Accurate reference pressure.


I've built an arduino controlled pressure chamber (sealed tank + compressor + pressure sensor).

I've no reason to doubt the accuracy of the pressure sensor, but still I'd like a way of positively confirming it's right, without relying on the accuracy of another pressure gauge (which may or may not be correct).

The chamber is designed to work between 25 - 45 psi, and uses a normal schrader valve (like on a car tire), after charging the chamber I then measured the pressure using a selection of tyre pressure valves I use in my workshop (one pencil type, one digital and one brass bourbon gauge) - all of which gave completely different values (different from the arduino sensor and different from each other!)

Is there anyway I can test the pressure using some indisputable physical constants - or at least some way that I can be sure of the results?

When I've worked with temperature probes I've been able to calibrate against the melting point of ice and the boiling point of water. When I've worked with low pressures (<= 1psi) I've made myself a u-tube manometer filled with water. Is there anything similar I can do in the 25 -45 psi range? At the moment my only option seems to buy/borrow another pressure gauge and hope it's calibrated correctly.


look at a u-tube manometer.

get 200-300 or so feet of clear tubing, or add 10 feet to a garden hose
hang the hose from the top of a 10 story ? building so you can see the clear tubing.
need to be about 120 feet of the ground for 45 psi

raise the other end of the hose until the liquid is in the correct level.
or change the pressure until it is at the desired point.
measure with a tape measure.

1 PSI = 27.6704523 inches of water

it will take some time as you have to provide the actual volume of air to displace the water.

very accurate.

get a cylinder with piston with a 1 inch total area of the cylinder.

put your piston in the cylinder
add the weight on top.
the weight will compress the air to whatever the weight is.

error is caused by friction

get 200-300 or so feet of clear tubing....
hang the hose from the top of a 10 story ? building

I live in the country side! Seriously, if I'm going to measure 45 psi, then the tubing needs to cope with 45 psi.

I'd be better off with a mercury manometer, but I'd prefer not to work with something potentially so toxic.

Best to calibrate one of your pressure gauges. This is normally done with a deadweight tester. Expensive to buy but fairly easy to make.


cliff ? barn roof ?

get as high as you can

now you can see why people pay to have a calibrated meter.

I too would have to travel 20-30 miles to a tall building. and even then, I do not think there is a 10 story building that close.