Capacitance

[PNachtwey]

I can't believe it. Why do you put up with manufacturers of hose and piping that don't have specifications for the capacitance of the hose or piping?

 

[hydromech]

Peter...

Has a hydraulic systems engineer done you an injustice somewhere?

Please define capacitance...

Adrian

 

[PNachtwey]

Has a hydraulic systems engineer done you an injustice somewhere?

No. Do they exist? That is the problem isn't it.

Please define capacitance...

This really should be common knowledge. It is the change in volume/change in pressure. Hoses have much more capacitance than metal tubing. The capacitance of each part of a hydraulic circuit between the valve and piston should be known. Can you imagine electrical engineers putting random size capacitors in their circuit boards?
Those that sell you the hose expect you to find out what the capacitance is of their tubing by trial and error.

 

[hydromech]

It has only recently become the case that end users and regulatory bodies are aware that faulty hoses are lethal.

I remember hundreds of cases where people would want to re-end old hoses to save money.

A single wire hose will expand if it sees excessive pressure, a multi spiral hose that sees the same pressure may not expand at all, but the pressure drop across will be the same. How is that quantified and accounted for in realistic terms?

Capacitors store electrical charge, accumulators are the hydraulic equivilent. The manufacture, supply and application of accumulators falls under the european pressure equipment directive. There are strict controls governing thier use.

Hoses and capacitors are not analogous. A hydraulic hose cannot store energy.

Adrian

 

[ScottyUK]

I agree with Peter - if the hose expands under pressure then it is behaving as an accumulator, albeit one of very small capacity. I suppose the electrical analogy would be the capacitance between PCB traces.

I think I can see where the train of thought is leading: the system needs to charge the hose 'capacitance', something which will affect the overall performance of the system. Whether it is significant enough to be a practical problem I'm not sure - I get the impression that Peter's controllers are signficantly faster than the turbine valve actuators I work with.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[hydromech]

If a hose expands enough to make an appreciable difference to the operation of the system, then the hose is likely fail or the system design is too marginal.

A hose may well expand and hold an extra 50cc of oil. A hot system with an old pump and several pilot operated valves will lose 10 times that amount in one minute.

If a hose is long enough to make a difference to the system then it is too long and the system will be performing badly anyway.

Too many people use hoses as a cheap short cut to get around bending pipe or to get around bad machine design.

A hydraulic hose cannot store energy in a realistic and useable sense.

Adrian

 

[PNachtwey]

On a servo system capacitance is a performance killer. It reduces the natural frequency of a system in a significant way. Valves should be mounted on the cylinders to minimize the capacitance due to piping and the bulk modulus of oil.

Yes, hoses can store energy just like electrical capacitors do. A lot has to do with the kind of fluid in the hose. A cubic inch of air can do more work than a cubic inch of oil when compressed to the same pressure.

Also note, the pressure waves take time to go get from one point to another so long lengths of hose or piping cause pure dead time as seen by the controller. How many milliseconds does it take for an increase in pressure at the valve to get to the piston? Have you ever thought about that?

 

[hydromech]

I don't wish to offend...but that is a typical EE response.

If pressure waves and mechanical systems behaved like electromagnetic systems the universe would be very different place.

Obviously a pressure transducer and controller can measure and respond to a pressure signal very much quicker than a hose can disipate presssure.

Hoses may well be a pain as far as control systems go but that does not make them useful as a pressure storage device.

A hose and a flow control valve in series does not make an RC circuit!

 

[ScottyUK]

I don't get the impression there was any consideration of using them as a storage device, more the opposite where stray 'capacitance' acts to roll off bandwidth in a system where high bandwidth is desirable. Knowing the value of the 'capacitance' would allow the rolloff to be modelled more accurately.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[PNachtwey]

Hydromech, you are missing the point. Hoses do cause phase delay and dead time. I didn't say it was desired. I am totally against hose between the valve and the piston.

ScottyUK, interpreted my meaning correctly.

So how can you guys design anything? There are few if any specifications for piping and hose. There aren't many software packages that can be used to simulate a hydraulic system even if there were numbers to plug into them.

EE's can buy parts and there are all sorts of specifications for each piece.

I am in the process of writing a hydraulic simulator for simulating servo systems. At first I was going to assume the pump, accumulator and valve are connected by a 'virtual hose' or pipe that has no capacitance or resistance. Now I am rethinking that. The capacitance would mean that when a cubic inch of oil is added to a pipe that one could get a little more or less out depending on the changes in the pressure. Also, the pressure will not instantly change across the length of the pipe. It should be noted that you don't need to add or subtract much oil from a cylinder to make a big change in pressure/force.

It takes a EE to think like this.

 

[hydromech]

You EE's are spoilt.

Maybe I did miss your point...oops!

I agree, although to me as a designer hoses are not a source of power, to a control system that are a black hole.

We use Amesim here at work and it seems pretty accurate, but we don't use hoses!

 

[IceStationZebra]

Yes, hoses do add some capacitance to hydraulic circuits. This is not a problem unless you need really tight control. I have an MTS system at work and the servo valves are mounted directly on the cylinders specifically to counteract this. And there is an accumulator just upstream of the servo to keep the supply smooth.

Obviously this is a concern to you, so mount the servos as close as possible to the cylinder and use mostly hard lines. You should also use the highest pressure rated hose you can to minimize the hose swelling. (stiffer hose = less capacitance)

One issue with specifying a capacitance is that it will vary with age and temperature.

As a side note, hoses often change length much more than they change diameter. When pressurized the diameter expands, which pulls on the braid causing the hose to shorten. This was a pain in my side at a former employer that made off-road telescopic forklifts.

------------------------------------------------
Quote PNachtwey: A cubic inch of air can do more work than a cubic inch of oil when compressed to the same pressure.

Yes that is true, but it also shows your lack of fluid engineering. You are compressing the air to almost 1/7th its original volume, but the oil is only being compressed ~1% at 3000psi. It takes more work to compress the air, thus you get more work back (minus the thermal losses).

ISZ

 

[PNachtwey]

You EE's are spoilt.

Yes, but only because we are smart enough to know when a supplier is selling us garbage and a bunch of hype.

We use Amesim here at work and it seems pretty accurate, but we don't use hoses!

I am not familiar with that one but good for you for joining the 21 century.

Yes that is true, but it also shows your lack of fluid engineering.

I know that all about the energy in air and oil. I can show you the calculus to prove it if you want. It is the tools and components that I have trouble with. Mostly because they are so crude and have no model. Not even for a simple piece of hose or pipe!! How dark ages.

BTW, did anybody figure out how to calculate the hydraulic system gain? ( meters per second per volt ) or ( meters per second per milliamp )?

 

[PNachtwey]

You are compressing the air to almost 1/7th its original volume,

How did you compute that?
ftp://ftp.deltacompsys.com/public/PDF/Mathcad%20-%20EnergyInAir.pdf
I had modified the worksheet to show how much energy is lost as a cubic inch of air at 3000 PSI to 14.7 PSI. Notice the volume increases by 44.6 times not 7

but the oil is only being compressed ~1% at 3000psi.

The percentage is more like -1.5%. I assumed the bulk modulus of oil is 200,000 PSI which is on the high side which means the compression will be normally be even more.

See the equation for V0
ftp://ftp.deltacompsys.com/public/PDF/Mathcad%20-%20EnergyInOil.pdf

It takes more work to compress the air, thus you get more work back (minus the thermal losses)

Yes, and I can prove how much more.

 

[budt]

HYDROMECH asked Peter:
Has a hydraulic systems engineer done you an injustice somewhere?

Peter replied:
No. Do they exist? That is the problem isn't it.

Now that certainly makes sense No, Fluid Power Engineers or Maintenance persons do not exist, AND THAT IS THE PROBLEM.

Been saying that since 1988 but I thought no one was listening.

Thanks for joining the fray for the implementation of TRAINED/DEDICATED Fluid Power Persons.


Bud Trinkel, Fluid Power Consultant
HYDRA-PNEU CONSULTING