Synchronisation of Series of Hydraulic Cylinders

connect them in parallel?

IMHO, if you want all four to extend by the same amount synchronously, the only reliable way would be to have a proportional valve and feedback for each, and have a controller for each, all fed with the same input setpoint signal.

With no individual control, the minor differences in force between them will inevitably mean that, under load, the 'strongest' will extend first, then the next strongest & so on.

Ditto.

In case of a similar application (Hydrostatic baring with multiple pads) what we do it, as in post #1, we connect in parallel.

Our use is a multi port positive displacement pump, (equal volume for each port for our case and will be for you too assuming your cylinders are same size).

The another way is using a flow divider from tha main pump (gear type flow dividers are there with equal/ unequal division of flow) that case you ensure the set quantity reaching each cylinder irrespective of the load or other type of restriction.

Only all these being positive displacement, there can be no adjustments, ie you can not control the speed of the cylinders individually later or time lags either. If properly calculated the flow division (easy if all are equal, a basic mathematics of the cylinder crossection and strokes if unequal) these can be forced to work simultaneously.

The combined speed of course can be adjusted by varying the speed of the flow divider, but can not be individually (since orifices etc will be almost useless in these)

I'd say all comments here are strictly correct. You should choose from an individual control with feedback or a flow divider with flow control for each cylinder, depending on how much is your "act at the same time".

Some amount of difference between each other will always occur, no matter the control system chosen.

I'd say that, if it's a really critical synchronization, you may even choose to add a linking bar between them, as a fixture with guides and all those things, to assure the correct movement. Of course, I'm speaking without knowing your requisites, and size and weight constraints.

To avoid any misunderstanding could you please make a rough sketch?

You should consider that although mentioned in a thread fluids ARE compressible so that it can be a reason for not synchronisation. But if the air was not purged correctly there could be compliant volumes to increase the stroke differences.

The flow divider is a rough solution since flows are different, slightly, but different and if your requirements are high precision it will not work properly. An other aspect is the leakage of different parallel pumps which will as well lead to different flows depending on load. Cylinders have different dimensions and the displacements are for same transferred volume proportional to piston area.

What you can do is NOT to connect them in parallel since load difference will avoid synchronism but use one of them as master and the other 3 as slaves either with a mechanical controlled valve which is sensitive to the stroke difference or do it electrically with stroke transducers and proportional valves.

Consider please that valves control flow and stroke is the integral of flow/area so that you have, using correct devices, a zero follow error.

With your sketch it will be possible to go deeper in the analysis and counsel.

There is a valve named "Flow Divider" for this purpose

http://www.hydraulicspneumatics.com/200/TechZone/HydraulicValves/Article/True/6410/TechZone-HydraulicValves

The idea came from raising grain silo vertical structures level on the farm with 8 hydraulic cylinders connected to one pump.... if one was not proportional, the tall silo would tilt and fall over when they were raising it for installation on tall permanent legs.

I use flow dividers to control exactly the same hydraulic oil to two separate devices in the case one device fails (seal or broken hose) and bleeds all that oil of that circuit, the other circuit isn't robbed of its oil because the flow divider still gives that remaining good circuit its share of oil.

Hope this helps you ....

The link contains a VERY important information which says that such devices CANNOT be used for precise flow controls.

As long as you control synchronism by the flow it is possible to have a limited error which depends on the loads and on the time the stroke requires since the cylinders INTEGRATE the flow and slight differences can give big stroke differences.

This was and remains the reason I recommended a STROKE/DISPLACEMENT difference as input for the control loop.

It is of course possible to use flow regulators (2 ways) and connect their metering orifices in balance with the displacements differences for an automatic compensation.

This is an odd sounding solution but in some cases (depending on the application) I think it would work.

You connect the two lines from the valve this way: one line to cylinder 1 top side and the other line to cylinder 4 bottom side.

The bottom side of cylinder 1 connect to top side of cylinder 2. Bottom side of cylinder 2 to top side of cylinder 3. Bottom side of cylinder 3 to top side of cylinder 4.

The "top side", "bottom side" designation is just a way to differentiate between the two ports on the cylinder. The cylinders would all need to be "rod both ends" type to maintain proper flow volumes between the cylinders.

A leaks impact on cylinder position could be a problem.

If I have time later I will provide a sketch and and do a little more thinking about it.

Dave,

Could you please give me the sketch?

Look at the flow divider.

https://www.jahns-hydraulik.de/pdf/kat-mto-05-e.pdf

This type of flow divider (driven by the positive displacement pumps) has a high control of volume flow and is in general used for the synchronization purposes.

i had a similar problem once. Turns out the answer was mechanical. i.e. A large diameter steel bar mounted in bearings. The bar had arms welded onto the ends. The arms were connected to the ends of the rams. Thereby one ram could not go up without taking the other along for the ride. Sorta like a bucks party involving handcuffs.