Hi Andrew

Yes a lick with a surface grinder will produce a useable surface plate, however because of the way surface plates are made it needs to be clamped to the table of the grinder, it cannot be pulled down on a magnetic chuck due to the three small feet on which it sits (the mag chuck would not have enough area to hold it down) Thus it will not be cheap. Scraping is probably a non starter as you will need two more surface plates and the three would then need to be scraped together. I am afraid you are between a rock and a hard place,unless you have a friendly local engineering Co. I wish you luck

If you cannot hold the surface plate to a magnetic chuck because of not enough area of contact can you first bolt it to a steel plate and then use this plate for mounting to the magnet.

Edited By Paul Lousick on 28/02/2013 20:52:12

Andrew how are you measuring the "bruise" depth at 0.0003? If you are to use it as a true surface plate then flatness over the whole area is the real measure, in simplistic terms I might add. Whole papers have been written on the subject and frankly you will never get anywhere near the accuracy levels measured in micro inches. If you are model making, or even doing other stuff, I would suggest that a good clean is all you will need. Surface grinding, unless done carefully and professionaly, may now cause deflection in the prime surface and then you would be far worse off than you are now.

In reality you need to decide what you are going to use it for. Perhaps you would let us all know as it may then change some views on what you should do with the plate. After all if you only intend to use it as a marking out table then a simple clean will do you I reckon. HTH.

Hi Andrew ,

Surface plates are graded according to their accuracy . Several different definitions of accuracy have been used in the past but easiest to understand and use is unilateral deviation from a true plane . Typical value for a medium grade workshop surface plate would be 0.0004 inch .

A modern day surface grinder will produce a surface so accurately flat that no further correction work is usually nescessary .

Surface plates for general use were I worked were routinely reground . Usually the big ones were just mounted on the grinder deadweight – no fixing at all apart from a couple of safety dogs fore and aft ..

Surface plates do not have to be exactly flat everywhere . A surface plate which is generally flat but with small pits is quite ok for most purposes . There is a type of surface plate for special purposes which has a pattern of deep cut slots in it deliberately .

Perfectly possible to scrape surface plate true but you would need use of a large master plate to test against . Using the alternative three plate method would probably take you the rest of your life .

Best advice would be to carefully remove any actual upstanding burrs and then leave yours alone .

Ask any questions you wish .

Regards ,

Michael Williams .

Michael mentions a surface plate with slots cut across its surface at right angles to each other, its a, 'Lapping plate' and I have used one to dress up a surface plate.It was along tedious process and dirty to boot. Using diamond paste and some Kero. It was rotated in figures of eight and regularly reversed in position and motion.It cleaned up the surface and using auto collimation to check the surface it was found to be very flat and even.While doing it it showed all the pockets that were worn.The only thing was, it left it in a Matt surface condition.This made moving devices across it a bit slow.

Initially, the Lapping plate was used for dressing the steel standards of theodolites to get the tilting axis true.

Clive

..Colin Hawes …."Note that a ground surface plate is likely to be unsatisfactory due to measuring equipment "wringing" to it and difficult to slide. This is a key reason for scraping surfaces such as this and machine slides. Colin" ..

Guess I am lucky our height guage hovers over the plate when needed to move….oh joy…… probably just as well since the plate is less than prestine

.

Andrew,

For one reasonable specification … see para. 2 of this

MichaelG.