My 875/0 square is an Inox, best thing to do when you buy a reference square is send it for calibration then you know you have a good reference device. Rub it with a Lanolin substance to prevent it rusting and put it in a box. Now I only this for checking other squares, it does not see machine duty.
-
Welcome back Guest! Did you know you can mentor other members here at H-M? If not, please check out our Relaunch of Hobby Machinist Mentoring Program!
Help for a newbie looking for measuring tools?
- Thread starter Mauser lover
- Start date
- Joined
- Feb 13, 2017
- Messages
- 2,138
A comment here from a (retired) maintenance man might help a noob getting started. The maintenance man is looking to get it (the target machine) back to running. Absolute accuracy is secondary. I usually didn't have the drawings anyway. When I make a (singular) part, it is to fit the machine as worn in. It might be three months wear, it could be thirty years wear. To Fit is the primary target.here. As such, the accuracy(absolute) of a measuring instrument is not so important. The precision(repeatability), on the other hand, is. Highly so.
A caveat here; Most of my professional background was in electrical and electronics. Two decidedly different crafts there. But precision is precision, whatever the craft. BTW, my calibrations were long before computers. And my calibration standards were a factor of ten, usually NBS tracable. The Cal Lab I used was one generation removed from the NBS. When I had to make a mechanical part, it usually had the same accuracy as the electrical parts had. A much too tight tolerance. But that's how I was used to thinking.
I realize that professionally I don't fit in well here. But in my day, basic knowledge of machinist's skill was a part of my profession. Building from scratch a D'Arsenval meter movement was as precise as that of a machinist.
A caveat here; Most of my professional background was in electrical and electronics. Two decidedly different crafts there. But precision is precision, whatever the craft. BTW, my calibrations were long before computers. And my calibration standards were a factor of ten, usually NBS tracable. The Cal Lab I used was one generation removed from the NBS. When I had to make a mechanical part, it usually had the same accuracy as the electrical parts had. A much too tight tolerance. But that's how I was used to thinking.
I realize that professionally I don't fit in well here. But in my day, basic knowledge of machinist's skill was a part of my profession. Building from scratch a D'Arsenval meter movement was as precise as that of a machinist.
Bill Hudson
My last full time job was in the repair centre of NATA accredited calibration lab. I was also authorized to perform calibration on pressure, temperature, RF and anything electronic. I totally agree with your precision (repeatability) over accuracy.
We had a couple of standards that were out of tolerance for absolute accuracy (due to drift over time) but we did not adjust them as we had 15 years of data and knew exactly what the drift was, so we could easily correct for the absolute accuracy, where as you could not correct for repeatability.
We had a couple of standards that were out of tolerance for absolute accuracy (due to drift over time) but we did not adjust them as we had 15 years of data and knew exactly what the drift was, so we could easily correct for the absolute accuracy, where as you could not correct for repeatability.
- Joined
- Feb 9, 2017
- Messages
- 5,244
While I noted that in what Bill said, this is a very curious statement to me. I never would have thought accuracy and repeatability to be somehow separate. When Bill mentioned it I put it in the old memory banks as something that might become clear at some further point. But with the agreement by warrjon I guess if someone would care to elaborate I would think it a worthwhile addition to my and possibly others understanding.
- Joined
- Apr 12, 2011
- Messages
- 3,472
What are the differences between accuracy and precision?
In other words, accuracy describes the difference between the measurement and the part's actual value, while precision describes the variation you see when you measure the same part repeatedly with the same device.
In other words, accuracy describes the difference between the measurement and the part's actual value, while precision describes the variation you see when you measure the same part repeatedly with the same device.
I need repeatability, most of the parts I make are fitted to another part or 7 etc. I use pin gauges instead of a dial caliber to determine actual hole sizes, they are consistent and give me a good base line always. Brown and Sharpe depth gauges work for me, bought used on eBay. But any decent brand of older tools is fine from my experience.
I have not liked the Asian machinist levels from Shars though, I prefer the Sterrett levels myself. But, you don't need anything super expensive for a hobby lathe IMO. I know guys who produce fine things on old worn out machines without any fancy levels. You gotta get your machine cutting straight regardless of level.
I have Harbor Freight measuring devices and most of the name brands. My sterrett and my HF micrometers actually read the same down to .001 to .0005. I have one HF micrometer that is a "premium version" and it woks good. The HF dial calibers are too cheaply made for me, I have tried them a couple times lately and they were not durable, not smooth. The QC is not there. The electronic ones fell apart and ate batteries.
I like the Shars better for a budget dial caliber, like this one, its cheap but not junk IMO: https://www.ebay.com/itm/SHARS-6-00...610458?hash=item45ff75241a:g:howAAOSwCHBbBDyZ
But, you just need what you actually are going to use, most of my measuring tools go unused for months. Then I have the ones I use almost daily. And you need the measuring tools for daily quick use around coolant and oil and the tools for fine clean work that you take perfect care of. I dropped a dial caliber a few months ago almost full open, bent it bad. Thankfully it was a $30 no name type. Replaced it with the Shars Premium one, haven't dropped that one yet.
I have not liked the Asian machinist levels from Shars though, I prefer the Sterrett levels myself. But, you don't need anything super expensive for a hobby lathe IMO. I know guys who produce fine things on old worn out machines without any fancy levels. You gotta get your machine cutting straight regardless of level.
I have Harbor Freight measuring devices and most of the name brands. My sterrett and my HF micrometers actually read the same down to .001 to .0005. I have one HF micrometer that is a "premium version" and it woks good. The HF dial calibers are too cheaply made for me, I have tried them a couple times lately and they were not durable, not smooth. The QC is not there. The electronic ones fell apart and ate batteries.
I like the Shars better for a budget dial caliber, like this one, its cheap but not junk IMO: https://www.ebay.com/itm/SHARS-6-00...610458?hash=item45ff75241a:g:howAAOSwCHBbBDyZ
But, you just need what you actually are going to use, most of my measuring tools go unused for months. Then I have the ones I use almost daily. And you need the measuring tools for daily quick use around coolant and oil and the tools for fine clean work that you take perfect care of. I dropped a dial caliber a few months ago almost full open, bent it bad. Thankfully it was a $30 no name type. Replaced it with the Shars Premium one, haven't dropped that one yet.
To give you an example.
I calibrated a lot of Fluke 65F's for defense, the calibration standard used was a Raytek MX4 which was calibrated by NMI (National Measurement Institute), the Raytek was IIRC about 3°C high at 100°C but its repeatability (precision) was always within 0.5°C which is about 1/2 its tolerance. So we knew if the Raytek read 103.5° the actual Blackbody temperature was 100.5°C ±0.5°C uncertainty, this uncertainty is better than the specification of the Raytek because we had years of empirical data.
Now if the absolute accuracy was 100.0°C (remember in calibration this accuracy is to put it simply the mean of multiple measurements). But the precision (repeatability) was ±3.0°C then the uncertainty would be somewhat greater than 3.0°C and the instrument would unsuitable for a reference standard.
Any measuring instrument repeatability is by far the most important spec.
I calibrated a lot of Fluke 65F's for defense, the calibration standard used was a Raytek MX4 which was calibrated by NMI (National Measurement Institute), the Raytek was IIRC about 3°C high at 100°C but its repeatability (precision) was always within 0.5°C which is about 1/2 its tolerance. So we knew if the Raytek read 103.5° the actual Blackbody temperature was 100.5°C ±0.5°C uncertainty, this uncertainty is better than the specification of the Raytek because we had years of empirical data.
Now if the absolute accuracy was 100.0°C (remember in calibration this accuracy is to put it simply the mean of multiple measurements). But the precision (repeatability) was ±3.0°C then the uncertainty would be somewhat greater than 3.0°C and the instrument would unsuitable for a reference standard.
Any measuring instrument repeatability is by far the most important spec.