Meter In or Meter Out for hydraulic cylinder control

[Arlin]

I have a question about using flow controls with a one-way check bypass (like this

http://www.hydraforce.com/Flowcont/Fcfcflo.htm

and

http://www.hydraforce.com/Sandwich/Sand-pdf/7-150-1.pdf)

Should the valve be setup so that it meters the oil traveling into the cylinder, allowing free flow of oil exiting the cylinder? OR should the valve be setup such that it meters the oil exiting the cylinder?

I am inclined to set-up the flow control to meter the outflowing oil from the cylinder. I think this would reduce the chance of cavitation in the cylinder and provide a smoother cylinder movement.

Any rules of thumb for this?

Thanks

 

[MadMango]

As you can see, it can be done both ways. It depends on what function/operation you are trying to control with the flow control. For what it's worth, I've only used these types of flow control for the fluid leaving the cylinder.

[green]"I think there is a world market for maybe five computers."[/green]
Thomas Watson, chairman of IBM, 1943.
Have you read faq731-376 to make the best use of Eng-Tips Forums?

 

[hydromech]

Where possible you should always install a flow control valve to control the fluid leaving the cylinder.

The reasons for this are...

1) It stops the load from pulling the cylinder I.E running away.

2) It helps to maintain a higher working pressure in the cylinder. Hydraulic systems generally perform better at higher pressures.

Hydromech...

 

[dtberry]

The previous are good replies. There is a phase that is used in the fluid power industry. "If in doubt, meter out".

 

[jrnsr]

If you meter in, the cylinder may see lower pressure and and not leak as badly (I deal with lots of customers' junky equipment).
This could be deadly as in lifting equipment (double acting hilo or dump bed for instance). Throttling a load on the suction side can result in oil vaporizing and dropping the load in a hurry.

 

[danhelgerson]

There is a Major Danger in “meter out” applications in hydraulics. Because of the relative incompressibility of hydraulic fluid, the inlet side of a cylinder will be at maximum system pressure. The outlet side will have to develop a resistive load large enough to overcome the driving force. F=PA. The area of the annulus of the rod end of the cylinder is smaller than the area of the piston end. Under no load, the pressure in the rod end of the cylinder will be equal to the system pressure times the piston area divided by the rod end area. If there is a load pulling on the rod, this will add additional pressure to the rod end.

For example; if you have a cylinder with a rod area equal to ½ the piston area mounted vertically with the rod pointing downward to lift an object that requires 3,000 psi on the rod end, to lower the load with a meter out circuit would develop 9,000 psi in the rod end; 3000 from the load and 6000 to resist the force pushing on the piston. With no relief in this line, something is going to fail.

As a rule, in hydraulics, it is always better to meter in. This requires some planning and some knowledge but will reduce heat and waste and make a safer system.

 

[hydromech]

danhelgerson is not wrong...neither was I...you see that's the thing with hydraulic systems...there are several ways to get it right and even more ways to get it very wrong.

There are major dangers when metering in and there are major dangers when metering out. Each application must be assessed and the engineer must apply the correct solution to ensure the safest operation.

Good hydraulic engineers don't make the rules they simply apply them in the best way they know how.

With any hydraulic system...apply what appears to be safest and best. If you don't...know ask for help.

He who does not know but asks is a fool for 5 minutes. He who does not know and does not ask is a fool forever.

 

[kcj]

Good points made above.

I made a post in the previous thread about overrunning and past center loads.

Meter out keeps the load in control from overrunning, but the intensification can be serious and must be checked. It applies full system pressure to closed side, and can produce even higher pressure on the rod seals.

Meter out must account for dynamic stopping conditions and shock, or it can produce high rod side pressures.

Meter in can lose control of overrunning load, and can't be used to slow or stop a high momentum load.

Counterbalance valves are a version of meter out used for load control during deceleration, but usually piloted to be open when the load is moving in a resiting direction.

As noted above, the skill is to identify the load and control issues, then decide which 'rules' to apply to the unique situation at hand.

kcj

 

[fIEROWISEGUY]

All are great answers. But I would also add; An engineer should look to avoid the need for metering oil in a circuit, by properly sizing all equipment. If you have flow controls, where does the excess oil go? In a fixed displacement pump circuit, it goes over the relief valve.

I cringe when I see circuits where there is one gpm going to the motor or cylinder, and 9 gpm is going over the relief valve. That is 9gpm of pure heat! A total waste of energy.

On one consulting job, a battery powered hydraulic boom crane had a 2 hour battery life. This was even with 48 volts and 300 amps of batteries. The end customer was very displeased that they could not use the equipment for a full shift. We increased it to a full 8 hours of operation with a circuit change, and just $90 worth of valves.

If you have no load counterbalance requirements, try to size everything to avoid the need for any flow controls or needle valves.

"fIEROWISEGUY"

 

[kcj]

I agree with your thesis that wasted energy is to be avoided. However, motion control by defintion is an inefficient process. That is a separate issue than waste from dumb circuit design.

-Most efficient is hydrostatic control. This produces just the amount of energy required by the load. In braking or load control, energy is regnerated back into the pump (now acting as a motor) and back to the prime mover engine or motor. HST pumps can be used to drive double rod cylinders for position control, but unless the loads are massive and the energy savings worthwhile, it is usually a complex and expensive way to control the load. Also limited to fairly low response speed.

-Next is open loop control by flow controls or countebalance valves. By definetion there, decelerating and controlling the load converts load enrgy into metering heat across some type of orifice: flow control,counterbalance, valve lands, etc. There is no way around that if the load is to be controlled without running away.
Usually these circuits have a variable pump so minimal excess energy is used, it is just that once the energy is put into the load (even at 100% efficiency) it can only go to heat on deceleration, for 0% efficiency.

-Servovalve system are quite inefficient, maybe 2/3 at full power delievery down to 0 efficiency at null when they are wasting null flow heat. But servo goal is control, not energy efficiency, and its the only way to have high response and high frequency circuits.

-Worst of all is what the prev post refers to: dumb circuits that create high energy fluid then dump it across relief. That is a separate issue from good load control circuits.
We've probably all seen bigger than normal gear pumps rtunning some crude motor circuit with a needle valve speed control. The wasted flow goes across relief, or across a bleed off flow control, and the builder/blacksmith wonders why it overheats.

kcj

 

[hydromech]

Ideal tool kit for Hydraulic systems Engineer.

1) Crystal Ball
2) Zero Point Energy Supply
3) Harry Potter

Job Done...

 

[kcj]

Don't forget the vendor who has wierd stuff, off the shelf, at no cost.
k