Pin Gage Question

[SouthernChap]

7um over nominal (about 0.0003”) in the worst case doesn’t seem that terrible. Clearly this is a plus set (all measured values were at or over nominal).

For the record, I generally machine to the closest thou (I have much less metric tooling). But working to the closest thou means I need gauges and standards that are held to tighter tolerances.

For the reason most concisely expressed in comment #42, I prefer minus toleranced pins for gauging a hole (“measuring” is common parlance but even micrometer measurements are relative to standards).

Yeah, I read that and as I said above, I wasn't even aware that plus and minus sets were a thing! I didn't know you could choose the direction of the tolerance!

I'll probably look to get a 0.5 - 5mm minus set from a supplier that shows the tolerances explicitly (so, not AliExpress) at some point.

 

[dbb-the-bruce]

The hole diameter will be 2.1547 x the middle pin diameter.

Thank you! I love this site. I've seen the three pin hole measure trick come up before, specifically with math explaining how to calculate the measured size, and my reaction was "that's a good trick, I may use it someday for holes bigger than my pins"

Now that you explained using a +.001 and -.001 leads to a constant multiplier - WOW. The trick is a whole lot more useful and I'm likely to use it. I'm going to write 2.1547 in sharpie on the inside of my gage pin case.

The other thing is we seem to get around to talking about gauge pins in a ---long--- thread at least once a year. Maybe it has something to do with cabin fever.

 

[stupoty]

Right I said I'd put the numbers up from my AliExpress sets. Here they are (late as usual, I'm crap at time estimating, just ask my boss!):

(n1 and n3 are either ends of a pin and n2 is the middle)

Specified Dia.	n1	n2	n3
0.500​	0.504​	0.504​	0.505​
1.150​	1.152​	1.153​	1.153​
1.800​	1.801​	1.802​	1.803​
2.450​	2.454​	2.454​	2.454​
3.100​	3.104​	3.104​	3.104​
3.750​	3.751​	3.751​	3.752​
4.400​	4.404​	4.402​	4.403​
4.950​	4.953​	4.953​	4.953​
5.000​	5.001​	5.001​	5.001​
5.300​	5.301​	5.303​	5.302​
5.950​	5.951​	5.957​	5.951​
6.600​	6.603​	6.603​	6.603​
7.250​	7.253​	7.252​	7.252​
7.900​	7.901​	7.903​	7.904​
8.550​	8.553​	8.552​	8.553​
9.200​	9.202​	9.203​	9.203​
9.850​	9.854​	9.853​	9.854​
10.000​	10.003​	10.003​	10.004​

I've highlighted the one big deviance; that might have been my measuring, though. I did have a 5 minute break after doing that particular pin as I think I was getting a bit bored (and thus maybe a bit sloppy!).

One thing I don't like is the little tubes for each pin in the sets. That's one extra bit of messing around before you have the pin in your mitts. Not sure if I can face making nearly 190 correctly sized wood pieces from square dowelling, and then drilling a hole in each to the nearest mm, though!

Given the very small size of the pins do you think that just holding them might have made them change size a bit ?

I know people say this about micrometers but the pins must be only a fer grams of steel.

My head is spinning at the thought of doing 190 mini holes in dowels

Stu

 

[stupoty]

All the discussion about accuracy to .0001" or tighter is interesting, but I'd like to know exactly what members are building that requires that level of accuracy. By "exactly" I mean I would like to see the actual project(s) requiring such accuracy. I understand it's nice to be able to work to those levels, but is it necessary, or are you just doing it because you can? I have so many projects either in progress or waiting in the wings that I'm rarely going to spend the time to get to that level unless it's actually needed.

I worked in a shop for over 20 years designing and building food processing and packaging machinery. 90+% of the parts only needed accuracy to .005". Tight tolerance parts would be to .001", and precision parts would be to .0005". I would guess less than 1% of the parts required an accuracy of .0001"

In the real-world time is money. I've mentioned several times in the past that we were often warned not to specify tolerances tighter than necessary. As the tolerance tightens the price increases. We were often told to remember that for each place the tolerance decimal point moves to the left the cost decimal point moves an equal number of places to the right.

Sounds silly but sometimes even the most basic thing can need a ridiculous size tolerance.

I made a replacement link for a doer-upper drill sharpener I bought (two parts were missing) , due to the way it moves it's got to have a spherical plain bearing. I finished the bores in the link with lapping as a couple of 10th's to small and the bearing was binding up quite badly when pressed and if the hole got a few 10'th to bit the bearing would just fall out.

It's much quicker as you say when you work to the loosest tolerance possible.

Stu

 

[Rex Walters]

I don’t think any of us are really arguing too strongly about working to tenths in the general case, but I will say that bearing seats and shafts in a hole do tend to require the most precision in my projects. The high speed spindle in my Quorn grinder comes to mind.

Even 0.0005” of radial clearance in a bore (a hole one thou oversized) is enough for a tiny but noticeable wobble in the fit of a shaft.

A hole that is exactly the same size as a shaft is a (light) press fit, but clearance is clearance. Working with plus size pins isn’t the end of the world, but I do believe it leads to slightly oversized holes, and it’s really a pain to put metal back on (I prefer taking some Emory to the shaft).

Even though I use pins on one thou increments and aim for on-the-thou targets, there’s a difference between drilling a clearance hole for a bolt and boring a bearing seat. I absolutely aim for better than +/- 0.0005” for the latter.

My only real concern with three pins vs two when I need to gauge a hole larger than 0.500” is complexity, not accuracy. If I need to sneak up on a 1.378” bore, I know exactly which pins to grab without reaching for a calculator, much less CAD software (I’d screw up the math even with a non-round constant multiplier).

 

[Rex Walters]

Given the very small size of the pins do you think that just holding them might have made them change size a bit ?

FWIW, I kinda divide my work into woodworking, general machining, precision machining (bearings and scraping machine ways), and optics.

IMO, body temperatures are really only a concern at optical tolerances (not my realm). But when my metal parts are uncomfortably warm to the touch, I absolutely wait for them to cool if I’m doing anything precise. It’s easy to measure a few tenths difference in even quite small parts if they are too hot to touch vs room temp.

If you care about to-the-micron level accuracy then gloves and temperature controlled labs are required, I think.

 

[stupoty]

FWIW, I kinda divide my work into woodworking, general machining, precision machining (bearings and scraping machine ways), and optics.

IMO, body temperatures are really only a concern at optical tolerances (not my realm). But when my metal parts are uncomfortably warm to the touch, I absolutely wait for them to cool if I’m doing anything precise. It’s easy to measure a few tenths difference in even quite small parts if they are too hot to touch vs room temp.

If you care about to-the-micron level accuracy then gloves and temperature controlled labs are required, I think.

Oh yeah , measuring something just after a few passes and it being a bit warm is a quick way to goof , honestly never done that , ha ha hah

Stu

 

[verbotenwhisky]

Unfortunately Accusize aren't available here in the UK. Given what I hear about them (and the fact that I have an sentimental fondness for Canada), I'd have an awful lot of stuff bought from Accusize by now, if they were.

Amazon UK didn't have much in that limited space between dubiously cheap and north of £500 for a single set when I was looking. It's better now I notice, but the sets I could justify, bear a striking resemblance to what I have, so probably from the same factories.

Consciously choosing not to buy from China is an admirable stance, I'll grant though.

I'd guess one very good reason to buy from a specialist vendor (and that means not China) though, is that you can choose whether you want a + set or a - set (which until I read this thread, I didn't know was a choice).

Looks like I ended up with + sets. At some point I guess I'm going to need to get a - set and from somewhere I can specifically choose a the direction of tolerance. Still early on in my journey as a hobby machinist but I'm guessing the 0.5 - 5mm set would probably be the one to get.

You cannot order from the Accusize website?

 

[verbotenwhisky]

Unfortunately Accusize aren't available here in the UK. Given what I hear about them (and the fact that I have an sentimental fondness for Canada), I'd have an awful lot of stuff bought from Accusize by now, if they were.

Amazon UK didn't have much in that limited space between dubiously cheap and north of £500 for a single set when I was looking. It's better now I notice, but the sets I could justify, bear a striking resemblance to what I have, so probably from the same factories.

Consciously choosing not to buy from China is an admirable stance, I'll grant though.

I'd guess one very good reason to buy from a specialist vendor (and that means not China) though, is that you can choose whether you want a + set or a - set (which until I read this thread, I didn't know was a choice).

Looks like I ended up with + sets. At some point I guess I'm going to need to get a - set and from somewhere I can specifically choose a the direction of tolerance. Still early on in my journey as a hobby machinist but I'm guessing the 0.5 - 5mm set would probably be the one to get.

I end up buying Chinesium often, more so than I want;but, I do my best to stay away from Alibaba and other sites which are actually Chinese sites. its a thing with me personally...

 

[woodchucker]

All the discussion about accuracy to .0001" or tighter is interesting, but I'd like to know exactly what members are building that requires that level of accuracy. By "exactly" I mean I would like to see the actual project(s) requiring such accuracy. I understand it's nice to be able to work to those levels, but is it necessary, or are you just doing it because you can? I have so many projects either in progress or waiting in the wings that I'm rarely going to spend the time to get to that level unless it's actually needed.

I worked in a shop for over 20 years designing and building food processing and packaging machinery. 90+% of the parts only needed accuracy to .005". Tight tolerance parts would be to .001", and precision parts would be to .0005". I would guess less than 1% of the parts required an accuracy of .0001"

In the real-world time is money. I've mentioned several times in the past that we were often warned not to specify tolerances tighter than necessary. As the tolerance tightens the price increases. We were often told to remember that for each place the tolerance decimal point moves to the left the cost decimal point moves an equal number of places to the right.

I'm building a time machine... I want to go back to my child hood and ... oh crap... I'd probably make the same mistakes all over again...