It is actually pretty common to turn thin walled parts in a three jaw that will show round when the part is actually triangular. A plug gauge and a two point gauge will show this. If it gauges larger with the two point gauge than it does with a plug gauge then there's a good chance it's a three sided bore. These shapes are called REULEAUX traingles. You can draw one by first drawing an equilateral triangle. Then draw three intersecting arcs using as the radius the length of one side. This will give you a a rounded triangle with a constant width. ERS
I've always liked Standard gauges. Not as good as Sunnen though. I had problems with several Standard gauges being sticky. Part of the problem was wear from being used by the hone dept. There were scratches on the moving parts as well. After sleeving (with aluminum nickel bronze) and polishing out the scratches they were once again accurate enough to measure to .0001". But before repair you really had to be careful. ERS
There's a variation of this technique in Guy Lautard's Machinist's Bedside Reader #1, using a gage rod inside the bore that's just a few thou shorter than the bore. The side-to-side movement of the rod is measured with a ruler, in sixteenths (!) right at the bore's edge--to calculate the difference between the actual bore, and the gage rod's length in thousandths. That sixteenths measurement can be done to the nearest 64th, just expressed as decimal--i.e., 15/64ths = 3.75 sixteenths.
Instead of a rod, I'd use a flat plate cut to that same top hat shape, whose edges are milled parallel and beveled to almost a point. Place it in the bore snug to one side and scribe a thin line with a knife on the outside face, then tilt the plate over to the other side of the bore and scribe another line. Get the measurement in sixteenths, and plug it into this formula:
C= A squared, divided by 2B
where A = number of 16ths of an inch of the plate's side movement at one end (to the nearest 64th, but expressed as a decimal sixteenth)
B = width of the plate, in inches, measured as accurately as possible
C = the diff in thousandths of an inch between the actual bore and the plate's width
You're supposed to get to a half thou accuracy with this, assuming the plate's edges are parallel to that tolerance as is your measurement of it's width.
A number of years ago I had occasion to work with a few guys doing bore distortion tests. The indicating mechanism was attached to the cylinder block. It advanced down the bore while it rotated, took data, and plotted out the bore shape. This was done at 2 temperatures. I can't recall all the details except we had to swab the bores with glycol to keep them from frosting up before they were measured...
The Enco gages are too sticky for repeatable precision. Get a Starrett in the size you need. Then practice with it.
I once knew an old machinist who could do .0002 repeatedly with a simple spring-leg inside caliper and a mike. I asked him how he did that. He said, "feel", thus exhausting his conversational quota for the month.
Guy Lautard describes a "rocking" technique in one of his Bedside Readers that works pretty well without relying so much on feel.
That would make a huge difference. I've handled such gages and they do make measuring more difficult.
I'm partial to the sets that telescope from both ends, keeping the handle towards center. My Starrett gages are so built, and the lock is made to function properly. I trust them for close work and don't get let down.
A Moore & Wright or
Yet you claim they won't work to .0002"? My expectations from my tools are much higher, and I"m not disappointed.
I, too, own several dial bore gages, all of which are Sunnen. Lovely tools, but limited for use in shallow bores.
I have an expanding small hole gauge. A split
I have a similar set, again, Starrett, but mine are the half ball, so they are affective in very shallow bores. My claim to fame was in producing small work when I was in business, so I tooled up appropriately. Again, I trusted my small hole gages when working in tenths. They are as good as the hands in which they are placed. If you learn the procedures, they are capable.
Perhaps, but are these better ways in the future for many of us? It's like a CNC versus manual machines. Knowing how to run manual machines and having the skill to do it properly is something that will serve you for ever, and will serve one well without spending huge amounts of money. That, to me, is far more attractive as a home shop type, which I now am.
You can't be serious! When I worked as a grinder, I was expected to read a micrometer within .000050", and did so routinely-----and could verify my readings with a Sheffield Shadowgraph. I recognize that the surfaces involved were far better than machined surfaces, which makes a huge difference.
and well beyond what most people would
I claim nothing more than having mastered my trade, at a level that allowed me to survive in the world I knew, which I did. If one can't read T gages to .0002", it's a lack of skill, or poor gages. Decent gages are capable.
Understand that there is a huge difference between building race engines and producing tooling for the aero-space industry. That's not to diminish in any way the importance of your work, nor the level of skill involved, but they are two very different arenas. In order for me to have been successful (which I was) I had to master certain things, and master them well. You, no doubt, have skills that I have yet to master, and may never. I wouldn't expect an automotive machinist, *any* automotive machinist, to step in and do the work I did, no more than I'd expect to step in and do the work they did. We are not trained in the same way, and in many instances the work bears little resemblance when compared.
When running a centerless grinder the same thing can occur when running on center. It's important to be above, or below, at all times, which eliminates the condition. A part so ground will mic perfectly round, yet be three sided.
As simple as his reply was, it spoke volumes. It's something one learns------and for those that have the capability to apply it, it's no big deal. From reading these comments, I'm beginning to wonder if it's something that many may never master.
Dunno, Don. The rocking technique, to me, is a part of the "feel" that makes it work. What Eric had to say about applying telescoping gages is pretty much right on. It's a matter of doing it enough times until you understand what you are "feeling".
It was mentioned earlier that it is rather difficult to use an inside mic with the extension handle in deep bores for one can't reach the adjuster. I've always set the inside mic to an outside mic, maybe a thou undersize, then see if it slides in. If it's loose, I'll open it a half thou, and try again. If it's too tight, dial down half the original adjustment. Repeat the process until you get the desired feel. I always thought of using this process as turning the mic into a go/no go gauge, and for a half thou tolerance, this is certainly accurate enough. With practice, a couple of tenths is possible, for those bearing fits. Ron
the problem with all indirect measuring techniques is that you are accumulating error. If you have a tenth of potential error in the measuring end of telescoping gauges, a tenth in the mics, unless you are using better mics. Cheap gauges have other sources of error.
IF it is really important buy or borrow good equipment. Learn to use it so you can get repeatable results. IE measure a bore until you can get the same number 10 out of 10 times. That gets you repeatable. Compare those measurements to a standard and you get accuracy.
Telescoping gauges are ok, but do require that 'feel' twice; that takes much time to develop.
Ken G. described the "Lautard Rock" above. It's a bonk-bonk method, and you measure the side-to-side displacement between bonks. It gets accuracy by virtue of geometry: for a small rocking angle about 90 deg, horizontal displacement is much larger than vertical displacement. For total rock of +/- one degree, vertical measurement error is about 0.87% horizontal error.
I don't think good quality telescoping gages are all that hard to learn to use, but cheap ones (Enco imports) are useless. Ya can't get them snugged without disturbing fit and feel, and they're sticky so they don't work smoothly but jump several tenths or more when trying to get a fit. This is one place where a beginner should use good tools so he's overcoming deficiencies in technique rather than being thwarted by tool deficiencies.
I've used these, I have a small selection of them. I get the impression that they are somewhat less accurate when fully expanded. For example, the .3 to .4 inch one seems like it's really opened up too far at .400 and that the ability to measure at that limit might be impaired.
Also the finish on the ball ends seems so-so.
Did you ever polish them to improve their function?
What other tricks can you suggest for using them to the fullest accuracy?
I've never touched mine aside from using them, Jim. What it takes is considerable practice, and something against which you can verify your readings until you know the feel. Plug gages are quite helpful if you have them at your disposal, something like a set of Deltronic pins, graduated in tenths.
Regards how they often are somewhat wrong for the bore at hand, I agree. When you have them at the top end, you become *very* reliant on the feel, which is very light. You must transfer that same feel to the mic when making your reading, and that takes considerable practice. Again, it's a matter of doing it until you know when you're there, or not.
I'm of the opinion, especially now, after reading some of these comments, that some people are just plain too damned ham handed to ever get it right. Doing fine, precision work involves more than just understanding how to run machines. It's something that comes from within. I've worked with some good hands that struggled with tough projects, yet were fairly good on the machines. They seemed to lack that killer instinct that sorts out those that can from those that can't.
Some time ago one of the readers here posted that they had spent a half day in the shop of a guy that was a toolmaker, watching him grind the ways of a machine, if I recall. His comments rang true. It's not what a good machinist/toolmaker does, but *how* he does it that makes the difference. It is that quality that many lack, and will likely *never* have it. You don't learn it, *it comes from within*.
Right on! That's exactly how to use them. If you measure over the inside mics, you can do work within .0002" and know you're there. Because the inside mics have a tendency to not repeat exactly (because of removable members) it's a good idea to not trust the direct readings, although mine seem to be pretty reliable.
In essence, it's damned hard to do tenths work on a machine that has thousandths of error. Very nice comment, Don.
I've always tried to encourage guys to learn the basics, to not take shortcuts. That way, when things don't work out as planned, you can evaluate the results and come to a conclusion as to why they went wrong. By taking every short cut known, you add so many variables that it's often impossible to determine the reason. And so it is with poor tools. Was it you, or the tool? A great example is measuring precision fits with a caliper, *any* caliper, digital, vernier or dial. They simply are not precise enough for precision measurements.
Buy good measuring instruments. You'll have them for your lifetime and use them with pride. However, an inexpensive import micrometer is better than
*no* micrometer. Some folks can only do so much with their limited funds.
When I worked in production, that was the inevitable result when a new guy was hired on. He would invariably measure the ODs on parts as too small, and the bores would be measured larger than they really were.
The amount of error told one how long they would last working at that job. If it were a thousanth or less, they would be long-term candidates. If it were three or more, they would be gone inside a month.
They had the feeling that the outside micrometers were some kind of clamping device, crank that sucker down till you get the reading you need.
The bore mikes were treated like some kind of elaborate tailpipe expander. If the bore wasn't big enough, you *make* it big enough!
This is one reason I semi-jokingly suggested to the original poster that he *try* a spring-leg caliper. Those things teach
*feel* more than anything else. Old time machinists didn't even measure a number when fitting a shaft, they just used an ID caliper to check the bore, transfered that to an OD caliper, and used that to check the shaft as it was being turned.
I keep a set of inside/outside spring leg calipers here at work. It's suprising to see how there are certain jobs for which they are still ideally suited.
Another reason to measure over the ID mikes is as a sanity check on the reading. I've always found the starrett ones read right on, but because the thimbles read the other way around, it's easy to get a number that is 25 thou off. Putting them inside an OD micrometer does a nice job of catching that mistake. Most folks can read the OD mike much more reliably than the ID one.
On Sun, 13 Feb 2005 22:40:31 -0800, the inscrutable "Harold and Susan Vordos" spake:
How precisely, in your eyes, can a dial caliper measure, Howard? I've always used them for comparative measurement vs. interference fit measurement, and that has always worked for me. What do you see as the slop factor for, say, a 6" import dial with a 0-100-0/.001" face?
Wha? Some people have actual _funds_? ;)
---------------------------------- VIRTUE...is its own punishment
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.