front design validation [Archive]

View Full Version : front design validation

Ezio_Auditore

12-16-2011, 11:47 PM

i have tentatively designed a front suspension and steering setup..
front track - 1250
front KPI - 6deg
front caster - 6deg
front scrub - 19 mm
front camber - 2deg
front RC height - 38.16mm
FVSA - 2910 mm
lower control arm is parallel to ground, front view length - 367.92mm

this the the front view pic - https://www.facebook.com/photo.php?fbid=10150443958342849&set=a.214630362848.128075.705167848&type=1&theater

the setup was modelled in optimum k with a pushrod setup and kinamatic analysis was done.

front view pic - https://www.facebook.com/photo...67848&type=3&theater (https://www.facebook.com/photo.php?fbid=10150443975122849&set=a.10150443974997849.352321.705167848&type=3&theater)

top view pic - https://www.facebook.com/photo...67848&type=3&theater (https://www.facebook.com/photo.php?fbid=10150443978497849&set=a.10150443974997849.352321.705167848&type=3&theater)

the steering geometry was designed as such - https://www.facebook.com/photo...67848&type=3&theater (https://www.facebook.com/photo.php?fbid=10150443983347849&set=a.10150443974997849.352321.705167848&type=3&theater)

the given steering geometry gave 29.274 mm as the rack travel for 45 deg of steer angle at the inner wheel.
due to lack of tyre data things had to be assumed and then analysed for results...

following are the simulation graphs -

roll 2 deg, 120 deg angle at steering wheel (45 deg steer angle at the inner wheel) pic - https://www.facebook.com/photo...67848&type=3&theater (https://www.facebook.com/photo.php?fbid=10150443987142849&set=a.10150443974997849.352321.705167848&type=3&theater)

the bump-steer comes around a maximum of .3 degrees in 3 in jounce...

roll 2 deg 45deg steer angle at inner wheel - and - bumpsteer analysis (https://docs.google.com/spreadsheet/ccc?key=0AkFjiTdwlKAndHZhZkFlaUEweXU3RWhGUDVERGhDb 2c)

if anyone could plaese help me out with validation of the results...
thank u...

Ezio_Auditore

12-16-2011, 11:47 PM

Ezio_Auditore

12-17-2011, 12:05 AM

the first analysis excel sheet was for bump steer
this is the analysis for 2 deg roll with 120 deg steering wheel angle corrosponding to 45 deg steer angle at the inner wheel -

https://docs.google.com/spread...kNkNGeVZIR1FZU3hxSkE (https://docs.google.com/spreadsheet/ccc?key=0AkFjiTdwlKAndFZNcDJVWFpkNkNGeVZIR1FZU3hxS kE)

woodsy96

12-17-2011, 12:53 AM

I'm hardly an expert but it looks like you are on the right track overall.

If this is a first year car I would do some side view stuff (anti dive etc), then stick with that and go build it. Then learn from it, and build a better one next year.

Ezio_Auditore

12-17-2011, 12:55 AM

even i hope im on the right one... well, did u check the excel files for the analysis??

12-17-2011, 04:29 AM

Ezio,

Why are you raising the CG as high as possible (ie. spring-dampers very high up)?

Why are you stressing the steering components so highly (ie. very short steer-arms)?

Z

Ezio_Auditore

12-17-2011, 04:44 AM

it is our first year car , and in absence of the Tyre data, i'm designing for 100% ackerman.. so if i increase the steer arm length, the bump steer increases as the steer arm moves further into the wheel and in the front view the relative lengths of the LCA front view can the tie rod changes increasing the bump-steer..

currently it is about 0.3 degrees for around 3 inches of bump

what do u suggest, should i be continuing with 100% ackerman?

we're working on the packaging of a pull-rod setup so that the spring and dampers come down....

any suggestions or changes that are evident???

Luniz

12-17-2011, 05:29 AM

from the current setup of the bellcranks, have you got a plot of wheel travel vs. piston travel in the damper? It looks like this geometry is very degressive, which you may not want it to be...

Also, look at what Z said, if you induce a lot of stress in the steering linkage, you will likely also get a lot of compliance. So you have to find a tradeoff between bumpsteer and ackermann. Maybe you can also find a way to make the bumpsteer actually help your cornering...

In addition to this, the angle of the front lower a-arm seems to be a bit narrow, which leads to high stresses and high compliance again.

Just my 0.02$... Overall you are on the right way, jst remember, engineering is all about finding the best compromise for your application, especially in suspension design.

12-17-2011, 05:39 AM

Ezio,

First thing you should do is get an overall view of what sort of car you need to do well in FSAE. Have you read the "Reasoning..." thread? If so, what are the results of your "top level" (ie. Level 4 & 3) design processes?

What I am getting at is that it seems that you are diving in and simply copying some suspension designs that you have seen on other FSAE cars. Unfortunately, those are bad designs. You can be much more successful in FSAE if you greatly simplify your suspension. Double-wishbones are not necessary. Push-or-pull-rods-and-rockers are a complete waste of time and money. Etc., etc.

To be more specific, tyre data is not necessary to design a good suspension. Whatever suspension you do design should have adjustable toe, camber, and spring rates. But really that is about all you need. The "fine tuning" of those variables AFTER the suspension is built (not lengthy initial design) is enough to satisfy whatever your tyres want.

Likewise, ackermann has very little to do with tyre data. It is mainly related to the kinematics of a car going around a tight corner. Put simply, you want as much "ackermann" (or dynamic toe-out) as you can possibly get. It is a simple matter, and preferable for structural reasons, to do this with longer steer-arms than you currently have.

FSAE has negligible bumps, so bump steer is not a big problem. Only if you do a spectacularly bad design will you have problems. Also, the suspension should have some final adjustability to cater for manufacturing tolerances, which you can use to fix bump steer.

But back to "top-level" design, why did you choose double-wishbones? What is your justification for using this unnecessarily complicated type of suspension?

Z

Ezio_Auditore

12-17-2011, 07:52 AM

thank you so much Z for ur suggestions,

i know the steering arm is too short, but i can compromise on bump-steer and ackerman, but increasing the steer arm length increses the rack travel as-well. the titan data sheets for fixed center rack doesn't have that high a rack travel (ie rack speed).

also i would like to ask, what would be the maximum steer angle (for the inner wheel if i'm running ackerman) that would be enough to negotiate the tightest corners in the track?
with the current wheelbase and track what i come up is around 40 deg, but it sounds way too high! (is it reasonable enough?)the lock to lock steering wheel angle what i assumed is -120 deg to 120 deg (is it a descent assumption?). with this value, the rack travel what i am getting is 29.274mm for 45 deg inner wheel angle.. now total rack travel would come out to be 29.274X3 = 87.822mm for 1 pinion revolution.

now if i increase the steer arm length, this value would rise... do i go for a custom made rack then? suggestions would be very helpul...

thank u so much guys...
thank u again 'Z' and 'LUNIZ'

http://fsae.com/groupee_common/emoticons/icon_smile.gif

12-17-2011, 05:49 PM

Ezio,

You are getting lost in the details. Eg., front CA length = "367.92mm", rack travel = "29.274mm".

At this stage you only need to be talking about "a longish grumlink" or "a shortish whatever".

So I ask again;

What are your team's conclusions drawn from the top level design processes?

That is, what is the "big picture" of the car that you formed at the beginning of the "Vv..." design process? (Do you know what I mean by "Vv..."?)

Do you accept that if you "optimally" polish a turd, then it is still a turd?

Z

Adambomb

12-17-2011, 06:35 PM

40 deg. seems a tad on the high side. I think I remember ours going to something like 25-35 or so. On the very tightest turns at competition you're almost on the stop with that. 40 deg. wouldn't hurt. Worst thing about it would be clearance, that could be a pain. The courses only seem to be getting tighter over time.

I'm a big fan of the Stilleto racks. Rack ratios come in either 4.7 in/rev or something like 6.5 in/rev. They are lightweight and good quality, and we just got one for like $240. They generally come with unfinished ends, and you can just cut to length and throw some clevises on there and it's good to go. They don't have a website, but you can do a google search for Woodhaven industries in Illinois and they list a phone number. They're used to dealing with FSAE teams too.

CameronBeaton

12-19-2011, 01:33 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Z:
Put simply, you want as much "ackermann" (or dynamic toe-out) as you can possibly get. </div></BLOCKQUOTE>

I'm not convinced by this. I think too much emphasis is put on ackermann and its only really 'useful' in low speed, commertial vehicle applications and or if you have the miracle the is a completely neutral steer car.

I don't think the level of ackermann sould be a major concern for most teams and particually new teams, parralle steer work fine (some might say better than 100% ackermann). Like you said later, a simple suspencion with no major flaws thats tunable is the way to go!

shark.ashwa

12-19-2011, 02:34 AM

Hi Ezio,
I think you should listen to what Z has to say. There are far too many other things to be decided for you to be getting into details. First prioritize your list of needs from the car and also put in constraints. This will be a good first step in starting the design concepts for the year ahead. This has always been our plan and does yield results if you stick to it! http://fsae.com/groupee_common/emoticons/icon_biggrin.gif
P.S. Yeah I checked the solidworks sketch, the rates seem to be digressive as Luniz pointed out...

Regards,

Sharath

12-19-2011, 02:51 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by CameronBeaton:
...I don't think the level of ackermann sould be a major concern for most teams and particually new teams, parralle steer work fine (some might say better than 100% ackermann)... </div></BLOCKQUOTE>
Cameron,

I don't think you understand "ackermann". Not many people do.

Can you justify your claim that parallel steer might work better than 100% ackermann, for FSAE type racing?

Z

Ezio_Auditore

12-19-2011, 08:47 AM

thank you so much guys for u suggestions..
thanks - Z, SHARK, CAMERON...

the rack-speed what i am arriving at is around 131 mm with a 70mm steering arm. isn't it too fast a rack speed?

CameronBeaton

12-20-2011, 10:54 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Z:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by CameronBeaton:
...I don't think the level of ackermann sould be a major concern for most teams and particually new teams, parralle steer work fine (some might say better than 100% ackermann)... </div></BLOCKQUOTE>
Cameron,

I don't think you understand "ackermann". Not many people do.

Can you justify your claim that parallel steer might work better than 100% ackermann, for FSAE type racing?

Z </div></BLOCKQUOTE>

Lets say we had the best, or atleast some very good, drivers driving in FSAE who could keep the car on the limit. In a corner we the outside wheel in the loaded more than the outside and therefore it's maximum grip is a lower slip angle. So there for you want less ackerman angle, maybe even some antiackerman (for example most of the F1 grid) to get the full poteshal out of the tyres.

The reason I said 'some may say' is that this is had to do and so when you do the hairpins at less than optimum speed your likely just to be scrubing the inside tyre and losing speed as a conciquence.

12-20-2011, 05:55 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by CameronBeaton:
...In a corner we the outside wheel in the loaded more than the outside and therefore it's maximum grip is a lower slip angle. So there for you want less ackerman angle, maybe even some antiackerman ... </div></BLOCKQUOTE>
Cameron,

I have carefully read your post, and in the above section (I think???) you are making the same mistake that too many other people make when considering this issue (I blame the education system! http://fsae.com/groupee_common/emoticons/icon_rolleyes.gif).

Specifically, how big a difference, in degrees, do you think there is between the slip-angle-for-max-Fy of the inner and outer wheels? My guess is maybe a few (1-2?) degrees. Anybody else, perhaps those with TTC data, want to offer a value?

BUT, much more importantly, why focus on this mouse, when there is an ELEPHANT breathing down your neck???

More later (unless someone wants to point out the elephant in the room...).

Z

Goost

12-20-2011, 06:17 PM

Z,
I completely agree. I can say from analysis I did with tire data, weight transfer considered, that you are correct: the best geometry in our case was only a few percent (deg. steer left/deg. steer right) away from 'true' ackermann. Of course it depends on which tires you're using, but not as much as going from true ackermann to parallel steer. But even knowing the theoretical values, the hard part is of course actually packaging that geometry....

BMEP

12-20-2011, 08:20 PM

Maybe I'm just another dumb idiot but how are yall defining what is %100 ackermannn or a certain percentage. Sure there is the classic KPI/tierodBJ/Center of rear axle trick but that doesn't account for rack positioning, etc. Are you basing this off of the inner vs outer wheel plots for a given steering input or???

Goost

12-20-2011, 08:44 PM

Ah, that's a fair question. if you assume near zero slip then true ackermann would besomething like: from a top view (vehicle X forward, Y rightward)

Given a horizontal (X=Constant) line M intersecting both rear contact patches
Given line N perpendicular to and intersecting the left front tyre, and
Given line O perpendicular to and intersecting the right front tyre,
The steering is true ackermann if at any given steer angle all three lines M, N, and O intersect at a single point.

Of course, the tyres do slip, and rarely all the same amount. Handling that is the trick.

12-21-2011, 12:37 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by BMEP:
... how are yall defining what is %100 ackermannn ...
... there is the classic KPI/tierodBJ/Center of rear axle trick but that doesn't account for rack positioning, etc. </div></BLOCKQUOTE>
BMEP,

You are right that this subject is usually, and wrongfully, discussed with next to no definitions (again, I blame the education system http://fsae.com/groupee_common/emoticons/icon_smile.gif).

Likewise, you are right that the "steer-arms-pointing-at-the-centre-of-rear-axle" is meaningless nonsense. Fore-aft R&P position has at least as large an effect on dynamic toe-out as the steer-arm angles. (A better gauge is the angle between steer-arm and tie-rod.)
~~~~~o0o~~~~~

Goost,

Your definition of "true ackermann" is good enough for me.

Another version, which allows for rear slip-angle, would be as follows. In a (simplified 2-D analysis) plan-view, find the Instant Centre for the car's motion wrt ground. For "true", or "100%", or "perfect", (or whatever), ackermann, the left and right front axle lines pass through this IC.

Repeating this for emphasis, at ANY and ALL times, the car moves about a certain IC in plan view (straight running has this IC at infinite distance sideways). This is a fact of life, and is regardless of momentary tyre forces or slip-angles. (It is in fact due to the integral of all forces (tyre, aero, gravity, inertial, etc...) on the car up to that moment.)

Typically, in steady cornering, the rear wheels are generating a centripetal force and hence running at a slip-angle. So the IC is now forward of the rear axle line by a few degrees. At zero rear slip this definition reverts to Goost's above.
~~~~~o0o~~~~~

Anyway, the important point is that for a typical FSAE car going around a hairpin bend, this true ackermann might require;
30 degrees steer angle for the outer wheel, and
45 degrees steer angle for the inner wheel! (Draw a plan-view picture!!!)

If the tyres are to generate cornering force, then they have to be steered a few more degrees, say 35 and 50 degrees!

As Goost says "the hard part is of course actually packaging that geometry...".

That is, most FSAE cars going around a hairpin bend have the front inner tyre pushing OUT of the corner. Fortunately, that tyre is only lightly loaded due to LLT. But even with only 10% of front axle load on the inner tyre, that 10% is working against the 90% on the outer, for a net cornering force of only 80% of what is possible.

So, design in as much dynamic toe-out (ie. "ackermann") as you can, because even that won't be enough.

Z

BMEP

12-21-2011, 06:07 AM

Alright another idiot question. I vaguely understnad what is being talked about, (too early to think properly) but what about when people are claiming they have 150%,200%, 50% etc. ackermann. The more I think about it, the more they are pulling these numbers out of their arse.
I'll admit I could care less whatever this mythical ackermann percentage is but rather just look at the wheel plots of both wheels for a steering input..

12-21-2011, 05:26 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by BMEP:
... what about when people are claiming they have 150%,200%, 50% etc. ackermann. The more I think about it, the more they are pulling these numbers out of their arse. </div></BLOCKQUOTE>
BMEP,

You are 100% correct. This "percentage" nonsense usually refers to the point on the car centreline where the steer-arm-lines intersect. At rear-axle it is 100%. At mid-point of car it is 200%, etc. But as you noted earlier, this is meaningless given the influence of R&P fore-aft position.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">... just look at the wheel plots of both wheels for a steering input.. </div></BLOCKQUOTE>
Yes, much better. If you have to quote one number, say in quick conversation with the Design judges, try something like;

"Oh, ackermann percentage?"
"Well, we have almost 13 degrees dynamic toe-out at full-lock."

Then add;

"Err, that's 34 degrees for the outer wheel, and almost 47 for the inner. We will try for more next year, but would you like to see this year's curves?"

Z