Most of the steering failures I have seen are from running out of angular
misalignment capability in the rod ends. Check for adequate travel through
full suspension travel at all steering positions. Most rod ends (including
the stock "high misalignment" type) will have less travel than your "OEM"
balljoints, the outer tie rod ends need to have more. A tie rod is usually a
two force member, so it can be surprisingly small, until you consider when a
joint bottoms out, or it gets hit with a stick or other object. Design the
tie rods to not buckle even if they have been bent in a previous incident
where the car was still drivable. Also look at the strength of the tapped
holes in the rods.
How far can a tie rod deflect before it will go overcentrer with the
knuckle? That would be an embarrassing problem.
The clevis should be in good contact with the ends of the rack, and the bolt
needs to be pretensioned so that the mating surfaces (and not the bolt)
handle the bending load. Consider how small the end of the rack is and how
it will distribute load into the clevis.
Also if the mounts for the rack are not adequately rigid, the extra load
will be taken up by the steering shaft, making the wheel very hard to turn.
Design the rack mounting system to distribute the load across more than one
part of the frame. Design the steering shaft to take up some misalignment
without feeling sloppy.
Some of the wheelstops I've seen at comp. are just to appease the safety
judges and would not actually hold up. Since they are so close to the
steering axis, they may cause the A-arm to bend. In most cases they
themselves will bend.
I would normally stay away from analyzing the stock ATV knuckles, but one
year I did hear of a team that broke the spindle right off. Just inspect for
wear and fatigue in that case. I've never heard of a balljoint failing
except where it was not mounted properly or worn from dust collection. Parts
from the ATV market are pretty durable if you use them the same way they
were intended to be used.
As for loads, your worst case will be a one wheel nosedive landing, with the
steering all the way to one lock. It is an impact load against the rack,
with the tie rod is as close to going overcenter as it will get. For this,
assume your wheel stops have failed. Also assume that the landing angle is
such that the suspension doesn't really move from the full rebound position.
Just distribute the loads through the system using statics. When you run
these numbers, you may find that what you and other teams have isn't nearly
strong enough, this is called a safety factor of less than one. It is
acceptable to expect failure in such a worst case scenario, as long as you
benchmark off the safety factor which got you or other teams though
competition and real world testing. No car can be failure proof. The tie rod
is the most easily replaced part of the system, so it should be the weakest
link in the chain. Fabricate carefully so that both sides are the same and
carry a spare.
Eric Hutchenreuther
Kettering
It would be more friendly to the other teams if you post your questions on
the forum, so that discussions can get started.
http://forums.sae.org/access/dispatch.cgi/minibaja_pf
----- Original Message -----
From: "millerp" <millerp@cadetmail.uscga.edu>
To: "'Valerie Isles'" <valisles@sae.org>; <mini-baja@Autox.Team.Net>
Sent: Friday, November 05, 2004 10:59 AM
Subject: engineering analysis
> Teams,
> I'm conducting some engineering analysis for the steering system of our
> vehicle and have a few questions. We're mainly looking at the stresses on
> the tie rod, spindel connection, clevis, and ball joints. Does anyone
have
> any suggestions as to how to conduct this the most accurate way? I was
> looking at taking some stuff from Shigley's Mechancial Engineering Design.
> Is there any component that I'm missing in my analysis? Any help would be
> much appreciated.
> /r
> Paul Miller
> US Coast Guard Academy MINI BAJA
/// unsubscribe/change address requests to majordomo@autox.team.net or try
/// http://www.team.net/mailman/listinfo
/// Archives at http://www.team.net/archive/mini-baja
|