Question about counterbalance valves 1

[freddyj]

My question related to the back pressure produced by the valve. From what I've learned so far on my own is that the Pilot ratio determines how much pressure is going to unseat the pilot and along with the bias spring determines the amount of back pressure that is maintained while the pilot is opened, correct? My application uses a counterbalance valve and the generated back pressure is taking away much needed net force. I have a 6:1 valve, would a 10:1 reduce the back pressure? If so by about how much compared to the 6:1?

 

[hydtools]

Use a meter-out pressure-compensated flow control with bypass check valve for free reverse flow for the cap end flow instead of pressure control. At high strip load, low to no speed, the cap end pressure will be very low. At low strip load, higher speed controlled by the flow control, the cap end pressure will be higher but when you don't need the high initial stripping force; just controlled speed.

Ted

 

[PNachtwey]

Hydtools, that will make the system non-linear and interfere with the motion controller's ability to control accurately.
The only 'active' valve in the system should be the servo valve. Other valves interfere with the PID control.

 

[hydtools]

Then he needs velocity feedback on the retract stroke to get max open valve for the cap end at zero to low velocity in order to get max pressure differential across the piston when first stripping the press.

Ted

 

[PNachtwey]

I don't know what you mean by that. The motion controller should always have position feed back. The motion controller can compute velocity from the changes in position. This is always happening.

A 2:1 spool will allow double the flow to or from the cap end of the cylinder for the same pressure drop. Ideally a servo system works best when the valve spool is cut in proportion to the surface areas on the piston.

I would follow kcj's advice about working through the VCCM equations to calculate the steady state speed when retracting. There is a term in the VCCM equation that takes into account valves with spools cut to provide 2:1 flow. If the calculations show the system can't attain the desired speed then a bigger valve is required. Other key points that haven't been brought up. Is there an accumulator? The VCCM equation assumes the supply pressure is relatively constant which will only happen if there an accumulator big enough to smooth out the pressure ripples. This is like having a big capacitor on a power supply.

This should be a reasonably easy press to control as the surface areas are about 2:1 and not 10:1.

 

[EdDanzer]

Another way to control this uneven area cylinder is to use cartridge proportional valves with different flow capabilities from HydraForce and meter out. Having individual valves and coils for each end of the cylinder may be much less expensive and easier to tune than other suggestions. You will need to install bang-bang valves to provide the flow and pressure to the non metering end of the cylinder so you only have to meter out for speed and position.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[PNachtwey]

Another way to control this uneven area cylinder is to use cartridge proportional valves with different flow capabilities from HydraForce and meter out.

Will it still be a servo system able to precisely control position and or force?

 

[EdDanzer]

A proportional valve can be just as accurate as a servo but may not be as fast of operation. Any control function in fluid power ends up metering flow for position. The moving force is the maximum of the supply pressure but a system can hold more force than the supply pressure generates if pilot operated check valves are utilized. If a meter out system is utilized pilot operated check valve will not affect operation if sized properly.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[PNachtwey]

A proportional valve can be just as accurate as a servo but may not be as fast of operation.

Yes if you get a linear or servo grade spool, but you said CARTRIDGE proportional valve. So I did a search and found this

http://www.hydraforce.com/Proport/Pvd4w.htm

I didn't see anything suitable for closed loop control. Hydraforce makes no claims as to the linearity or response.
The valves aren't even that big.

So now that we have agreed that a proportional valve can be used somewhat interchangeably how does that help?

The moving force is the maximum of the supply pressure but a system can hold more force than the supply pressure generates if pilot operated check valves are utilized. If a meter out system is utilized pilot operated check valve will not affect operation if sized properly.

I didn't follow what you are trying to say except that I saw another key word and that is pilot. Counter balance valves have pilots too. When pilots open or close valves they change the way the system works during operation. This makes it hard for the controller to do a good job of controlling position or force. The person tuning the system will go nuts trying to figure out what is happening. This doesn't sound like a good servo system to me.

I would stick to the 2:1 asymmetrical spool idea. When designing hydraulic servo system remember that there can be only one active valve and that is the servo valve. As soon as freddyj said he was doing closed loop control the design rules change.

 

[EdDanzer]

Typically no servo motor accelerates linearly so why does a hydraulic valve have to be linear?
Freddyj has not stated the maximum speed, acceleration velocity, pressure or accuracy required of the system. Many presses can have a high speed low tolerance primary movement that is controlled by large flow on-off valves and slower speed higher accuracy finish movements. Many of the proportional valves are more linear as the flow decreases so the tuning of the control is easier if a high response speed is not required. If the system does require high flow (40+ GMP) the response of a large servo valve may not be any better than a small proportional valve at low flows.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[PNachtwey]

Typically no servo motor accelerates linearly so why does a hydraulic valve have to be linear?

The is a big difference between have a linear system and the non-linear ramps!

What a control person wants to see is that the actor moves at 10% of maximum speed with a 10% control output, 20% of maximum speed with a 20% control output. 30% of maximum speed with a 30% output etc.

Hydraulic controllers should never use linear ramps! They need to use s-curves or s-ramps. A linear ramp assumes the acceleration can instantly change which it can't. It takes flow to change the force and that takes time so s-curves are the way to go.

Freddyj has not stated the maximum speed, acceleration velocity, pressure or accuracy required of the system. Many presses can have a high speed low tolerance primary movement that is controlled by large flow on-off valves and slower speed higher accuracy finish movements.

That is possible but not with a simple PID controller. You are expecting a control person to figure out the gains at high speed and then at low speed. Most control people have a hard enough time just trying to figure out how to tune a PID. If you make life difficult for the control guy you will not make any friends.

Many of the proportional valves are more linear as the flow decreases so the tuning of the control is easier if a high response speed is not required.

Response times and linearity are two different issues. Freddyj is controlling a press. High response times are always required on a press. Fortunately most valves dither well enough to control pressure. The ability to accelerate and decelerate is limited by the response of the valve, the diameter of the cylinder and the mass of the load.

If the system does require high flow (40+ GMP) the response of a large servo valve may not be any better than a small proportional valve at low flows.

As you said, freddyj didn't state the flow or speed required but if he wants to close the press at 10 inches per second he will need a very big valve. Moog has some valves that can control this easily. The issue is whether there are 2:1 valves for big valves.

Why do you resist? What is wrong with my solution?
I have explained what is wrong with other non-linear solutions. If you want to remain friends with the control guys you will heed my advice.

I have many different press customers. Some are controlling four corner presses. In these cases the ability to synchronize the actuators within a few thousands of an inch is critical. In the past there were Metal Form shows. 1/3 to 1/2 of the hydraulic press manufactures there are our customers.

Hydromech wants to know what qualifies me to post on this forum. I would say that our company has been involved with more hydraulic presses than all of you combined. As soon as the words PID, closed loop or servo control are mentioned the design rules change. If you don't design the press to be controllable you will not be asked to design another press. Is that simple enough? Why do you resist?

 

[freddyj]

FYI, the solution I'm eventually going to try out as soon as the parts arrive is very similar to what EdDanzer is talking about. BTW, my speed is relatively slow, which is why I didn't really need a stand-alone motion controller and probably why tuning isn't that difficult. For now I'm satisfied with the velocity and acceleration characteristics.

I guess I wasn't absolutely clear on what type of press I'm using. Hydraulics are only used to fill the die cavity and remove the pressed part, not to press the part. We all know that there's more than on way to skin a cat or in my case a powdered metal slug.

PNatchwey, you're all over the place man, almost every hydraulics forum I come across. Question for you is, how do I tell if a valve has a 2:1 spool, namely proportional valves. In most of my findings this is not something that is explicitly stated under any specifications. Definitely not on the specs for the parker valve I'm using. It does mention that Qa=Qb and that valves are available with Qa>Qb and Qa<Qb where Q is flow, no specific ratio but I presume this is related to spool ratio. Just curious because I may have to head in that direction.

http://www.utronkinetics.com/technology.html

 

[maytag]

Contact your local Parker distributor and let them run the model code of the valve. The Parker website is not very user friendly. Rexroth and Eaton/Vickers are both easier,
at least for me to navigate.
Maytag

 

[PNachtwey]

PNatchwey, you're all over the place man, almost every hydraulics forum I come across.

More systems are going to closed loop control and that is what I do. Unfortunately there aren't many schools or hydraulic classes that that teach hydraulic designers how to design hydraulic servo systems so they are often uncontrollable. We sell controls for thousands of hydraulic actuator each year. Many of them are installed with a counter balance valve and we get the tech support calls when the system doesn't tune right. Hopefully those that read this thread will learn from your experience and my warning.

Question for you is, how do I tell if a valve has a 2:1 spool, namely proportional valves. In most of my findings this is not something that is explicitly stated under any specifications.

I can't tell by looking at the valve. I must see the specifications. The last valves specifications I saw for a 2:1 valve is on the Eaton/Vickers site. I just looked a few days ago at the start of the thread and couldn't find anything about 2:1 spools otherwise I would have posted a link to an example. I get frustrated trying to find things on the hydraulic web sites like many others.

A good distributor can help find the right valves.

Freddyj, I figured you had your solution. I just wasn't going to let the thread die without first making the point that the design rules change when doing servo controls. I hope the hydraulic designers now know that counter balance valves are active devices that ruin servo control and make life harder for the control guy. At one time I thought is was a plot by the hydraulic designers to make life miserable for the control guys but I soon realized it was just ignorance of what it takes to make a good servo hydraulic design. Some bigger companies have had me give training sessions. At first the hydraulic designer don't like my ideas because they are more expensive but they found that in the long run they saved the money because the systems tuned up more quickly. You can use auto tuning on a well designed servo hydraulic system but you can't auto tune an actuator that has a valves with dead band, dual gains or has a counter balance or other piloted valves that may open or close by them selves with out the knowledge of the controller.

A counter balance valve is kind of like a zener diode. You wouldn't put one in a linear circuit.
BTW, I really should have (control) after my name. I have EE degree but don't do any electrical design. I concentrate on hydraulic controls.

 

[kcj]

still on this topic, but sort fo related
fairly often I run across significant momentum or overruning loads where the valve must be remote from the load, i.e hose in between. No positon feedback, no real tolerances or accel or decel requireements. In the past we used soft shift or bang bang bangs with cylinder mounted CB just fine.

Now, I have several applications I'd like to go to prop valves, but need some sort of load blocking valve for hose break protection, and the valve cnnot be at the actuator.

A solenoid blocking valve won't know when hose is broken.
I have other servos with PO checks activated by the main P line before the valve. That is to lock the load upon shutdown when pump pressure is gone. Also won't work for hose break protection.

POCB would work for now as the applicaiton is not closed loop. Hwoever I can see adding position feedback later (+ or - a half inch, not in the thousandths of an inch range)
POCB I already new not the preferred choice, but what other ideas does anyone have? will never be a servo valve application.
What do boom trucks or aerials, etc applications use for protection?

kcj

 

[PNachtwey]

Use your pilot operated check valve for now. I don't think you will have any problems with the POCV as long as the pilot is connected to the supply pressure. In this case the check valve is always open and never shuts. The motion controller just can't handle the POCV opening and closing during normal operation.

If you read what I wrote about the blocking valves above you can see that blocking valves must be energized open and opening any one of a number of relays will cause the blocking valve to close. The motion controller can detect errors. It wouldn't take long ( milliseconds ) for a motion controller to detect that it can't control the actuator because a hose broke and pressure is gone. The motion controller can thengenerate a fault which causes the block valve to de-energize. There may be other methods that don't require a motion controller too like a simple pressure switch.

However, with a motion controller you should have smooth moves and shouldn't need to worry about the hose breaking. At least the hose will be stressed less. As I pointed out above, a good motion controller uses s-curves for the acceleration and deceleration ramps. S-curves can be achieved with smooth pressure changes. Linear ramps require and instantaneous step change in acceleration ( infinite jerk ). This can only be achieved if there is an instantaneous step change in force which requires an instantaneous step change in pressure which we know can't happen because it takes flow to change pressure and that occurs over time even if it is milliseconds.

 

[maytag]

Electric arc furnace regulation uses the PO check (actually a logic cartridge valve) and pilots it open so the Moog valve can control the movement of the electrode. A shuttle valve is connected to both supply and load induced pressures for the pilot controlling the PO check. Although we don't use a velocity fuse, I have seen these used on this application.
Maytag

 

[EdDanzer]

Peter,
Not every system can justify the expense and maintenance of a servo. Meter out position control will provide a way to use PO checks to stop over running loads, broken hoses, fittings or pipes and will reduce spring if the oil has air entrapment. If a load sensing pump is used it can reduce energy consumption and cooling requirements. With some work regeneration can be implemented to further reduce energy requirements.

Even though energy prices have gone down they will go up and then system efficiency will become a major priority. Servo valves are not the most energy efficient way to position hydraulics.

Making the motion control supplier or the person tuning the system job easy by increasing the cost of operations is poor business practice for any manufacture or distributer.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[PNachtwey]

peter,
Not every system can justify the expense and maintenance of a servo.

You really have missed the point. Some of the design techniques or considerations change when and if you ever do design a hydraulic servo system.

BTW, do you know that regeneration reduces the natural frequency of the system.

I have noted that that use servo valves isn't very 'green' but sometimes speed and precision are more important.

 

[kcj]

In these applications a motion controller/servo is out. They will be just joystick controlled prop valves for now at least. I am thinking ahead to simple plc control crudely closing the loop. Will never have the need or justification for a full stand alone controller.

Even with control, granted less stress on hoses, but still need some sort of safety backup.

The PO checks from the supply pressure work great, but those are for sensing pump shutdown (another application that does use a servo). If the pump is on, even with hose break,the load would move. In that case it doesn't matter so the load holding checks work great for shutdown time.

Granted it is not the recommended choice, but we may try 10:1 pilot motor POCB valves. I will get data acq on the cylinder pressures first to find load pressures and servo/prop pressures.

maytag, hows the steel mill business down there? pm me.

k

 

[maytag]

The steel business at the present time pretty much sucks as does a lot of other business. We are down to running 40-50% capacity. Worst downturn I've seen-I've been with Nucor since Jan. 92.
Kevin you have a pm on the IFPS board.
Maytag