Surface plate support quick question

There's no such thing as an earthquake proof building. :grin big::grin big:

....but we do keep trying.

Oh I know. I lived over there in the 90's and early 2000's. It's more of a let's try to over do it comment. :abnornal:At 10 Richter scale nothing survives. Probably much lower down the scale even.
 
Actually I worked with a company near me that had a design that used a cross meshed array of pure Boron wires that functioned as bearings under each column supporting the building. The idea was to allow the foundation piers to move around while the building slide around on the bearings allowing the inertia of the building to keep it more or less stationary.

And as far as all the "big guys" using it, they don't always go for the ideal solution due to cost constraints. That and ease and speed of installation. And of course people are creatures of habit, so once a method is established, it can be hard to change. I'm not saying that method would be perfect, but I think I see some advantages to it.
 
There are very good reference works on surface plates and mounting them. If you 'fully' support the bottom by potting it in the floor compound, it will not be as per the calibrated accuracy. If you only care about thousands, and don't mind your plate being a few thousands out, then got for it! A good surface plate (not shop grade) will be locally accurate in tens of millionths of an inch, and over the entire surface in tenths of a thousand of an inch.

I am trying to answer the question -correctly-, there are a lot of ways to 'make do'. A pivot is on one side of the plate to ensure that all four contact points have exactly the same pressure on them. This way the plate will experience minimum flex if they are on the 'Bessel' points. I prefer to build once, the best I can, so I don't have to do it over. Oh and yes, an improperly supported plate will flex and you can see it on a tenths indicator.
 
Last edited:
Actually I worked with a company near me that had a design that used a cross meshed array of pure Boron wires that functioned as bearings under each column supporting the building. The idea was to allow the foundation piers to move around while the building slide around on the bearings allowing the inertia of the building to keep it more or less stationary.

(sorry OT)
i worked on a crew doing a few seismic retrofit projects.
on one job in San Francisco at the PG&E building, on Market St, i was out there for eight months.
we took a 17 story building and one at a time chopped out a 6' hunk out of the base of each column and inserted seismic vibration dampening devices , kinda like giant hockey pucks.
the work was the most interesting stuff i have ever been a part of.


as far as supporting your surface plate on 3 points, it's a sound principle and i use the method myself
 
Last edited:
Dabbler, I was only thinking out loud and rambling. I'm a firm believer in following the Federal guidelines for mounting plates. I've been around them quite a while and some large enough to park cars on. And even cases where more than one plate had to be aligned with another accurately enough to be used as one. I'm very familiar with them.

If one were to be bedded/potted, it would have to be lapped back into flatness specs anyway since that's not how it was held for manufacture, and I doubt temp variations in an average shop would let you get away with it to any great precision. A cleanroom maybe, but since there are established support standards that everyone accepts, that's what should be used. They all move when you load them anyway, it's just a matter of how much and whether it makes it possible to maintain the desired accuracy. I've done some testing with instruments on plates with and without loads and found that they move more than most people probably think.
 
Tony,I misunderstood your previous post as a recommendation - I apologize for sounding heavy-handed. The friend in my above comment is a highly skilled toolmaker that ignored the recommendations that you endorse. It is all too easy to follow the recommendations, if a little picky.

I agree that if you do an alternative, some recalibration will be required. I've heard of aligning plates to use as one, but have never seen it in any shop I've worked.
 
It might make for an interesting experiment someday, if I get the instruments again. I kind of doubt it, but in an odd sort of way it makes a bit of sense to me to give it 100% support and have it lapped. I may shoot an email to a couple of old friends and see what they say about it. Obviously it's nonstandard, but there is something nagging in the back of my mind (where it probably ought to stay) that if you used some high durometer/Shore RTV about 1/2" thick to set the plate in, it would spread any load over a larger area than the standard points. I know for a fact that if you load the corner on the single pad end of a plate, it moves the entire plate. It would seem to me that there would have to be a positive effect to have some support under the whole plate. Setting it in while the potting is still setting would prevent uneven pressure from underneath, so how could it not help? What am I missing?

But, I'm just rambling again.
 
If you were to try the potting experiment, it would be vital that the foundation under the RTV would support the RTV evenly as well... Perhaps a network of vertical ribs?
 
Well, yes and no I think. That's where I get a little unclear about it. Since the objective would be to distribute the weight away from the typical 3 (or 4) points, you would have to gain some ground on how much the plate distorted under load by spreading the weight. It would definitely be different from needing a strong support under a single one of the 3 (or 4) points. In my mind, a reasonably heavy steel plate with perhaps 4 equally spaced angle, channel or I-beam cross bars would be needed, depending on the size of the plate, of course. I could see that quickly the construction complexity and expense would exceed the normal approach. What I am mulling over is whether it would actually be superior in any way. I'll have to keep thinking about it.

And then there is another angle. Instead of "soft" (relatively) RTV, use something self leveling, but pourable, like concrete or maybe just portland cement. Sort of like grouting in a machine. That wold give 100% support, conform to the underside of the plate, and not be very flexible.

Mentally, I may be simply trying to get around using soft supports on the common points, knowing that they do allow some weight driven distortion. The obvious answer is to just get a thicker plate.:)
 
I once found a 'free' surface plate on the internet.... > 2000 miles away, weighing 12t. --but it was 16" thick! Of course I passed. Having had a little experience in ferrocement boats, concrete flexes more than you might think. Perhaps a solid concrete base, say 30" high? Now that would be rigid!
 
Back
Top