12" machinist level

I have mostly disassembled mine as IN ONE DIRECTION, I was out of any more adjustment. I found 2 screws (one on each side at the opposite side of where the vial pivots) you screw them out and vial drops, in, it raises. It gave me more adjustment, but same thing, dead on center one way, 180* and it favors one side. Tried splitting that difference, but it will always heavily favor one side. It really makes no sense. Is it possible with all the hours (6-8?) just this time around AND Aukai doing much the same, that something is WRONG with both our levels? We have watched many videos, fine tuned the he11 out of our level surfaces. We've gotten all kinds of help from this forum.... I am ready to throw in the towel. For my mill/lathe, my digital level is good enough.
I know its probably futile, but I even put a DTI along the vial to see if it was matching what was happening. Getting it level that way has no difference in result. I'll stop jumping in here, Good luck Aukai, I'll still be reading this thread. Go enjoy that beautiful playground you live in!
 
Please stay here, a lot of good coaching in one place I think with all of the leveling I/we have been doing, the comment was made that the table is not level enough. I looked at my level, and the resolution is 0.0002in/10in, per the legend on the level. My take is that my table level is outside of the .0002in/10in for my level to be adjusted. To get in the adjustment window, I would have to make the bubble levels off kilter. That is what I'm interpreting from one of the comments. I hope someone can set me straight.
There is a nice Sterrett 98-8 on Ebay, they are .005 per foot.
 
Well, hopefully you can answer this, but lets say your table is off perfect level, by say... .002 over it's length. You shouldn't NEED that much perfection (?) I've seen plenty of videos of people zeroing it out, just making sure that amount of error is equal both directions. I've been down this road (and I believe you as well) Say one division off one way (and this matches more/less the error on table) but the other direction it's a train wreck. I'm just trying to add less as it's a bit chaotic as it is...making popcorn and tuning in regularly however!:p
Aukai said:
Please stay here, a lot of good coaching in one place I think with all of the leveling I/we have been doing, the comment was made that the table is not level enough. I looked at my level, and the resolution is 0.0002in/10in, per the legend on the level. My take is that my table level is outside of the .0002in/10in for my level to be adjusted. To get in the adjustment window, I would have to make the bubble levels off kilter. That is what I'm interpreting from one of the comments. I hope someone can set me straight.
There is a nice Sterrett 98-8 on Ebay, they are .005 per foot.
Click to expand...
 
See Dabbler's response above, I may be interpreting it wrong.
 
Aukai said:
See Dabbler's response above, I may be interpreting it wrong.
Click to expand...
I did read it before as well, but that is disappointing. They are SO sensitive that you really need something in the same ballpark of sensitivity (and accurate) just to get a surface level enough to adjust. As I have only been doing these maneuvers all day long and thinking bout it in between, here is where I can make it add up as to why we are having these problems...
If our surfaces are say .001" low on the left, we effectively raise the vial that much on one side. Remembering that table is low .0005" on left side and .0005" HIGH ON THE RIGHT. Now we rotate the level and that vial we just raised on left side is now on the right and remember, that side of the table is .0005" high...we have just combined those two errors on top of each other. SO we are level on one axis, but (for a level of such sensitivity .0005" per 10") it's way off in that rotation. Yes? No? Maybe?
The world is in shambles with Covid...I don't need this ulcer. I'm going to use the techniques the Egyptians used to make the pyramids, I just saw a youtube about it! ;)
 
I think that is the gist of it.
 
OK let me be clear. On a .005 div per foot level, you can level a surface to around a thou per foot quite nicely. (If you can't read a quarter division, you need better glasses). This is plenty flat enough to calibrate a .0002 level. I've done it 3 or 4 times for people.

Fact is, an engineers level with .003 to .005 per foot /division is plenty for any machinist. those super sensitive levels really aren't useful, except on freshly surface ground surfaces. Even a spec of dust on you surface can render it off scale. The glass is NOT flat enough to do a good job.
 
Tell you what, guys. You do not need a surface plate to level a precision level. The table you use simply has to be flat and it has to be able to be leveled to the world (ideally with a tripod adjustment) and it has to sit on a solid surface that you do not lean on or touch. That's it. This is not rocket science or magic.

Basically, you have a precisely curved sealed vial filled with liquid and a bubble. One end of that vial will usually be the adjustable side and the other side will hinge. That's all there is in there.

The basic idea is to level the table with a bullseye level first. This will not get the table perfectly level within a millionth of a degree but it will get it close enough to do a preliminary adjustment of the precision level (PL). Let me tell you what I do and I'll make it as simple as I can. I am assuming you have a flat table with three adjustable feet - two feet on one edge and one foot in the middle of the opposite edge. The table should be at least a foot square and it must be flat. Plywood, Melamine ply/mdf will work. If you don't have such a table, go make one. If you choose not to then you're wasting your time.
  • Put a small framing square or two flat pieces of something a precise 90 degrees to each other on the table and clamp it in place. This will be used to precisely flip the PL 180 degrees and reproduce its position.
  • Then use a bullseye level and get the table level. Get the bubble as dead center as you can.
  • Determine one orientation for your PL that will be the primary position. In Aukai's instance, we called this the Front. Put the PL in this position so the long side and one end touch your orienting brackets or whatever you chose to use.
  • Look at the bubble of the PL and note its position, then flip the level 180 degrees and note the bubble position again. There are only 4 possible outcomes:
    • The bubble is dead center, in which case the table is level and the PL is calibrated.
    • If the bubble moved to the same side when you flipped the PL, the table is off.
    • If the bubble moved to the other side when you flipped it then the table is okay and the level itself is off.
    • It is possible that the bubble position will be aberrant and not consistent, in which case both the table and the PL are off.
It appears that at least in Aukai's case, the last outcome is what we're dealing with. To address this, knowing that the table is at least roughly level because your bullseye level says it is, you start by adjusting the level to get it to the point where the bubble is not pegged or buried at either end. Then you adjust the PL until the bubble moves to the same side when flipped. This allows you to finely tune the table so turn one of the table adjusters until you get movement and repeat the flip procedure. Adjust the level and the table position until you get the bubble centered when flipped.

This is the only possible way you're going to get the PL calibrated. At this point, you have touched the PL adjusters so you don't know where you are so you'll have to go through this until you get it adjusted.

I know you guys are frustrated but none of this is hard. You just have to be patient and persist until you get it adjusted. I've done this multiple times and haven't yet failed to get one calibrated.
 
I got rid of the glass, I'll be moving to a less sensitive level, I'm glad that you pointed this out, it makes sense now why it was being a challenge. Thank you for adding to the discussion, I did not consider that, that resolution was adding to the situation, never even thought about it until I stared at the numbers. Having the knowledge confidence to deal with it is experience, working on that....Tilting the table out to get the bubble to respond,,,, never thought of it. Thank you guys for sticking with this.
 
Last edited:
I guess this thread lead me to realize....I have a Chinese (Grizzly) lathe and a Chinese (PM) mill. The tolerance they are made at is slop compared to what the level can detect. WAY overkill. I will get a better setup to level this damn thing, when I need that time to just sit and focus, I will pull this thing out again. So what is a far more realistic level (less sensitive) to get? I keep hearing a Starett, but I also read they have seriously dropped off from what they were. I don't want to end up with 3 of these children of satan....
Thanks everyone for all the pointers!
 
You must log in or register to reply here.
Back
Top