Hydraulic Motion Control - When is a PLC not good enough?

Obviously it depends one what you are trying to do.

So what were you trying to do with the HACD and Control Logix? It must not have been a very demanding application. Did you program the Control Logix do support feed forwards? Did you change the gains automatically when the actuator went from extending to retracting and back? Did you even use the derivative gain?

There are no cut off points where one can say a Control Logix will not work and only a hydraulic controller will. Obviously I can make a Control Logix work better than most others in this application because of my knowledge of how to do these kinds of applications. The control logix has enough tools where one can do a lot that can't be done in many other PLCs. However, this would take time and still not be as good as what I can get off the shelf.

If you need to coordinate position pressure control you should look at the RMC150. This is the best you can get now for your next application. You can coordinate 4 actuators with position/force control in one box so there are no delays between axes caused by trying to control using multiple controllers.

A good hydraulic motion controller should have hydraulic application "knowledge" built in. This is what you are paying for. This is what makes it better than just using a servo motor controller or a PLC.

I have many Mathcad work sheets that show the effect of resolution, sample time, sample jitter, PI, PID vs PID2D control, with and without feed forwards and jerk feed forward etc. We wouldn't be concerned about those problems if they didn't make a difference in somebody's application. We must make a controller that can handle all hydraulic applications and you are only concerned about one so the hydraulic controllers may be over kill for your last application. You had better get a hydraulic motion controller for your next position/pressure application though.

I haven't even talked to the Rexroth folks (or anybody else, yet) but I'm not sure how one can coordinate the motion with 4 HACDs.

Click to expand...

I think you can do this with the new controller they have because its up to 4 axis. However the older one's are only 2 axis and I don't know how you would pass the position from one to the other. Also the last time I checked the new controller was still in Europe. No one in the states could tell me anything about it.

The application was to move an 18 inch bore cylinder at a specified velocity ranging from 15 to 0.1 inches per minute.

Click to expand...

I don't know your application but this is very slow in my world. That would make a difference in the control as well.

I guess another reason for my question is that our salesperson tells me he is going to talk to a potential customer about a project that will require co-ordinated position and pressure control of four cylinders. He wants to know if I think we can do it, and, if we can, what will be required...

Click to expand...

You could do this easy with a the RMC controller. Will you be doing position & pressure control?

agarb said:

I sure hope my original post didn't come off as sour grapes.

Click to expand...

No, I thought it is more of an off handed slam on the competitors.

One last question, does Delta have any vested interest in the 1746-QS, 1756-HYD02, or 1756-M02AS modules?

Click to expand...

Yes, Delta Computer Systems, Inc makes these controllers for Rockwell Automation. I wrote the code for the QS and we modified M02AE code to make the HYD02 and M02AS. That is why I had better know how they work. We make the modules and we make some profit. We don't do tech support for these products. However, I will answer questions on this forum. I have done so within the last two weeks.

If I can't sell and RMC I will try to push the Rockwell products because we still make money. When it comes to hydraulic motion control my opinions are biased.

For us, our biggest concern was achieving the low end of the required speed range as it equated to about 4 times our SSI encoder resolution per second.

Click to expand...

I did not see this part the first time. I may do some test tomorrow to see how my machine moves at low speeds. About 0.250" per second is about as slow as I go.

You could download RMCTools and connect to an RMC70 on the internet and simulate your project. The 70 is only 2 axis but it would give you an idea on how the gearing works.

Peter, do you have any 150's connected to the internet? I only found links for 70's & 100's.

Peter,

I really didn't mean for any of this to be a slam on the competitor's products. This is the first project I've done any closed loop hydraulic motion control and I enjoyed it. Your competitor's product worked well, and I'm sure their controllers can do much more than I required of them. There are a bunch of features I didn't even mess with. In the past two years, all but 1 of our power units have came from our Rexroth distributor, so it is natural that Rexroth controllers get pitched pretty hard to us. Not that we couldn't or wouldn't buy something else; just takes more effort and some justification to overcome the natural momentum.

If you are interested, I'd like to correspond more about our potential application privately. Or is this something that is better conducted through local distribution or through one of your regional sales offices? I have a card for Bruce K who came into to visit us a couple years ago.

Charles,

Good idea on connecting to an RMC via internet. I actually started to do this a couple days ago when I registed for their free design guide, but got side-tracked because I had to document some PanelView screens for a customer.

CharlesM said:

I did not see this part the first time. I may do some test tomorrow to see how my machine moves at low speeds. About 0.250" per second is about as slow as I go.

Click to expand...

This can be caused by low resolution feed back but SSI feed back should be more than good enough. It could also be caused by a slip stick action. The pressure builds up enough to over come the static friction and then the cylinder moves. When the cylinder moves the pressure drops so the force drops too. When it drops below the force for static friction the cylinder stops and the pressure must build again before the cylinder moves again. ANY closed loop controller that is just using a PID will aggravate the problem because when the cylinder does move, the controller adds more oil than is necessary. When the cylinder does move it will probably overshoot the target position or velocity so the controller will back off too much. The HACD card does have the ability to use pressure sensors on either end of the cylinder. The controller will look at the rate of increase in the pressure while the cylinder is "stuck" and output a negative control signal to compensate for the positive control signal caused by the follow error. This will keep the cylinder from over shooting to much or at all when the cylinder does break free. Likewise, when the cylinder does break free the rapid drop in pressure is also multiplied by a gain that will add to the control output. This will compensate for the negative output caused by over shooting the position or velocity. When these gains are in balance then the cylinder will move smoothly even at low speeds.

Active Damping is a must if this slip stick action occurs in open loop mode too. Just give the valve a small control signal like 0.1 volt and monitor the velocity on a graph. The motion should be smooth.

I know the HACD has this feature because the Bosch Rexroth sales guys like to make a big deal about it. This technique works extremely well but the technique is not good if one must follow a motion profile so it shouldn't be used of cylinders must be tightly synchronized.

I have movies that show the effects of active damping
ftp://ftp.deltacompsys.com/public/pneu/Preparing for IAAPA.MP4
This is a pneumatic system which is very difficult to control at a smooth rate. There are graphs that show how the pressure affects the control output in the same directory but they may not make sense without an explanation

Peter, do you have any 150's connected to the internet? I only found links for 70's & 100's.

Click to expand...

I will check into it. I think we do. I know I can make it happen. Brad can also do a quick tour just to get agarb started.

agarb said:

Peter,

I really didn't mean for any of this to be a slam on the competitor's products.

Click to expand...

I know you didn't mean it that way but I would at least have a graph showing the RMC could out perform the PLC by a significant margin even if it isn't necessary and embarrassed if it couldn't. Similar performance is not good enough but that is just the way I am.

In reality, I don't see the RMC competing with Control Logix but rather complementing it. The Control Logix should do what it does best which is to handle a lot of I/O. The RMC can off load processing the few time critical I/O and motion that must be handled in a synchronous and deterministic way.

If you are interested, I'd like to correspond more about our potential application privately.

Click to expand...

You should call Delta if you want a lot of information quickly and want to talk about the application.

Or is this something that is better conducted through local distribution or through one of your regional sales offices? I have a card for Bruce K who came into to visit us a couple years ago.

Click to expand...

I know what fourth of the country your are in. I still don't know who the distributor is. You can call Bruce too he has seen many applications since he worked for Moog for many years before coming to Delta but you need to call Delta to get the on-line demo.

Originally posted by Peter Nachtwey:

...but I would at least have a graph showing the RMC could out perform the PLC by a significant margin even if it isn't necessary and embarrassed if it couldn't.

Click to expand...

The thing I don't like is when this is taken in a vacuum. It should be intuitively obvious that a dedicated motion controller will perform motion operations better than a general purpose plc. But that doesn't help me as an end user on an application specific case when I am looking at my application. I have not dealt with Delta directly and based on responses this probably doesn't apply to them. But many vendors, in the search for raw market share, will start making blanket statements about the superiority and value of their device without considering the application requirements.

Now, I'm down with the whole caveat emptor thing. But this is why I VERY seldom take anything a supplier tells me at face value.

Keith

Take this for what it's worth but I've been told by the RR people that they have an appliation corrdinating 4 axis with HACD cards that are actually running independantly from each other but recieving the same target signal. All ramps are set the same in the HACDs and they control this specific application very well.

That said, I had a similar opportunity and didn't feel comfortable with doing this. I can't say if it worked or not becuase we seem to be in an eternal quote stage with them or waiting for other issues to settle before we can go back to this one.

RR has two controllers that are capable of multi axis coordination. One is called the MAC8. I have no experience but know it programs in C and that's enough to make me want to stay away from it. I'm very comfortable in C but it's not a generally accepted language on the factory floor and I don't want to be tied to a system like that. I believe the new controller someone spoke of is the 3rd generation HNC controller or HNC-3X as it would be called in Rexrothese.

If coordination in the HNC-3X is like previous versions, you can have a single master with slave axes or you can do an averaging type of synchronization. Again, I haven't used this version so I'm not positive.

If things are moving slow enough and your tolerance is open enough, you can use simple directional valves and just turn them on and off as each cylinder starts getting ahead of the rest. This isn't really motion control but for some applications it is all you need.

I did my little speed test on my machine with good results. I have a 5" cylinder with the RMC70 SSI feedback moveing at 0.001 inch/sec. Trends looked smooth I did not see any ruffness at all. I would also add that this is a quality cylinder with low friction seals.

CharlesM said:

I would also add that this is a quality cylinder with low friction seals.

Click to expand...

That is good. If you have lower quality cylinders then you can run the test again a look for the cylinder to slow down or stick at the same locations. This would show the rough spots in the cylinder.

My $.02

If I have an application where I simply need to roughly control the speed of a cylinder with a proportional valve then I will use a PLC. If closed loop position or velocity control is needed I buy a motion controller.

I am sure a lot of my applications could be done successfully with a PLC. I could drive a bus to work every day as well but I prefer my little sports car.

If you want to do motion in a PLC you need some hardware. You need either analog in or SSI for feedback and analog out for the valve. You might have spare analog in but SSI is ideal and chances are you won't have a spare SSI input laying around. You may or may not have spare analog out.

So if you choose the PLC route you need to spend money on hardware. If this is a new install you have already spent money on a cylinder, a rod, a valve, accuulators etc.... This money needs to be spent either way.

At this point the hardware cost of using either a motion controller of a PLC pretty much turns into a wash. What would you rather work with?

With a PLC you need to set up your own PID's. You either won't have or need to set up your own feed forwards. None of what you do in this area will be close in performance to what is canned in a motion controller. What about the scan time of your PLC? Is that going to be an issue? Do you have the time or knowledge to figure out what scan time is going to do to your application? Maybe when you initially install the hardware it works good but you expand the control system and all of a sudden your motion isn't as good as it used to be. PLC's are good at what they are designed to do, control I/O. In the controls world 10mS is generally plenty of time to control a process. 10mS in the motion world is an eternity.

If you go with a motion controller a lot of what would be unknowns with a PLC are well known with a motion controller. The scan time is fixed, it is never an issue. The software is designed for motion control.

Peter is not supposed to try and promote his product here because he is the president of the company. But I can because I am not affiliated with Delta or sell any product to anybody.

The RMC series of controllers is like motion control for dummies. The product is very powerful underneath but the interface and programming language is ridiculously easy to use. It is designed for motion control. I can program and tune a run of the mill hydraulic cylinder positioning system in an hour with a delta. I'm sure it would take me days with a PLC and the results would never be as good.

The last point is diagnostics and support. The easiset way to tune a system is by educated trial and error. With a Delta you make a move with the system and it will draw a graph for you showing what happened. You adjust the tuning paramaters and keep making the same move watching the graph to see if what you have changed is making things better or worse. You can't do that with a PLC (easily). And at the end of the day if you are not getting the results you desire you can email a graph to delta and they will look at it and tell you (or at least point you in the right direction) as to what could be wrong. Maybe your valve is to big, your cylinder is too small, you have inadequate presure, the mechanics of your system is too loose. With this type of control there are a litany of things that can make your motion not work properly. What is the guy at tech support for a PLC manufacturer going to tell you when your hydraulic motion system isn't performing well enough?

If any of you actuall read all the way to the bottom, congratulations this is my longest post yet. And to be clear I am not an exper in hydraulic motion contol, these are just my honest opinions.