Surface plate

So maybe the way forward is to make up an sturdy angle iron frame (With upstanding corner pieces, and plenty of reinforcing between the side) before floating a bed of plaster onto a substantial sub base. Then place a piece of polished marble or granite. The thicker the better - twice as thick is eight times stiffer. (An off cut from a local stone / funeral mason?,) onto the wet plaster.

Hopefully, this will result in a surface plate of reasonable flatness.

Ideally the angle iron frame should have adjustable feet so that it can be levelled..

A friend of mine excommunicated, (Rightly ) another friend, who borrowed his surface plate and used it lap flat the flanges of a engine manifold!!.

Howard

I follow a Brazilian YouTube channel in which a user is building a homemade lathe without the aid of heavy machine tools. All based on a file, rasket and some small cutting machines. And he instructed me to start with a surface plate so that I could build the guides and every part of the lathe that required a certain precision. Using the withworth method and with three 15mm black granite slabs, he built a surface slab with an accuracy between 0.005mm and 0.01mm and I also wondered if I could reproduce this method using the porcelain tiles I already have available and then use engineer's blue to flatten the lathe guides. An accuracy of .01mm would be fine for me.

Thanks for that, Wesley … it helps a lot

If you could post a link to the Youtube channel, I think several of us will be interested to see it.

MichaelG.

That’s great, thanks Wesley

Possibly the best news is that he’s not making 800mm square plates !!

I will watch it with interest.

MichaelG.

.

Link: __ https://youtu.be/htzvYEzEzjc

Edited By Michael Gilligan on 24/11/2022 23:28:28

I also believe that using 80mmx80mm porcelain tiles, despite using a very rigid structure, would not support it and would flex at some point. But I intend to cut it to a smaller size, something around 60x60 or 40x40. Perhaps the fact of saying that I have the piece of such size suggested that I would use the piece in its dimensional totality.

No offence intended, Wesley … but let’s try to work consistently in millimetres if we can

When your tiles are in the bathroom, centimetres are commonly used, but ‘Engineering’ doesn’t recognise that unit and we poor old things get confused.

MichaelG.

Sorry my friend, here in Brazil we don't use inches very much, if that's what you mean. Even in engineering we use the meter metric. So, I mentioned that the measurements were around 23" (60x60cm) and (80x80cm) 31".

No, Wesley … what I meant was that you first referred to your tiles as 80cm

Then I referred to them as 800mm

Then you last wrote 80mm

Over here, at least, centimetres are not a ‘preferred unit’ in engineering and science

It honestly doesn’t matter though.

MichaelG.

.

https://en.wikipedia.org/wiki/Centimetre

Edited By Michael Gilligan on 25/11/2022 00:15:16

Here we use kg, meters and km/h and in the USA they use pounds and inches and mph.

Sorry … I should not have mentioned it

I seem to have caused confusion instead of preventing it

Obviously time for bed.

Zzzzz

MichaelG.

Sleep well my friend and thank you for your contribution.

.

For the wider forum membership:

I recommend this as being probably essential viewing

… many will likely learn a lot.

MichaelG.

Whitworth's method depends on being able to detect high-points with engineer's blue, and then remove them.

Detecting high points depends on the plates being rigid enough to not bend whilst being compared. Therefore the plates have to be above a certain thickness and stiffness - more the better. Otherwise, when two thin sheets are blued and pressed together they bend enough to squeeze the blue so it doesn't show the hills and valleys. You might not think a porcelain tile bends, but I fear they move more than enough to spoil a process aiming for an accuracy of 0.01mm. To apply the Whitworth process the tiles would have to be supported to stop them bending.

The second problem lies in the scraping process. I assumed Brazilian porcelain tiles to be like many British tiles, that is made of a hard ceramic, usually with a glazed surface. Small ones are about 8mm thick, big ones about 12mm. The type I imagined are too hard to scrape flat, but it should be possible with others. Again, the problem is avoiding bending and cracking whilst scraping; the answer is also to support the tile with a stiff backing.

A great deal depends on the porcelain tiles you have in Brazil! I think the only way to find out for sure is to try it. The problem is the Whitworth process is a lot of work, and - if it's going to fail - it will do so at the end. At some point no amount of blueing and scraping will improve the surface if the tile bends. It only has to bend a tiny amount because the target accuracy is 0.01mm. That's a challenge!

15mm Granite is a better bet than porcelain I think, but as mentioned the traditional UK answer to making a basic cheap surface plate is a well supported sheet of float glass. Float glass is very flat, so the main problem is to stop it bending when items are laid on it. Perhaps the best method is to fill a sturdy frame with 30mm or deeper wet plaster, lay the glass on top, and allow the two to dry and stick together. In use, the frame is also supported on a sturdy bench.

Fascinating project. I'd be a proud man if I successfully applied Whitworth's method to porcelain tiles. Good luck!

Dave

.

I agree, Dave … which is why I suggested earlier that Wesley should use Whitworth’s logic to select the best one of his big porcelain tiles, and use that as his marking-out table … If he wants/needs a more accurate reference plane then there are much more suitable starting points, and generally no need for such a large area.

In that position … I would be looking for three thick ‘pyrex’ discs [like telescope mirror makers use] as my starting point.

MichaelG.

.

But … being, at the time, in a rather more fortunate position … I saved myself a lot of work by buying this:

.

.

.

The important point being that I understand how to do the job, if the need should ever arise.

MichaelG.

A query slightly off piste regarding granite surface plates. My daughter was very pleased with her black granite kitchen work surfaces. I was however rather taken aback on being told to be carefull not to get it wet. And true enought the result of leaving a blob of water on the surface overnight was a darkened patch in the morning. Two thoughts ocurred to me. One - Is the granite used for surface plates a special waterproof grade. Two - if it's not waterproof does water absorbtion result in dimensional changes. Any ideas

Sounds like her worktop was not sealed. Whenever I have granite fitted on a job my sub contractor leaves a care package that includes additional sealer that wants applying approx once a year. This is the usual sealer, though it won't stop things like lemon juice marking the stone.

I also use this on other natural stone such as tiles where I don't want an excessively shiny surface or an enhancer rather than a sealer if I do.

Edited By JasonB on 25/11/2022 14:51:37