cibachrome
Automotive
- Jan 16, 2007
- 512
Regarding thread60-87973 on the Subaru effort gradient, why didn't anyone suggest changing the torsion bar in the steering valve assembly (NOT the valve pressure angle profile but the torque sensor)? This is fairly easy to do, especially after practicing with junkyard parts. Steering systems all use a torque sensor based on a metal torsion bar element running in parallel with the boost system's moment reaction element. These t-bars have rates of about 1 Nm/deg +-0.3 Nm/deg. The assist is modulated by how much displacement this tbar twists. If you put in a "big" tbar, then almost no assist occurs. Small tbars give lots of assist for the same valve curve. Punch out the tbar, put in a stiffer one and more mechanical feel and less assist force from the rack or piston occurs. The problem with adding more caster is that rack load twists the tbar more to get the assist balanced, hence more understeer. Slowing down the pump output cuts the assist level and increases the tbar windup to get the necessary counterforce. Again much more understeer. The valve is a diverter for open-center hydraulic systems, so effort is tbar moment + power assist moment - the net kingpin moment. Anything that interferes with the hydraulic control in this case causes a degradation in feel, increases in compliance and understeer. Increasing the tbar stiffness just dials down the ratio of tbar moment to assist moment. Thus less compliance and understeer. Remember, the steering input you apply is really a moment, not an angle as far as the vehicle is concerned. The input steer moment is balanced by the kingpin moments reaction, which is the tire and steer geometry additive effects. There are tire and even drive torque transient responses to include in your analysis. The brand of gear (Koyo, Saginaw, Delphi, TRW, , Toyoda, Visteon, etc. sets the torque sensor and the valve body design. They have dozens and dozens of torque sensors available to punch into an existing valve body, housing and steering gear able to cope with the many different vehicle masses, tierod loads and market segment tuning level. Find some junk parts and start playing development engineer. Manufacturers have kits full of these parts used to evaluate varius levels of steering effort by steer angle and lateral g gain. Its a 3D balancing act. BTW: a soggy steering shaft isolator above the pinion will now soften the system because the increased steer moment will increase the twist in the column or ishaft. Remember all these parts are running a constant moment balancing act. The weakest part defines the system response. This causes some fascinating dynamic impressions on the track and going straight if you should be so unlucky as to have the tbar NOT be the softest element in the system. This may be the root cause of this original post.
