An Ultimate Hand Scraped Surface Plate

[cazclocker]

Thanks for sharing, jmhoying. I've always been fascinated by the process of scraping. It must be incredibly tedious and exacting. Scraping and checking against your reference plate (or edge) over and over and over again. Until you've achieved a state of nearly perfect flatulence. Sorry, I've been waiting to use that joke for years! Yeah, I know, I shoulda kept waiting...!!!
...Doug in Arizona

 

[intjonmiller]

I thought that might be the case, Jack. I just wanted to make sure. Thanks for the new (for me) info on machinist lingo.

I had that same confusion a couple of times when I first encountered that terminology. By now I'm accustomed to it so I don't notice anymore, but I remember that feeling. It clicked for me when I realized that nobody would mention "accurate to four tenths" (or whatever amount for whatever situation). Home center plywood is often flat within four tenths of an inch. Four tenths (of the usual 0.001" standard unit) is orders of magnitude more impressive.

I want to see the process of printing the plate in the scraping process. I can't imagine they had another 4'x8' plate, already comparably flat, and they put one on the other with some prussian blue and slid it around to print the high spots. How does one approach such a large piece?? I can think of several possible ways, all of which have potential failings.

 

[Dubbie]

Yeah that was what I was curious about. I can't really see how you can make something planar with just a straightedge.

 

[Dresden]

At Mpls Honeywell we had three identical plates about 3x5 feet, we had them at three different locations, you could always check with a straight edge but we would get the other two and scrape them in as a set using the three plate method
if needed. At one location we had a six by eight pink granite about a foot thick
that was used for rebuilding.

 

[johnh]

I used to work at a university that had a steel optical table, probably a little larger than the one shown (perhaps 10' x 6'), that was scraped flat using optical interferometry techniques (back in the 1960's). Eventually the table became superfluous for the laser optics research team and a few years before my time it ended up in the workshop I managed - as a welding table!
It's an amazing set-up surface for welding but what a waste of a lot of some poor sod's time spent scraping.

 

[benmychree]

Yeah that was what I was curious about. I can't really see how you can make something planar with just a straightedge.[/QUOTEI think one would use bridge straightedges for overall accuracy and surface plates for the local flatness, or, an autocollimator for accuracy of any part of it. Incidentally, those of us who have taken Rich King's scraping classes can appreciate the fact that the knee on the Bridgeport should have been finish scraped and NOT flaked as this one has been "decorated"; the flaking allows grit to be pulled into the interface. If flaking is done, and it should be done, it is properly done on the places that you cannot see, such as the underside of the saddle surfaces and gib, where it holds oil, but does not invite grit to be drawn in. U&lmadoc can testify as to this issue as he has also taken Rich's class.

 

[fmj1911]

I'm an old (maybe 52 isn't old, old) machine tool rebuilder/retrofitter that started a year out of high school. I was fortunate to apprentice with an old(er) gentleman (His name was Gene - we called him Gene, Gene the dancing machine) who had been doing it since the 50s. He was a true craftsman and artist. Together we scraped a number of projects so he could teach me the principle of 'because you can'. Once we scrapped the heads and block of an old v-8 so flat that headgaskets were not required. It was typical to scrap some parts like gears and bores not just to tenths but to microns. You can easily get two parts so flat that adding a drop of oil causes a hydraulic lock that 'glues' them together. You could only get them apart by sliding them apart sideways and shearing the bond.

That was all done by hand scraping. The half-moon flaking seen in the pictures above used to be done by hand but most people do it with a power flaker now because of the time involved - it's just not cost effective to hand scrape machines any more other than to touch up spots or gibs (or flaking for cosmetics and picking up oil). Machines have become more accurate now as well with CNC controls so grinding is also more accurate which also eliminates the need to scrape as much.

It's sad to see that skill going away. More and more folks like myself are aging out of it and true Artists like poor Gene - are long gone and with them special skills, tricks and tips that will never come back.

I've scraped in so many machines over the years that I've lost count but here where I live - I don't know of another single person that still does it. So while I've scraped in my machines in my garage, kept all of my hand scrapers, power scrapers, camel back straight edges, indicators, masters and even my hidden supply of red lead... my sons, now grown chose different paths so I imagine when I die, my wife will give them to someone who doesn't know what they're for and will scrap it all. Well, that's progress I guess - sniff, sniff!

Thanks for those great pics though. It's nice to know the art is still alive in a place or two.

 

[intjonmiller]

'm currently mourning the loss of tooling for my grandfather's roughly WWII (I think) era tool & cutter/surface/universal grinder because my father or brother didn't recognize them (thinking this machine was just a surface grinder) and tossed them. It pains me to think about it, especially when I wonder what else may have also been thrown out.

So if you're honestly concerned that your family will toss them feel free to send them my way. I promise I will put them to use learning and preserving those skills.

 

[Ulma Doctor]

'm currently mourning the loss of tooling for my grandfather's roughly WWII (I think) era tool & cutter/surface/universal grinder because my father or brother didn't recognize them (thinking this machine was just a surface grinder) and tossed them. It pains me to think about it, especially when I wonder what else may have also been thrown out.

a cold chill ran down my back at the thought...

 

[intjonmiller]

So according to my calculator this plate should weigh over 7100 pounds (assuming it's solid, not webbed). I really have trouble imagining this or any comparably sized plates getting flipped and rubbed/printed multiple times in the scraping process. But I don't have any scraping experience so what do I know? Can a localized approach with a smaller reference plate achieve this grade? I would love to understand the technique even though I can't imagine ever using it.