Making a silk purse out of a sow's ear, or...

Topics include, Machine Tools & Tooling, Precision Measuring, Materials and their Properties, Electrical discussions related to machine tools, setups, fixtures and jigs and other general discussion related to amateur machining.

Moderators: GlennW, Harold_V

Torch
Posts: 1684
Joined: Wed Mar 17, 2010 7:58 am
Location: Muskoka

Making a silk purse out of a sow's ear, or...

Post by Torch »

... tuning a depth gauge?

I bought a used Scherr-Tumico depth gauge on eBay. It came from a seller who mostly handles clothes, but was only $25 so I decided to risk it. Got it yesterday.

It came in a nice wood box with old certification labels attached and magic marker labelling suggesting it dates from 1932. Also, it is stamped "U.S. GOVT. PROPERTY" on the inside of the lid. Outwardly in fine shape for an old instrument.

On closer inspection, the instrument has been mixed up with another set. The instrument and the installed rod both bear serial number 67597. The remaining rods all bear serial number 87915. The 8" long rod (for 4" to 5" measurements) is duplicated. The 7" long rod (for 3" to 4" measurements is missing. The installed #67597 rod is within 0.001" of a true reading (against two 123 blocks on a surface plate). The #87915 rods are all approximately 0.110" too long for this instrument -- either the specification was changed or this base was reground for some strange reason. If it was re-ground, it was highly polished afterwards and is perfectly flat, although the barrel is 0.002" out of perpendicular from the base from top to bottom, as checked with feeler gauges against a square.

I contacted the vendor, he does not have the other set.

The rods are adjusted to the instrument via a threaded end and pair of locknuts. There appears to be sufficient thread to take up the 0.110" excess. However, the cap of the barrel clamps down on the protruding end of the rod, so adjusting the locknuts to shorten the rod reduces the number of threads that the cap will engage and by the time the full 0.110" is take up, I don't think the cap will catch even the first thread! The rod with the matching serial number has a shorter threaded section, suggesting someone ground off the excess threads in the past.

My thoughts thus far:

Option 1: pitch the thing and consider it a $25 lesson in eBaying. The cost of shipping it back wouldn't be worth it.

Option 2: As the #87915 rods I have are all too long, I can re-grind them shorter and recalibrate them to work with this instrument, including shortening the 8" rod to 7". I wouldn't trust myself to hand grind this, I'm thinking chucking them in the lathe and mounting a tool post grinder to keep things square. I would use a dremel to lop off an inch from the 8" rod first.

Option 3: As the #87915 rods have enough thread to adjust them, I can make the adjustment and grind off the excess thread on the back side. The 8" would have to be shortened first, as above.

Option 4: Whomever ended up with instrument #87915 happens to read this before I do anything, and I swap him for his short rods.

Thoughts and suggestions?
P7110643-800.jpg
P7110644-800.jpg
P7110645-800.jpg
PeteH
Posts: 1065
Joined: Mon Jun 11, 2007 10:49 pm
Location: Tidewater Virginia, USA

Re: Making a silk purse out of a sow's ear, or...

Post by PeteH »

Interesting. I bought a set of those - also from a FleaBay vendor, quite possibly the same one -- and there are no serial numbers on any of the parts; also, the date on some documentation in the box was "1980". And the box itself is fairly rough.

I wonder how many of these things our gubmint bought, anyway ?
Pete in NJ
JackF
Posts: 1616
Joined: Mon Jan 26, 2009 3:56 pm
Location: Caldwell, Idaho

Re: Making a silk purse out of a sow's ear, or...

Post by JackF »

In my unprofessional opinion option 3 sounds best to me as long you have some way to accurately measure your results to your tolerance requirements. :)

Jack.
Torch
Posts: 1684
Joined: Wed Mar 17, 2010 7:58 am
Location: Muskoka

Re: Making a silk purse out of a sow's ear, or...

Post by Torch »

PeteH wrote:Interesting. I bought a set of those - also from a FleaBay vendor, quite possibly the same one -- and there are no serial numbers on any of the parts; also, the date on some documentation in the box was "1980". And the box itself is fairly rough.

I wonder how many of these things our gubmint bought, anyway ?
The serial number on the rods is tiny and hard to read. I needed good light and a magnifier to read them. The one on the instrument itself is right on the base and easy to spot. If yours doesn't have one, then I suspect they stopped applying them by 1980. BTW: they still make the same set, but at 10x the price I paid.

I circled the locations on this photo, but by the time the resolution is reduced to comply with the forum rules, they are impossible to read:
serial_no_locations.jpg
User avatar
BadDog
Posts: 5131
Joined: Wed May 17, 2006 8:21 pm
Location: Phoenix, AZ

Re: Making a silk purse out of a sow's ear, or...

Post by BadDog »

I think you are correct in your guess that the base has been ground down. Normally they have a beefier cross section. And Sherr-Tumico made good stuff, so I would comfortably wager their QA would not have let out a tool that far from square. And, for truly accurate (with expectations for such instruments) measurements of the distances implied by the rod length, that much error would yield unacceptable error in measurement. That said, the cosine error would be tiny enough for most work to be unaffected.

So as far as fixing it, I would ask myself a few questions. First, is it a rotating rod design. If so, I wouldn't even start to consider fooling with it. Next, will the error present a problem for you? If so, do you have the skill or (more importantly) interest in trying to square it? If I made it this far and had not rejected it completely, I would say that #3 is a viable approach and could be quite satisfying, though probably not practical considering the cost of acquiring a complete and unmolested set.
Russ
Master Floor Sweeper
Torch
Posts: 1684
Joined: Wed Mar 17, 2010 7:58 am
Location: Muskoka

Re: Making a silk purse out of a sow's ear, or...

Post by Torch »

BadDog wrote:First, is it a rotating rod design.
Yes.
If so, I wouldn't even start to consider fooling with it.
Ok, why?
User avatar
BadDog
Posts: 5131
Joined: Wed May 17, 2006 8:21 pm
Location: Phoenix, AZ

Re: Making a silk purse out of a sow's ear, or...

Post by BadDog »

Imagine what happens as you approach touch down. Your not only dealing with "feel" of the thimble, now you are dealing with rotational friction at the tip (which may vary largely by finish and material), and possible torsion effects of the long rod. Whether it makes a difference in the real world or not, when I was equipping measuring gear I noticed the different types, and decided to avoid the rotational models. By paying attention, the rotating models vs non-rotating are about the same price anyway.

Also, I have one I bought cheap that has the rods ground for slots. Obviously that wouldn't work if it rotated, but that wasn't the reason for my comment.

Anyway, if it was good-to-go and cheap, I wouldn't worry too much about it, particularly for my needs. But to do that much work on it compared to the price, no way.
Russ
Master Floor Sweeper
Torch
Posts: 1684
Joined: Wed Mar 17, 2010 7:58 am
Location: Muskoka

Re: Making a silk purse out of a sow's ear, or...

Post by Torch »

BadDog wrote:Imagine what happens as you approach touch down. Your not only dealing with "feel" of the thimble, now you are dealing with rotational friction at the tip (which may vary largely by finish and material), and possible torsion effects of the long rod.
At the moment, I have the rod that sticks out of the end of my calipers. I think the best that can be said is that it's non-rotational. :lol: Ok, seriously, I can see your point, but given the instrument has a 1 thou resolution would the torsional effects be significant (compared to something graduated in tenths)?
Anyway, if it was good-to-go and cheap, I wouldn't worry too much about it, particularly for my needs. But to do that much work on it compared to the price, no way.
Good-to-go and cheap would have been ideal. Actually, I'm thinking I'm going to attempt to tune these up one way or the other. If I am going to consider the money wasted and pitch them in the scrap metal bin, then I can't hurt anything and I could learn something in the process. If I screw them up, I'm not out any additional money over what I already wasted. If I do happen to be successful, then it would be like making my own, but without all the extra work.
User avatar
BadDog
Posts: 5131
Joined: Wed May 17, 2006 8:21 pm
Location: Phoenix, AZ

Re: Making a silk purse out of a sow's ear, or...

Post by BadDog »

Yes, there was a time I would have thought differently. But these days my shortest resource is time and energy. With some time on a tool or surface grinder, or perhaps even by hand (though I would hate to think how much), you can likely correct the angle. Or just choose to live with it. Shortening the rods shouldn't be bad at all, particularly if you can borrow some time on a tool (or surface) grinder.
Russ
Master Floor Sweeper
Torch
Posts: 1684
Joined: Wed Mar 17, 2010 7:58 am
Location: Muskoka

Re: Making a silk purse out of a sow's ear, or...

Post by Torch »

BadDog wrote:And, for truly accurate (with expectations for such instruments) measurements of the distances implied by the rod length, that much error would yield unacceptable error in measurement. That said, the cosine error would be tiny enough for most work to be unaffected.
I was using feeler gauges, so the 0.002" difference could actually be almost 0.003". Using 3 thou as a worst case estimate, over 1", that's a deviation of 0.17189°. In the case of the 9" long rod, the theoretical error at the centre of the rod would be -0.00004". In the case of the 3" long rod, it would be -0.00001". Of course, that is to the centre point of the rod, which is 1/8" diameter. Half of that is 0.0625" and if I managed to make the tip truly perfectly square, the tilted tip would introduce an error of +/-0.000188". Let's call the compounded error +/- 0.0002", on an instrument with a resolution of 0.001". It could actually be in spec. Maybe I should just leave the base alone?

Ok, now I'm getting obsessive, but I just thought of something: the box says it was originally US Government property, so it must have met a US Government spec. Hang on.

Google pulled up "GGG-C-105C". I scanned down to page 20 and found Table XII. A 0 to 6 depth micrometer must have an overall accuracy of 0.0004". But perhaps more importantly, the preceding page specifies "surface of the rod seat on the barrel end of the base shall be parallel with the face of the base". This instrument does not have flat mating surfaces for the rods -- they are spherical: a ball and socket joint. And the hole in the base has 0.002" clearance for the rod. I just realized that the majority of the error can be eliminated by assembling the rod with the instrument vertical -- in fact, assembling it lying horizontal on the table as might otherwise seem natural could introduce some error.

A bit of experimentation with a pair of 123 blocks on a surface plate confirms there is little or no difference when orienting the base in different directions compared to the error I get staying in one spot. The measurement varies between about 1/2 and 3/4 of the way between 0 and 0.001" even when I don't move anything. In other words: at the moment, I am the source of most variation! :lol: I see what you mean by dealing with the feel of touching down -- it's a bit different than inside or outside micrometers. More like a height gauge. It is very easy to lift the tool (or flex the rod?) slightly. I will need some practice.
dalee
Posts: 68
Joined: Sun Mar 11, 2007 8:54 am
Location: Minnesota

Re: Making a silk purse out of a sow's ear, or...

Post by dalee »

Hi,

Scherr-Tumico did not exist until the early 1940's, so it can't be from the '30s. You have a tossed together mashed potato set. I wouldn't even say the wood box is matched to the base. At least my S-T box from the late 80's doesn't match that style. And the base looks to be a newer style also. Maybe 70's?

S-T still sells replacement rods for your mic if you are interested.
More Speed, More feed!
Inspector
Posts: 721
Joined: Mon Feb 23, 2009 4:25 am
Location: Saskatoon, SK, Canada

Re: Making a silk purse out of a sow's ear, or...

Post by Inspector »

You might find that a gentle tap of the barrel with a small soft face mallet might be enough to knock it back to vertical. It may have been dropped at some point in its life.

If you decide to look for a new, better one, the base can be modified by attaching a dial indicator to it and making a dial depth gauge out of it. :wink:

Pete
Post Reply