The Home Machinist!

Standard industry specifications for accuracy of metrology equipment is + - one count. Actual measuring and marking out are totally different. And you said measuring. Adding a 10ths indicator zeroed to a surface plate will still only allow height measurements to the accuracy level of the tool and it's vernier scale. Using an independent method of setting the indicator to a known zero height like gauge blocks and then comparing your part height to that is then a factor of the accuracy of the surface plate, those gauge blocks and the indicator. The height gauge is then just an easy way of holding and adjusting the indicator. It's a bit more complex than that because of temperature variations, uncertainty of measurement etc, etc, etc. But ignoring those to make things simple, with those additions to your height gauge you can do measurements far more accurate than the tool was designed to do. But you wouldn't then be using the tools markings to do it.It becomes just a transfer stand. Better than a thou just using the tool as it is? Maybe on a really good day with everything perfect, but I wouldn't bet money on which half thou increment is the correct one. Or maybe even to a thou if you read the next part.

Not to steer this thread off course since it's unrelated to what you were asking, but it's still very much related to the tool you have. And as Harold mention there wasn't a chance of marking out to better than a thou. I was re-reading one of my books a few day's ago written by T.D. Walshaw and ran across something that should have caught my attention the first time I read it. He gave a set of numbers obtained from a test run by a technical college in the U.K. (I'm guessing that test might have been run possibly in the 1960's or early 70's?) Anyway that test was supervised by a "very skilled instructor" and the 12 apprentices "had some years of experience". They were asked to lay out, drill and ream 6 holes on a pitch circle. Unfortunately the pitch circle size or hole size wasn't listed. 4 were to use normal marking out methods with a scale and dividers, 4 were to use co-ordinate marking out methods, (I'm assuming a vernier height gauge much like your Mitutoyo) and 4 were to use a compound table to mark out, drill and ream the holes. The AVERAGE error was obtained but not detailed with "very sophisticated measuring equipment".
Normal Marking Out .0112"
Co-ordinate Marking Out .0080"
Co-ordinate Drilling and Reaming .0033"

Walshaw goes on to mention that by average standards these were skilled practitioners doing similar work all day long "so it may be assumed that the amateur will be worse not better". He also states that the machine used to do the drilling and reaming was well past it's prime, and that with decent condition equipment and some care many with a home shop could do even better with the co-ordinate drilling and reaming. But those are the numbers Walshaw gave. They surprised me as I would have thought the average errors of the first two would have been quite a bit less than they were. I've long thought that marking out was worthwhile to reduce machining mistakes, but gave up center punching drill locations a long time ago. It's accurate enough for some items of lower accuracy on a drill press, but I'll use co-ordinates on the mill if the part has a decent level of accuracy required. Those test numbers seem to indicate that's correct. And remember those numbers are average's. All 3 test's are also based partly on the accuracy of the machine used to do the drilling and reaming.

earlgo wrote: ↑Tue Dec 04, 2018 1:14 pm vhg.jpg
Your height gage must be different from this. The bottom of the scriber holder is most likely not able to be zero if it were able to reach the surface plate. Adding a tool bit or gage block would still not fit your description. Turning the scriber over would not work either.
Am I not able to understand your explanation? Maybe I pulled an incorrect picture of a VHG.
Perhaps a picture would help?
--earlgo

OK, went and looked at mine, which appears to be a bit more elaborate than the one pictured. Certainly my scriber's offset is much less than the one pictured and, the bottom of the measuring beam (fixed horizontal part of the vernier slide) cannot be set onto the surface plate. My baaad.

However, the main vernier scale on my instrument can be adjusted up and down by 1/4" or so in the main body so that it can be set to read zero to wherever the measuring beam happens to be within the capability of the instrument. On my vernier, with the beam at an elevation of 1.000" above the surface plate, the instrument is easily set to read zero.

Therefore, I can lower the measuring beam onto a 1" gauge block or a 1" precision parallel (I just did that) and adjust the main scale position up and down such that the vernier reading is exactly zero. Having done that I simply deduct 1.000" from any measurement taken as described earlier, above.

Alternatively the work piece may be set onto 1" parallels to eliminate that step.

Come and think of it, this may be preferable (less error) if your parallels are not exactly 1.000". In which case use the same parallels to set the measuring beam to zero, and to raise the work piece. w

After looking at my instrument's scale with a magnifying glass I must admit that guessing at 1/4 thou is wishful thinking but, reading the scale to 1/2 thou is possible. Whether the instrument warrants this is another question. If someone were to try this and compare the readings to a gauge block stack, is would be appreciated if you were to report here. w

Inspector wrote: ↑Tue Dec 04, 2018 1:22 pm Wolfgang that may be okay for you in your home shop but it wouldn't fly if you worked as a quality inspector like I used to. You had to use the appropriate tools. No hair splitting permitted. I still do it this way for the stuff I hack out for myself.

When I worked I used height gauges and an indicator to get a zero and then transfer to gauge blocks or a height master to get an actual number. That sort of thing is usually done now with a CMM but there are times when it is quicker to do it the old fashioned way. It's a good technique to have in your skill set.

Pete

I taught metrology in college and am fully aware that this would not fly in any kind of business. We are, after all, talking about a home work shop, right? .) w

The problem with splitting is that you don't really "know" what you're seeing, so the resulting calculated dimension may or may not represent the actual measurement. If you're working to tenths, it simply wouldn't be adequate.

This is much the same as using a caliper, be it vernier, digital or dial, for close measurements. You can assume you know the size, but it's not necessarily representative of the actual dimension.

There was a bit of a rumble in regards to using calipers in inspection where I was trained. Because they are not truly definitive, parts could NOT be accepted or rejected based on such measurements. They were used routinely, of course, but if a dimension was questionable, it had to be determined by other means. That was a good lesson to learn, so critical dimensions weren't established based on the use of calipers. That still applies today.

H

W:
If one could conjure up a clamp to affix a tool bit or gage block to the bottom of the scribe, then your idea would work. I failed to imagine that before I posted. Sorry I got off track.

Just for fun, here is a pic of my Po'mans Height gage. It works for my amateur efforts. --earlgo

With a magnifier (my eyes are shot) it's easy to read a good Starrett or other vernier height gauge to half a thou.

I'm glad your eyes aren't too shot.

Harold, thanks again for your explanation/
Yesterday I got back to Kelowna after a trio
down to Oroville with a sleigh load of goodies.
One of them is a Fowler height gauge from maybe
the 50's or 60's in excellent condition for its age.
I'm at a computer at work right now so I can't post a pic.
Will do that later tonite or in the morn.
I put it on our surface plate,here.
From what I can see it reads .025" per rev. exactly
like a micrometer. But ... each of the 25 has 10 lines between
which are the 10 ths. AND ... on top of the 250 lines vertically
on the thimble there are another 10 lines (vernier) on the body.
Could that be 100 thousandths?

Ok. Here's some pics.

That's very much like a Cadillac!
In regards to the vernier scale, I'm inclined to think it is for hundred thousandths (tenth of a tenth) but I don't know that I've ever seen one like it. I rarely use my Pla-Chek, and I don't recall it having the vernier. I'll take a look when I'm in the shop next time to see.

Nice acquisition, by the way. While you rarely need that level of precision, it's nice to know that you can deal with it when necessary.

H

A nice score by the looks of it Wally. :-)That Fowler gauge was very likely built by one of the better known high end metrology company's for Fowler. Back then the Fowler name seemed to have had a much better reputation than the re-brander of mostly lower end off shore goods they now slap there name on. That said there still seems to be "some" Fowler branded goods today at the upper price range that are of pretty good quality and accuracy. But if I'm going to pay that much for new I'll go with Mitutoyo, Starrett, Mahr etc. and know who to deal with for spare parts or service if needed.

CMM's seemed to have almost replaced a lot of the high end inspection equipment like these Cadillac type gauge's, squaremasters etc. But for a small job shop / home shop a height setting gauge like yours is a great replacement and time saver of having to building up a stack of gauge blocks to do the same. But if you do have a very good and trustworthy set of those gauge blocks then double checking your new gauge against maybe a half dozen gauge stacks would help verify if it's still within it's allowable tolerance range. At least that would let you know if you can or can't trust what the gauge say's it should be. If anyone would know more about that height gauge or who may have made it possibly an email to Long Island Dial Indicator Repair might get you some further information. Or maybe a post on the PM Metrology forum would get someone who would recognize it. Even doing a Google image search shows there were possibly dozens of company's who made gauges like that. Sending it in to a calibration lab for a re-certification and accuracy check is I've heard quite expensive for these types of gauges due to the time involved to check it's full height range.

50 years ago I was a newbe tool designer and the tools, mostly part holders with sensitive electrical connections, were built by very qualified tool and die makers. A good share of what I know about machining was learned by watching them. Every granite table they used had one of those gauges on it. I've learned to make measurement by other techniques but always wanted one of them. Are they just called a "Fowler Gauge". I searched for that and got lots of hits including one for the Fowler tool company Produce Catalog. However it did not show a picture of anything that looked like that. A long the same lines it seems the term "Cadillac" is being used. Is that term somehow related?