Is that REALLY the size?

[elf]

Elf, is there a link to a procedure or a video?

.... not that I know how to use any stiching software.....

I can't find a link to it now. There is another technique using a scale bar and measuring the pixel size of the scale bar, then transferring that to the subject image. Micron scale bars are available.

[polaraligned]

I stumbled on this video yesterday. Measuring flatness to microns with laser and webcam.

https://youtu.be/hnHjrz_inQU

[Mcgyver]

One can make excellent fits with nothing more than spring calipers. If you can show what units those are calibrated in, I'd love to know.

Absolutely. if by chance the clearance needed just happens to be the extent of what you can do with calipers. Like a couple of thou on a model with piston rings for example.

However.....suppose you need a 1 thou clearance, not 1/2 thou, not 2 thou. Or .0002", or about any interference fit. No one is doing that by feel. Those realities make a blanket statement about size does matter with fit rather silly. He's made a play on words, typical Doozer style. "If you control the fit size does not matter". Obviously you can't control the fit unless you pay attention to size between the parts, are able to measure it and use that measure in machine one or both parts. Time wasting silliness imo so I'll got back to work now

[Richard P Wilson]

Absolutely. if by chance the clearance needed just happens to be the extent of what you can do with calipers. Like a couple of thou on a model with piston rings for example.

However.....suppose you need a 1 thou clearance, not 1/2 thou, not 2 thou. Or .0002", or about any interference fit. No one is doing that by feel. Those realities make a blanket statement about size does matter with fit rather silly. He's made a play on words, typical Doozer style. "If you control the fit size does not matter". Obviously you can't control the fit unless you pay attention to size between the parts, are able to measure it and use that measure in machine one or both parts. Time wasting silliness imo so I'll got back to work now

'No one is doing that by feel' How do you think we did it (in some cases still do it) when all we had were callipers?

[RSG]

My last commercial customer ran a business at his home which was an eye opener to most folk who ever visited. His business supplied about 1/3 of the world.s supply of carbide drills for drilling brake linings for trucks. He and his staff had the equipment and knowledge to manufacture and measure to tenths.

Regards David Powell.

LOL, wait a minute....weren't you the guy the started this thread saying it wasn't possible for the home shop guys? So it turns out some "home shops" can turn out work to the tenths.....just because it's at home doesn't make it impossible.

[Doozer]

Absolutely. if by chance the clearance needed just happens to be the extent of what you can do with calipers. Like a couple of thou on a model with piston rings for example.

However.....suppose you need a 1 thou clearance, not 1/2 thou, not 2 thou. Or .0002", or about any interference fit. No one is doing that by feel. Those realities make a blanket statement about size does matter with fit rather silly. He's made a play on words, typical Doozer style. "If you control the fit size does not matter". Obviously you can't control the fit unless you pay attention to size between the parts, are able to measure it and use that measure in machine one or both parts. Time wasting silliness imo so I'll got back to work now

Inking, printing, and scraping.
No one is measuring anything.
Yet parts are fit enough to make
precision machinery. Hmmmm...

The only play on words is
you can prick your finger
but you can't finger your prick.
RIP George.

-D

[Mcgyver]

'No one is doing that by feel' How do you think we did it (in some cases still do it) when all we had were callipers?

and yet I noted exactly that in my post, you even quoted it!

You response is taking my statement "No one is doing that by feel" out of context. Be fair. It directly followed "suppose you need 1 thou clearance, not 1/2 thou, not 2 thou. Or .0002", or about any interference fit." If you can do each of those things by feel or sprung caliper, I stand corrected and my hat is off to you.

Offering instances were an erroneous blanket statement is correct is a bit like trying to sell the the broken clock based on it being right twice a day.

[Mcgyver]

LOL, wait a minute....weren't you the guy the started this thread saying it wasn't possible for the home shop guys? So it turns out some "home shops" can turn out work to the tenths.....just because it's at home doesn't make it impossible.

My interpretation of it was along the lines of don't get all bent out of shape or let it stop you doing stuff as its not necessary for the vast majority of work, maybe all the work for any particular person. I agree with that, always have. It's just that that is very different than claiming it can't done or is never needed (not that you did), a erroneous conclusion so many seem to have leapt to. As AK points out, all shops are not identical, nor are their requirements.

[eKretz]

Hmm, I think there's a miscommunication here somewhere, maybe between relative size and absolute size. The problem is nailing an absolute size in an environment with wildly fluctuating temperature like, say, in a garage. On something very small with a direct comparative measurement like an indicator on a stand comparing to a gage block, probably can be hit pretty close. *If* the guy doing the measuring knows what he's doing. Measuring to tenths and below requires very careful and meticulous work. Are you sure your reference surface is flat to within that level so you are getting an accurate comparative measurement? Because even a pretty small temperature differential from top to bottom, like one might find in a shop with stratified air layers from recently running something like a heater can screw that up and make a surface plate go concave or convex. Are you sure your parts are clean and sitting perfectly flat on that surface? So on and so forth.

That's just the thing in a home shop. The average home shop guy may not know the intricacies of any of this. A person that's a skilled machinist that has worked at it for a career should obviously be capable of working to a higher level of precision than an average home shop guy who has a shop solely as a hobby.

[BCRider]

Absolutely. if by chance the clearance needed just happens to be the extent of what you can do with calipers. Like a couple of thou on a model with piston rings for example.

However.....suppose you need a 1 thou clearance, not 1/2 thou, not 2 thou. Or .0002", or about any interference fit. No one is doing that by feel. Those realities make a blanket statement about size does matter with fit rather silly. He's made a play on words, typical Doozer style. "If you control the fit size does not matter". Obviously you can't control the fit unless you pay attention to size between the parts, are able to measure it and use that measure in machine one or both parts. Time wasting silliness imo so I'll got back to work now

Like the replies above this one I would very much dispute that. It becomes a slow process if doing it without any special tools but it is certainly possible. But with some clever methodology to allow it to be done relatively it could be a fairly rapid and mundane task.

Just one possible option is an old version of the micrometer. Various examples of tapered flat or round gauges have been in use for centuries. And if someone is looking for this degree of a fit they could make up their own lightly tapered cylindrical gauge that is more or less nominally the proper diameter at roughly the middle and tapering down by a thou or two to the narrow end up up by a thou or two at the bigger end. This can be used to gauge a bore or to set the feel of a set of spring calipers for measuring a pin to that size. And so a bore and pin can be set to a very fine fit or some degree of press fit. All without ever knowing the exact dimension to within any number of tenths.

And I'm sure that there are other options for doing such things.

Now aside from measuring the other issue is removing stock in a controlled enough manner. These days it's all about grinding. And even in this thread more than once folks referred to something along the lines of "this is tool post or surface grinding territory" or " It can't be done other than by grinding" or similar. But there ARE other ways of kissing off a tenth or two (or a few microns in metric) other than by grinding. Methods that seem to be either ignored or seemingly forgotten for this discussion.

[BCRider]

Hmm, I think there's a miscommunication here somewhere, maybe between relative size and absolute size......

I wondered about that part way into this thread too. But David never came back with any clarification on that idea. Or I missed his post?

[J Tiers]

Like the replies above this one I would very much dispute that. It becomes a slow process if doing it without any special tools but it is certainly possible. But with some clever methodology to allow it to be done relatively it could be a fairly rapid and mundane task.

Just one possible option is an old version of the micrometer. Various examples of tapered flat or round gauges have been in use for centuries. And if someone is looking for this degree of a fit they could make up their own lightly tapered cylindrical gauge that is more or less nominally the proper diameter at roughly the middle and tapering down by a thou or two to the narrow end up up by a thou or two at the bigger end. This can be used to gauge a bore or to set the feel of a set of spring calipers for measuring a pin to that size. And so a bore and pin can be set to a very fine fit or some degree of press fit. All without ever knowing the exact dimension to within any number of tenths.

And I'm sure that there are other options for doing such things.

Now aside from measuring the other issue is removing stock in a controlled enough manner. These days it's all about grinding. And even in this thread more than once folks referred to something along the lines of "this is tool post or surface grinding territory" or " It can't be done other than by grinding" or similar. But there ARE other ways of kissing off a tenth or two (or a few microns in metric) other than by grinding. Methods that seem to be either ignored or seemingly forgotten for this discussion.

I think the point has been missed. If you need to take off a particular amount, then you are going to end up with some sort of "units" involved, even if only to check that you DID remove that much.

If the 5/8" shaft needs to fit the hole with "some clearance", OK, make it so. Only rough units needed, if any.

If you need 0.0015" clearance, you are involving actual units. Dick around with spring calipers all you like, but you are going to use some calibrated unit measuring device or you will not know if you hit the dimension. Just having the "dimension" means there are units involved.

OK, you can measure the shaft with outside calipers, with a 0.0015 feeler against the shaft (feeler is calibrated in units, you see). Set the caliper to the "stack".. Transfer to inside calipers, and bore until calipers just fit with the right "feel". You never had a number, quite............. and maybe you got there.

Does that make you happy?

As for grinding, There is a "YOU can DO that" effect.... Yes you "CAN" do many things taking enough time and trouble. Skew cutters, etc, etc... they "almost" get to a grinding finish. With an industrial lathe and appropriate carbide cutters, you can attain near-grinding finishes. You got that setup in your shop? Some here do, by no means all do.

But, grinding is a cutting process that takes cuts far smaller than any lathe tool. There are a huge number of cutters, so thousands of small cuts per second add up to serious removal of material. For production, or reasonably efficient part making, grinding is very effective and even attainable by anyone, if you expand the definition slightly. Hold an oilstone, or a stick covered in decent sandpaper up against the spinning work. Who cares which part is moving?

[BCRider]

If we need a set value of clearance then yeah, we need to measure in some way and use numbers. But for a one off it's also possible to do it purely by feel too. One little trick being to leave some extra material on the end and machine it down or bore it up by a whisker and try the other part in that little starter end. If it fits or doesn't we base the next full cut on that fit.

There's no doubt and I fully agree that in a home shop with wildly variable temperature and non calibrated measuring tools that any numbers are only relative to the day and items at that moment. I guess it comes down to the machinist's ability to use the measuring tools and to select the cutting tools that permit working to very small increments and then on to give a good finish that works with the methods being used to size an item.

It would be interesting to start a new thread about actual techniques used by all of us for working an item to size that is within a tenth or two.... Relatively of course. I've got a couple of tricks that work for me and I'm sure others have their own. It would be interesting to see some of them and learn more options. After all, not everyone can or wants to set up a tool post grinder for some small step of a project. Not if some other slow but controllable method can do the same or near enough.

[J Tiers]

If measuring tools and work are the same material (steel) and at the same temp, then the measurements should be nearly perfect. If not the same material, then the tempcos need to be taken into account, but good measurements can be made.

Labs are at a standard temp to take care of every possible question or error that can be dealt with. Eliminates variables.

[BCRider]

That's right. I'm taking a break while a rough turned part cools off before final work to size. And the parts I measured the other day still measure the same. But it's probably 12 to 15C warmer in the shop today. Almost T shirt warm in fact where the other day I was considering grabbing a toque to keep the noggin warm.