Expert: Ernest (Ernie) Kenward - 6/2/2011

Question
I had my power steering pump replaced, and the mechanic said that it had a bubble that they just couldn't get out.  When the car is running and my foot is on the gas the steering performs pretty normal.  When I am at idle with my foot off the gas the steering wheel feels very loose.  
The mechanic said that the bubble should bleed itself out in less than 500 miles, is this true? And if so is there any issue with driving it with a small bubble?

Answer
Power steering is not my specialty by any means, although I have had some fluid-power experience in other projects.  For that reason, I was about to elect to send your question on to the group-at-large in the hope that one of the others could give you a specific answer.

However, I was caught by your comment that "... at idle ... foot off the gas, the steering wheel feels very loose" - now, looseness is relative, and it is only in consideration of that my sense of due diligence suggests that this should be clarified.  In fact, a "little looseness" can be expected, for the reason outlined in the next few paragraphs.  It could also mean a problem, which should be checked out.  Your mechanic, by the way, very likely is no stranger to due diligence, which is the mark of any professional.  Chances are very good that he did a comprehensive check of your system, and will again if you are concerned.  (My point here then is not to second-guess him, but to provide some insight as to how fluid-power assist works so that, if nothing else, you and he may discuss this in more detail.)

The basic structure of the majority (if not all) of the automotive power steering systems in use over the past several decades is one in which a solidly-coupled mechanical system is fitted with a fluid-power assist.  What that means is that the mechanism portion of the steering system provides positive control from the steering wheel, through the ball-and-nut (or rack-and-pinion) assembly to the control arm and tie-rods which ultimately provide directional control and stability to the car.  The assist, as such, merely removes some of the steering workload by supplementing manual force with controlled fluid force.  If the assist should fail, positive steering control would remain in effect (as it must) although at greater effort.

Again, you indicate a "loose wheel" when the vehicle is at a stand-still.  What would matter here is the nature of that looseness!

On one hand (and the most likely one, I hope and expect), this could simply be as a result of air lock keeping a portion of the control valve "high and dry" and, if that problem is a matter of an air lock and not simply too little fluid, then that problem may resolve itself in due course.  (If the fluid should happen to be low, then it would have to be topped up, but that is easily done, but don't do it unless it actually IS needed.)

The fact is that power-steering control valves are like any other feedback-based control system, in that any corrective action or controlling force that they provide is first based on their response to an "error" signal, in this case, the displacement of the control valve spool in one direction or the other; the valve must move in order to direct the fluid to the actuator which "corrects the error" by assisting the control arm in moving in the direction that helps to re-center the valve.  Note that, should the assist system NOT being working, the mechanism will still work to provide safe steering, but there will be a small amount of play caused by that valve moving that small distance without the assist being available to compensate for it.

On the other end, looseness can be indicative of any one or more of a number of worn bearing surfaces, such as idler arms, tie-rods, ball-joints, all of which do wear over time, and this is why one inspects these as a matter of course.  Aside from wearing out both tyres and driver fairly quickly, worn suspension parts can catastrophically fail, and so I would say again that your mechanic has most probably done this, and would be happy to discuss it with you.  And, if that wasn't done, now it can be.

Effective communication is key!

<sorry for the longish answer, but my workload has just eased with the end of an engineering term, and I guess I am still working at high output!>

Best regards ... EGK