

in 03, we were racing a 69 camaro at sebring which had come from england, there was masses of toe out and the car was undrivable before we reduced it considerably and also reduced the camber somewhat, let me add that the driver was very experienced, Jacky Oliver, this may have been due to a change from goodyear slicks to the mandated hoosier grooved slicks.

i know that some toe out results in better turn in but we are dealing with autocross, so i want a car that is precice from lock to lock and am willing to sacrifice the turn in for rapid left to right, right to left ability.

the roll center below ground is to make the vehicle roll rather than "twist" thus activating the suspension geometry from one side to another quickly.

right now, the restraints of the particular spindle/upright which is triumph alford & alder but gives a scrub radius of over 2" so any help with the steering wheel is good.

at this time i will be experimenting with these for springs and posibly no shocks.

Attachment:
311-19-1701.jpg
311-19-1701.jpg [ 8.77 KiB | Viewed 10141 times ]

_________________
this story shall the good man teach his son,
and chrispin chrispian shall ne'er go by,
from this day to the end of the world.
but we in it shall be remembered.


Top
 Profile  
Reply with quote  
PostPosted: December 28, 2014, 10:15 pm 
Offline
Toyotaphobe
User avatar

Joined: April 5, 2008, 2:25 am
Posts: 4805
Location: Fort Worth, Texas
You peaked my interest with those the last time you mentioned them. I realize there must be some tradeoffs or else we'd all be riding on them.

I'm presuming the trade off is for street use and lack of travel.

I love the idea of not having to find room for shocks and a little smaller packaging issues with the shocks too.

I'm not sure I like the idea enough to actually try them, but I'm more than willing to watch someone else do it. :o

Now I'm going to have to watch your build and this one where he's using Nascar torsion bars.

http://www.ozclubbies.com.au/index.php?/topic/8558-the-we-designed-for-tony-rear-engine-subaru-powered-clubman/page-58#entry168629


Attachments:
1.jpg
1.jpg [ 550.29 KiB | Viewed 10137 times ]

_________________
mobilito ergo sum
I drive therefore I am

I can explain it to you,
but I can't understand it for you.
Top
 Profile  
Reply with quote  
PostPosted: December 29, 2014, 2:03 am 
Offline

Joined: October 19, 2009, 9:36 pm
Posts: 2139
Location: meadview arizona
well they were $21.00 for the four delivered so i thought it would be worth a try, i will build a simulator to test the rate.

i will be using pushrods and bell cranks to achieve 2" of suspension compression at ride hight with a further 2" of available bump, the "secret" will be in the bell crank ratio.

alas, i have no idea of the front corner weight as i don"t have a car to weigh but i would assume the front unsprung weight will be around 200 - 250 lbs. but there are other factors to take into account such as push rod angle and the outside location of the push rod to the lca and the chassis pivot position.

_________________
this story shall the good man teach his son,
and chrispin chrispian shall ne'er go by,
from this day to the end of the world.
but we in it shall be remembered.


Top
 Profile  
Reply with quote  
PostPosted: December 29, 2014, 10:12 pm 
Offline
User avatar

Joined: December 1, 2014, 3:57 pm
Posts: 8
Location: W.Poland, ME
john hennessy wrote:
it would appear that i made the assumption that the toe figure was overall, wheel to wheel, is this not the case, the degree figure is 1.1 degrees.

The toe figures are for each wheel, so for total toe you add them together... or at least I thought it was that simple.

Apparently I made an incorrect assumption too... if you measure your toe the same way I do (front & rear of the wheel & subtract the smaller) then you have to double them again... so in your case you've got a static setting of 1/2".

It seems that the program only calculates the inch toe value as the difference between the axle centerline width & the front edge of a 13" wheel. (plug what you see in any right triangle calculator using 6.5" for one side)... I've never measured toe or bump-steer that way in my life.

Given this wierdness... for bump-steer use a value of 0.06 degree/inch or less... that's the same ballpark.

Either that or remember that you need to double the value you see for each wheel

john hennessy wrote:
in 03, we were racing a 69 camaro at sebring which had come from england, there was masses of toe out and the car was undrivable before we reduced it considerably...
...this may have been due to a change from goodyear slicks to the mandated hoosier grooved slicks.

Extremely likely... in addition to other potential varibles

Most all autocrossers I know (front or rear steer) use static toe out... as do most short-track oval racers & many other race cars... as it helps with the initial turn-in... front-steer cars will use more of it due to the inherently poor ackermann... back when I ran a Camaro front clipped Late-Model I've used as much as near 1/2" out on the little flat 1/4 mile bullrings.... even rear-steer cars (if based on most street-driven vehicles) usually don't have enough ackermann built into them... on purpose... you don't ever want average Joe driver to oversteer if he goes into a corner too hot... so even rear-steer AXers will start with static toe-out... just less of it then the front-steers as a general rule.

Even if they don't know that ackermann isn't just some guy's name, most limited class oval track racers understand from experience... the shorter & tighter the track the more static toe-out they need to run & lap times get better... up to the point the car starts to get loose on entry or feel twitchy under braking.... an AX course isn't Road America, so similar thinking would apply.

john hennessy wrote:
i know that some toe out results in better turn in but we are dealing with autocross, so i want a car that is precice from lock to lock and am willing to sacrifice the turn in for rapid left to right, right to left ability.

Think about what you're saying there for a minute... you state that toe-out helps turn-in, but that toe-in would be better for lock to lock transitions?

A transition from one direction through straight ahead back to the other is no different than the transition from straight ahead into a turn... other than it being more extreme... so how can toe-out help one and impair the other?

This is especially true for a front-steer... a front-steer is already on the losing end of the ackermann battle & only gets worse as you approach the extremes of steering lock

Next time you're at an AX event eyeball the static toe on what most of the top guys are running (ingore anything you're told, use your own eyes)... most of 'em are likely to be obviously toed-out... there's a reason for that.

Now, rear-steer on something like high-banked ovals & road racing where speeds (& tire loading) are higher, can (& should IMO) start with a small amout of toe-in & let the ackermann do it's thing... so long as there's enough of it designed in, & you have a car that's relatively "neutral"... without a lot of inherent understeer or oversteer... this is one of a very few situations where I'd depart from starting with static toe-out.

The thoery behind that has to do with tire loading & slip angles (which is another of those terms that I hate because it draws the wrong mental picture... like "roll center").... "Slip angle" actually has nothing to do with what you'd normally think of as "slipping"... in the case of the front tires, having the tire skid sideways (AKA: "push") has nothing to do with the slip angle.

Basically, "slip angle" is the difference between the direction the steering is attempting to point the tire and the direction it's actually moving in... that slip angle is going to depend on variables like grip, load, sidewall flex, contact patch deformation, etc.

When you have very little slip angle (think: wide high speed turn, AKA a "sweeper") the turn center of the rear axle is actually ahead of the front axle (rear wheels track outside the fronts)... in this case you'd actually want a little toe-in... which you'd get from the static setting.

As turn radius decreases & tire loading increases, the slip angle gets larger... the turn center of the rear end moves back toward the rear axle & the rear tire's track transitions to inside the fronts... if you've ever seen an ackermann diagram you know this requires some toe-out... which you get from having the ackermann designed in.

The "gotcha" is that when you're generating high slip angles (on the outside front vs the inside) you don't need (or want) 100% ackermann in the steering geometry.. the difference in slip angles between the inside & outside tire is creating some of that for you already... as the outside tire is operating at a much higher slip angle than the inside... you may have turned the outside front wheel 3 degrees but it's operating like it's only at 2-1/2 or less.

This is why a lot of oval track cars with sticky tires and relatively tall sidewalls actually work well with very little (or even reverse) ackermann... the tires are generating it anyway... a heavy car on a track like Martinsville or NHMS can have the RF operating with a slip angle of up to 5 degrees.

One of the better explainations of this concept I've read published online was by Mark Ortiz...

http://www.auto-ware.com/ortiz/ChassisNewsletter--August2011.htm

With front-steer however, it would be very difficult (if not imposiible) to build enough ackermann into the geometry to warrent static toe-in... unless we're talking about your average grocery-getter where we want Mom to feel nice & comfy when she starts to make a lane change on the highway or hit an off-ramp.... Mom likes mushy.

If you doubt this, go into Wishbone & keep increasing the "Y" value of your outer tie-rod pivot (for a front steer) until the program says your static ackermann is a perfect 100%... now look at the diagram & ask yourself if you could physically put it that far into the wheel.... I doubt it.

If you go back to your base numbers, you probably notice that your static ackermann percentage for a front-steer, is likely a single digit number at best.... more than likely it's a negative value... nature of the beast... If you want to try racing a front-steer car with static toe-in, God bless you & I wish you luck... but my money would be on handling only slightly better than your average school bus.

The design of any front-end has compromises built into it somewhere.. all the way up to and including F1... there's no such thing as the theoretically "perfect" front suspension in practice... the trick is knowing what compromises are better made than others.... which is why there are guys that get well into 6-figures for doing it... and there's no computer program, regardless of how much it cost, that can tell you that.

And what looks good on paper doesn't always work so well out on the track... compromises exist because they have to... just like theoretically perfect ackermann doesn't work anywhere except a parking lot.
Quote:
In theory there is no difference between theory and practice; in practice there is.

At some point you have to stop the mental masterbation & build it... otherwise you'll just be tweaking a "paper car" forever... chasing prefection that can never come.

There have been plently of races won with front-steer, rear-steer, pro-ackermann, reverse-ackermann... you name it... it's all about picking the best set of compromises that works best with the specific package you've got.

john hennessy wrote:
well they were $21.00 for the four delivered so i thought it would be worth a try, i will build a simulator to test the rate.

i will be using pushrods and bell cranks to achieve 2" of suspension compression at ride hight with a further 2" of available bump, the "secret" will be in the bell crank ratio.

alas, i have no idea of the front corner weight as i don"t have a car to weigh but i would assume the front unsprung weight will be around 200 - 250 lbs. but there are other factors to take into account such as push rod angle and the outside location of the push rod to the lca and the chassis pivot position.


Using those as springs can work... our F500 cars use a piece of rubber 2" diameter by 1" thick as "springs" & no shocks either (all as per the rules)... but, if you're not going to use shocks then remember that you'll want to build in some friction points somewhere as a dampeners.

Also, since you're using what is effectively a rapidly rising rate spring, it's true that the motion ratios & the bellcranks will be key... but remember too that they're a double-edged sword... the motion ratio can reduce the amount of compression travel of the "spring" per inch of wheel travel, but this also "preloads" the spring more which uses up some of your available travel from the get-go.... you can't just go from 1:1 to 2:1 on the bellcrank and expect double the wheel travel and/or to be cutting the wheel rate in half as the spring now has to support twice the weight, it's going to be preloaded more.

_________________
Image
http://F500.us


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 160 posts ]  Go to page Previous  1 ... 7, 8, 9, 10, 11  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
POWERED_BY