Vernier caliper accuracy

I'd avoid them unless you really need them. Full disclosure -- Mitutoyo was my client for many years, and I wrote all of their articles during those years and some of their instructional materials. They're quite honest about what they claim. Just be aware that things like interchangeable-anvil micrometers sacrifice some assurance for the sake of convenience. They can be as accurate as any mike; you just have to be a bit more careful.

One thing to be very careful of with all calipers, is the comparative accuracy of the inner and outer jaws. On one set cheapie Chinese digital calipers I have, the difference between an inside and outside measurement (of identical dimension) was about 0.003"

Pete

Save that one for woodwork.

On Mon, 22 Feb 2010 03:58:39 +0000, the infamous Christopher Tidy scrawled the following:

Now that I'm no longer doing greasy mechanical work, I tend to wear gloves more often, to keep my finnernails clean. Barrier cream is great, except for that.

Fair to high. 2 thou to start, 1 thou after you get used to it, 5 tenths once you're good and comfortable with your dial calipers. But for critical measurements, micrometers are the way to go.

When a gage block isn't thick enough, you stack another on top of it.

Remember that the COE is different for steel and aluminum, with steel expanding less by half. Aluminum pistons are wobblier by nature.

Don't go by the piston, go by the bore. It sounds like your engine is in fairly good shape, but check for elongation perpendicular to the crankshaft. Bores become oval from wear, where the connecting rod pushes 'em up and drags 'em down the opposite sides. If you bore it out, you'll need all new pistons by default.

-- "Just think of the tragedy of teaching children not to doubt." -- Clarence Darrow

On Mon, 22 Feb 2010 04:09:32 +0000, the infamous Christopher Tidy scrawled the following:

Then your 2 thou is likely more hopeful. Get a pair of dials, boy! They're less easy to misread. I forgot you were talking about verniers. Got a good magnifying glass? ;)

-- "Just think of the tragedy of teaching children not to doubt." -- Clarence Darrow

Yes, or plumbing. I think that all calipers should be guilty of this shortcoming until proven innocent. It's something to beware of anyhow.

Pete

The Starrett 123 series calipers have adjustable scales and are as good as verniers get. The 50 division scales with the vernier and main scale in the same plane make them very easy to read. I can read them without glasses in a pinch, though I don't recommend it.

It is pretty amazing what you get for your money when purchasing bearings. But even though the vast majority of bearings will likely be better, the bore of a bearing in this size range could be close to .001 low and another .001 out of round. As you might expect, thin section bearings have a wider roundness tolerance than the standard

Indeed -- you need to develop a feel for the sliding force on the caliper jaws once closed. (And, of course, the jaws are not carbide faced, so they will wear more rapidly than most micrometer jaws.)

Probably adequate -- especially as you are closing the calipers to a given setting (using the adjustment screw with the extra traveling head clamped down) and then using it as a reference for the dial gauge. If you had a set of gauge blocks, you could close it to a light sliding fit on those (to free yourself from the possible errors in the vernier and in reading it) and hold it in a vise to eliminate the problems from hand heat causing expansion of the beam.

My first one from about 1960 (long gone, and I don't remember the brand) had only a small diameter spinner, not a ratchet or friction thimble, and I developed a feel for letting my fingers slip on the knurled portion of the man thimble.

I've got others with various designs, including a 0-6" Brown & Sharpe set which still have no torque limiting features, along with others with very good friction thimbles.

Remember that you have to zero them against a standard each time you change the anvils -- or for that matter if you pick it up a week later and want to measure -- the room temperature might have changed between the zeroing and the time of the intended measurement.

My main consideration of the problems with the multi-range micrometers is that you have the large C frame so there is more metal to expand as your hand warms it, thus more error from the thermal expansion.

If you can prevent this (including hand warming of the standards you use when you zero the micrometers) you should be OK. Often a bench clamp stand for the micrometer is indicated.

And potentially good for better than that as you will be using them in zeroing the bore gauge. If you use gauge blocks to set the calipers, you will be better than the caliper.

But then, if you have a stack of gauge blocks, and two longer blocks at the ends, and you have a setting reference which won't need the calipers at all. There are devices designed for just this purpose

-- a rectangular tube to hold the blocks, a pair of long blocks for the ends, and provisions for clamping it down firmly.

Good Luck, DoN.

They *should* come with standards for zeroing the micrometer after changing the anvils -- as many standards as you have anvils. These should be enough to assure accuracy at the zero point at least, and you can use gauge blocks to check for errors in mid spindle travel, but this should not be a problem with new micrometers anyway -- only after a lifetime of use by you, or someone else.

Enjoy, DoN.

Well ... the ones which I have are quite old. These days, most people buy either dial or digital calipers. Dial is nice if you can keep chips out of the rack, but Digital (if good) is far more convenient, including the ability to zero at any point (such as on what you are fitting to) and then measure how much more needs to be removed to reach the proper dimensions. Also, measuring the distance between the centers of two holes of identical size by zeroing it while measuring the ID of one hole, then shift to measuring the distance between two opposite sides of the two holes in question. This subtracts the diameter of one hole from the final measurement which is the sum of the center distance and half of each hole's diameter, giving the actual center-to-center distance.

O.K. But it misses three-lobed bores, which can be in a shape which actually measures the same between any two opposite points, but still is potentially seriously out of round. (This is more likely to be formed by certain machining operations -- including centerless grinding for OD shapes, and in a used automobile engine, the ovality is more likely, and for that your bore gauge is probably quite adequate.

To show that kind of error (the three-lobed one) with the three-point micrometer, you would want to do at least two measurements one rotated 60 degrees from the other -- and ideally yet another half-way between them just to be sure.

For measuring this sort of thing for outside diameters, there are V-anvil micrometers which duplicate the three-point contact. (Actually, there are two styles -- one for the three-lobed measurement, and another for measuring the diameter of five-flute milling cutters.)

Good Luck, DoN.

O.K. And from the image which you posted and I saw after my question, it looks as though it has a very long bearing surface on the moving jaw assembly, so there should be very little jaw rotation, especially if the gibs are set properly.

As your later posted image showed. (Or was that someone else's image? I lose track from time to time. :-)

I asked because in other discussions -- here and elsewhere -- I've been led astray by people using "vernier" as a term for any slide bar calipers. I just wanted to be sure that we were talking about the same thing. :-)

Well ... having the setting device more accurate than what is being set is always good practice. As I just mentioned a bit ago (and you of course had not seen when you posted this), a stack of gauge blocks with two long blocks sticking out to one side will make a more accurate setting reference than your calipers.

Note that with the calipers, or with the gauge block stack, if the angle of the gauge is a bit off, the reading will be too high (thus the setting will be too small), while with a ring gauge, error in following the diameter will be too small, while error in keeping the probe shaft parallel to the axis of the bore will cause it to read too large.

Well ... this one is for setting an air gauge, so it has to be pretty close to the diameter of the probe. An air gauge is hollow with both ends sealed, and air feed in through one end, with a hole of a specific size (or maybe more holes? -- I haven't actually seen one -- just read about them in books). The air goes out the hole, and flows out between the probe walls and the ID of the bore being checked, and the pressure built up in the interior of the gauge is measured and is a very good indication of the difference between the OD of the gauge and the ID of the bore -- over a fairly narrow range. So -- this ring was for setting something quite close to 16.8349 mm --- maybe 16.8350 mm is the actual target size. Anyway -- the rings are made too small, hardened, and then ground to a diameter near what is needed, and then marked as measured. They may be lapped as a final finishing pass.

If the image which I saw was yours, the traveling body is long enough to minimize the jaw tilt which I was worrying about. I've seen others which were much shorter (including the digitals which I have) which make the gib setting more critical.

Enjoy, DoN.

Gloves are a *very* bad idea when operating machine tools, but when measuring with micrometers or calipers are a good way to reduce thermal transfer resulting in errors. (Of course -- the thermal errors can be in the workpiece as well -- if you have just turned a steel workpiece in the lathe, it will measure significantly larger than after it has cooled for a couple of hours -- which can lead to fit problems when fitting a backplate to a lathe chuck for example. Finish the rough turning, go eat dinner and read a book or watch a TV show, and come back later to measure it and take the final passes to get the desired size.

Multiple gauge blocks wrung together to make up a precise size.

A good set of blocks will have something like:

And -- in a separate box, I also have:

"Wringing" two blocks together involves cleaning the surfaces about to mate, crossing them at right angles, and rotating them until the two blocks are parallel. A good block will cling to its neighbor, and you can progress to a fairly long stack of blocks which can be lifted by the top-most block and they will all stay together. (The cheap Chinese sets, supposedly accurate to 0.000050", tend to not have a good enough finish to bond together properly -- at least my set does not. But I have a good B&S/"Jo"hansen block set which does this well, and which comes with a certificate of calibration listing the actual size of each block.

The combination of multiple blocks will allow you to build up almost any dimension as long as it is shorter than the multiplicity of ones which are just above '0.10000"' (they don't often make thinner ones because they can warp more easily). As a result, some sine plates have an extra section ground 0.20000" or 0.30000" below the main surface, so you can build a stack of blocks which will achieve the small angle you want to make.

Then again, softer materials tend to embed abrasive particles so they wear away the harder material against which they are run. This is the principle of a "lap".

At a guess, I would expect the measurement down into the bore at right angles to the crankshaft axis to measure larger than the dimension at that height parallel to the axis. (The forces from the crankshaft and connecting rod would tend to apply more force along that cross axis.)

Understood -- and also to discover whether you need to bore

Note that the skirts of some pistons are designed to move with temperature to fit well at operating temperature. The solid part near the top remains more constant in dimensions than the bottom of the skirts.

Good Luck, DoN.

Might do that, but I'm going to try taking the measurements first. I have the urge to get into the workshop :-).

Interesting. Why are pistons intentionally made elliptical? I've heard of pistons where the top land is of a smaller diameter (apparently it reduces wear because the lubrication is poorer at the top), but I've not heard of an elliptical piston. I'm surprised it doesn't increase the wear on the sides due to the reduced surface area, and I'd have thought it could allow the piston to vibrate in an angular sense about the gudgeon pin. Any more information, Ed?

Best wishes,

Chris

Thanks. I'll remember you're the guy to ask about Mitutoyo gear! The interchangeable anvil micrometer I saw advertised doesn't include all the calibration standards Don mentioned, so I'll avoid it.

Best wishes,

Chris

The piston is wobblier perpendicular to the crankshaft that it is in the parallel direction, which suggests to me that there's some significant wear. Also, the compression isn't great. But the bore itself looks fine, hence the bore gauge purchase.

Chris

Thanks for the explanation. Won't a stack be somewhat less accurate than a single block manufactured to the right size? Or is the tolerance on gauge blocks so small that it doesn't matter?

Just briefly, what's a sine plate used for?

I've not seen +0.100" oversize pistons. For this engine the options are

I haven't removed the piston yet. The wobble I can see is adjacent to the solid part, at the top.

Best wishes,

Chris

Thanks. I'll bear them in mind if I ever need another set.

Chris

I chose the vernier calipers because I wanted both imperial and metric on the same tool, and I wanted them to last. I hate replacing tools.

I've not seen a three-point bore gauge intended for measuring engine cylinders. All the gauges I found on the market were two-point with a centring device. I picked the one I mentioned because it's complete with all the anvils, locking nuts and the spanner, and is in good condition. There were more accurate gauges available (measured down to 0.0001" instead of 0.0005"), but they were much more worn and had missing parts. Also, I figured that for investigating wear as opposed to reboring, having a 40 thousandths range on the dial might be more useful than a 10 thousandths range. If people are interested, I can take a picture of the bore gauge and calipers.

Best wishes,

Chris