Granite Surface Plate Stand

admin

This thread will chronicle building a 3 point support stand for a 36x48x6 inch grade A granite surface plate from Precision Granite in Whittier, California.  The plate weighs in excess of 1000 lbs.  The plane is flat to within 0.0002 inch as measured by autocollimator, and the repeat reading is 0.000035 inch (35 millionths).  Per the federal specification, the plate must be supported in the same 3 locations as it was in manufacture.  To be geologically precise, I don't think the plate is actually granite, but another igneous formation known as gabbro.  Stay tuned for more.Attachment 961251/.kjj Attached Images
Reply:That is one helluva kitchen island! Looking forward to seeing the build.GravelThe difference between theory and practice is that in theory there is no difference.
Reply:First things first, need to get it off the skids and up on jackstands.  The plate isn't fastened to the skid so it has to go up fairly evenly, a few inches at a time.I already had a ton of these 1/4 inch steel shims laying around.Just keep shoving stuff underneath it until it's high enough.
Reply:Probably not necessary, but I cut some thin plywood to make a pad for the jackstands under the plate.  Also, you can see the red "x" marking a support point under the plate.Instructions along with a diagram straight out of the federal specification for how to mount the surface plate.
Reply:I did a little quick and dirty beam analysis to determine what cross section and size material to build the frame out of.  This calculation is by no means exhaustive of all considerations when designing a beam, but it gives a rough idea of how the beam will perform.  After several iterations, I decided on 3x3x1/4 angle for the support beams and the superstructure as well.This formula seems to indicate 0.0055 inch deflection on the beam with the single bearing point, given a load of 350 lbs.  Each support will probably see something more like 400-450 lbs, but this doesn't change the result by much.  It's still well within my goal of deflection of less than 1% the span of the beam.  I could have way overbuilt this, but it's not really an engineering problem unless cost is a consideration.So on to cutting the components.Here's the frame's top side inverted and snugged down on the 1.5 inch thick jig table.  She's starting to look a little worse for the wear.  In the months since I built it, it's had a lot of cycles of filing and stoning applied to it.  She sure was pretty when she first came back from the blanchard grinder and I poked all the holes in her, but it's a tool to be used, and anything but precious, and the stains and grinder scars are to be expected.  Still loving the big stronghand table mount clamps.  Those things are beastly!  The mill clamps work pretty good too and actually don't take that long to set up either.On with the welding.  Not shown in this picture, but I was having some major problems with porosity on these parts.  I already clean my weldments of mill scale and wipe with acetone, but I ground it all out, cleaned some more to no avail.  Changed my spool of wire, mig liner, contact tips, tried a different bottle of c25, until I noticed a torn o-ring on the mig gun pin.  I lost a half day because of a .74 cent part.
Reply:Congratulations! That looks like a great plate. I'm interested in your build. I looked over their prices and they appear fair.
Reply:One of the support beams coped to the frame.  Even on the corner joints of the main structure, I preferred to cope angle frames rather than miter.  It's a personal choice, and it may only be academic, but with a coped joint, there is more surface area of weld in the joint.The leveling pads at the three points on the cross beams will be 3/8-16, so nuts were located and welded accordingly.  To keep the nuts from lifting on edge, I put a bolt through them to draw them tight to the beam while tacking, but I must have shrunk one of the nuts too much during tacking.  Got the bolt stuck and snapped the head off.  I was able to drill out the bulk of the material, and then use a tiny ball file in the high speed to get the rest of it out without a scratch on the female threads .  Yeah I know, just don't break the bolt in the first place, right?!Not the leveling pads I would use, but they worked for a mock up.On the side of the frame that has the one support, I welded in two nuts in the corners, so that an anti tipping bolt could be installed and run up to within 1/16" or so of the plate just in case I do something less than intelligent.  You know, like say, putting a 300+ lbs bridgeport knee on the corner of the table.  Should never need the anti tip bolts, but I've had my fair share of bozo moments.
Reply:The plate is 36x48 so I built the top frame 36 5/8" x 48 5/8" to give some breathing room once it's installed.  The diagonals are equal to within 1/16".  Got her clamped and tacked to the columns and lowered to the floor.  This is why you need a million cheapo vise grip clamps.  I blow one up every 3 or 4 weeks, but they're easy to come by.  I love my stronghands and besseys, but when I'm holding a 70 lb frame square to a leg, you can't beat a one handed clamp!Then built a lower frame section identical to the top section minus the cross support beams.  It was perfectly on size as well as I can measure with my tape, so getting everything square with the columns was easy.  I just inverted the frame and dropped in the lower section and tapped it in until I got the depth (height?) the same.  The height of the lower section off the floor wasn't critical except in the respect that it had to accommodate the legs of an engine hoist under it so that I could do things like put a bridgeport knee on it's corner and tip it over!  Oh, and no big deal, but so I had room to put the darn plate on the frame in the first place!  I can't remember, but I think it was something like 8 1/2" with the casters added.This view shows better the triangulated layout for the supports called out by the spec.Welded out all the joints, attached the caster plates, then the casters.  They're 5 inch non locking phenolics.  It seems like phenolic doesn't tend to pick up trash in the wheel as much as rubber or polyurethane.Most shop equipment I've built in the past I either left unpainted or painted black.  I wanted to change it up, so I went with something called satin lagoon.  I think it turned out well.  There are a few runs and sags here and there but I never claimed to be much of a painter.  It wasn't a consideration, but I guess the color won't show hi-spot blue as bad, which my hands will often be stained with while working with the plate, not that it matters much.  Also, you can see that 3 standoffs were added to the lower frame section.  These are to accommodate 3/4-10 leveling feet that extend to take the load off the casters when the plate needs to be stationary.
Reply:Coming along nice, looks pretty heavy!
Reply:You can see the leveling feet standoffs better here.No forklift or gantry here, we're low tech.  There are 4 slings per hoist for a total of 8.  They're supposedly rated for 2000 lbs in a straight pull.  In a simple basket they should be good to 1000 lbs.  They're not long enough to basket the entire plate to the hook, so we have a lower strap basketed to an upper strap.  The slings are being used at approximately 30 degrees, so I de-rated them to 500 lbs each.  This probably would have been barely sufficient but under dynamic loading with shock loads, it becomes pretty sketchy, so I doubled the number of slings so that I could feel more comfortable with the lift.  I also stuffed some rags under the sharp lower edge of the plate to prevent unnecessary wear to the slings.She's airborne!  We ended up having to go up 42 inches so that we could position the frame under the plate (hoists remained stationary).My assistant hoist operator/911 dialer.  Before we started, we formed a clear, cogent plan for the lift, including contingencies for undesirable situations.  I also performed a thorough inspection of all the lifting hardware before we started.  I had to borrow the second hoist from my buddy, and it's a good thing I gave it a good looking over, because one of the two big bolts that attaches the hoist's upright (mast?) to the base was missing a nut on the bottom.  That couldn't have ended well.  It's hard to see, but there was a digital level on the plate and we kept it within 1 degree of level all the way up.Got the stand under the plate, and in a few minutes from when I took this picture, I'll be breathing much easier.
Reply:Touchdown!Detail showing the leveling feet.  They have a rubber anti skid material on the bottom surface.  The foot itself is actually threaded onto a 3/4-10 stud.  This is so to facilitate maintenance so that if I ever wipe one out, I just have to jack the frame an inch or so instead of 6 or more inches if the stud was integral to the pad.  The pad looks (and is, for that matter) pretty close to the floor.  I don't think the low clearance will be an issue, it's not like I'm taking this thing off roading!View under the plate of the 3 supports I used.  They are similar in construction to the leveling feet for the frame itself, except that they are on a 3/8-16 integral stud.  With the levelers ran down to minimum height, the plate is approximately 1 inch off the surface of the frame.Detail of the anti tipping screws in the corner of the frame on the data plate side of the plate (single support).  They are set between 1/32" and 1/16" from the plate and secured with jam nuts.Thanks for looking!
Reply:Congratulations again! It looks first rate all painted up. Glad your lift went well.
Reply:Thanks man!  And yes, precision granite has very affordable prices, and I felt their sales people were great too.  I think the price may be further evidence for my suspicion that the material is gabbro and not technically granite.  The only downside I can see to the gabbro is that it doesn't contain quartz, so it may wear a little more quickly than actual granite.  I will strive to minimize the wear by following diligent cleaning and storage practices, and by forcing myself to not work in the same area of the plate all the time.  Even though my primary use will be machine tool reconditioning, I'm a one man band, and I expect the plate to provide me many years of use before it needs resurfacing.
Reply:Originally Posted by GravelThat is one helluva kitchen island! Looking forward to seeing the build.
Reply:That looks great!!
Reply:Man, that is an awesome site! You do some really fine work!GravelThe difference between theory and practice is that in theory there is no difference.
Reply:F me. that is killer. Great work !!!  You got a California Car Cover for that thing?
Reply:Originally Posted by InsanerideLMAORExpat, looks good.  I like the fact that you calculated deflection of your stand. What good would a surface plate be with a stand that flexed? If I remember correctly, black granite weighs approx 57 pounds per cu. ft.
Reply:a lot of years ago,a friend and I indulged in a long liquid lunch. when we returned to his pattern shop,i showed  him the finish nailer I had just bought.he promptly hooked up his air hose and nailed two pieces of 3/4 " wood together before either of us noticed the gun was loaded with 2" nails and the wood was resting on his black granite surface plate. after the initial heart attack,we inspected the plate and couldn't find a mark. since he was feeling a teensy sheepish,we didn't celebrate.miller thunderbolt 250vlincoln square wave tig 175 prolincoln idealarc mig sp250everlast tig 210EXTeverlast power plasma 50chicago electric (hf) 130 tig/90 arcchicago electric 90 amp flux wire3 sets oxy/acet
Reply:Thanks for all the comments y'all.  Jimmy pop, I ordered the vinyl cover with the plate to keep dust and grinding grit off of it.  As boatbuoy's anecdote illustrates, granite is resilient, but I have a small area to work in, and the metrology dept is right next door to the grinding and welding dept.
Reply:With the suspense and the way you told your story I think you should consider writing industrial novels. I'd buy one, lol. Nice work btw. Company I work for had a metrology lab that was built to isolate all outside vibrations from cranes and large machinery. The surface plate was freaking huge and every so often would get rescraped by hand by high paid pro's. They calibrated all there own precision tooling. Cool stuff
Reply:Thanks man.  Industrial novels eh?  I didn't know that genre even existed!
Reply:As I try to hide my inadequacy and lack of exposure to precision tooling/repairs/fabrication, could you please elaborate on the use of such a stone/plate in your shop?  I can't quite visualize it's practicality.30+ yrs Army Infantry & Field Artillery, 25 yrs agoMiller 350LX Tig Runner TA 210, spool gunLincoln 250/250 IdealArcESAB PCM 500i PlasmaKazoo 30"  vert BSKazoo 9x16 horiz BSClausing 12x24 lathe20T Air Press
Reply:Originally Posted by storemanAs I try to hide my inadequacy and lack of exposure to precision tooling/repairs/fabrication, could you please elaborate on the use of such a stone/plate in your shop?  I can't quite visualize it's practicality.
Reply:yes. ^^^^^ this.....Dynasty 200 DX with Coolmate 3211 mvpThe tool room where I was an apprentice had a granite plate about twice that thick . It was a monster , Foreman used to get pissed when we played card on it at lunch time with out the cover . We made all are own tooling for the B/S screw machines .
Reply:The surface plate is the reference datum for precision measurements.  It all goes back to the relative nature of accuracy.  What is accuracy, precision?  If you're leveling a field, 1 ft over 100 yards is flat, for a carpenter it may be 1/4" over 10 ft.  The stand I fabricated for the plate ended up within 1/16" square, I think that's accurate for a weldment of it's size and function.  Many machinists speak of tolerances in terms of 0.001 inch or some multiple thereof.  For a precision scraped surface, accuracy and precision are spoken of in the scale of 0.0001 inch.  This is where the precision surface plate shows it's worth.  When reconditioning the cast iron bearing slideways of a machine tool, the geometry of the slideways must be held to very high standards of flatness, parallelism, and coplanarity, for the precision of the parts the machine produces can only ever approach (but not equal nor exceed) the original accuracy built into the machine.  The surface plate is used to, as gxbxc alluded to, create your own tooling.  It will be used to qualify the respective geometries of cast iron straight edges, parallels, angle plates, 123 blocks, and other tooling, either created or acquired.  Before WW2, surface plates were made (and some still are) of high quality cast iron, machined, high temperature stress relieved, and then scraped for flatness.  In the 1800s, Joseph Whitworth wrote on the automatic generation of gages, later to be known as the "3 plate method", where one could scrape in 3 cast iron plates to true flatness without having an external reference to use as a master.  If anyone is interested in the exact technique, I can provide a link to the paper.  During the material scarcities of WW2, various manufacturers began building surface plates out of granite and other igneous formations.  It was soon found to be a superior material due to it's relatively low coefficient of thermal expansion, hardness, it's tendency to chip rather than raising a burr, and the fact that corrosion is not an issue (among other advantages).  Below are some basic examples of how a surface plate might be used.Say you wanted to prove that the 45* on your nice starrett square head was actually 45*.  You could compare it with other square heads, you could even compare it with a 45* master, were you so inclined.  But these are not quantitative measurements, meaning they don't yield any useable data about the actual condition of the object we are measuring.  Here's how you would measure this quantitatively.  This is a sine bar, so named because you use it in conjunction with the trigonometric sine function.  It is exactly 5 inches from center to center of the precision rolls attached to this bar.  If we tip one side up using a stack of jo blocks properly wrung together (a process of assemblies the jo blocks so that they actually stick together due to atmospheric pressure), we can see that we have constructed a right triangle.  The sine of 45* is 0.707106.  Multiply by the hypotenuse (distance between the rolls), and we have 3.5355.  This is how thick the jo block stack must be to create a 45* angle on the sine bar.  So we wring together 3.5355 inches of jo blocks from our weber set, then we carefully place the square head on the sine bar, sweep the height gage and watch for indicator deflection.  This square head was pretty good, within a couple thousandths.  More than good enough for a rough layout tool.Similarly, proving that 2 faces of a set up block are parallel.  Same principle, sweep the gage and look for deflection.These are rough castings of a dovetail straight edge and a parallel straight edge that will be machined, stress relieved, and then scraped in to a high bearing quality for use in rebuilding.  Basically, they will become the surface master for the instances when you can't bring the work to the plate, you (indirectly) bring the plate to the work.Another, perhaps more specific use of a surface plate is when bluing a part to check for flatness.  I don't currently have any pictures of this process but essentially, you apply a marking medium to the surface plate, and then rub your part on the "blue-ed" plate, and inspect the part for the "high spots" which will show themselves by transference of the marking medium.  Here's a good video on youtube that illustrates the principle much better.  Thanks for the comments and questions!
Reply:Thanks. I understand the rationale and use of the plate as a reference surface.  What do you do in your shop that requires that degree of accuracy?  Or do you just desire to have such a tool in your shop?30+ yrs Army Infantry & Field Artillery, 25 yrs agoMiller 350LX Tig Runner TA 210, spool gunLincoln 250/250 IdealArcESAB PCM 500i PlasmaKazoo 30"  vert BSKazoo 9x16 horiz BSClausing 12x24 lathe20T Air Press
Reply:ARexpat thanks for posting your build. I enjoy reading your explanations.
Reply:^^^^^^^ +230+ yrs Army Infantry & Field Artillery, 25 yrs agoMiller 350LX Tig Runner TA 210, spool gunLincoln 250/250 IdealArcESAB PCM 500i PlasmaKazoo 30"  vert BSKazoo 9x16 horiz BSClausing 12x24 lathe20T Air Press
Reply:Thanks y'all.  In the upcoming months, I will be re-scraping the bearing slideways of a 1960's era bridgeport mill for bearing quality and alignment.  Then other machines will follow, a lathe, additional milling machines, surface grinders, and perhaps a jig bore or two.