[iaknown]

Hello all, So I am involved in a never ending argument with a few guys I currently work with and wanted to get your thoughts on the subject. It is regarding whether bolting cnc equipment down to the floor is unheard of and if it is somewhat common for heavy machines or better precision, or any reason for that matter. At my last job, I worked at a machine shop where we made tooling for injection molding and I remember our Mazak 6-axis coming in and the anchors for it being drilled and epoxied into the floor. And I could have sworn the other 5 Mori's were also anchored into the floor. The floor was all concrete but I figured this was done for better rigidity or the like. Did I dream all of this up or is this done when setting up machines? Is this at all common? The guys at my current job all have equipment and say it's ridiculous, they've never heard of it. Anyone have any input? Thanks in advance!

[andyb4]

Every machine should have an installation manual that specifies the type of foundation that is reqired for the machine. It usually involves a much thicker depth of concrete than most floors. The larger the machine the deeper the foundation. The purpose of this heavy foundation is to hold the machine "square" once it has properly been set up and verified. If a machine is just sitting on the floor, then as the floor shifts, and they all do, the machine will shift. If it is not properly anchored then you take your chances how it shifts. From my experience bolting down is much better and easier to maintain machine geomerty.

[RICHARD ZASTROW]

Machines are secured to foundations for isolation as well. The foundation should not be connected to the floor itself when high precision is required. A fork lift driving by a grinder will show up on the ground surface.

Old rule of thumb: foundation should weigh 3 times the machine weight. Concrete weighs 150# per cubic foot. Concrete should be reinforced as well.

Dick Z

[mactec54]

Hi iaknown

The answer is yes, any machine can & should be bolted down for better precision & to maintain better accuracy/stibility of the machine, as said above in the other posts should be on a seperate concrete block

[Geof]

Unfortunately mactec54 is incorrect; Haas machines have no provision for being bolted down and with some of them it is completely unnecessary.

You are going to find this issue argued up, down and sideways.

On big machines which are very precise a separate isolated foundation block is essential to attain the full accuracy and precision the machine is capable of. The isolation is from the building floor pad, this is cut out, a large hole is excavated, with luck to a good solid base, and then this is filled and rammed with roadbase material and a concrete block poured. As mentioned the mass of the concrete should be heavier than the machine. It is a moot point whether the machine should be bolted down to the block. If the machine is capable of very high speed rapids with large masses of fixturing and parts on the table then it probably should be bolted and will probably have bolt pads not just leveling pads.

At the other end of the spectrum with a machine like a Haas MiniMill it does not even need to be levelled; it does not even need to be sitting on all four feet! However, if it is not it will rock and make annoying noises.

I have gone through this thing about bolting machines down back and forth with people for over forty years. I ran turret lathes ranging from 3500lb weight up to 20,000lbs and none were bolted down. Once I moved a big one to a new location while it was set up for a particular job, merely shimmed all the floor contact points so it didn't wobble without levelling the machine and the first part made after the move was within 0.0005" of the last part before the move.

On some big machines like big shaft turning lathes where the workpiece may be approaching the weight of the machine they have to be bolted down to concrete foundation blocks otherwise the machine might start walking around when it is spinning an out of balance shaft. In this situation the foundations really become part of the machine, and because they settle the machine must be re-aligned and levelled frequently. I live on a river delta and one big machine shop here could only run their biggest lathe at its best precision when the tide was at a particular level because the ground moved with the ebbing and flowing water.

But, having said all this, you are bound to find people who will argue dogmatically one way or the other.

[Andre' B]

At the other end of the spectrum with a machine like a Haas MiniMill it does not even need to be levelled; it does not even need to be sitting on all four feet! However, if it is not it will rock and make annoying noises.

Some very high presicion mills are now being made with only 3 pads so they do not rock and you can not twist them when setting them up.

[Geof]

Some very high presicion mills are now being made with only 3 pads so they do not rock and you can not twist them when setting them up.

One of my medium size turret lathes was supported like this and that was the only machine I ever saw with three pads. It stands to reason that for something smallish this is the way to go.

[RICHARD ZASTROW]

There are very accurate machines on 3 points. (SIP,Hauser etc.) These machines are over-designed to be so rigid they will not deflect. However I was referring to modern machines which tend to be lightweights.

Also, as Geof mentioned, heavy parts in motion will move even large, heavy machines. On those glorified drill presses (aka turret milling machines) isolation pads probably will suffice.

A 50hp milling machine WILL move about when heavy cuts are taken. One of our operators used to use a big pry bar to put the mill back in line at the end of his shift. He used to crank up the feed till it stalled then back up one notch. (6'10" 320 pounder named of, course, "Tiny") LOL

Dick Z

[Caprirs]

Be careful when bolting machines to the floor that the machine is not permanently twisted/deflected. You are forcing the base to follow the contours of the concrete which can cause as many problems as it might cure.

I have laser calibrated some big horizontal mills (Niigatas). I am amazed how small turns of the leveling screws can affect the accuracy of the machine. Running a Renishaw ballbar tester for circularity calibration also reveals how small tweaks to the leveling screws can affect the roundness of an interpolated hole.

[iaknown]

Thanks guys, I appreciate all the input....it seems its not unheard of depending on the application and machine. I'm guessing that Mazak may have specified that the Integrex I referred to be anchored down or I assume they wouldn't have done it, but I could be wrong.....Feel free to share any additonal information!

[Geof]

The Integrex I probably falls into this category; If the machine is capable of very high speed rapids with large masses of fixturing and parts on the table then it probably should be bolted and will probably have bolt pads not just leveling pads.

[whitis]

One of my medium size turret lathes was supported like this and that was the only machine I ever saw with three pads. It stands to reason that for something smallish this is the way to go.

While 3 point support is wonderfullly kinematic, it creates other problems. Most machine tools are fundamentally rectangular designs. Rectangles do not support well on 3 points - the stresses tend to get weird. Tipping can be a problem. Rigidity can be reduced.

Imagine a 2' x 2' square made with 2"x4" beams on all 4 sides and a flat top and put a heavy load dead center, such as a gantry base. Now put feet in the two back corners and one front center. First thing is that the front foot bears twice the weight of the back ones. The two side rails are supported by their end in the back but in the front, they are supported by cantilever on the front rail. Hardly symmetric. The front center of the table is bowing upward.

Now, machine tools often have uneven weight distribution. The headstock end of a lathe is heavier and either the column or the knee is heavier on a mill. If you put the two feet on the heavy end and the one foot on the light end it isn't as bad as the situation described. But if you have one foot in front of a mill on a knee mill, and you run the table all the way forward and to one side, things could get a little tippy. Due to the weight distribution and/or the machine being bolted down instead of sitting on feet, the machine may not actually tip over but there is a bunch of metal that was intended to support this torsional force that is now doing nothing. Now, a knee mill often has a centered support cylinder in the front which helps bear the weight so it isn't all bad but the twisting moment is still intended to be supported by the corner of the base through the Z ways.

It is possible to design a base to support a rectangular bed on three points but it will be a rather exotic looking design. Something you might see in an high end CNC machining center but not in a traditional machine design.

Support on 4 points and you risk distorting the machine if the levelers aren't properly adjusted. Support on three points and you will probably distort most mills regardless of how well it is leveled. But it all boils down to the geometry of a particular machine, plus the quirks of the flooring, plus the way the load from a heavy piece being machined is distributed.

To suggest that a large class of machines should be supported on three points without considering the geometries typical of the class of machines let alone the geometry of individual designs is not a good idea. However, I suggest counting the number of feet or mounting holes and considering their position - those are where the designer expected the machine to be supported, if they did their homework, and are where the machine probably should be supported unless you have done your homework. If you haven't done the math or the measurements, don't deviate from the original design.

[Geof]

[QUOTE=whitis;673518....To suggest that a large class of machines should be supported on three points without considering the geometries typical of the class of machines let alone the geometry of individual designs is not a good idea.....[/QUOTE]

With all due respect to a long post I did say "smallish'" machines which in the case of milling machines could be very well supported by two points at the front where the table and workpiece mass moves sideways and one point at the back under the column which remain stationary.

[Switcher]

In the amount of time it took to post/read this thread, the machine could have been bolted to the floor (lol).

Then you wouldn't need to take time thinking about should I do it, or not?

My $0.02 -

[RICHARD ZASTROW]

Securing a machine to the floor or what is the meaning of life? Aprox. the same debate. LOL

Dick Z

[Caprirs]

Bolting the machine down because others have done so is foolish without knowing why and how. Being able to measure the changes before and after is simply good engineering/science. I cannot imagine spending the time and money to go through the process without being able to quantify improvements in the end usage of the machine. Unless, of course, it's not my money and I'm working on someone else's time.

Changing things for the sake of change is hardly a rational approach. The reality is most shops cannot measure their machine's accuracy with sufficient precision to know if bolting the machine down has improved the finished part.

Clearly, there are applications where bolting the machine down is mandatory. However, that does not translate to "Everyone run out and bolt all the machines down!"

[Geof]

On a different forum I did see the argument made that machines should be bolted down in some areas for seismic protection in case they fell over or slid around the floor; although I think it is more a case of the floor sliding under them while they stay still. The fear was expressed that one could get crushed between a machine and a wall as the machine slid into it.

I couldn't help but think that if the floor was thrashing about fast enough and far enough for machines to 'move' several feet you would have a lot more things to worry about than getting bumped by a machine.

[MadMax]

I had some issues adjusting the levelling pads on an older Daewoo lathe. 6 pads and I just couldn't get within the desired 1 increment of levelness in across both axis of travel and keep all the pads on the floor. One of them was always lightly loaded so I guess it might have helped to be able to crank that one down onto the floor. It may have caused other issues, but in the end I gave up and called in a tech (my first levelling job is this lathe). He moved things around and said that my levelling job wasn't too bad and went about making adjustments to the spindle alignment and got things cutting nice and right.

Maybe a more interesting question: how often is everyone checking the consistency of their machine levelling in their PM plan?

Have you noticed that the flex in a levelling tool can be varied by moving a level around? I had to mark where I was putting my level on the turret mounted levelling plate in x and z directions because I found that moving a mass of metal around on the plate made a measurable difference on my level indicator.