Author Topic: DIY gauge block tolerance verification  (Read 4304 times)

Offline PekkaNF

  • Hero Member
  • *****
  • Posts: 2523
  • Country: fi
DIY gauge block tolerance verification
« on: August 30, 2019, 11:58:42 AM »
I bought a set of gauge blocks that I though would be useful of verifying accuracy and tolerance of measuremets that has be pretty exact.

Gauge blocks do that nicely. I have a 88 part set that came with a measurement sheet. It is all nice, but the 0,001mm steps are in 1.001 - 1.009 series and often slighly thicker blocks and in down direction series would be more usefull. I found one that looked pretty usefull at 10 blocks from 1,991 to 2,000 mm on 0,001 mm steps.

However, when I got it the blocks were loose in the box and covered with heavy grease, but no wraping in them and no individual test report....

They really did not wring together, only two of them weakly. First strike.

I measured them with a mitutoyo micrometer and um indicator, results were not consistent and I lost faith in them.

One guy I know lend me a comparator that should do the trick - when carefully used and read with the loupe.

I used known good 2,000 mm gauge block and made few test runs. It took a little practis to get consistent results, read the needle postition reliably and while measuring avoid breathing into the system. Few mitutes of being close to the comparot and berathing to it would make the needle to move.

Final runs and all results in Excell showed that only one gauge block was close enough, most of them off and few way off.

Is there any particular name for these 10 piece constant step block sets, that I could search and maybe buy? I did find Mitutoyo block set, but closed the page, because price was a bit steep.

I did get good data and coross check several times during the measurement that the zero (2,000 mm) hasn't changed and measured the blocks in random order to minimize psycological factor and random errors. I think I established that I can't trust these blocks and that good blocks are worth having, bad ones are just shiny pieces of metal.

Offline efrench

  • Full Member
  • ***
  • Posts: 213
Re: DIY gauge block tolerance verification
« Reply #1 on: August 31, 2019, 01:47:27 AM »
Can't you make them the right size?

Offline PekkaNF

  • Hero Member
  • *****
  • Posts: 2523
  • Country: fi
Re: DIY gauge block tolerance verification
« Reply #2 on: September 01, 2019, 05:38:08 AM »
Can't you make them the right size?

My primary consideration was "Can I trust these new blocks or not". All indication is that I can't trust them and it's probably not worh of trying to usem them as they are, even with home made correction table. Maybe if they had near correct values, but when the measured thickness is most likely more than a step (next or previous block nominal value ) away they would be very awkward to use.

I have no idea how gauge blocks are made, but I would guess that the accaracy and planar trueness (if there is such a expression?) are pretty much un attainable at HMS, not to mention that it probably would become a lifetime quest - by the time I porbably would have forgot what was my initial need for them.

Grade 1 tolerance with smaller than 10 mm gauge blocks is 0,2 um. I have no idea how to lap these close to plane that is even near that number.

Offline richardwilde90

  • Jr. Member
  • **
  • Posts: 6
  • Country: gb
Re: DIY gauge block tolerance verification
« Reply #3 on: September 01, 2021, 10:59:43 AM »
The safest route is to get those what you already have UKAS calibrated.

There is a LINK REMOVED that do it for cheap.

Alternatively if you buy some Mitutoyo "micrometer test blocks" they will require calibration anyways, and without it youll be in the same boat.




I bought a set of gauge blocks that I though would be useful of verifying accuracy and tolerance of measuremets that has be pretty exact.

Gauge blocks do that nicely. I have a 88 part set that came with a measurement sheet. It is all nice, but the 0,001mm steps are in 1.001 - 1.009 series and often slighly thicker blocks and in down direction series would be more usefull. I found one that looked pretty usefull at 10 blocks from 1,991 to 2,000 mm on 0,001 mm steps.

However, when I got it the blocks were loose in the box and covered with heavy grease, but no wraping in them and no individual test report....

They really did not wring together, only two of them weakly. First strike.

I measured them with a mitutoyo micrometer and um indicator, results were not consistent and I lost faith in them.

One guy I know lend me a comparator that should do the trick - when carefully used and read with the loupe.

I used known good 2,000 mm gauge block and made few test runs. It took a little practis to get consistent results, read the needle postition reliably and while measuring avoid breathing into the system. Few mitutes of being close to the comparot and berathing to it would make the needle to move.

Final runs and all results in Excell showed that only one gauge block was close enough, most of them off and few way off.

Is there any particular name for these 10 piece constant step block sets, that I could search and maybe buy? I did find Mitutoyo block set, but closed the page, because price was a bit steep.

I did get good data and coross check several times during the measurement that the zero (2,000 mm) hasn't changed and measured the blocks in random order to minimize psycological factor and random errors. I think I established that I can't trust these blocks and that good blocks are worth having, bad ones are just shiny pieces of metal.
« Last Edit: September 02, 2021, 05:13:34 PM by spuddevans »