Long Cylinder Synchronization

I’m working on an application that requires the synchronization of (2) long cylinders. Each has approximately 35’ of stroke. They are connected to the ends of a gate that is approx. 20’ wide and moves up and down. The design currently calls for a bang-bang valve to control direction and a separate proportional throttle valve that will bleed off flow between the valve and cylinder. Each cylinder has it’s own dedicated directional and throttle valve. The motion controller would accept position feedback from each cylinder and use it to control how much flow is bled off. The cylinders are about 12" in diameter and max speed is about 1 ft/sec.

[] Given the long cylinder stroke, I’m interested in hearing some recommendations on the type of feedback device that would work best.[/]
[] If the control valves are mounted on the cap end of the cylinder they will still be 35’ from the rod end. Do you have any recommendations with regard to valve placement?[/]
[] Does the bleed-off style system make for a good design, or would a proportional directional valve be any better?[/]
[] Any general wisdom?[/]

I’ll greatly appreciate any feedback you got.


We have control cylinders as long as 30 ft using 30 ft Temposonic rods but I bet these are expensive. If the system is vertical you should be able to get by with a string pot system. I prefer string pots with SSI outputs so the output is absolute instead of incremental. Incremental encoders is not a good idea if you need to keep two axes synchronized. If all else fails you can use an SSI with analog out put that is a last resort because of the noise and the resolution.

Something that many don’t consider is to use a rack and pinion to synchronize the two actuators mechanically and a SSI absolute encoder can be attached to the pinion to get position.

On the cap end is as good as any. Especially if the cap end is bearing most of the load and pushing the gate up. If the gate is heavy enough to push the actuators down you may only need to control oil to the cap side and let the rod side fill and discharge to a tank at atmospheric pressure. Unfortunately I don’t know of any good quality 3 port proportional valves.

It would work but it there are hassles dealing with directional valves. The directional valves can’t switch quickly enough for good position control when stopped but you probably are not that concerned. The big hassle is that you must always be making sure the directional valves are open correctly to go the direction you want and they will open and close when at the set point. The RMC can be programmed to open or close the directional valves(s) based on the error or control signal. It is yet another thing you have to do and all the extra things add up. I prefer simple . It is easier to control one proportional valve per cylinder. Also I would have blocking valves between the proportional valve and the cylinder that must be energized open. If power goes off or any error occurs the blocking valve are de-energized and shut so the actuators will stop moving. Many hydraulic designers want to use counter balance valves. Avoid them as they make the system much harder to control.

We retrofitted a 1000 ton break press that had a bleed off system with digital controls and it was a bear. It had 36" bore cylinders at each side of a 20 ft. long break ram.

Besides the control of the directional valves as Peter mentioned, everything is backward in your thinking. Bigger command signals to the proportional bleed off valves slow the cylinder down and vice versa.

However, I can understand why the original designers did it. They sent the majority of the flow (around 1000 gpm per cylinder ) through logic elements and just had to control enough of the bleed off the keep the cylinders at either end of the break ram in sync. Back when that press was designed, there was no standard device to control that much flow proportionally. It is still expensive to do so.

1 ft/sec. on 12" diameter is about 350 gpm. A lot but still doable in a size 32 valve