Author Topic: Accuracy of engineers level  (Read 840 times)

eskoilola

  • Guest
Accuracy of engineers level
« on: March 21, 2018, 12:42:21 PM »
I would like to reinforce the base of my lathe (the table it is standing on) from particle board to become something more rigid.
The particle board is by no means weak as it is a composition of two 30 millimeter thick boards which are epoxied together. It is just somewhat flexible and does not make the lathe as rigid as it could be.

A 25 mm (one inch) thick iron plate would certainly add some rigidity when bolted under the lathe. As a matter of fact, that lathe has the holes for the bolts already in place. That plate would then in it's turn be bolted into the particle board.

There seems to be several methods to check the levelness (not absolute but relative) of the ways. When attached to a rigid base like that the ways could be easily straightened by shimming the bolts.

So -assuming that I would check the ways with a machinist level, what would be the accuracy needed here ?
0.3 mm/m would tell 0.1 mm differences between the ways (they are about 15 cm apart). Is this accurate enough or do I need to go down to 0.01mm range in order to get any benefit of this procedure?
Are there any alternate procedures to check the way straightness ?

And Yes, I searched the forum with N/A.


Offline chipenter

  • Hero Member
  • *****
  • Posts: 778
  • Country: gb
Re: Accuracy of engineers level
« Reply #1 on: March 21, 2018, 01:23:40 PM »
An enginers level is usualy half a second  thats 120th of a degree , put across the ways at each end of the bed to check and ajust for twist .
Jeff

eskoilola

  • Guest
Re: Accuracy of engineers level
« Reply #2 on: March 21, 2018, 02:55:37 PM »
Actually - getting the lathe level with itself. The easiest (by far) way to do this is to make it level with the world. This is because the levels (not the digital ones, I hate them) usually work on a very limited range and tend to measure levelness with the wold ...

After browsing the World Wide Wait I figured out that nobody seems to be using more accurate device than about 0.4mm/meter. So 0.3mm/meter should be enough. Those are German millimeters, not the chinese ones ....

So, I went and bought one of these.

It is a machinist level with 0.3mm/m accuracy.

Offline AdeV

  • Madmodder Committee
  • Hero Member
  • *****
  • Posts: 2055
  • Country: gb
Re: Accuracy of engineers level
« Reply #3 on: March 21, 2018, 03:58:14 PM »
You should be able to adjust the angle of the vial on the level; so, put it across the ways at one end, adjust until the bubble is in the middle; move to the end of the ways, and any discrepancy should be in the lathe, not the world (unless you happen to be somewhere near an inconvenient gravitational anomaly - see numerous episodes of Star Trek to see how to deal with those).
Cheers!
Ade.
--
Location: Wallasey, Merseyside. A long way from anywhere.
Or: Zhengzhou, China. An even longer way from anywhere...
Skype: adev73

Offline Pete.

  • Hero Member
  • *****
  • Posts: 883
  • Country: gb
Re: Accuracy of engineers level
« Reply #4 on: March 21, 2018, 04:42:30 PM »
An enginers level is usualy half a second  thats 120th of a degree , put across the ways at each end of the bed to check and ajust for twist .

I think your maths is off Jeff. There are 3600 arc seconds in a degree so half a second is 1/7200 of a degree. A half-second level would be extremely sensitive - it would move as you walked around a concrete floor in your workshop. Two seconds is still very sensitive and your basic engineer's level is more likely to be 4-10 arc seconds.

eskoilola

  • Guest
Re: Accuracy of engineers level
« Reply #5 on: March 22, 2018, 01:21:31 AM »
it is just mathematics:
If the sensitivity of that machinist level is 0.3 mm per meter this means that it can tell a 0.017 degree difference in the angle which is HUGE when compared to an arc second (0.00027 degrees).

Offline chipenter

  • Hero Member
  • *****
  • Posts: 778
  • Country: gb
Re: Accuracy of engineers level
« Reply #6 on: March 22, 2018, 02:18:38 AM »
Thanks Pete I fogot about the minutes for a minute there , oure Eclips level has 0.5 second accuracy stamped on it .
Jeff

Offline mattinker

  • Hero Member
  • *****
  • Posts: 1103
  • Country: fr
Re: Accuracy of engineers level
« Reply #7 on: March 22, 2018, 05:05:15 AM »

So -assuming that I would check the ways with a machinist level, what would be the accuracy needed here ?
0.3 mm/m would tell 0.1 mm differences between the ways (they are about 15 cm apart). Is this accurate enough or do I need to go down to 0.01mm range in order to get any benefit of this procedure?

I think your missing a 0, I have a Moore and Wright that is 0.03m/metre, 0.3m/metre would be good for an ordinary level!

Regards, Matthew

Offline MetalMagus

  • Jr. Member
  • **
  • Posts: 31
  • Country: gb
Re: Accuracy of engineers level
« Reply #8 on: March 22, 2018, 08:13:29 AM »
Google
Rollies dads method.

This is a method for aligning a lathe that does not require a level.

Regards

Sean

Offline awemawson

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6273
  • Country: gb
  • East Sussex, UK
Re: Accuracy of engineers level
« Reply #9 on: March 22, 2018, 09:45:38 AM »
And works well  :thumbup:
Andrew Mawson
East Sussex

Offline philf

  • Hero Member
  • *****
  • Posts: 847
  • Country: gb
Re: Accuracy of engineers level
« Reply #10 on: March 22, 2018, 10:00:19 AM »

.......

I think your missing a 0, I have a Moore and Wright that is 0.03m/metre, 0.3m/metre would be good for an ordinary level!

Regards, Matthew

Matthew,

I think you're missing more than the odd 0!  :scratch:

I assume m/metre should be mm/metre?

Phil.
Phil Fern
Location: Marple, Cheshire

eskoilola

  • Guest
Re: Accuracy of engineers level
« Reply #11 on: March 22, 2018, 10:31:44 AM »
...
I assume m/metre should be mm/metre?
...

Actually making a machinist "Level" that is "Accurate" to 1m/1m would be difficult. Mathematically this is next to impossible but as we are machinists we deliver difficult things right away - miracles take some time....

Offline Spurry

  • Sr. Member
  • ****
  • Posts: 278
  • Country: gb
  • Norfolk, UK
Re: Accuracy of engineers level
« Reply #12 on: March 23, 2018, 11:10:21 AM »
The standard Starrett levels are shown as 1 division = 0.005"/ft. The Master version is shown as 0.0005"/ft or 0.04mm/m.
I have some of the standard types spare, if anybody needs one.
Pete

Offline mattinker

  • Hero Member
  • *****
  • Posts: 1103
  • Country: fr
Re: Accuracy of engineers level
« Reply #13 on: March 23, 2018, 02:46:52 PM »

.......

I think your missing a 0, I have a Moore and Wright that is 0.03m/metre, 0.3m/metre would be good for an ordinary level!

Regards, Matthew

Matthew,

I think you're missing more than the odd 0!  :scratch:

I assume m/metre should be mm/metre? Yep, double typo!!

Phil.

Cheers, Matthew

eskoilola

  • Guest
Re: Accuracy of engineers level
« Reply #14 on: March 24, 2018, 12:48:40 PM »
The Wasserwaage arrived!

Today the post delivered the machinist level. This is a 2 kilogram piece of iron. The scraping is still visible on the bottom surface so maybe this has not been used too much. If someone has an estimation what is the age of this thing - this bit of information would be greatly appreciated.

Someone said that these things could be calibrated. In this case that would probably mean breaking the seals on either end. I already tried whether I could get the same reading by turning the thing 180 degrees. The only surface that is stable enough here in my germany apartment is the ceramic stove. By using the chopping board as a rest I could determine that this thing is not too far from being calibrated.

When the device was on the table I could center it by choosing a suitable place to stand. This apartment has wooden floors....



Offline Pete.

  • Hero Member
  • *****
  • Posts: 883
  • Country: gb
Re: Accuracy of engineers level
« Reply #15 on: March 24, 2018, 02:06:04 PM »
That's a nice level!

Offline mattinker

  • Hero Member
  • *****
  • Posts: 1103
  • Country: fr
Re: Accuracy of engineers level
« Reply #16 on: March 24, 2018, 03:18:19 PM »
i agree with Pete! That's a nice level!

I was wrong about the moore and wright level that I have, it isn't 0.03mm/m it's like this one, 0.3mm/m!

Regards, Matthew

Offline AdeV

  • Madmodder Committee
  • Hero Member
  • *****
  • Posts: 2055
  • Country: gb
Re: Accuracy of engineers level
« Reply #17 on: March 24, 2018, 03:21:59 PM »
Oooooh, nice!

No, that one doesn't have any adjustment on it, I guess they must set them in the factory & be confident they'll never change (which, let's be honest, is probably a fair guess when it comes to a machine level!)
Cheers!
Ade.
--
Location: Wallasey, Merseyside. A long way from anywhere.
Or: Zhengzhou, China. An even longer way from anywhere...
Skype: adev73