That must have weighed over 1000 lb? A little more than I'd want to be moving around.
How did you lap that down to being flat? I've read a bit about such things but it is a bit beyond me as to how it is done.
John, It ran about 1200 lbs. The frame I made for it would collapse or expand with a compound lever similar to those used to cock crossbows. The collapsed frame fit on a pallet jack. I have not been able to make a move without a hydraulic liftgate truck since I was 15. I have two 2500 lb shop cranes and a pallet jack. If all goes well (i.e. I win the bid), I will begin making an automated assembly line in a few weeks. Each "station" will weigh about 1800 lbs and there will be nine "stations" in the total assembly. I will probably rent a forklift when it comes time to load everything on a truck.
Starrett used to have a video about how they made surface plates. My process was about the same, just more handwork. I started by lapping the entire surface with a CI lap and (220 grit, as I recall) carbide paste. I worked up through 1200 grit paste checking my work with straightedges and .0005 feeler gages. When everything checked out that way, I used a 6 ft X 1 ft CI certified plate to control a split coherent beam to an optical flat. At the time, I was limited to a 2 inch diameter flat (since then, I managed to get NASA to buy me 8 inch optical flats), so I was quite limited in the area I could directly verify (I told you there was a lot of hand work involved). However, I was really only going for a .000050 RMS overall flatness.
In terms of lapping, I made a 4-bar driver mechanism that I hooked up to an electric motor (5 HP as I recall) for gross lapping. I applied counterweights to the lap to correct wear patterns. It's the same way that Johansson gage blocks are made -- just turned upside down.
You then set your certified flat next to the plate. You have to pick your reference points carefully to minimize hand work, but once they are tied together, you count and mark interference fringe patterns relative to the leveled reference plate (where the split mirrors are mounted). The rest is like any other scraping/grinding job. You whittle away at the high spots until the fringe patterns all go straight. I polished the whole thing with a piece of felt stretched over the (plastic coated) reference plate and imbued with one of the "ultra-bright" type toothpastes (the least expensive find abrasive you can find). When the surface is completed, check the reflection of something with a long straight pattern to it. As I was learning how to do monochromatic light measurements at the time, I got to use this for part of my required time for my inspection license.
http://emtoolbox.nist.gov/Publications/NISTMonograph180.asp shows some of the information collected by NBS (later, and worthlessly, NIST) on gage blocks that might help if you haven't been lucky enough to work in such industries. Optical flats and monochromatic light help, but fantastically good flats have been generated with nothing more than making a really true reflection in the surface. It is a bit more complicated than that, but that is the basis.
Does this help?