Controlling two analog valves for force control

Vorian

Member
Join Date
Jul 2012
Location
Athens
Posts
12
Hi, this is my first PLC project and I need some help.


The goal is to apply steady force on a random surface and for that I control a piston on the z-axis using two analogue valves, one pushing from above and one from underneath.
Can anybody suggest any strategies? I tried using PID control on the upper valve and a steady pressure on the one underneath but it does not work very well.

Do I need to control both with a PID control? Or two simultaneous PID controllers working together? Any suggestions will be appreciated.


Thank you for your time and excuse my use of English.
 
Are you using one hydraulic cylinder and two proportional (analog) valves? (one for each port)
Or, more likely, two hydraulic cylinders, each controlled by a proportional valve, operating on the same axis but in opposite directions.
Is the "random surface" gravitationally locked?
 
Last edited:
Can you supply more details. What are you using for feedback. Is this a hydraulic or pneumatic application. What processor are you using. I did a hydraulic project this summer doing metal forming using hydraulic pressure control what worked quite well.
 
I'm a little confused. If the two cylinders are pushing against each other by definition they will each exert equal force. This is Newton's first law. Otherwise there will be movement.
 
Sorry for the wait...I was away from pc for a few days


Details:

a)It's a pneumatic application
b)I am using a pressure sensor
c)I use two analog valves to move a "piston" (don't know the exact translation in English) up and down.
d)the random surface is gravitationally locked.
 
let me start off by saying that i worked for an oem for 11 years and we did several projects like this.

there are several issues that you must deal with.
1. how long do you need to clamp? seconds, minutes, hours? if for a few seconds, a pid would not work, its to slow. you would have to reverse engineer (for lack of a better term) how the pid valve works and create 2 math statements in the plc. #1 for the pressure to the valve, and #2 to determine the pressure applied.

2. in my opinion, you want to apply equal pressure on each side of the object. the best way to accomplish this is with only one valve. from the valve, use a tee and pipe to each cylinder. the air lines need to be the same length.

even though you are applying the same pressure to each cylinder, the force will not be the same. the math is the same for each cylinder, but in the real world, each air cylinder is different, forget the math, their different. the reason is that air cylinder seals create friction in the cylinder walls.
this friction (sticktion is the term on air cylinders), the seals are slightly different in size, the cylinder walls are slightly different in size. all this will add up to 5 to 25 lbs of force lossed. Many will laugh, but we had to special order air cylinders on several jobs because of this. pain the air cylinder to test the loss of force on several cylinders so we could get the ones with the lowest sticktion. cost us $500.


3. you need to describe your mechanical process better. i do not understand how you wish to hold the part.

regards,
james
 
The way I interpret this is that there is one cylinder with a proportional valve on either end of the cylinder.
This is an extremely difficult way to control the net force for clamping a object. It would be easier to put about half the system pressure on the rod side and then control the position and force by controlling the valve on the cap side of the piston.
 
The way I interpret this is that there is one cylinder with a proportional valve on either end of the cylinder.
This is an extremely difficult way to control the net force for clamping a object. It would be easier to put about half the system pressure on the rod side and then control the position and force by controlling the valve on the cap side of the piston.


Cap side you mean the upper side or the side tha faces the surface?
This is the installation...I move the random surface below by hand and the piston is supposed to track it with a steady force i choose. I am supposed to do it with two analog valves.
47560538.png
 
Another project from hell.

Do you have a pressure sensors for the top and bottom of the cylinder? What you really need to do is control the net force. That is the pressure on top x area of the top of the piston - the pressure on the bottom x the area of the bottom of the piston.

Vorian, I am serious, this is extremely difficult to do. I do this kind of stuff for a living and if this is some project you have from a pointy haired boss or stupid professor then I think you need to have a talk and ask them if they have figured it out because I can ask question about how to solve this problem that they probably haven't even thought of and yet they just dumped this problem on you.

As it turns out I have a similar project where I must maintain force on an uneven runway. The advantage I have is that there is a laser that is measuring the elevation of the runway before the wheels get to that spot and I can 'anticipate' whether the wheels will need to go up or down instead of just react to the elevation changes like you must do. Pneumatics are not the best way to control pressure if the pressures or positions are always changing. I am using hydraulics and one servo valve.

When you did your PID test with the pressure on bottom side being constant, how constant were you able to keep the pressure? I bet you couldn't keep it that constant because the volume on the bottom side is always changing therefore the air is expanding and contracting.

The project is not a good one for a PLC.

There must be a 'pointy haired' boss involved.

If I were doing this problem I would get a small DC motor with a roller on the tip of the linear actuator and simply provide a constant current to the armature so the motor would apply a constant torque or force.
 
I have one pressure sensor at the top. The experiments i did by controlling the top valve and keeping steady pressure on the bottom one worked relativley well (after a lot of tuning), though the error is huge at some points.

Thankfully my professor does not require a perfect result since the equipment isn't top notch anyway, so as long as I can track the surface with relatively steady force I am good.
I am going to try cascade control next and see what happens.

PS: He does have pointy hair though
 
Am I missing Something? (Probably). Assuming that the friction of the seals is < maximum error (use a rolling diaphragm cylinder) then why not use a standard E/P regulator such as the SMC ITV2050; it is a closed loop pressure regulator with very fast response.

Nick
 
Am I missing Something? (Probably). Assuming that the friction of the seals is < maximum error (use a rolling diaphragm cylinder) then why not use a standard E/P regulator such as the SMC ITV2050; it is a closed loop pressure regulator with very fast response.

Nick

I am thinking along the same lines. Seems like a quickly relieving pressure regulator fits the bill. Not certain what specifications are trying to be achieved. If an air reservoir is used between the regulator and the cylinder where the volume in the cylinder is small compared to the reservoir and the restrictions are minimized is should track very well at a constant force without any control necessary.
 
The point of the project is to see IF it can be done using PID control of two analog valves. As long as I can track it sufficiently I am good. I didn't choose the equipment so I can't use those suggestions, even though they would work easily.

My professor is generally interested in pneumatic systems, position cotrol, force control etc so I guess that's why he gave me this project.
 
Lets back up and again.

What controller are you using?
What are the valves are you using? (saying "analog valves" narrows it down to only a few million options)
What is the source pressure?
What is your setpoint force range?
What pressure sensor are you using?
What analog modules are you using and what is their resolution and conversion times.
How long are the lines between the valves and the actuator and what size are they?
What is the bore and stroke of the actuator?
What is physically pressing at the tip of the actuator? Is it a ball tip? If you are using a cheap actuator with bushings taking the side load as opposed to bearings you can loose a lot of your linear force by translating it to side load on the shaft.
Is pressure your only feedback?
The picture is too small for my beady little eyes to make any of this out.
 

Similar Topics

Hi everyone i have a lenze VFD D-31855 it uses a potentiometer to control the speed of the motor 1-is how the potentiometer is connected to...
Replies
4
Views
1,542
Hi How i can program Analog input 4-20mA to control digital output module with setpoints? I need to create control circuit example for heater...
Replies
4
Views
3,193
Hello men Hardware: 1769-L32E (Processor) 1769-IF4XOF2 (Low resolution Analog Input (4 point)/Output (2 point)card) VFD: PowerFlex 40 -...
Replies
14
Views
10,289
Hello I have a -10V to +10 V signal from a device which I would to control a VFD with. The problem is that I need it to work like this. -10V...
Replies
10
Views
5,866
I have to take a slc 5/03 and use an analog out card ....number is 1746-n04v. To speed up or slow down a motor for a scrap belt. I know I can...
Replies
16
Views
6,664
Back
Top Bottom