quote:
Originally posted by M.B.DOC:
If you are racing the car then the self-leveling system should go, but for street use it should be able to perform fine. As you surmised the system is for maintaining ride height & not racing. By removing the system the car will weigh less & pick up a bit of HP (pump).
Just wanted to say thanks for all those that responded. I appreciate the input . I posted my question to the Mercedes Realtime mailing list as well and thought I should share this very enlightening response from David Bruckmann:
----------
(Note that my frame of reference is the self-levelling system from the W123
T-series, although I believe that the 201 uses essentially the same wacko
scheme.)
Wael El-Dasher wrote:
>>>I see the reference to the relief valve in the shop manual, which is set to
>>>open at 143+/- 10 bar. However it seems the pressure in the feed line is
>>>directly related to the exhaust camshaft as the hydrolic pump is driven by
>>>the exhaust camshaft. So am I to deduce from this that the rate and pressure
>>>vary with engine rpm?
No. The excess pressure generated at higher RPM is bled off at the pressure
regulator. The fluid is simply routed back to the reservoir.
>>>Although I see no reference to nitrogen spheres in MB's shop manual, I am
>>>equating it to the leveling valve, or should it be the pressure reservoir?
The nitrogen-filled spheres are the "bombs" that are located near each
strut. MB calls them "pressure reservoirs". It may help to think of the
spheres as coil springs (which are really also "pressure reservoirs" in
their own way), and the hydraulic fluid as a steel bar or other
incompressible member. Fluid doesn't compress or expand (much), nitrogen
gas does.
>>>Could the rpm relation between the hydrolyic pump/leveling valve cause the
>>>pressure reservoir to over pressurize the rear shocks leading to less
>>>travel? (ie such as at slow speed, low gear but high rpm <2nd gear 5000rpm
>>>steadystate for eg.> ).
There's no such relationship between RPM and system pressure, except when
there is an internal leak and the hydraulic pump can't keep up with
pressure loss (in a worn-out controller, for example). The pump should put
out at least the nominal system pressure at idle, although I believe it
reaches full output at around 2000 rpm. So it will take longer for the
system to compensate for a heavy load added at idle than when the engine is
revving higher.
>>>So what you are saying is the rear spring's compression fights the
>>>overpressurizing of the spheres (pressure reservoir).
You'll have to ask the MB engineers why they felt it necessary to leave a
spring there when they could have just used system pressure to support the
car. As I mentioned before, I rather suspect they simply didn't want to get
nasty phone calls from Citroen's patent lawyers.
Let me explain a little more about the principles at work as I understand
them. Firstly, you have a steel sphere (well, more oblong than spherical,
but let's call it a sphere for now) with a rubber diaphragm bisecting the
two halves of the sphere. When there is no fluid pressure in the system
(i.e. sphere is in a box on the parts counter at the MB dealer while the
customer is being given smelling salts after hearing the cost), the
nitrogen gas fills all available space in the sphere, forcing the diaphragm
against the walls of the sphere. In operation, fluid is introduced under
pressure into one side of the sphere. It presses against the rubber
diaphragm, thus compressing the nitrogen gas (on the other side) and taking
up as much space as it needs until reaching equilibrium with the pressure
of the nitrogen gas, let's say roughly in the middle of the space. Since
the nitrogen is still gaseous, it can continue to act as a perfectly
progressive spring.
So let's say you hit a bump with normally-inflated spheres. The strut
piston forces fluid through the line into the sphere. Fluid doesn't
compress but gas does, so the nitrogen takes up the additional load. As the
gas is compressed, it's pressure progressively increases until the gas
pressure balances the load. The diaphragm stops, the strut travel is
arrested, and the wheel is stabilised. To prevent overcompression/bounce
etc., a flow restrictor retards the movement of fluid between the strut and
the sphere, thus achieving an effect similar to a conventional shock
absorber. Small, low-amplitude bumps are passed through to the spheres
because the fluid is flowing relatively slowly through the flow restrictor.
Larger bumps invoke larger and more sudden flow, which is restricted more
noticeably. This is where different fluid viscosities will make a
difference. Note that it is two-way damping: you hit a bump, the wheel
moves up and fluid is forced from the strut through the restriction into
the sphere. You move off the bump, the wheel returns downwards, and the
spring effect of the sphere is damped by restricting hydraulic fluid flow
to the strut.
So, if there's too MUCH nitrogen, the diaphragm will remain biased towards
the fluid side in terms of its "home" location, since the gas will be
compressed to equilibrium with the fluid earlier than expected. If there's
too LITTLE gas pressure, the fluid pressure will push the diaphragm further
into the space normally occupied by the gas, again until equilibrium
pressure is reached. If the diaphragm has ruptured or all the gas is gone
(it will leak through the rubber over time), there will be no spring effect
at all, since fluids don't compress (much).
The pressure of the fluid and the resultant equilibrium pressure of the gas
do not change in either case: what changes is the volume of nitrogen gas.
So if you have a lower initial nitrogen fill pressure resulting in, say,
half the normally acceptable gas volume once compressed to equilibrium, you
will have twice the spring rate (or thereabouts; the actual relationship is
likely different but that's the idea) und thus HALF THE TRAVEL before
reaching a given compression. So hitting a big bump in a car with low gas
pressure could cause wheel travel that results in very high pressures in
the sphere/strut, and this would happen in a time shorter than would be
required to vent the excess through the relief valve. The risk of damage is
not so much a blown line as it is blown strut/piston seals (larger bore,
moving part, etc).
When you add a load to the car trunk, the car becomes heavier. The weight
forces the strut piston further into the strut cylinder. This causes fluid
to move from the struts into the spheres as the nitrogen gas compresses to
balance the load, causing the car to sink. The level controller reads the
change in wheel height and introduces more fluid into the system, which
takes up the displacement of the now-compressed gas, causing the struts to
extend again. The level controller is designed to admit fluid slowly so as
not to react to every bump in the road. Relatedly, the controller reads
wheel position in the middle of the stabiliser bar so that cornering
doesn't induce a false reading.
>>>I can also see were such a rate rebound quicker, relegating the rear shocks
>>>to play a much smalled role. Please correct me if I am wrong.
As to the damping effect, it is important that the diaphragm rest in such a
way that it can move in both directions. This is why there's a nominal
pressure in the system even without additional load in the car. There must
be enough pressure in the system to overcome the effects of gravity AND the
steel spring when, for example, the car goes over a deep pothole or
suddenly becomes airborne (hey,