Scraping Projects

[RayG]

Hi Brian,

Good post, you covered many important issues.

I remember Marko, saying to set up at the least used section of the ways, there is always some part that has had little use. Like the tailstock end, the annoying thing is that you have to scrape the only "good bits" ... If it was me, and trying not to be overly fussy about it, I'd split the difference...

On the indicator set up, I think you need flat tips for gauging round objects, and round tips for flat objects... that said I like your **patent pending*** solution, cheaper than multiple DTI's...

Regards
Ray
PS hoping to do some scraping over the weekend, gotta keep in practice..

[Bryan]

Am I reading this right? This is Mcjing's granite surface plate page: McJING Tools Online. To save you clicking the link, here's what it says at the

Accuracy "00", Error Range: 3.5 Accuracy "0", Error Range: 7

Graduation: 0.001mm/m

All the plates listed are grade "0". I have the 600 x 400. To me that says my plate should be flat within .007mm per metre. So over a 600 length that would be .6 x .007 = .0042mm. Does that make sense?

[Machtool]

So over a 600 length that would be .6 x .007 = .0042mm. Does that make sense?

I’d concur with your maths. Those 2 micron clocks you were playing with, here back in October, its a whole two divisions on those, anywhere over 600mm. Perfectly adequate for mere mortals like us.

Phil.

[.RC.]

. Does that make sense?

Yes, but flatness is not the only requirement.... The standard also says you cannot have a deviation of more then x over x distance...

Meaning you cannot have a point say at 0 in one spot, then 100mm away have a point that is 4 microns higher...

You could have say a point 2 microns higher within 100mm of the 0 spot and then another point 2 microns higher then the 2 micron higher point, but not 4 microns within x distance of 0...

They used to use a device called a repeat-o-meter to measure this.... They probably just use super sensitive electronic levels in this day and age...

[Bryan]

Thanks guys. RC, I was reading the link you posted on PM, that's what got me thinking about it. That article is hardly a glowing endorsement of Chinese plates BTW. With no spec for repeatability I don't know what to think about that. It's an academic question; it's most likely the only surface plate I'll ever have. However I need to use it to spot other references so it's worth knowing what I'm dealing with I guess.

[Machtool]

With no spec for repeatability I don't know what to think about that.

Although they don’t lay it out, if they are claiming Grade 0, local repeatability is part of the same test as over all flatness. It wouldn’t be right to claim it as a Grade 0, if they have only checked flatness, you have to do local deviation also.

This is the US Spec, that everyone seems to copy. For a plate that size it should be within 60 microinches local repeatability. (1.52 micron)

http://www.tru-stone.com/pdf/Fed%20Spec%20GGG-P-463c.pdf

Whether we trust that Grading is another argument. That small plate of mine is Chinese, but I paid over $1,000 bucks for it 15 years ago. I’ve tried my own version of the repeat test, hang an indicator off a height gauge, zero it to a gauge block, then try it at different positions all over the plate. I reckon its within about 2um. Its a division on a good clock.

There was a Rahn-O-meter on US Ebay years ago, it went for about $1,200 if I recall correctly.

Regards Phil.

(On Edit)

That article is hardly a glowing endorsement of Chinese plates BTW.

I’m not buying that, did you see who contributed to the article. What else were they going to say.

Our thanks to Rahn Precision, Rock of Ages, Starrett and Tru-Stone for their help in this story.

“They say they’re traceable to the federal specification, but they aren’t,” says Coalson. “We bought some and we could never get them flat.” The main problem is that the plates are not sold with mounts, says Coalson. Because the mounts represent an integral factor of plate accuracy and there’s no way to know the mounting conditions in which they were produced, there’s no real way to calibrate them.

That argument doesn’t make any sense to me. That Federal spec those guys live and die by, tells you exactly were the supports go. ¼ to 1/5 of length - in etc.

So even if they don’t come with mounts, surely you can put them on some at the correct position and make it flat. That cant be any harder than starting from a rough sawn stone block.

[Stustoys]

Hi Guys,

Phil you beat me to it.

One questionthat applies to what you said and what they said "Repeatability measurements are taken across the entire surface to ensure that there are no local peaks or valleys that fall outside the repeatability specification."
Aren't we really interested in what the peaks are doing? Assuming they are close enough together does it matter how low the low spots are?

Has anyone tried the DIY laser pointer test on a plate?(I'll try and dig up a link)

Stuart

[Machtool]

Has anyone tried the DIY laser pointer test on a plate?(I'll try and dig up a link)

Is the one you are thinking of Evans, over at HSM, several years ago? Something about bouncing a laser beam off a plate at a low angle, and seeing what the dot does on the wall 20 feet behind the plate. If that’s the one, don’t bother.

Phil.

[Stustoys]

Something about bouncing a laser beam off a plate at a low angle, and seeing what the dot does on the wall 20 feet behind the plate. If that’s the one, don’t bother.

Hi Phil
It is.
Ok I wont.
Stuart

[Machtool]

Ok I wont.

You can if you want too, but I saw no merit in it. Have you ever looked at the beam that comes out of those laser pointer things. Its hardly round, or a dot, how you read that on a piece of paper gaffer taped to the wall, half a cricket pitch away is beyond my technical capacity.

And your talking about measuring itty bitty 1/4 of a finger smell test deviations like 1.52 microns max. How do you move the laser pointer or the plate to a new spot, so that you’re bouncing the beam to the same spot. Personally I think that was just a Theoretical exercise.

You might have to go and find it. I recall reading it, and seeing pictures. There was talk about this Master Japanese precision straight edge. And you lay that on the plate, and bounce laser light between it and the plate. Light goes between the low points. Then I saw the photo’s, the Master Precision Straight edge. Whilst its true it was Japanese, you and I would know that as a 300 mm Engineers Ruler. It was just laid up on edge, engraving and all.

Phil.

[.RC.]

Just pick up one of these brand new still in the plastic items for $33 like I did and use it for your home-made repeat o meter...

[Machtool]

Just pick up one of these brand new still in the plastic items for $33 like I did

I’ll give you $34 dollars for it, and even pay for the postage.

I got my money back for that Electronic one. I just have to add that bloke to my , “guys I’m going to burn down one day” list.

Phil

[Stustoys]

The one I saw used a "beam splitter"(?The thing that makes a laser draw a line). The line was then set up to cover the entire surface (with at least some not touching the surface and hitting the wall behind the plate directly. Then there was some math involved that I didnt take the time to work through. The theory went something like, There is a straight line on the wall direct from the laser then a "line" some distance above that line reflected from the surface of the plate, the width of the top line tells you how flat the surface is".

Thinking about this more while I type you end up with what I was talking about before, it will show up all the low spots as errors.

I'm sure I have a link somewhere, didnt have any luck with google.

Stuart

[RayG]

Ahem. speaking of DTI's.... Don't laugh... I put in low bid on a few lots of Dial Gauges... guess what..




0.02 .. 0.01... 0.005 and 0.002..

Regards
Ray

PS.. None of them as nice as .RC's Tesa...

[Bryan]

Just pick up one of these brand new still in the plastic items for $33 like I did and use it for your home-made repeat o meter...

Tinny bastard. I just ordered a chinese one for the same price!

Ahem. speaking of DTI's.... Don't laugh... I put in low bid on a few lots of Dial Gauges... guess what..

Bloody. That's what I get for banning myself from ebay.

PS: Phil, thanks for clearing that up about the plate specs.