Thread: How flat can I surface grind 2x18x27" plate A36?

you need to get it milled than bolt holes drill and tap

before you get it stress relieved

after stress relief than mount rails/ trucks inspect with a level

than decide what is next if anything

taping the holes will blow your expected tolerance

than you need to deal with how the base is mounted and how what gets

mounted to the trucks effects alignment

I think you are miss guided, just machine the mounting pad points and drill tap the holes where you need them. And use the heaviest plate you can or make it a welded part that has been stress relieved. Right now I am machining a cnc mill for a firm that builds cnc routers. Heavy deep weldments and just machining were things mount...Phil

What about a granite surface plate? Can be machined to accommodate mounting and is already plenty flat. Can be had pretty cheap brand new.

EDIT: my apologies, just saw Thermite already mentioned granite

Gawd, I hope this isn't for one of those horrible little 7x12 lathes

>001 per foot is the standard but a decent horizontal or vertical surface grinder should be able to get .001.

Chuck would need to be near dead flat, it's not uncommon for a chuck to be out .001 per foot..Chuck needs be checked with a straight edge or better, not just grinding the chuck and checking with a mounted indicator.
2" thickness part would need to not be sucked down with high Mag.
Super checking for bugs, burs, and grungies on the part or chuck.

Wheel course grit and on the hard side, with running wet.
Might have to spark from center of part a few times.
Might take twice as longn and flip flop a few times.
Surface finish 64 or so as a high finish likely would/may heat the part.

Plus all the other stuff good stuff Pm guy mentioned.

I workr=ed for Interlakes based back in the 60s and we had a special technique for flatness, But it would not be fair if I told. We were building special machines but the base part really took off and took over the company..

Originally Posted by Mcgyver

Gawd, I hope this isn't for one of those horrible little 7x12 lathes

NFW!

Linear rails on A36?

Far more clever .... auditioning for a government job as Joe-Buyed-in's Pentagon spokeperson ... and public-speaking coach:

Space Force Weapons Systems

Back when everything was a casting....
everthing except the mounting pads was relieved, and left as cast.

The pads are all then milled or planed in one set up. changing the stress pattern in the base minimally. Not only that, but if the pads were "distant" from each other, they were more likely in a plane than a whole flat surface of similar size would be. e.g. hollow, crown, or twist in the entire surface is less, or not, apparent if the only reference points are pads at the 4 corners.

Not knowing the architecture of your part, with a cnc mill it might make sense to relieve everything except the mounting surfaces, drill & tap them, then stress relieve, and surface them.

Or plane & scrape the whole surface.

smt

I know the difference between flat and parallel. As Ekretz said, it takes shimming, varying the magnet strength, flip flopping, etc. but it CAN be done.

edit: We had an 4' x 8' surface plate to check these larger plates as well.

And who is going to pay the Bill for all this, some fed grant for the op to build a home machine come on its not going to happen...Phil

Most all my stuff gets inspected with a Faro arm (at the customer) for whatever that is worth. I check for flatness here on whatever table is handy with a straight edge.

Same customer just about unilaterally accepts blanchard ground plate. I'm tempted to buy one, but dont know all the pitfalls.

Originally Posted by stephen thomas

Back when everything was a casting....
everthing except the mounting pads was relieved, and left as cast.

Well... there you have it.

A 10EE's bed sits atop the base casting that way. One pad at each end.

The pads are all then milled or planed in one set up. changing the stress pattern in the base minimally. Not only that, but if the pads were "distant" from each other, they were more likely in a plane than a whole flat surface of similar size would be. e.g. hollow, crown, or twist in the entire surface is less, or not, apparent if the only reference points are pads at the 4 corners.

Not knowing the architecture of your part, with a cnc mill it might make sense to relieve everything except the mounting surfaces, drill & tap them, then stress relieve, and surface them.

That makes eminent sense.

Or plane & scrape the whole surface.

smt

THIS . .is where it is cheaper to just mount it onto a granite surface plate.

Because they are mass-produced at attractive price-points. Or are a drug on the market, surplused, and can easily be sliced-up to size.