Hi there
I am COMPLETELY new at this
I want to know if there is a component that you can use to compare the level/height of 2 points. The component needs to be accurate within mm but does not need to give actual GPS height...
Also it should be cheap

Many thanks

 

How far apart are the two points?

A piece of hose works very well.

 

How far apart are the two points?

A piece of hose works very well.

How exactly would this work?

 

You almost fill the hose with water and place the two ends at the two different locations. You might have to pound posts into the ground to hold the ends up.
Measure from the water level to the ground at both ends and calculate the difference.

I did this on my house and found that the north west corner was 1 3/4 inches higher than the south east corner. Great way to figure out how to apply siding without coming up with a gap at one corner of the building.

 

I don't see how that produces an unambiguous result. Doesn't it require the ground to be at the same height at both ends? What am I missing?

 

The description I wrote in post #4 is a method to find the difference in height of the ground from one point to another. On my house, I measured from the water level to the first grout line between the concrete blocks to find the difference in height of the concrete slab.

For clarification, I'm using transparent tubing and calling it, "hose". The height of the water in the hose will be the same at both ends. It doesn't matter how many loops or curves there are, as long as no part in the middle of the hose is above the water level at the ends. At the ends, you want a few inches of air in the hose above the water level. The distance from the meniscus to the ground might be 20 inches at one location and 22 inches at the other location. You have just found that the difference in the altitude of the ground is 2 inches, and at which location the ground is higher by 2 inches.

 

The height of the water in the hose will be the same at both ends.

That's the key. You can establish a very accurate level over quite a long distance this way. I can't think of a more accurate method. And yes, a clear plastic tube is far better than a garden hose but most folks are more likely to have the latter.

 

You only need a clear (vertical) bit of tubing at each end (actually just at the low end), the bulk of the hose can be any type.

 

The hose trick is used in tanker vessels to adjust the trim to a certain value by transfering cargo between tanks and not going outside to read drafts on the ship's sides.

The marking is previously done with the vessel at different known conditions including that of "even keel".

 

Got it. Thanks for the clarification.

 

The Egyptians used the water level concept to lay the foundation for the pyramids. They would have loved having a nice long piece of Tygon tubing.

The water level and the plumb bob were probably in use long before the pyramids, and remain "the right tool for the job" to this day. Pretty amazing.

 

I used to work on auto-leveling lasers, and they were not as accurate as the water hose method! The lasers were guaranteed to 1/4 inch at 100 feet. The water hose method is accurate to as well as you can measure at unlimited distance. For me, that is tighter than 1/8th of an inch (3 mm) at several hundred feet.

The RB suggestion that you only need some clear tubing at one end also reveals a method. You set up the hose first, then fill it with water until it overflows at one end. Then measure the height of the overflowing hose end and compare that to the meniscus at the other end.

Fun conversation. I wonder if the OP will ever come back.

 

The water hose method is accurate to as well as you can measure at unlimited distance.

Well, sort of. It is sensitive to air pressure, so I imagine there is some potential error due to local barometric variations. Think wind and whatever causes airplane turbulence.

And over long distances, curvature of the earth comes into play. Are huge buildings built straight or level? Not the same thing. A laser "straight" floor of sufficient size would drain to the center, since the edges are farther away from the earth, ie. higher in elevation. One thing I'm sure of is that water flows downhill! The error could be as much as 3" or so for the largest buildings. Maybe it just gets lost in larger "noise" from other error sources.

 

When you get a hose long enough to get into a different barometric pressure, be sure to take into account the number of time zones you have crossed.

 

I guess my hose must be bigger than your's.

 

Interesting point about, "laser level" causing a drain to the center. I can see that, but I haven't done the math for how large the footprint would have to be for, say, an inch of error from the curvature of the earth. It really makes me wonder if the pyramids are damp in the middle.

I tried doing a triangle with one side being 4000 miles and the hippopotamus being 4000 miles plus one inch. My calculator can't resolve things that small. Can you suggest a method?

 

The rule of thumb is 8 inches of drop per mile although it depends on the square of the distance. Here and here too.

The world's largest building (Boeing factory) might be a half mile long, so I have to think a 4" error had to be dealt with. I have no idea how.

 

Interesting point about, "laser level" causing a drain to the center. I can see that, but I haven't done the math for how large the footprint would have to be for, say, an inch of error from the curvature of the earth. It really makes me wonder if the pyramids are damp in the middle.

I tried doing a triangle with one side being 4000 miles and the hippopotamus being 4000 miles plus one inch. My calculator can't resolve things that small. Can you suggest a method?

I don't know why you are using 4000 miles since the radius of the earth is less than that.
But anyways, 253440000 / 253440001 = 0.99999999605429294486153351932791 according to MS Windows calculator. And arctan (0.99999999605429294486153351932791) = 44.999999886963819049782710203113
or did you want arcsin (0.99999999605429294486153351932791) = 89.994910207826558570722955308271

 

Holy Smoke, those are long numbers!
So, the answer is? 89.99 somethings per inch?

 

Holy Smoke, those are long numbers!
So, the answer is? 89.99 somethings per inch?

No, the answer is in degrees, so that is almost a right angle.