SUSPENSION QUESTION

This is the Forum for all your Citroen Technical Questions, Problems or Advice.

Moderator: RichardW

Post Reply
SimonM
Posts: 32
Joined: 09 Apr 2006, 19:41
Location:
My Cars:

SUSPENSION QUESTION

Post by SimonM »

I, I have a basic understanding of how the suspension works on my C5. One thing I can't get my head around is this; When the car is fully laden, yesterday for example it was up to the the roof and you can fit a lot in the estate :lol: I noticed that there was no real difference in ride quality. I would have thought that due to the system pumping more fluid into the spheres this would make for a harsher ride. Anyone have the time to explain?
dnsey
Posts: 1538
Joined: 20 Oct 2004, 01:39
Location:
My Cars:
x 19

Post by dnsey »

The fluid volume determines the suspension height, but the ride and handling are controlled by the interaction of the compressed nitrogen and fluid either side of the diaphragm in the spheres. Springing is supplied by the gas cushion, and damping by the fluid flow (the spheres contain dampers to restrict flow as required).
For a given height, the amount of fluid pumped into the spheres is constant, although naturally the pressure is higher due to the increased load, but the system stays in equilibrium and behaves in (nearly) the same way as when it's lightly loaded, as the gas / oil pressure either side of the diaphragm is equal. That's the beauty of HP suspension!
The really clever bit is down to the way that the shape of the diaphragm changes under loading, but that's very difficult to explain without a diagram!
In practice, I suspect that most owners would agree that the ride tends to be better when the car is loaded than otherwise,.
SimonM
Posts: 32
Joined: 09 Apr 2006, 19:41
Location:
My Cars:

Post by SimonM »

Thank you for clearing that up! I would agree that the car does actually ride better when loaded, it can feel a bit unsettled sometimes when empty
Dave Burns
Posts: 1915
Joined: 14 May 2001, 05:30
Location: United Kingdom
My Cars:
x 2

Post by Dave Burns »

Nope, you can't have an increase in gas pressure due to heavier loadings and still have the same fluid volume in the sphere, its the increased fluid volume that raises the gas pressure within the sphere.
The ride height is then corrected by replacing fluid in the suspension cylinder that has been forced into the sphere, but yes of course the pressure has been raised accordingly.

Dave
dnsey
Posts: 1538
Joined: 20 Oct 2004, 01:39
Location:
My Cars:
x 19

Post by dnsey »

Agreed - I was (over) simplifying! I should have said that the overall sphere volume must remain constant.
I was trying to make the point that the ride depends primarily on the gas, the pressure of which is balanced by the fluid.
User avatar
Kowalski
Posts: 2557
Joined: 15 Oct 2003, 17:41
Location: North East, United Kingdom
My Cars: Ex 05 C5 2.0 HDI Exclusive 145k
Ex 97 Xantia 1.9TD SX 144k
Ex 94 Xantia Dimension 1.9TD 199k

Post by Kowalski »

All that is special with regard to a oleopneumatic Citroen is that it self levels and adjusts the suspension to match the load, its not particularly complicated but the effect is quite big.

Without self levelling the suspension has to be a compromise between fully laden and empty. When the car is unladen, you've still got the springs fitted that you need to cope with a full load. When you've got a full load, the springs have to be stiffer than on a self levelling car because the car has its suspension compressed by the load, i.e. the suspension travel is reduced so the springs have less travel to do their work.

What oleopneumatic Citroens can't do is adjust the damping dependant on load, so when the car is unladen it still have to have enough damping for when its fully loaded, so unladen it might be over damped.
alexx
Posts: 462
Joined: 19 Nov 2002, 02:42
Location: Slovenia
My Cars:

Post by alexx »

As Dave said, working pressure in spheres is higher on loaded car. I did some calculations a few years ago and the spring rate is rising with the 2nd potention of the load. So, you have relatively stiffer springs. But on the other side, damping force remains constant (although higher would be needed) and the ride is a bit different, but not much

Spring rate is rising in proportion to the load on the cars with pneumatic suspension (MB S class, VW Phaeton etc) and damping force is also controlled by the suspension computer
Xantia mkII 1.8 16v '98
previous: BX16 '89, Visa 1.1 '86 and father's GSA '81
User avatar
Mandrake
Posts: 8615
Joined: 10 Apr 2005, 17:23
Location: North Lanarkshire, UK
My Cars:
x 664

Post by Mandrake »

Hi Guys,

Interesting comments, although I have a slightly different take on it than some of you :)

There are two aspects to maintaining a ride that is more or less constant with load changes - one is the ride height - a conventionally sprung car obviously sags when loaded up, and this asymetric equilibrium point (relative to the limit stops of the suspension) upsets both the ride and handling substantially, so obviously hydropneumatic has the advantage there, but ride height is only half the story....

The other thing that happens is that the resonant frequency of the suspension is maintained almost constant over a wide range of loads. (It isn't with conventional springing)

The resonant frequency is primarily determined by the sprung mass of the vehicle on a given axle, the spring constant of that axle, and is also modified a small amount by the damping rate.

Optimum ride comfort is generally accepted to be a resonant frequency between about 0.8Hz and 1.2Hz. Go much below 0.7Hz and you're into "sea sickness" territory, above about 1.4Hz and you've got a fatiguing jittery sports car type ride.

Typical conventionally sprung saloons tend to be around 1.0 to 1.2Hz, while Hydropneumatic Citroen's tend to be around 0.7 to 0.9Hz. (In soft mode for Hydractive 2 models)

With conventional springing the spring constant only increases "somewhat" in response to an increase in weight, (depending on factors such as how the coils are wound in a coil spring type) but never in 1/1 proportion, which means the resonant frequency inevitably goes down with a weight increase.

A change in resonant frequency requires a different damping rate to give "optimal" damping, so it follows that the damping of a conventionally sprung car can only ever be optimal at one load weight.

For a saloon - especially rear suspension - that tends to be when lightly loaded, (giving a soft wallowy ride when heavily loaded) or for an estate when heavily loaded. (giving a harsh bouncy ride when empty)

With hydropneumatic, the constant quantity (not volume) of gas in a sphere gives a unique characteristic - the springing stiffness increases exactly in proportion to the load weight.

(At least until the point where the diaphram is nearly at the top of the sphere, which happens just beyond maximum rated load, unless the spheres are low in gas)

alexx - I'm not sure where you get your "2nd potention of the load" from, the characteristics of Citroen spheres can be worked out easily using Boyles law - gas volume = 1/pressure. (A classic 1/x curve with an asymptote on the veritcal axis)

The precharge pressure of the sphere is the gas pressure when the gas is allowed to exapand to fill the entire inside - about 400cc or 450cc on most spheres.

If you apply exactly twice the oil pressure the gas will be compressed to exactly half the volume. Because the springing rate is determined by the change in pressure vs the change in oil displacement, half the working gas volume gives exactly twice the springing constant, therefore the spring constant is proportional to load weight over the operating range of the suspension.

Because the spring constant gets stiffer in direct proportion to the weight increase, the resonant frequency stays the same.

Also because of the very soft springing rate, the full travel of the suspension (in terms of oil displacement by the hydraulic rams) only moves the diaphram in the sphere over a small percentage of its range, which means the springing rate also doesn't change much over the travel of the suspension. (Good linearity)

When you load the car up with more weight, the "working point" of the diaphram inside the sphere shifts further up for the same ride height.

Kowalski - the damping is not really an issue as the required damping is related more to the resonant frequency of the spring/mass system than it is to the specific load weight - so by keeping the resonant frequency the same the amount of damping required is also the same.

There is a small change in ride quality with increased loads however, and that is due to the change in the sprung to unsprung weight ratio...while the sprung weight increases with load, the unsprung weight always stays the same.

So when the car is lightly loaded the unsprung weight is a greater percentage of the total weight, therefore the ride is not quite as good.

Regards,
Simon
Simon

1997 Xantia S1 3.0 V6 Auto Exclusive in Silex Grey
2016 Nissan Leaf Tekna 30kWh in White

2011 Peugeot Ion Full Electric in Silver
1977 G Special 1129cc LHD
1978 CX 2400
1997 Xantia S1 2.0i Auto VSX
1998 Xantia S2 3.0 V6 Auto Exclusive
alexx
Posts: 462
Joined: 19 Nov 2002, 02:42
Location: Slovenia
My Cars:

Post by alexx »

So, let's do some simple math

Let's assume front sphere on my car has 450 ccm and pressure 55 bar (when new). Piston diameter is 2.2 cm (maybe it's more on Xantia mkII, but doesn't matter). Weight over one front wheel is about 400 kg (3924 N)

According to F = p A (force = pressure x area), we need 103.3 bar pressure on the piston to produce the force equal to the supporting weight. This is done by the pump, filling a part of the sphere with LHM and compressing the nitrogen in the sphere

According to p V = const (pressure x volume = constant), nitrogen is compressed from 450 to 239.7 ccm

Now, let's push the car down 1 cm. It would fill another 3.8 ccm of LHM into the sphere, and pressure will rise to 104.9 bar, producing force 3986 N. From that, we can calculate spring constant - 62.2 N/cm

If we double the weight to 800 kg, according to the same formulas we get spring constant 257 N/cm, about 4 times more

During the drive, effective spring constant is higher, because suspension movements are quick and the process in the sphere is not isothermal, so we should use p V^k = const, where k is somewhere around 1.4. Spring constant is higher for quick movements than for slow movements, which is good

In the case of pneumatic suspension, volume of the gas remains constant - you maintain the height by pumping the air into the air spring causing its pressure to rise, so spring constant is proportional to the weight (for the real air spring, this also somewhat depends on the shape of the spring)
Xantia mkII 1.8 16v '98
previous: BX16 '89, Visa 1.1 '86 and father's GSA '81
sub42
Posts: 70
Joined: 18 Oct 2005, 20:56
Location: UK
My Cars:

Post by sub42 »

I just wish they fitted it to my Synergie!
Previous convictions for ferret and giraffe rustling.

Synergie 1.9td SX loaded spec.
Peter.N.
Moderating Team
Posts: 11563
Joined: 02 Apr 2005, 16:11
Location: Charmouth,Dorset
My Cars: Currently:

C5 X7 VTR + Satnav Hdi estate Silver
C5 X7 VTR + Hdi Estate 2008 Red

In the past: 3, CX td Safaris and about 7, XM td estates. Lovely cars.
x 1199

Post by Peter.N. »

My word, a really in depth discussion. Something that no one has mentioned (unless i've missed it) is the sprung to unsprung weight ratio, which of course is applicable to any type of suspension. The heavier the load in the vehicle, the 'lighter' in proportion becomes the unsprung part of the suspension, wheels, brakes, hubs etc, so less movement / vibration is transmitted to the body. The tail wagging the dog syndrome.
User avatar
Mandrake
Posts: 8615
Joined: 10 Apr 2005, 17:23
Location: North Lanarkshire, UK
My Cars:
x 664

Post by Mandrake »

Peter.N. wrote:My word, a really in depth discussion. Something that no one has mentioned (unless i've missed it) is the sprung to unsprung weight ratio, which of course is applicable to any type of suspension. The heavier the load in the vehicle, the 'lighter' in proportion becomes the unsprung part of the suspension, wheels, brakes, hubs etc, so less movement / vibration is transmitted to the body. The tail wagging the dog syndrome.
Yep, you missed it, I mentioned the change in sprung to unsprung weight ratio :lol:

Regards,
Simon
Simon

1997 Xantia S1 3.0 V6 Auto Exclusive in Silex Grey
2016 Nissan Leaf Tekna 30kWh in White

2011 Peugeot Ion Full Electric in Silver
1977 G Special 1129cc LHD
1978 CX 2400
1997 Xantia S1 2.0i Auto VSX
1998 Xantia S2 3.0 V6 Auto Exclusive
dnsey
Posts: 1538
Joined: 20 Oct 2004, 01:39
Location:
My Cars:
x 19

Post by dnsey »

Does anyone have a link to that nice Citroen diagram which shows how the diaphragm area varies with displacement? It's an important factor in the design, but very difficult to discuss without something to look at :)
User avatar
Mandrake
Posts: 8615
Joined: 10 Apr 2005, 17:23
Location: North Lanarkshire, UK
My Cars:
x 664

Post by Mandrake »

alexx wrote:So, let's do some simple math

Let's assume front sphere on my car has 450 ccm and pressure 55 bar (when new). Piston diameter is 2.2 cm (maybe it's more on Xantia mkII, but doesn't matter). Weight over one front wheel is about 400 kg (3924 N)

According to F = p A (force = pressure x area), we need 103.3 bar pressure on the piston to produce the force equal to the supporting weight. This is done by the pump, filling a part of the sphere with LHM and compressing the nitrogen in the sphere

According to p V = const (pressure x volume = constant), nitrogen is compressed from 450 to 239.7 ccm

Now, let's push the car down 1 cm. It would fill another 3.8 ccm of LHM into the sphere, and pressure will rise to 104.9 bar, producing force 3986 N. From that, we can calculate spring constant - 62.2 N/cm

If we double the weight to 800 kg, according to the same formulas we get spring constant 257 N/cm, about 4 times more
Hmm..... I'm surprised by that.... but I can't see anything wrong with your maths when I double checked it... :?

Although one thing that you have possibly done wrong is calculated outside the normal working range of the suspension - 800Kg on one front wheel would require a pressure of 206.6 bars, which is well beyond the 170 bars maximum pressure, and would be pushing the sphere diaphram well up into the non linear region near the top of the sphere...

If the spring constant goes up by 4 times when the weight increases by 2 times, how come the resonant frequency of the suspension doesn't increase by 1.4 times ?? (Since mechanical resonance is the square root of spring constant over mass....)
During the drive, effective spring constant is higher, because suspension movements are quick and the process in the sphere is not isothermal, so we should use p V^k = const, where k is somewhere around 1.4. Spring constant is higher for quick movements than for slow movements, which is good
Do you have a reference for this ? I find it hard to believe that the almost massless gas and diaphram can't keep up with the movement of the heavy unsprung weight of the suspension....if the gas couldn't follow up the movement instantaneously you'd have problems with cavitation in the damper valve. (Exactly the reason why conventional gas filled shock absorbers have gas in them)

Also suspension movements aren't nearly as quick as people think - because of the large inertia of the unsprung weight most typical suspension is limited to movement rates of no more than 10-15Hz, the so called "wheel hop frequency".

Any road oscillations faster than this and the suspension simply can't respond due to this inertia. The only thing that can respond fast enough is the flexing of the tyre (since very little inertia is involved there) and if the tyre can't flex enough the tyre will simply lose contact with the ground.

Regards,
Simon
Simon

1997 Xantia S1 3.0 V6 Auto Exclusive in Silex Grey
2016 Nissan Leaf Tekna 30kWh in White

2011 Peugeot Ion Full Electric in Silver
1977 G Special 1129cc LHD
1978 CX 2400
1997 Xantia S1 2.0i Auto VSX
1998 Xantia S2 3.0 V6 Auto Exclusive
Peter.N.
Moderating Team
Posts: 11563
Joined: 02 Apr 2005, 16:11
Location: Charmouth,Dorset
My Cars: Currently:

C5 X7 VTR + Satnav Hdi estate Silver
C5 X7 VTR + Hdi Estate 2008 Red

In the past: 3, CX td Safaris and about 7, XM td estates. Lovely cars.
x 1199

Post by Peter.N. »

Oops Sorry! :oops:
Post Reply