|Tim Stevens||21/07/2019 18:16:41|
1259 forum posts
It is fairly easy to check the flatness of a slab of eg granite worktop, before purchase. Hold the slab at eye level and about horizontal, in a god light, and look carefully at the reflections in the whole of the surface. As you move the plate slightly, the reflection should not distort at all as it moves across the surface. Any wavyness or hollows etc will show as the reflection changes shape and/or size as the plate is moved.
This is a useful check on any bits of worktop left over or sink cutouts etc, when it is not possible to do a proper engineering check in a builder's yard.
And although worktops can be fairly thin (12mm rather than 50mm for a 'proper job' a slab of thicker MDF glued on makes a good reinforcement.
|Michael Gilligan||21/07/2019 18:17:45|
16200 forum posts
It does, of course, rather depend upon the 'aspect ratio' of the plate
My Optical Flat is relatively thick in relation to its small diameter.
But 900 x 600 x 6mm [for example], whether in Glass or Granite, would be a very different matter !!
P.S. ... some useful information about granite, here:
Edited By Michael Gilligan on 21/07/2019 18:30:41
|Pete Rimmer||21/07/2019 19:23:38|
|773 forum posts|
If a piece of glass or worktop is out of flat by 0.0002", a flat part will spin on it and you'll think that it's not flat. Part of the analysis of parts when scraping is a 'hinge test'. It's used for identifying the high and low spots on a part. If your part hinges in the middle then it's humped. If it hinges on the end then it's dished or bowed.
I have a small home-made dovetail straight edge that I use for scraping internal dovetails and it's slowly creeping into a bend over time. If I leave it 6-12 months I have to be sure to check it before I use it because it will bend a couple of tenths and spin like a top on my plate. It only takes a quick scraping pass over it to get a good hinge and make it useful again but if I didn't have a plate that was at least that flat I couldn't check it and any slide ways I scraped would be bent also.
You really do need a proper plate for scraping.
|Neil Wyatt||21/07/2019 19:30:04|
18139 forum posts
You should be able to tell if it is as smooth as the face of the waters...
|Neil Wyatt||21/07/2019 19:35:55|
18139 forum posts
Which 'flat' surfaces do you mean and how do you know they need improving?
You might questionably improve its smoothness by scraping some of the working surfaces, but a surface plate would be of little help with those.
|Barrie Lever||21/07/2019 20:25:39|
|688 forum posts|
That is naughty, you have only taken part of what I said about the differences in bending and flex between granite and cast iron. I did not say they did not bend or flex.
I think you might be reading too many Sunday papers and picking up bad habits from the journalists!!
I would expect a granite surface table of 900x600 to be a minimum of 100mm thick, like you say area to thickness ratio is also important.
|Michael Gilligan||21/07/2019 22:28:46|
16200 forum posts
Not wilfully naughty, I assure you !
... and if you believe that I misrepresented you, then I apologise unreservedly.
I simply quoted the words which I wanted to discuss [in the context of granite vs glass] and omitted the reference to iron as being irrelevant to that.
The fact is that a granite table can 'bend & flex' ... which is made quite clear by Starrett in the text about using the correct support locations. [ref. 12. in the FAQs]
My simple point was that the aspect ratio is very significant: Your granite table is very stiff ['though we must remember that nothing is actually rigid] largely because it is well-proportioned ... and my glass flat is also very stiff, for the same reason.
I am labouring this point a little, because Andrew probably does not need a large area table for his job ... and by minimising the area he will get a stiffer plate.
Again ... Apologies if I inadvertently misrepresented your post.
Edited By Michael Gilligan on 21/07/2019 22:50:00
|Barrie Lever||21/07/2019 22:56:12|
|688 forum posts|
I agree with everything you say regarding flex, just about everything moves, it just depends how much force and how far the movement is.
Apologies not really required, you are one of the most civil and respectful persons that I have ever communicated with but it is just worth pointing out the perils of taking a snap shot out of a quote.
Edited By Barrie Lever on 21/07/2019 22:56:46
|duncan webster||21/07/2019 23:30:08|
2734 forum posts
There is a reasonable size surface plate on homeworkshop at a very good price.
|jaCK Hobson||22/07/2019 08:32:57|
|171 forum posts|
My cheapo granite surface plate is a lot flatter than my granite worksurfaces or granite tiles. My conclusion comes from the close surface effects of sticky/floating you get between two 'flat' surfaces.
|Michael Gilligan||22/07/2019 08:41:51|
16200 forum posts
As one might reasonably hope
|Pete Rimmer||22/07/2019 09:46:39|
|773 forum posts|
That one does look in fair condition but it looks like a base from a piece of metrology equipment like a comparator. What you have to watch out for is that it might have had the same part slid on the same spot thousands of times in it's working life. There could be a dip in it. Easy to check though and it is cheap.
6189 forum posts
I'm intrigued as to what high-precision flat surfaces are used for in a home-workshop? When exactly does extreme flatness matter and why?
In a tool-room or metrology lab it's obvious. A standard needs to be about 10 times better than the gauge or jig being calibrated or set. Usually the goal of manufacturing is making interchangeable parts within tolerances. That's a good reason for owning a good surface table, but meeting that requirement it's unwise to buy second-hand off ebay and hope it's OK. When interchangeable accuracy is required, surface tables are carefully set-up, kept clean, maintained. and periodically recalibrated. There are, I think, three different grades of surface table, and the better the grade the more fuss is made about keeping them in good condition. High grade surface plates are laboratory instruments, completely over the top for most tool-room purposes.
In my workshop, which admittedly may not be typical, I don't work to tolerances. Instead I use the older technique of 'fitting'. For example, making a cylinder and piston for an engine, I would first bore and finish the cylinder to nominal size. Then I'd make the piston slightly larger and finish it to fit the cylinder. Using the cylinder as a gauge allows me to make an accurate fit, highly accurate if I put the effort in. But it's not a precision fit because the piston and cylinder are matched relative to each other, not to absolute dimensions. If I make two cylinders/piston pairs, it's unlikely that swapped pistons would fit properly in the other cylinder. They are not interchangeable.
The point about fitting is that good results can be achieved without elaborate measurement, hence there's no particular need for me to own high-end measuring gear of any type. In earlier times craftsmen did extraordinary work with very simple measurement aids - it's amazing what can be done with a pair of dividers and a home-made lathe. In practice, being able to measure accurately saves time. On the lathe I mostly use a digital caliper; it measures with reasonable accuracy, and the locked blades can be used as a size comparator. When more precision is needed I reach for my very ordinary micrometer, I have no need for microns. I have DRO on my mill, but it's a time saver rather than essential to accuracy. I could get the same results with primitive methods.
Flatness isn't top of my list; an ordinary lump of work-top is good enough for setting-up most of the time, then I have a milling table, and - rarely - might get the float glass out.
As I don't do scraping and rarely transfer dimensions with a scribe block, what's my Business Case for buying a surface plate? I think Andrew's in much the same place as me except he wants to improve his mini-lathe. His resources are a budget DTI and some steel-plates, and he said ' They mostly seem to involve a surface plate, which is not a cheap item.'
In that case MichaelG's immediate reply suggesting float glass seems spot on to me.
Of course, we've all failed to answer the exam question which was "Many people must have improved their mini lathe performance. How have they done it?"
Based on my mini-lathe experience, I suggest it's a mistake to rush to the internet and review all the existing advice for ideas. This can be highly misleading because what people have done depends on their particular lathe. For example, when mine arrived I expected to find it plastered in chicken fat, poorly assembled, and with fairly obvious improvements to do like replacing the steel gibs with brass, plastic gears with steel etc.
Actually mine arrived in good order. It did have a few issues but these only became obvious after I'd learned how to use it. My advice 'If ain't bust don't fix it'. Instead, use the lathe to make several test pieces, facing, turning, drilling and boring. Bear in mind that the characteristics of materials, cutting tools, and cutting speeds all have to be learnt too. The lathe may only one third of the problem! Avoid jumping to conclusions until the operator can tell the difference between trainee mistakes and machine problems. But, if consistently getting a poor result, it's time to focus on what might be causing it. Asking on the forum is a good place to start. Describe what you are trying to do, and what's wrong with the result. Photos help enormously. Doing that will produce a list of potential causes and how to test them. Focussing also helps filter internet searches and avoid information overload.
I think most self-taught beginners go through a dazed and confused stage. Don't worry, it passes. My epiphany came when I realised turning random scrap-metal was a bad mistake. Far better to invest in alloys intended to be machined. Suddenly my lathe stopped struggling and started cutting properly...
|Andy Carruthers||22/07/2019 12:09:26|
279 forum posts
Thanks Dave, very helpful
I'm making 6 similar items which do not need to be absolutely dimensionally accurate but must pass visible muster, so far I don't need a high precision flat surface and caliper measurements are just fine
I'm glad I am not the only one who uses the 'fitting' approach, I haven't yet needed to machine anything to tight tolerance but that's probably because I don't make models - but I have enormous respect for those who do
However, I did pick up an unused granite cutting board yesterday "just in case", I hate Sod's Law...
|Barrie Lever||22/07/2019 12:21:05|
|688 forum posts|
Everyone's workshop is different as you elude to, but uses that I have put the granite surface table to so far have consisted of the following.
The above are additional to any normal marking and measurement jobs that crop up.
All the while it is very unlikely to cost me money in the long run, I might even make money who knows.
Have your fingers been burnt buying things on Ebay? I buy lots of really top quality items on Ebay. Latest being a Dremel 3D printer in as new condition with only 3 hours run time on it, got it for 40% of the price of the most heavily discounted new Dremel printer of the same spec. Same thing with many airbrushes and spray guns.
|old mart||22/07/2019 14:03:02|
|1913 forum posts|
You don't need anything too special for home workshop use, the 2 foot square cast iron table I bought has a brass label on it "not for inspection use", but the 0.0005" error it has is more than good enough for hobby use.
|Michael Gilligan||22/07/2019 14:23:11|
16200 forum posts
Just for info.
Starrett usefully states, in the FAQs:
3) How is surface plate flatness defined and specified?
A) Flatness can be considered as all points on the surface being contained within two parallel planes, the base plane and the roof plane. The measurement of distance between the planes is the overall flatness of the surface. This flatness measurement commonly carries a tolerance and may include a grade designation.
The flatness tolerances for three standard grades are defined in the federal specification as determined by the following formula:
Lots of good stuff on those pages
|old mart||22/07/2019 14:32:18|
|1913 forum posts|
In our fitting shop at work, we had a 2 foot square black granite 4" thick just sitting on a sheet of rubber directly on the bench. It was tested every year and given a grade A every time.
|602 forum posts|
I don't know what you would be doing surface table wise on a lathe, bus you definitely need some form of reference flat for taking measurements, using a height gauge for example.
I was lucky enough to obtain (not going to say 'pick up' a 3' x 2' cast iron table on its own stand - hadn't built the workshop at that time - just the concrete base, literally built everything else around it.
We put it in the middle and its stayed there ever since, no way I'm trying to move it, and I can get all around it for access.
FWIW, I have a cheapy Clarke CL300M for work that is too small for the bigger lathes, mostly small valves and pipe fittings certainly never needed to put it on a surface table.
|Jeff Dayman||22/07/2019 15:33:43|
|1853 forum posts|
Getting and using a surface plate (whatever grade and type) to improve an inexpensive mini lathe - the "million dollar shine" on the two dollar pair of boots.
You can do a lot of work on most mini lathes without doing anything to the flat surfaces. If you aren't sure exactly what you are doing with "improving' the flat surfaces you may do more harm than good. I'd try the lathe first to make a bunch of different parts and see how it goes.
Just my $0.02 worth - your mileage may vary.
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.