Power steering behaving badly

[balimatharu]

This is continuation of the topic started by me in Oct. 2020.
The steering system was refitted during the restoration of my 250sl. The power steering didn’t kick in when started the engine. Repeated bleeding was performed and even after driving for 150 miles, no power steering.
Strange behaviour noted.
When the steering is in centre position and start the engine there is no power steering . Turn the steering wheel to the left about 30 degrees and Rev the engine the steering starts turning all the way to the left in aided. Same again bring the steering wheel to the rest place, turn the steering a bit to the right and Rev the engine, it will turn all the way to the right lock on its own.
Time is come to sort this PAS.
Is it the pump or the box.
Please help me to pin point the culprit and how to fix it.
Bali

[MikeSimon]

I am a little confused about your description:
Is the symptom the same turning the steering wheel both ways?
The pump does not know which way you turn the wheel.
If there is a difference, the problem is with the steering valve or the box.
The pump reacts to drive input/engine rpm only.
If there is lack of support at lower rpm, the pump may have an issue with the flow control valve inside.
Many times, problems are caused by minor contamination. Dirt makes the valve stick.
Any good service dealer (more often independent ones, who do not work on brand new cars and are familiar with hydraulic steering) will have a test device for easy diagnose.
It is a T-connection which taps into the pump output line and has a flow meter/pressure gauge attached to it.
This will eliminate the pump.
If the issue is with the steering valve or the piston in the box, this will have to come out and be checked.

[balimatharu]

Thanks for the reply.
I am confused myself after reading what I wrote.?
I will try to describe again.
There is no power in the steering. It is stone dead. Doesn’t make a difference whether the car is moving or stationary.
Now this is the difficult bit to explain.
Now what I am describing happens when the car is stationary.
Turn the steering to the left or right ( engine revved up or idling) no power steering.
Now the strange bit. Turn the steering about 20 degrees to the left or right, takes the hands off and now rev. the engine, magically the steering wheel turns to the lock position.
If I am turning the wheel and rev the engine no power steering.
It is like a scene from a spooky film.
I do not understand the working of of the power steering system.
Does this mean that the pump is producing power but the power is no applied to the box at the right time.
Back ground. I had major work done to the car.
The pump and the box were removed, covered and stored for about a year.
Thanks for taking interest.

[MikeSimon]

Trying to explain how the system works:

There are two valves in a power steering system: The flow control valve in the pump and the steering valve at the entry of the box, controlled by the steering column.
The pump produces flow as soon as it turns/the engine runs. Up to a certain rpm, the flow increases proportionally to the rpm until it reaches a point where the flow remains constant - usually around 1000 rpm - the so-called flow control. Regulated by the valve in the pump.
The fluid flow produced by the pump is routed to a valve at the entry to the steering box. The valve is rotated by the steering column. The opening of the valve is controlled by a device called the "torsion bar". The harder the resistance to turn is (the more you want to turn the wheels) the more the torsion bar gets twisted and the the more the valve opens.
Depending to which side you turn, the steering valve sends the flow to the respective side of the piston, where a pressure on the piston is created, helping to move the steering. Thus power steering.
From what you are describing, I would diagnose - from far away - your steering valve in the box sticks in place after you turn the wheel by 20 degrees. This will cause the wheel to turn to the lock position. I assume the wheels on the car turn also to the full turn! Normally, if you let the wheel go, it should try to return to center. This will not happen completely when the car is stationary because the geometry of the front axle does not produce the forces it would if the car was rolling. If you let the wheel go, there is no force anymore to twist the torsion bar and there should be no flow to either side of the piston in the box - unless the steering valve is stuck. This could be caused by a broken torsion bar or some other part in the valve.
My suggestion is to take the box out and have it checked.

[WRe]

Hi,
here can see some insides: https://www.mercedesscheune.de/portal/restauration-w111-coupe/step-5-motor-restaurieren/servolenkung-restaurieren.
Over the years, a lot of greasy, rusty oil sludge accumulates there, which not only clogs valves, but also hinders movement.
You can try to flush the system through or ultimately to dismantle it.
...WRe

[balimatharu]

Mike, very clearly explained and understood the workings of the PAS.
I think first I should check the efficiency of the pump. I will have to fabricate the pressure testing tubes with the pressure gauge. No garage around here with this setup.
Thank you for your interest. If you come up with any other advice please let me know.
There is one more observation. From the center point the steering takes about 1/4 circle extra when turned towards left side as compared to when moves to the right side.
It seems the steering column may not be in the centre position? When the steering wheel is in the center position, the road wheels are in correct position.
Is there a problem with the alignment of  the steering column with the box I wondered.

[WRe]

Hi,
in my experience there is no need to check the pressure with measuring devices. It is sufficient to observe the flow rate, because the flow rate is very high when the steering is turned (up to some liters per second depending on steering position).
With this procedure, there should always be a sufficient supply of power steering oil available. The steering can also be flushed hereby.
...WRe

[balimatharu]

How do I check the flow rate. Could you please
describe the procedure.
Thank you.

[WRe]

Hi,
suck old oil out of the reservoir, use the bleeding screw to drain the old oil. When the engine is running (idle) and the vent screw is opened 1.5 turns, an approx. 600mm long transparent hose is put on the nipple and in this way the old oil gets into a used oil container. A large part of the old oil is pumped out of the oil circuit, while the oil is being drained, fresh oil is constantly topped up in the storage tank so that no air gets into the system. Since almost all of the oil is pumped out, you need about 1.5 liters fresh oil for changing but significantly more for flushing or you reuse it, when it's get clear.
...WRe

[MikeSimon]

The pump check I described is actually a flow check. the pump does not create pressure. It creates flow. Only if that flow runs against a restriction (a valve or a piston) is pressure created.
The meter device in the check line is a combined flow/pressure meter and it can be "throttled" to check flow. You could do a simple check by routing the "output" line of the pump back into the reservoir and see if there is any decent flow coming out.
I am a little concerned about your description of the different turning angles from center. The turns should be equal to both sides. It may affect the operation of the steering valve.

To check the pump, try to find a "hydraulic" shop, not necessarily an automotive repair place. They can check the pump easily.
You could take it off the car and take it there. They have test stands to test it.

[balimatharu]

Mike
I am going to pick your brain a bit more if it is alright.
The steering wheel turns much more when turned to the left as compared with it been turned to the right.
You expressed that this condition could cause some problem with the opening of the steering valve. Could you please help me to centralise the steering column. Is there a way to centralise the steering box and then I could move the steering column to the middle.
At the moment I have made sure that the marking on the column and the marking on the steering wheel coincide.
How can I check whether the steering column and the steering box are connected rightly.
Help would be greatly appreciated.

[Pawel66]

To find the center in the steering box:

1. Suck out some PS fluid from the reservoir
2. Put a piece of cloth around the bolt marked on the picture
3. Unscrew this bolt and look inside the orifice
4. As you turn the wheels (slowly, the fluid will be splashing out) somewhere close to the steering wheel middle position, you will see a round recess similar in diameter to the screw orifice (ca 8mm) on the piston moving front and aft under in that orifice. When you align this round recess with the screw orifice - this is the center of the steering box.
5. If you find 8mm screw, small thread (!), make a cone on the its tip and screw it in so that the coned tip goes into this recess - you blocked the steering box in its middle position. Of course there are special tools to do that, but in case you do not have them - I used the small thread screw.

Here is some more on that:
https://www.sl113.org/forums/index.php?topic=16504.msg114302#msg114302
https://www.sl113.org/forums/index.php?topic=27379.msg196556#msg196556
https://www.sl113.org/forums/index.php?topic=21950.msg156566#msg156566

And plenty more if you search "steering box center" or alike.

Please note:
 - there are marks on pitman arm and its shaft that need to be aligned
- the mark on the tip of the steering wheel shaft needs to point up

[balimatharu]

Today I checked the flow and the pressure from the steering pump.
I isolated the steering box from the pump. Connected the outlet of the pump to a set up
which consisted of a pressure gauge, a stop valve and a tube which would return the fluid back into the pump reservoir.
Started the engine with the valve open. At idle ( 800 rpm)the fluid was trickling back into the reservoir. As the engine was revved the flow increased progressively.
Then I closed the valve. The pressure registered to 5 psi at idling.
When the engine was further revved up to 4000 the pressure progressively increased to 80- 100 psi. Pressure would still go up.
Is this a healthy pump I wondered.
The pressure at about 1000 revs was 10 psi.
Please come in and help me to sort the power steering.

[balimatharu]

Today I checked the flow and the pressure from the steering pump.
I isolated the steering box from the pump. Connected the outlet of the pump to a set up
which consisted of a pressure gauge, a stop valve and a tube which would return the fluid back into the pump reservoir.
Started the engine with the valve open. At idle ( 800 rpm)the fluid was trickling back into the reservoir. As the engine was revved the flow increased progressively.
Then I closed the valve. The pressure registered to 5 psi at idling.
When the engine was further revved up to 4000 the pressure progressively increased to 80- 100 psi. Pressure would still go up.
Is this a healthy pump I wondered.
The pressure at about 1000 revs was 10 psi.
Please come in and help me to sort the power steering.

[WRe]

Hi,
in my opinion the pressure on your pump is way too low and should be at least 750 psi, see att. info of Vickers pump from workshop manual for 230SL.
...WRe

[balimatharu]

Hi WRe
Thanks for your reply.
Is this the right way to check the pump. I didn’t involve the steering box in the test.
From the outlet of the pump the pressure gauge, the stop valve and the fluid draining into the reservoir.
It is like bench testing. What’s your comments on this.

[WRe]

Hi,
in the workshop manual of the 280SL the checking of the power steering and the servo pump is described when installed, see attached file.

The test is carried out by connecting a test tool/pressure indicator with throttle valve on the pressure side of the high pressure oil pump.
The delivery pressure provides information of the function of the flow control and pressure relief valve.

Let the engine run at idle speed and take the following measurements:
- When the throttle valve is open the circulation pressure should not be higher than 60 psi. During this measurement, the steering wheel may not be moved, otherwise the control spool is deflected and the pressure increases.
- Close the throttle valve on the test tool and measure while accelerating the engine and hold at  approx. 1000 rpm.
The maximum pressure can also be achieved by turning the steering on the left and right stop.
The delivery pressure must be at least 800 psi and should be max. 950 psi.
...WRe

[balimatharu]

WRe
Thanks for your reply.
When the throttle valve is closed the pressure registered is 5psi and rises to 100 psi. at 4000 revs.
So best thing to do is to remove the pump and service it. Leave the box alone for now.
There are no leaks outside. Must be internal problem.
Thanks for your help.

[MikeSimon]

The pump has an internal "pressure relief valve". If you throttle the check device you are using completely closed, the pump will go into "pressure relief" which is the maximum steering pressure in the system. In the car, it is equivalent to turning the steering wheel all the way to either side stop, or blocking the wheels from further turn, e.g. against a curb. It is a feature to prevent other components from damage. Depending on the pump you have, the tag (aluminum plate) on it, should state the maximum pressure.
I would agree, that something is not right with the pump.
Your test method was O.K. This is like "bench testing" the pump. 65 bar equals 933 PSI. We use a factor of 14.35.

[balimatharu]

I am very pleased to say that the Power has returned in the power steering.
I removed the pump and tested the pump using a power drill machine to drive the pump. It would register 100psi max.
Overhauled the pump and cleaned out the relief valve. Clamped the pump back on the vice. Couldn’t believe the power generated by the pump. The drill machine had a hard time. The pressure gauge I have, would only go up to 400psi. I could feel that it could register even higher reading.
I made a makeshift pressure testing setup. Looks very crude but did its job. I made it out of home plumbing bits.
Fitted to the car and it works perfect.
By the way I don’t need to do any bleeding. Turned the steering lock to lock a few times. That’s it. Happy man.
Thanks everyone for the help.
Have a look at the photos

[MikeSimon]

If you took the vanes out of the pump rotor, I hope you made sure you put them in the right way. One side is flat, the other round! Round side goes to the outside, of course.

[balimatharu]

Yes Mike I made sure of that. The only thing that could go wrong while assembling the pump is the oval shaped component that encloses the vane rotor. It fits both ways.
Thank you all for the precious comments.

[WRe]

Great job! :-)
...WRe

[450sl]

Hi  Balimatharu ,

Could you explain what you mean by    "Overhauled the pump"

mark

[MikeSimon]

Yes Mike I made sure of that. The only thing that could go wrong while assembling the pump is the oval shaped component that encloses the vane rotor. It fits both ways.
Thank you all for the precious comments.

On later pumps, the ring (oval part) had a little arrow ( like this:  > ) stamped on it to show the direction of rotation. Not sure whether the early pumps had this.