Hydraulic accumulator & ram as supension 1

[kosheboy]

Ok, we have come up with a hydraulic suspension system for HiRailed truck, the suspension is needed to allow for twist in the rail network, and for any bad joints in the rail

I am struggling with one issue, that is thermal expansion of the gas and the affect this will have on ride height, and whether we have enough stroke for the suspension to do its thing, as well as allowing for a ride height change due to thermal expansion of the accumulator gas

Let’s say we set the accumulator up so that it balances the weight of the truck at mid stroke of the ram on a 20° day, say the day gets hotter and with the suspension working for a time the temp of the nitrogen in the accumulator increases to 40°

This will expand the gas in the accumulator and as a result the truck will rise, but as the truck rises it take more effort to lift (due to design)

How can I factor in the two variable to work out how much the truck will rise, & therefore how much extra stroke I need to allow for this (and meet the min twist requirements)

We have

D1 = 0, Ram position (Ride height) at 20°
D2, Ram position after pressure increase (due to temp increase)
V1, volume of accumulator gas at ride height
P1, pressure at ride height @ 20°
P2, pressure due to increase in temp (20° increase)

Can someone help with this
I can crunch the number just fine, but finding the right formula is my problem

Best regards Brendon

 

[MikeHalloran]

You need a ride height control valve in the hydraulic circuit, probably one at each corner, at least one at each axle.

See "Citroen DS".



Mike Halloran
Pembroke Pines, FL, USA

 

[kosheboy]

We dont mind if the ride height increases a bit, but I need to know it the increase is going to use up x% of my spear stroke, if it turns out that the increase is to much for the design to handle then, we will have to look at ride height control valves..

Brendon

 

[MikeHalloran]

Now figure what happens when a 300 lb gandy dancer jumps in or out of the truck.



Mike Halloran
Pembroke Pines, FL, USA

 

[kosheboy]

it's a 13ton truck.! so it will drop a mm or so when people get in,

we are going to plumb up a test rig to check if its a problem or not, My gut feeling is that it will be in the 10's of mm rise, but we will soon see, I'll keep you posted,

In the mean time If anyone has formula for me to crunch... It would be good to see how close we get....
Coefficient of Thermal expansion for nitrogen is 0.00753 1/C

Brendon

 

[kosheboy]

Oh, and the Payload is constant (will not change)

 

[LittleInch]

I really don't think you have much of a problem.

If you remember your boyles law of gases a perfect gas has a relationship P1V1/T1 = P2V2/T2.

The key is that p is in absolute pressure and T is in degrees KELVIN. Hence your 20C (I assume) rise in ratio terms is actually 313/293 = 1.068, say a 7% increase in pressure. This isn't quite precise as you need to put in the compressibility factor, but its close enough. Only you know what your initial pressure is but I guess it's fairly high. Clearly the ride might go up a bit, but in volume terms will be somewhere around 3 to 4% as the increase in pressure means an increase in volume which then reduces the pressure.

I can't see this having a big effect.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[HPost]

The polytropic index of nitrogen being 1.4, the variation in temperature won't make much difference. If you think it will, increase the volume of nitrogen to compensate.

I can't see an issue here to be honest.

HPost CEng MIMechE

 

[Compositepro]

"but as the truck rises it take more effort to lift (due to design)"
The answer to your question is determined almost entirely by this statement, and you have given no details. Lacking the details, what you have is a force balance problem and any out of balance will cause the ram to go full stroke one way or the other.

 

[PNachtwey]

I can't see a big problem with temperature either. The suspension will simply go up or down a little due to temperature.

I am more interested in the details. Is there one accumulator for all the suspensions? It seems to me this would be a good way to go. Is their gas on the top side of the piston? It will expand too. There must be a way to compensate for leakage.

Is this just a leveling problem or a damping problem too?

The big trick is the damping. I have seen some clever ways of increasing the resistance to movement as the piston nears the end of stroke but then the load must be centered again.

Finally, shock absorbers basically convert kinetic energy to heat. Then shocks will get hot if they are doing the damping. I have seen tests done on shock absorbers designed for ambulances, bus, fire trucks up to tanks. The shocks can get hot enough to burn the paint off the outside of the shocks.



Peter Nachtwey
Delta Computer Systems

http://www.deltamotion.com

 

[Oldhydroman]

I’m surprised that you are limiting your deliberations to such a narrow temperature band. You haven’t said where in the world you will be running this vehicle so let’s guess at somewhere in the temperate zones of the northern hemisphere. It is easy to imagine that, after running at high speed for a long time on a rough track on a hot day, the oil temperature could easily be as high as 70 degrees centigrade. Conversely, on a particularly cold day in winter (but not so cold that no work can be done) you might need to consider the vehicle having to operate with the oil temperature as low as -20 degrees centigrade. Now take these values as a ratio of absolute temperatures and you have 343/253 = 1.35. So that’s a 35% difference in gas volume for the same pressure at the two extremes in temperature. This may not be insignificant and, as you suspected, warrants further investigation.

Disregard the accumulators for a minute and imagine that the hydraulic cylinders in your suspension were all connected together and piped to a hand pump with a pressure gauge on its outlet. With the vehicle stationary and on level track you could gradually pump up the suspension and create a plot of pressure against ride height. This will be a function of the weight of the vehicle, the number of wheels, the size of your cylinders and the geometry of your mechanisms. The fact that there are no accumulators in circuit won’t change the pressures – it just means if you ran like this you would have a very hard ride and probably generate pressures high enough to damage something.

As the ride height varies so will the pressure needed to support the vehicle at that height – this is because you have a fancy geometry involved in your mechanism. It’s for you to work out the trigonometry, but, for the purposes of an example, let’s image that each cylinder had a bore of 50 mm and a stroke of 100 mm and that the pressure varied linearly from 80 bar (lowest ride height) to 130 bar (highest ride height). It’s also up to you to work out how the stroke of the cylinder varies with ride height but let’s represent the distance from the back of the piston to the rear of the cylinder (inside) by the letter X. So when the cylinder is fully compressed (lowest ride height) X=0 and when the cylinder is fully extended (highest ride height) X=100. The expression relating pressure (P) to piston position (X) will be P=80 + X/2. These will be gauge pressures; strictly speaking you should work in absolute pressures but that’s a complication which isn't worth the effort.

Now let’s think about those accumulators you are going to install to act as the springs (not the dampers). You wouldn't want the accumulator bladder to be fully expanded before the suspension cylinder was fully extended so we can assume you would still want ~10% of the accumulator volume filled with oil. We might also assume that you wanted the cylinders to be at 50% stroke at 20 deg C. So let’s just pick an accumulator and see how it works out.

A typical “10 Litre” accumulator has a gas volume of 9.4 Litres. You want there to be 10% of oil still in the accumulator when the cylinder is fully expanded so the gas volume will be 8.46 Litres (8640 cc) at this point. The example cylinder has a total swept volume of 196 cc so when at 50% stroke the gas volume will be 8542 cc. When at 50 mm cylinder stroke the pressure would be 105 bar (just using my example geometry and numbers). This lets you define the accumulator pre-charge pressure at 20 deg C. Your 8542 cc of gas at 105 bar will reduce its pressure by the ratio 8542/9400 and will be just 95.4 bar. (If we bothered to do the calculations according to absolute pressures the number would be 95.3 bar and even that’s not accurate because we haven’t included compressibility factors which is why it’s safe to ignore the “use absolute pressures” stricture).

Then the temperature changes...imagine it’s now 40 deg C. If the oil circuit were an enclosed volume the gas pressure would rise. But the volume isn't closed so the cylinder rod will extend a little. The complication is that your suspension geometry means any change of stroke causes there to be a change of pressure – and this will affect the gas volume. So here’s how you get round it:

In a spreadsheet enter your equation for pressure in terms of cylinder stroke (P=80+X/2) and arrange this so that you can input a value of X and get out a value for P. This will be the pressure needed to sustain the vehicle at that particular height. Call this “P(geometric)”

In a nearby cell enter the expression that allows you to calculate the gas pressure for any value of X and at any new temperature T (Kelvin). It goes like this (Boyle’s law):

P x V = N x T
[Where P is in bar, V is the gas volume in cc, N is a constant (for that size accumulator and that particular pre-charge pressure) and T is the temperature in Kelvin.]

We know that at 20 deg C (293 Kelvin) the gas volume is 8542 cc when the pressure is 105 bar. Plug in all these numbers and you get N = 3061 (can’t be bothered to work out the units here). Now the gas volume was 8542 cc when the cylinder was at 50 mm stroke, so we could say that the gas volume was:

V = 8444 + 196X/100 , i.e. when X = 50, V = 8542.

Now we can enter a formula on the spreadsheet to give us the gas pressure for different values of X (and T) for our current selection of accumulator size and pre-charge pressure:

P = (3061 x T)/(8444 + 196X/100)

This will be the gas pressure at any particular cylinder stroke at any gas temperature. When you plug in T=293 and X=50 you will get P=105. Call this “P(gas)”

Then, in another cell, put in “P(gas)/P(geometric)”. This is the ratio of calculated pressures and the answer will should be 1.000 because both pressures are 105 bar (actually it won't quite be 1.0 because the constant 3061 was rounded down a little). You can then correct 3061 if you want but I wouldn't bother – this is, after all, only an example.

Now put in a new value for the temperature (40 deg C = 313 Kelvin). The spreadsheet will calculate a new P(gas) value. Now – and this is the crafty bit - click on the cell with the ratio of calculated pressures and use the spreadsheet’s “goal seek” function to make the ratio equal to 1.000 by changing the value of X. You will then get a new solution where the gas pressure dictates a particular cylinder displacement which uses exactly the same pressure to support the vehicle. In the simple example I've used here the cylinder displacement changes to 63 mm. Remember that this is the change of cylinder stroke – you still need to calculate the new ride height based on that particular cylinder position.

No-one can tell you whether the change in cylinder stroke between 20 deg C and 40 deg C is significant because it depends on so many factors and also on your opinion on what change is acceptable. You might want to investigate a much wider temperature range though. If you are a little more sophisticated with the spreadsheet you can use it as a design tool to investigate the effect of different accumulator sizes, cylinder sizes, pre-charge pressures etc.

Hope this method gives you the steer you were looking for to get a feel for how your design will perform.

DOL

 

[kosheboy]

Thanks all, We have a spread sheet running with lots of numbers in it already that is giving us, stroke, shaft size, volumes, balance pressures and oil flows etc.., I'll churn over all this info, and shoe horn some of it into the spread sheet, I think, It could be a real problem, but if we look at it and its not then cool..

I'm just the mechanical designer, our Hydraulic Engineers, did not consider this, I’m just bringing it up out of concern..

I will keep you posted, I think we may need to install some type of ride height control

Regards Brendon

 

[EdDanzer]

Temperature will have a significant impact on you gas suspension if your accumulator gas volume is not large enough. Some years ago I spent some time and money on a delimbing machine that used accumulators in a closed loop system without a pump for knife pressure control. Even a 10° temperature rise had an effect on the knife pressure. After several weeks I gave up and put an engine on the machine to keep the pressure within 200 psi from -10°F to 110°F. It may be possible to have a large enough gas bottle connected to a smaller accumulator tank and be successful. The other choice is a small 12VDC pump, pressure switches and relief valves to control pressure.

In the last year we have done some work with downhill mountain bike suspensions using gas springs. My conclusion is the gas volume is not large enough and the spring rate changes through out the ride and not in a good way if you start on a cold morning. If you ride in the evening is does not seem to be as bad. On one front fork we added a small gas bottle and the change helped, but adding the gas bottle did not fix the lousy shock valving.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[LittleInch]

Part of the issue is that for a suspension system you are "working" the gas by compressing and expanding it on regular basis which will heat the gas up a lot more than maybe your 20 degrees. In that case as oldhydroman says, it could easily be 70 or more if the gas volume is quite small hence it could easily be compressed a lot more and hence rise in temperature more.

As soon as you introduce changes in pressure and tempearature on a varying basis, the simple calcualtions go out of the window and small volumes and small cyclinders really make any analysis very hard. If you want less impact then a larger gas volume in practice works better than smaller ones...

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[EdDanzer]

LittleInch is correct in that the calculations become very complex. To get good results you will need a good software. I have looked at doing this type of simulation and LMS is probably the best choice if you don't want to create the math.

http://www.lmsintl.com/LMS-Imagine-Lab-Platform

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[Oldhydroman]

There are a lot of systems that do use accumulators as a suspension spring (fork lift trucks being a common example). It is also common to switch an accumulator into circuit on the lift circuits of off road boom lift vehicles.

The business of the gas heating up when it is compressed (during running) is a bit of a side issue - the gas will cool down again when it is expanded a fraction of a second later. The heat I was concerned with will be generated by: friction in the cylinder seals, fluid friction in the pipework/fittings and Joule heating of the fluid (work done against its own viscosity). But that wasn't the original question and the dynamics of the system haven't been presented for comment. Some very high [instantaneous] temperatures can be generated if you compress the [ambient temperature] gas to a very small proportion of it's original volume. Similarly very low [instantaneous] temperatures are caused by extensive expansion of a volume of [ambient temperature] gas. You do need to do some sums in this area and decide if you need any special fluids or special formulations for the bladder/diaphragm.

There was an earlier comment about this being a "force balance problem" which is heavily reliant on the increasing pressure with ride height by virtue of the suspension geometry. This is, I believe another red herring. If the cylinder were vertical and there were no variable geometry interfering with the mechanism, then the accumulator itself would still act as a gas spring and create the necessary relationship: As your crane lowered the vehicle onto the tracks the gas in the accumulators would compress and the suspension cylinder pressure rise until that "pressure x area" just balanced the weight of the vehicle.

It is convenient to think of the gas volume as analogous to the spring RATE (a large gas volume changes pressure little with the stroke of the cylinder = a low spring rate, a small gas volume changes pressure massively with the stroke of the cylinder = a high spring rate). I don't think a large gas volume would help here because you would lose the spring effect for the dynamic operating mode of the equipment. A 35% change in absolute temperature will cause a 35% change in volume at the same pressure. A bigger volume to start with will induce a bigger incremental change in volume.

You might find you calculate to the ultimate in precision but then can't get the system to work as you expected because you are unable to define the cylinder friction. Get the worst case figures from your cylinder suppliers and then make them a bit worse still. Then perform a sensitivity analysis: Will it still work if cylinder friction were really that bad? Will it still work if cylinder friction were zero?

Also be aware that it is possible to "trim" the gas volume of standard size accumulators by introducing some oil into the gas side of the accumulator.

Let us know how you get on.

DOL

 

[EdDanzer]

I spent several hours riding downhill with a Fox 40 Float air front fork Saturday and had it get stiff from gas temperature change. The change was from 1+” sag to none in the front. The Vivid air in the rear did the same thing but I did not measure the sag difference. The ambient temperature only changed about 10 to 15°F which has never caused much stiffening if only riding an hour or two. The trails did have several drops and rough areas so there was a lot of higher speed suspension movement of several inches. Towards the end of the day the vibration into the handle bars when going 40 mph down the logging road was uncomfortable on the hands and wrists.
If the air is being compressed for ride it will get hotter and if the reservoir is not large enough you will see the spring rate and ride height change.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[MikeHalloran]

It's been almost fifty years since I drove a Citroen DS, but I remember the suspension stiffening up just a teensy bit as I drove it, really hard, on a really bumpy road.
The sensation of seeing big bumps go by, but hardly feeling them, was strange.
The car stayed level, and the ride height didn't change, because of the automatic leveling valves at each corner of the car, while the 'spheres' (accumulators) soaked up the bumps.
No other contemporary car could go half as fast on a bad road.
I'm not convinced that a modern car could match it either.

Driving a Citroen is something you should definitely do if the opportunity ever presents; it's like no other car. Okay, I don't know if the new ones are as impressive. ... or as weird in so many ways.


Brendon, you need leveling valves to compensate for thermal pressure changes in the accumulators, both from ambient temperature changes, and from self-heating of the oil flowing in and out of the accumulators, and of the damping valves if you're not fitting discrete shock absorbers.



Mike Halloran
Pembroke Pines, FL, USA

 

[kosheboy]

I agree, and now see the light,
I am a novice (hydraulically), but I did a simple P1xV1xT1=P2xV2xT2, and then P2xV2=P3xV3, (based on only 40° temp increase), or though this is very basic, it indicated to me that the increase in gas pressure was enough to fully lift the truck by an extra 200psi, (i.e. the suspension cylinder goes to end of stroke).

I’ll pass all this on to my boss and our hydraulic engineers, It’s all doable by us, but just hadn’t been considered in the project, so we will wear some of the cost, opps.

Again, Thanks all for you time

 

[kosheboy]

Correction, Ment to type P1xV1/T1=P2xV2/T2