Homemade surface plate - how flat is it?

F

FromScratch

Registered
Registered
Joined
Jun 18, 2019
Messages
15
I'm tackling a project that needs more precision than my wood shop tools can deliver (small vibrating solenoids). So -- down the road to precision, I:

* made three straightedges which show no light when joined (backlit with a window in a darkened room - best backlight I've found).

* made a wedge with two more straightedges spaced 2 mils apart at one end, so I can see what light shows through under a straightedge at various gaps. (0.1 mil shows light. My straightedges go together nice and dark - matching to better than 0.1 mil).

* made a 12" x 12" surface plate. I won't go into detail in this post, but I glued two 1" slabs of bluestone together (total cost $14), learned from experience (mistakes), and ended up with a plate that shows dark under straightedges put down anywhere in any orientation.

* bought three cheapo 6" speed squares and worked out the technique for squaring them to one another, which was an interesting problem. Now they seem to be good to about 0.1 mil too. (One was out a disgusting 11 mils!! I mean, really.)

So -- how flat is the plate? I've seen references that a straightedge might not tell you how flat a surface is, but can anyone give me an example of a surface that fools it? A twisted surface (imagine taking a capital H and twisting the two parallel strokes) shows flat if the straightedge is kept parallel to the twist axis, or at right angles to it. But held obliquely, it shows curvature gaps. I think I have a shop grade surface plate -- attractive pale blue color, too. I only worked it down to 1500 grit sandpaper finish, so it has a medium sheen, and it's sealed with grout sealer. It seems to be working.

BTW, I got a lot of experience trying to estimate small dimensions by eye, and it's amazing how your eye fools you. At the end of a long work session, you can get frustrated because those tiny errors start to look as big as potholes, so you think you aren't making any progress. Get a night's rest and -- hey -- they're much smaller!
 
My eyes play tricks on me everyday . :rolleyes: Old age .
 
What you really need is an autocollimator and and a repeat-o-meter, used together to properly check flatness. Both need to be calibrated and certified to an order of magnitude higher than what you want to test your plate for, and a skilled operator. But it sounds like you want to do it all yourself. You can make things flat to some extent, and parallel to some extent, but with your current resources you are guessing, perhaps well, perhaps not. Man with one watch knows what time it is...

Please keep us up to date on your progress, it will be interesting to see what it can be certified accurate to. There are people who go around calibrating metrology equipment for labs, manufacturers, and small machine shops and hobby machinists. If you really want to know how it turns out, you might pay the fee and find out how good you made it, within a quite narrow tolerance range.

To have standardized parts, tools, and other accurate things takes known reference surfaces and testing tools, proven beyond a doubt. For fun in YOUR shop, your plate is what you believe it to be, and is your shop standard. You are following in the large shoes of Joseph Whitworth in the 1830s, who developed and used the three plate method.
 
mmcmdl said:
My eyes play tricks on me everyday . :rolleyes: Old age .
Click to expand...
Weeellll.... I wasn't gunna say anything, because I'm only 65. But yeah, mebbe a little of that too. Gotta watch the coffee intake as well... not always the miracle cure for a boring work session that you'd like it to be!
 
@BobKorves:

I'd have to see how cheap that can be. Might throw for it. Meanwhile, it seems to turn in results at least good to 0.2 mil when I use it, which is all I need. I squared those three speed squares on it until they all mate with virtually no light showing through, and they do so no matter where I place them. It may not be full grade B, but I'm sure it'll get my solenoids project done.
 
Eyes and straightedges have limitations when used alone to determine accuracy. It is pretty easy for us to convince ourselves something is straighter, rounder, squarer, or whatever than it really is. I am talking about everybody, me too, not just you. You have figured out the means of using multiple parts in careful cycles to find accuracy, which is great. It took mankind about a million years to get to where you are. Don't know how much you studied and how much you used logic to get to where you are, but you can certainly be proud of your success so far. I also do not understand what flatness has to do with small vibrating solenoids...
 
You have a great start
You will have to do lots of work
Get some high Lapping grease dyes, a heavy straight edge ( it has to be very hard)
Lots of elbow grease) the dyes will show the low areas.
The finer the lapping oils the higher grade the surface
Start scraping do not go back and forth, constantly change directions and lots oil
How accurate all depends on your work.
REMEMBER in the old days it had to be done by hand.
There is NO short cuts. Slow and steady take your time
 
Bob Korves said:
Eyes and straightedges have limitations when used alone to determine accuracy. It is pretty easy for us to convince ourselves something is straighter, rounder, squarer, or whatever than it really is. I am talking about everybody, me too, not just you. You have figured out the means of using multiple parts in careful cycles to find accuracy, which is great. It took mankind about a million years to get to where you are. Don't know how much you studied and how much you used logic to get to where you are, but you can certainly be proud of your success so far. I also do not understand what flatness has to do with small vibrating solenoids...
Click to expand...
I used a lot of logic, always do. My solenoids must be fairly precisely made, with yokes made of stacked sheet silicon steel filched from a microwave oven transformer, that must fit very closely, and must leave a central core opening in which a shaft with some neodymium magnets vibrate. The smaller the gap between yoke and magnets, the stronger the solenoids are for a given coil current. All must be made pretty precisely and assembled pretty precisely. That's my motivation for all of this homemade precision work.
My aim is to see if I can get fossils called concretions (very hard iron carbonate) to open by vibrating them at their resonant frequency. Pounding with a hammer is a very deprecated practice, and leaving them out to freeze and thaw in winter is an exercise in patience to say the least. The solenoids will be driven with a special amplifier that uses a transducer to receive vibration back from the fossil, which is therefore the feedback loop that makes the amp oscillate at the fossil's resonant frequency. Much logic in that design!
 
FromScratch said:
My aim is to see if I can get fossils called concretions (very hard iron carbonate) to open by vibrating them at their resonant frequency. Pounding with a hammer is a very deprecated practice, and leaving them out to freeze and thaw in winter is an exercise in patience to say the least. The solenoids will be driven with a special amplifier that uses a transducer to receive vibration back from the fossil, which is therefore the feedback loop that makes the amp oscillate at the fossil's resonant frequency. Much logic in that design!
Click to expand...

Much logic indeed! That's a very interesting project.
My first thought for that application would be abrasive cord type of band-saw, or even water jet.

Is that within the capabilities of an inexpensive microcontroller (like arduino) or fully analog, or SOC?
Anything you feel like sharing will be greatly appreciated.

-brino
 
You must log in or register to reply here.
Back
Top