DIY measuring and lapping a worn surface plate

[DIYSwede]

-Anybody here on HMT using uncalibrated surface plates, worn for years with unknown flatness properties,
lacking the funding or willingness for transport, calibration and resurfacing by a certified techie?

-Anybody wondering why the professionals refuse to show how they lap worn plates back into spec?

-Anybody wanting to DIY the whole process down to the tolerances the techies won't reach,
as nobody wants to spend that kind of money?

-Anybody wanting an AA grade (w/o a 200 dollar sticker) plate, willing to waste 10-20 bucks and a few evenings of elbow grease?

Matthew Tinker has done it: Knocked together a DIY repeat gage, ordered some online diamond dust,
charged an old CI plate with it and started lapping a big granite plate himself.

Look for yourself as he's showing the entire process down to 1 micrometer of deviation:

[CharlesWaugh]

WOW
I mean: W O W !

[Toolmaker51]

re Mr.Tinker, His technique parallels that of onsite plate certifying specialists; though never saw any toting an iron plate.
Aluminum, especially the grade known as tooling plate is the norm. Aluminum secures lapping grit far better.
Any variety of lapping works the same way. The abrasive embeds into softer material and being harder than workpiece, creates the cutting action. Laps can be of copper, brass, antimony, cast iron, even soft steel [such as needle eye laps]. Some can be used many times, being adjustable by expansion.

https://www.google.com/search?client...+tooling+plate
https://www.google.com/search?client...le+copper+laps
https://www.google.com/search?client...eedle+eye+laps

[Bcampbell1989]

Here is another awesome surface plate lapping Video

[JoeH]

I appreciate your posting this as it introduced me to yet another tool I wasn't familiar with, the repeat o meter. Not that I need one, but the concept is interesting. After looking at Tinker's video I found this one that goes over the repeat 0 meter concept and helped me understand.

Beyond understanding the repeat 0 meter, the guy had interesting ideas about the feet.

[DIYSwede]

Further info on winging a DIY "Cheap-o-Meter" together,
for those who simply can't find 1600 bucks a bargain for the Rahn (sans indicator):

Repeat Reading Gage - DIYSwede's "Cheap-O-Meter"

A few improvements that went into my version:
1) 175 mm radius hinged probe arm - for even less angular error than the original (app 4,5" pivot length of its 10" total).
2) No coarsely sawed flexure, or any yield point limit stays needed- degree of freedom only by design of the "travel lock".
3) Ridiculously fine adjustment with a 508 TPI (0,05 mm pitch) differential screw.
4) Plastic handle (instead of the steel "butcher hooks") transfers less heat from the hand into the instrument.
Most of these inspired by Tom Lipton's & CatalinaRacing's thoughts on the RoM,
the plastic handle idea came to me when rummaging around, finding a 1 x 8 x 14" acrylic door handle.
5) Then, not even wasting any carbide inserts for the feet, I instead pressed in some salvaged 12 mm ball bearings,
ground them in the lathe's 4-jaw with a toolpost-mounted grinder*, honed them on some diamond disks.
This wasn't my idea, either - got it from Matt Tinker.
Only a bit of inspiration, material from the junk box and some 20 hours went into it.

Cheers
Johan

*always protecting my ways etc with oven foil when grinding, machining cast iron or case-hardened stuff.

[Paul Alciatore]

I am not putting down what he did. It is an excellent job and I am sure, and I mean that seriously, I am really sure that when he was finished that he had a very good plate. It was much, much better than when he started. And it will be a big asset to his shop.

But I do have a small word of caution here. The pros do not rely on just the repeat-o-meter. The repeat-o-meter does not measure flatness. It can only tell you if the surface deviates from a perfect sphere. That sphere can be concave, flat, or convex and, using only the repeat-o-meter, you can not tell the difference. If you want to verify flatness with a repeat-o-meter, you must first calibrate (zero) it on a known, flat surface. A surface plate that is known to be flat, even a small one, can be used for that calibration and then the repeat-o-meter can be used to check other plates, even larger ones. Of course, as the second video shows, other instruments can be used to check the plate for flatness or to find an area on a plate that is flat where you can then calibrate the repeat-o-meter.

I definitely learned something here.

[DIYSwede]

Thanks for stating what I perhaps silently assumed, Paul.
My quest started with three pieces of marble plate and different grits eventually leading to making
a few tools: first the spherical comparator, which then with an indicator became the "Spherometer".

These could only prove down to app 1-2 micrometres deviation, but not where the highs and lows were.
But that was good enuff for my workshop use - I didn't mind if it was a sphere with a 600 m radius.

Then I suddenly found the free, uncalibrated diabase plate, which were to become my "master plate"
Hence I winged the Cheap-o-Meter together, finding the deviations less than 1,5 mumeter in any axis.

If it is "flat"? - I still dunno - it can be either somewhat concave or convex within those 1,5 my of deviation, but it's NOT warped,
and absolutely good enuff for my walk-in closet workshop.
And it didn't cost me 250 bucks for some other dude's calibration of it either - but then I'm real cheap.

My initial fundamentalist quest for flatness took a real hit when I read a few old books on metrology,
realizing that I've been standing on a relativistic quagmire all the time...

Tip (p 24ff): https://pearl-hifi.com/06_Lit_Archiv...l_Accuracy.pdf

Just my 2 cents

[Bcampbell1989]

Tip (p 24ff): https://pearl-hifi.com/06_Lit_Archiv...l_Accuracy.pdf

Just my 2 cents

Haven’t made it all e way through but man that is an awesome book, that first chapter on flatness was a real eye opener for me. I almost felt enlightened after reading it haha

[Toolmaker51]

Over years, worked in shops with calibrated plates, naturally; but the surrounding conditions occasionally made measurements erratic. So the questions were eliminated or at least reduced simply. Hit three spots moving the instrument, then leave it parked and shift the part 3 x. I'd say whatever out-of-plane condition the plate had, influenced a dinky gauge base WAY more than a part occupying many square inches.
Which revealed another point. In shops that use plates heavily, if there is a low spot it's near the inspection cabinet...

The link disguised title of final destination. https://pearl-hifi.com/06_Lit_Archiv...l_Accuracy.pdf
Just the Moore Tool Co., that's 's all. If any company had grasp of mechanical accuracy it is them. Also pds CSIP, Deckel, Hauser, Dixi, Lucas and many others had copies, hidden maybe, dog-eared just the same.

To me, three US companies stand out equally when it comes to practical accuracy.
Machines that could, instruments to detect. Pratt & Whitney. Moore. DeVlieg.