Using equally sized single acting hydraulic cylinders to transfer motion (i.e. master and slave)

[ddowns46]

I'm working on tool that incorporates a machining slide that is positioned by a hydraulic cylinder. This hydraulic cylinder will be actuated by an identical hydraulic cylinder driven by a crankshaft ~20 feet away. If the master cylinder being run by the crankshaft compresses 1" per revolution, then the slave cylinder will cause the slide to move 1". There will be a return spring on the machining slide to ensure that the column of fluid always remains compressed. Therefore the system converts the rotary motion of the crankshaft to a reciprocating motion at the machining slide. Positioning accuracy is not a huge concern because i can adjust the 'run out' of the crankshaft to provide the desired motion of the tool slide, so it only needs to be close to 1-to-1 motion transfer within a few percent.

How does one go about filling and bleeding this system? Is this method common in industry? Are there any particular issues or risks with this approach?

thanks,
Danny

 

[MikeHalloran]

Move the crankshaft or substitute a 20' steel bar.
You will never get all the air out of the oil and the cavities within the hydraulic system, and therefore it will not be stiff enough to machine against.






Mike Halloran
Pembroke Pines, FL, USA

 

[ddowns46]

My thought is that if the return spring on the slide is strong enough, it will help to preload any air bubbles in the master and slave cylinders and in the hydraulic ram. I'm currently looking at approximately 800 pound return spring. With 2" hydraulic cylinders, this will create 250psi of static pressure. A lathe tool will be mounted to the slide and the depth of cut per pass will be approximately .030" on low carbon steel so the cutting forces will be relatively low.

 

[MikeHalloran]

Okay, don't believe me.

Find a way to prototype it for cheap before designing it into something important.

Don't stand nearby when it's running, and don't try it at all with positive top rake or an interrupted cut.


Mike Halloran
Pembroke Pines, FL, USA

 

[ddowns46]

Somewhere in the connection between the hydraulic cylinders i am going to install a tee. Connected to the tee will be a hand pump. I'll use the handpump to fill the hose and both hydraulic cylinders. I'll drill and tap the lower plate of each cylinder for a bleed screw and make sure that each bleed screw is at the high point when filling the system. In effect the handpump is analogous to a master cylinder in a car and the two hydraulic cylinders are like drum brake slave cylinders. I'll bleed the system just like a car's brakes.

I'll let you know how it works.

 

[3DDave]

I too wonder about the elasticity of the system. The fluid is compressible and the hydraulic lines will expand. Temperature changes in the fluid will also cause different output from input. If the travel is small and you use a large diameter piston then the pressures will be low, decreasing the compressibility and line expansion problems.

You can use a vaccuum pump to remove the air from the system and then backfill under the drive of atmospheric pressure with hydraulic fluid.

Hydraulics are good for transmitting force; not as good for transmitting position. The main problem is stiffness; cutting tends to notice small changes in engagement. A few tenths of a thousandth will represent a 10-20% change in cutting depth.

It's interesting to try, but avail yourself of the back of an envelope before opening the wallet.

How is the crank synchronized with the lathe?

 

[ddowns46]

The positioning capability is not important for this process. The only thing that is it important is that the tool slide moves 1-to-1 with the crank pin. In reality that isn't even important because i can easily increase or decrease the runout of the crank in order to induce the correct amplitude of reciprocating motion at the tool slide. Additionally, the machinist will only need to machine to cleanup and isn't concerned with hitting a diameter so thermal swelling of hydraulic fluid is not a concern. The machinist will simply keep an indicator on the slide to make sure that the reciprocating amplitude is correct and will keep taking passes until the entire OD cleans up.

The crank is actually mounted to the end of a very large and very heavy workpiece. If you haven't guessed yet, this tool will be used to eccentrically machine a feature of a part that is running concentrically. These parts are much too large to put between 4 jaw chucks for eccentric machining. The centerline of the OD that this tool produces will be offset from the part centerline anywhere between .050" and .150" for a diameter that is roughly 140".

As far as ridigity is concerned, as stated the the tool slide and the column of fluid will have 800 pounds of preload on it. I would bet that the radial cutting force (by far the lowest of the three cutting forces) will be less than 100 pounds for this cut. So we're talking about a spring that has an extremely high K-constant with already 800 pounds acting against it. Adding an extra 100 pounds is going to be negligible in terms of the displacement of the additional displacement of the spring.

The machining slide with the hydraulic linear actuator shown on the cover page is what i plan on using:

http://www.wainbee.com/suppliers/gilman/Gilman_slides_dovetail.pdf

My guess is that if hydraulics were not rigid enough for machining, then this slide wouldnt be available off the shelf.

 

[3DDave]

You know best.

 

[MikeHalloran]

>>> My guess is that if hydraulics were not rigid enough for machining, then this slide wouldnt be available off the shelf.<<<

You may not have guessed that hydraulic slides are typically equipped with linear position sensors, servovalves, and servo controllers or something equivalent.


Mike Halloran
Pembroke Pines, FL, USA

 

[PNachtwey]

I'm working on tool that incorporates a machining slide that is positioned by a hydraulic cylinder. This hydraulic cylinder will be actuated by an identical hydraulic cylinder driven by a crankshaft ~20 feet away.

I would listen to those that have doubts about the elasticity problem.
Why not control the slave cylinder directly?
Put a servo valve directly on the slave cylinder so the natural frequency is higher and the volume of compressed oil is lower.
What is the master cylinder responding to? If it is a contact system then it should be easy to gear to it electronically. The shaped being monitored would not need to exert much force on any feeler or gaging system. If you know the formula for the shape then just enter it into the hydraulic motion controller.
Do the gearing electronically. You will not have the air problems, temperature problems mentioned above.

You may not have guessed that hydraulic slides are typically equipped with linear position sensors, servovalves, and servo controllers or something equivalent.

ddowns46, heed what Mike just said.
ddowns46's machine sounds like something designed before position sensors, servo valves and servo controllers were invented.




Peter Nachtwey
Delta Computer Systems

http://www.deltamotion.com

 

[imcjoek]

Bog standard screw slide with big ol' stepper motor attached?
Why are you reinventing the CNC machine with Rube Goldberg's help?

 

[hydroman247]

A lot of dislike for a master/slave set up in here. Sure a servo operated system would be nice and more accurate but as the OP has pointed out several times, accuracy is not the most important thing in this application.

Master slave will work, just fit some PRV's and bleed points in a couple of places and you should be good to go.

 

[frrst245]

Look to automotive hydraulic brake components for inspiration.

Why not have some sort of cam driven off of the lathe head itself?

 

[berkshire]

This sounds like a home made " TRU-Trace " machine, those things went out of favor when CNC got less expensive.
B.E.

You are judged not by what you know, but by what you can do.