Hydraulic lathes?

Sometime recently I read about someone who acquired an older milling machine that had hydraulic feed on the table. I wondered about how these tables might be controlled and if any lathes have been designed using hydraulics to move the carriage or crossfeed instead of lead or feedscrews. (After all, there are mechanical, hydraulic and pneumatic linear actuators.) I don't know if you could control hydraulics as precisely as feedscrews (with respect to headstock spindle rotation) for cutting threads, but you'd never have to worry about worn screws and metric/ inch conversions. I imagine that the big problem is in designing the feedback system and getting it to respond well. I've seen descriptions of hydraulic systems made by companies like Enerpac that control the balancing of bridge sections to keep them level while being positioned with cranes. I know that hydraulics can be controlled with electronics and proportional or servo valves. Probably it's difficult to beat mechanical feed and leadscrews because it's a simple and accurate system, but I was just curious.

Reply to
Denis G.
Loading thread data ...

We have an older NC lathe, "Sheldon Tape Lathe" that uses Moog proportioning valves.

IIRC The valve needs a 0-20 mV signal to operate. Our resident electronical guru made us a pair of boards that turns +/-10V servo control signal into the signal that makes the Moog valve happy.

We retrofitted using a galil dmc740 motion controller.

This works great for really big iron.

Unless you NEED it, I'd try to avoid hydraulic servos.

Reply to
Jon

Hydraulic duplicating (tracer) mills and lathes were once common. Last I knew a customer of mine was still using a tracer mill to duplicate molds for shoe counters for which there was no CAD data available. Google True-Trace or Mimik and you may find some info.

Sinker EDM rams were often servo-hydraulic, but, I think, not so much for coordinated motion as for smooth motion at very slow speeds

All the large planers I've seen were hydraulically driven, though that's more for brute force rather than controlled motion.

Reply to
Ned Simmons

Small point, but the primary reason was that hydraulics could handle the quick reversals -- often several times per second -- involved in the EDM servo motion, and do it with relative simplicity. They just used a voltage-sensitive valve switch. DC servos didn't have enough "first-pulse" torque to do it until some advances came along in servo drive controllers. The first ballscrew-driven servomechanisms for EDMs that were successful used stepper motors.

Production milling at one time was mostly a pass-through operation, a lot like planers. Before CNC, production parts were, wherever possible, designed so that you didn't have to precisely control the start and stop positions of the axis traverse.

-- Ed Huntress

Reply to
Ed Huntress

This is something we do all the time and is very common on spinning machines because of the high forces required.

We can hold tolerances as well as if not better than a leadscrew because of no mechanical influences.

formatting link
Our website is
formatting link

If you need any further info let me know

Daveb

Reply to
DaveB

Over 30 years ago I ran Sheldon lathes that were hydraulicaly controlled. We could hold .0001" easily with these lathes. They were

15 inch swing machines, I think. They had an adjustment called "dither" which kept the valves always in motion. This was to avoid the problem with "stiction". I don't remember if it was the valves, the actuators, or both that needed the dithering. Interestingly, I have two older CNC lathes with Fanuc controls that also use dithering. The servos always move a little. The repair man had to adjust one of my lathes when the dithering was out of whack, it would lose precise position and the finish would suffer. He told me the dither was left over from hydraulic systems. My newer CNC controls no longer have a dither adjustment. ERS
Reply to
etpm

"Denis G." wrote in news:a4003630-b489-4b03-bd1e- snipped-for-privacy@79g2000hsk.googlegroups.com:

Hydraulic feed is an _old_old_ technology. Very inefficient also. Reliable control can also be a significant issue.

Reply to
Anthony

We do it all the time with applications that make leadscrews impossible.

The last five machines we built have had zero follow-up calls.

This must be an area that you are unfamiliar with. Daveb

Reply to
DaveB

I seen a special on Pro Baseball bat manufacture. Each player provides his favorite blank. They are numbered and stored. A hydraulic tracer lathe turns the bats with the master in the tracer.

Karl

Reply to
Karl Townsend

I don't think that a hydraulic system could retain anything approaching the rigidity that you get from lead screws, so I don't see it as being a good candidate. I think you'd get such a springy feed that you'd be constantly bouncing off of hard cuts, then digging in too far when the cuts got light.

Notice that the cited examples (the mentioned tracer machines, the film of the metal spinning, old old mill with hydraulic feed) were all things where some fixture provided the rigidity, or where the precision of the feed wasn't critical.

I'll believe it if I see it, and you can pay me my going rate to do a feasibility study if you want, but it's not something that I'd recommend off the cuff.

Reply to
Tim Wescott

We have built machines that free spin.......no tooling. This is how most cng cylinders are made to eliminate welding.

The last one we built was 75 hp (spindle) and we were forming .375 material cold.

example:

formatting link
On our website the pnc 75 shown also has a turning holders as they do

2nd operation work.

There are plenty of turret type machines that do both turning and spinning.

formatting link
Daveb

Reply to
DaveB

BTW Tim, if you can afford my rates I may show you how its done. If your at ITMS would like to show you what we have (for free) Daveb

Reply to
DaveB

On the tracers the hydraulic fluid transmits the cutting force to the tool. I don't follow what you mean by "some fixture provided the rigidity." A properly sized leadscrew is certainly stiffer than a column of hydraulic fluid, but the stiffness of the rest of the machine structure is a bigger factor than the stiffness of either a leadscrew or a hydraulic actuator of reasonable length acting in compression.

Eric attested to the feasibility by virtue of the fact that the machines he ran were able to hold tenths in production.

Reply to
Ned Simmons

I missed that post -- hmm. Having read it I'll count it as 'seeing', though.

I know that the hydraulic fluid itself isn't very springy, but even with all-solid lines you'd expect there to be spring as a consequence of the valving. I'd be interested in seeing just how the hydro-mechanical system is put together so that it's errors are correctable by the servo system.

Reply to
Tim Wescott

What sorts of precision do you achieve? Is there anything special done in the hydraulic system to make its action more positive, or is just long rubber hoses and spool valves?

I'm glad I put weasel-words in my assertions, it looks like I'd have to stand corrected otherwise.

Reply to
Tim Wescott

Tracers are a bit different technology than servo control. We used two true-trace tracer units in production for years.

I have one available for sale, by the way.

Reply to
Jon

Oops, I missed a detail in the original post, the sheldon tape lathes use hydraulic motors to drive ballscrews. However, as later posts have described, technology marches on, and high precision is feasable using cylinders now. But still, it usually is saved for applications requiring high forces

Reply to
Jon

One disadvantage of replacing the leadscrew on a lathe with a hydraulic cylinder would be that the length of the lathe would probably need to be doubled (close). I tried to find details of how the Sheldon tape lathes were made, but was unsucessful. Finding details about how such machines are constructed is difficult, but I was happy to find out that they exist.

As far as the criticism of stiffness between mechanical and hydraulic systems, I believe that one only has to compare mechanical and hydraulic presses. I think it's a matter of proper design to get what you need.

Finally, I found a 2006 description of a CNC machine made by a German company (Schutte) using electrohydraulic axes with a bed resolution of

0.1 microns (1 micron accuracy at the tool).

Here's a link:

formatting link
Here are excerpts:

... In 1996, Sch=FCtte made the decision to go with electro-hydraulic axis drives for all end-slide and cross-slide actuation, replacing cams. Speed, power and accuracy were the motives for this decision and fit into the company=92s multi-spindle philosophy.

As the name implies, electrohydraulic drives use a combination of hydraulic fluid and electronic control to move an axis. Hydraulic fluid provides the motive force, and electronics, in the form of valve control and linear feedback, provide precision control of the motion.....

A key to successful use of electro-hydraulic drives is precise control of the valve that allows hydraulic fluid into and out of the cylinder. It controls the force, speed and distance traveled.

The valve, in turn, is regulated by a processor. The valve and processor system was created in conjunction with Bosch to give finite measurement of fluid movement in the shortest period of time. In addition to the hydraulics and electronic processor, each slide is equipped with a linear scale to give actual position feedback to the processor. The digital feedback loop operates at 800 Hz, feeding back the slide position 800 times per second.....

=93Multi-spindle applications have always been associated with high cutting forces,=94 Jim says. =93Form tools and drilling operations are two prime examples of where feed power and cycle time intersect. The idea is to rough out the part as fast as possible and then bring in the semi-finishing and finishing operations. We believe the machine must be capable of power for the roughing and finesse, such as single-point turning for finishing. Electrohydraulic actuation gives us both.=94

Also, the system is capable of delivering the close-tolerance specifications necessary for shops to be competitive. =93Our electrohydraulic axes have a resolution of 0.1 micron. That resolution delivers a usable repeatability of 1 micron at the tool,=94 Jim explains. =93That tolerance is necessary because of the changing nature of the work being run across the multi-spindle. Close-tolerance machining capability, along with processing flexibility, is in demand to eliminate the need for secondary operations. Customers want to drop parts off the machine complete, and that ability starts with accuracy.=94

Reply to
Denis G.

Tim, some machines we build have stainless tubing from the hydraulic power supply and some use steel under and over braid hose.

The servo loop usally consist of just a standard cnc control +-10 volt error signal and the position loop is normally 20 micron scales (quadrature) with a zero reference mark.

We use moog servo valves and servo amps

In some cases we use a frequency to voltage converter driven by the scale feedback and use this as velocity feedback.and of sum this with the speed reference.

Some machines have rotary hydraulic motors with encoder feedback and ballscrews.

Only downside of hydraulic loops is rapid traverse speeds, we try to run 400 ipm in rapid and can repeat to .0001 all day long, due to using scales.

If you walked in and watched a machine run you would be unable to tell any difference.

Daveb

Reply to
DaveB

snipped-for-privacy@nt.org (DaveB) wrote in news: snipped-for-privacy@news4.dslextreme.com:

Dave, We used to have a floor full of hydraulic fed machines. They work, but can be troublesome. We still use hydraulics for one axis on one certain CNC machine type (It is a CNC axis - X axis specifically). Works great, adjustable in increments of 0.0002 microns _diameter_. Very stable. But....when it has issues...it has serious issues and downtime and tuning can be a lengthy and tedious process.

Reply to
Anthony

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.