when precision falls short

When Precision Falls Short: The Quest for a Properly Calibrated Automatic Level

Paul has recently bought a new automatic level, a Leica NA332. Paul is an Architect and a person that likes learning by doing. Being diligent the first thing Paul did with his Leica NA332 was to check Its accuracy by performing a two-peg test. And this is where his problems started.

And this is when I received an email from Paul, who resides in Texas, seeking assistance. Paul encountered accuracy problems with his recently purchased Leica NA332 automatic level. Below (with Paul’s permission), you’ll find the transcript of our email exchange as I provided guidance to help him resolve his auto level issues.

Disclaimer: This site, and in particular this page, contains affiliate links to products. You are under no obligation to but if you visit and buy something from their site from a link on this site, we will receive a small commission, the price you pay will remain the same. This helps to keep this website running and your help is appreciated.

Hi Ian,

I am writing to you from Texas, and hoping you can help me with the new Leica NA332 automatic level I purchased. I went ahead and performed the two peg test that you’ve written about to check the accuracy of the level. At 160’ (48m) I received an error of 0.25” (6.35mm). I shot the pegs again and received the exact same result.

Is this within the acceptable range for the Leica? It seems like a large amount of error to me.  I’m not really sure how to interpret the standard deviation Leica lists when it says 1.8mm for 1km double leveling. I am not a surveyor by trade, but an architect who likes learning by doing.



Hi Paul,

Thank you for your emailing me about your Leica NA332 issues.

I would say that the error you are getting is too great for Leica NA332 (certainly for my liking).

The first thing I would say is not to worry about interpreting the standard deviation of 1.8mm for 1km of double levelling. Double Run levelling is a technique that allows identification of the errors that will occur over very long levelling traverses and are not applicable to the two peg test for level checking.

Double Run Levelling is a technique where the Backsight and Foresight observations are taken twice, this aids the identification of errors. The Backsight and Foresight readings are taken from one setup position, the auto level is then moved and then the Backsight and Foresight readings are taken again. Think of this method as being able to do the entire levelling run twice, but quicker.

Going back to the accuracy of your two peg test. As you have consistent readings, I presume that you have performed the two peg test correctly. Do you know whether the NA332 is reading high (line of sight above horizontal) or reading low (Line of sight below horizontal)?

As the weather in Texas can be hot, if the reading is high then consideration must be given to potential for heat haze. Although heat haze would lead to inconsistent results, which you don’t have. So, this could be discounted.

I know that you said that the NA3332 was new and I would have expected it to have been checked before it was sent out. Unfortunately, most suppliers will only verify that it was correct at the time of checking but it might be worth taking it back and asking them to check it. If this is not possible then you can adjust it yourself.

You can adjust the Leica NA332 yourself if you wish (although you may need to get it calibrated (checked) if you are doing work for clients). There is a procedure detailed in the manual but if you look at the Leica Adjustment Screw picture attached this is where the adjustment takes place. When I am adjusting levels (and I don’t have an auto-collimator to hand) I setup to targets on fixed objects that make it easier to adjust the levels. But setting these targets takes time. (To see what I mean take a look at https://www.lssltd.net/kaiweets-kt-360a-3d-laser-level-technical-review/ Here I set up a level base line to check levels easily, although it does take some time to set the level to the right height).

If you are going to adjust the NA332 yourself (you can’t break it) then I would suggest the following to get then best results. Along with following the guidance in the manual and the method for the two peg test.

  • Use 2 levelling staffs and secure these in place, as level and plumb as possible, but also that they don’t move throughout the test/adjustment. Consistent level readings are important. If you don’t have two levelling staffs then a good tape measure secured to a wall or building will suffice (as long as it doesn’t move).
  • Check the level vial is in the centre and that the compensator is working correctly.
  • To check the compensator, look through the level at the non-moving levelling staff (or other fixed marker) and turn one screw to point the telescope end either up or down. There should be a few turns of one screw before the crosshair moves off the fixed marker. Try both directions. Look at the levelling vial once you spot movement of the crosshair, the bubble will no longer be in the centre. Remember to re-level the level before adjusting.
  • Adjust the NA332 using the longest distance sight as your reference. Turn the adjusting screw (see photo) to move the crosshair to the desired level (the reading that you have calculated). Take your time and the screw should move with little force but consistent force. This is because the crosshair lens is spring loaded for movement.
  • Repeat the whole two peg test again. This is your check that you have performed the adjustment correctly. I usually find I have some residual errors left but this can fine tuned by repeating the task.
  • When adjusting the NA332 use the finest readings you can. An E-Grad staff may not yield the best results like a tape measure with lots of graduations would.

Hopefully, these tips will help you get accurate levels with your Leica NA332. If you want to know what a level looks like inside then have a look at this article The Automatic Level – LSSLTD.NET

If you need anymore help please let me know.



Hi Ian,

Thank you for taking the time to write back and crafting such a thorough and helpful response. 

Regarding the methodology of the two peg test, here’s a quick run through of what I did.  I set two pegs 160′ apart and then set the automatic level at the midpoint, 80′ from either peg.  I recorded these two elevations (A1, B1), and then moved the automatic level to within 10′ of the first peg and recorded the subsequent two elevations (A2, B2).  I subtracted the first (A1-B1) and second set (A2-B2) of elevations from themselves and then found the difference between the two groups to be 1/4″.  Should I have done anything differently?  Texas is very hot at the moment, today we will hit 39C/102F!  I wasn’t even aware that heat haze could impact the results.  For reference, I conducted the peg test right before sunset.

I do not know if the Leica is reading high or low, but I will read up on this and see if I can answer that for you.

The Leica should be new, as it was purchased from Amazon, however I can only guess at what bumps it may have received along the way.  I have considered returning it, and trying to reorder, but there is no guarantee that the replacement will be better.  If there is not a risk of breaking the compensator through adjustment, I may try my hand at that, following the directions in your email.  It seems like it would be a worthwhile skill to have. 

Do you know roughly what it costs to have a unit calibrated?  I am not using this for work for clients, just for personal work.  That being said, as this is a precision tool, I would want to be able to use it with confidence and trust the readings I am taking.

All the best,


Hi Paul,

It sounds like you have performed the two peg test correctly.

With regard to whether the level is reading high or low, you can figure this out from the readings you have. Using the three readings A1,B1 and A2, you should be able to calculate what the reading at B2 should be.


If the reading at B2 is larger than the calculated value, then the level is pointing up (reading high) and vice versa. (B2 is the longest reading taken and therefore the one with the greatest error).

In my experience most levels start to point down rather than up.

All optical measuring devices will need adjusting (Calibrating) from time to time. Sometimes it can be within a few days, other times it can be years before it needs adjusting. It is just the way it is (and how well it is treated). This is why they should be checked regularly and before any critical measurements.

With regard to heat haze and deflection of light this is really an issue when sighting close to the ground or heat sources (think of a car or machinery). By close I mean sighting around 1’ over the ground or obstacle. Most of the time this is not an issue.

In the UK a calibration for the Leica NA332 would cost £35 upwards (about $45). This would be checked on an auto-collimator at an infinite distance.

Being able to level with confidence requires practice and being able to repeat the levelling points with the same answers time and time again. If you can level around a site and end up back where you started within 3mm (1/8”) then that is a successful levelling traverse. Very rarely do people level further than 100m (300ft) and that is why I think 3mm (1/8”) is acceptable.

It does sound like you are on the right path and do more than most people do when they are using a level.



Did Paul Resolve his Auto Level Issues?

Whilst I asking Paul permission to publish our email exchange I did find out that he successfully resolved his issues with his Leica NA332. Paul reached out to Leica in North America who put him in touch with a local dealer who adjusted and calibrated his NA332 as a warranty item.

Paul now has an automatic level that he has confidence in, and the knowledge that if he has issues that he can’t sort out he knows that his local dealer can sort it out for him.

What is the compensator inside an Automatic Level?

Below is a picture of the compensator inside every Automatic Level. This compensator is what differentiates an Automatic Level from a dumpy level. The compensator allows the automatic level to be accurate over a large range of from the horizontal plane of the instrument. Think of it this way. The barrel of the auto level does not need to be level, it can point upwards or downwards of horizontal, the compensator will automatically adjust to the horizontal plane and thus adjusting the path of sight to horizontal. As long as it is within its working range. Before placing an automatic level on a tripod you can often hear a clicking sound from inside the automatic level, this is the compensator moving between its limits.

Automatic Level Compensator

If you have any questions regarding auto levels, then please email me. I will do my best to respond quickly with an answer. But please don’t forget to search this website for your answer.

Leica NA324

Leica NA324 Automatic Level

Built to last but at a price you can afford.

Any of the Leica NA300 series of automatic levels will serve you well. The accuracy you need and the durability you want.

24x magnification allows the easy reading of the E-Grad staff at any distance. From under 1m to over 30m from the instrument.

Stadia constant of 100 allows for easy distance calculations.

IP54 rating so the rain won't stop you from working.


Related Posts

The Two Peg Test

The Automatic Level

Achieving Accurate Levels

Automatic Levels vs Laser Levels

The Best Levels for under £200