Re: [CAD_CAM_EDM_DRO] CNCing a big surface grinder ?

Hi All:
On a surface grinder, there's not a lot of value in adding a servo to
the longitudinal axis, unless the machine is rigid enough to handle creep
feed grinding.
A long stroke hydraulic cylinder is all that's needed on the X axis, since
the grinding stroke has to overtravel the ends of the work anyway.
It's a simple and robust system with few moving parts, which is why it's so
widely used in this application.
The Y axis and the Z axis can profitably be controlled with steppers or
servos, but again, there's not a lot of value in creating super motion
control.
About the only things you'll do with it are surfacing...Y axis incremental
stepping with occasional Z axis moves, plunge grinding which is Z axis
only, or side wheeling which is usually Y axis only.
My experience with surface grinding (injection molds mostly) is that CNC
would be of no use to me whatsoever.
Most of my complex tasks involve wheel dressing, or sine plate setups or
both.
The principal value of a surface grinder is in the creation of precision
surfaces and angular relationships with good repeatability.
The other big advantage is the ability to process hardened materials.
If you have this kind of need in a production setting, there may be value in
the CNC modification for you.
If you're doing toolroom type work, it's not likely.
Cheers

Marcus

----- Original Message -----
From: "Kim Lux" <lux@...>
To: <CAD_CAM_EDM_DRO@yahoogroups.com>
Sent: Friday, November 07, 2003 10:35 AM
Subject: Re: [CAD_CAM_EDM_DRO] CNCing a big surface grinder ?


> So do you think that it would be feasible to replace the hydraulic drive
> system with a high speed servo system ? This is quite interesting
> because most of these units even today are hydraulic drive...
>
>
> On Fri, 2003-11-07 at 11:27, Jon Elson wrote:
> > Ejay Hire wrote:
> >
> > >Alternately, what's the feasibility of spending a few bucks on
> > >electrical hydraulic valves? You'd still need some form of encoder for
> > >positional feedback, but no need to tear into the treadmills.
> > >
> > >
> > Ugh! A bad choice. Back in the relatively early days of CNC, many
> > machines were made with hydraulic motors for brute force, and
> > proportional servo valves as the amplifiers. This required a hydraulic
> > pump running into a pressure relief valve, so that the system drew
> > maximum electrical power all the time, and dissipated the heat from
> > the fluid in big coolers. (Why they didn't use variable-displacement
> > pumps, I don't know.) The noise, power consumption and maintenance
> > headaches were huge, and as soon as power electronics got to the point
> > that large electric motors could be controlled with electronic servo
> > amps, all these machines were retrofitted or scrapped.
> >
> > If you want a "project" to keep you busy for the rest of your life, go
> > ahead and do it this way. But, you'll be sorry!
> >
> > First, good Moog proportional servo valves will run somewhere
> > around $5000 each. Used and repairable ones have a core charge
> > of something like $1000! Then, there are all sorts of hairy problems
> > when using some tyoe of servo driver to operate a servo valve-controlled
> > load. there are dynamic effects like fluid compressibility (actually,
> > I think it is the hoses that bulge) and sticktion in the servo valve
> > spool that really confound servo amps that expect linear performance
> > like an electric motor. The sticktion is dealt with by adding
electronic
> > dither to keep the spool floating on oil. But, this wears out the valve
> > faster. The dynamic effects are dealt with mostly by lowering gain and
> > bandwidth a great deal.
> >
> > Jon