You will need a dead weight tester to check the gauges (expensive)
Gauges come in various accuracies, 2.5%, 1.6%, 1% etc. so you will need to check whether it is still within the original specification and supply a calibration certificate showing the deviation from the pressure shown both on rising and falling pressures.
Calibration of equipment is usually carried out by specialist companies due to the cost of the equipment required.
John
__________________
Best advice I can give as you get older "Never miss an opportunity to relieve yourself"
You need to be more specific as to what "uncertainty" is.
There is a wide range of variables that may need to be considered such as:
Unknown system dynamics?
Unknown physical state of the pressure gauge due to poor maintenance or exposure to forces beyond design specifications?
Determine all system dynamic(s) concerns, operating constraints, operating environment, gauge location in system, type of gauge, temperature, age, accuracy required, operating characteristics, and the requirements of the application.
In short you must know the process very well that the gauge is being used in/on before you can answer the question.
__________________
They said; "Brain size?" I heard; "Train size?" so I said: "I'll take a small one, thank you."
The calibration uncertainty will depend on the calibration method, and the uncertainties in the components you are using to perform the calibration. As jesw55 said, the "proper" way (giving least uncertainty) would be with a dead-weight rig, but buying one for only a few instruments doesn't make economic sense.
__________________
"Love justice, you who rule the world" - Dante Alighieri
Gentlemen I am sorry to say that you are by far not up to date with measurement techniques and definitions.
Uncertainty was introduced mant mant years ago as term to define just exactly what it means the fact that due to the stochastic influence of the factors which influence a measurement result this one is NOT a sure value but one in the "fog", a value related to a probability to be in a certain domain.
The uncertainty takes into consideration ALL possible influences as: nonlinearity, hysteresis, temperature, supply (if the gauge needs a DC or AC supply), humidity, environmental dynamics (especially for mechanical gauges) and so on...
According to modern secifications the probabilistic distribution of the different dispersions is to be used for some it is a normal distribution (gauss) for other it can triangular or even recatngular (same density all over the field). The final "uncertainty" is computed with those values.
As for the calibration of the pressure gauges I would suggest a much more economic approach if the number is not VERY important an if the application is ausual industrial one. Uncertainty of dead weight calibrators is too much for industrial user maintenance it is some times justified but not always since transducers are very reliable and compensated as weel in nonlinearity as in temperature drift.
This is a calibration by comparison with the output signal of a high precision pressure transducer (from a very good supplier) for which a calibration sheet is available with data points and own uncertainty. Further more you need a good supply ( volatge stable ) as well for the reference gauge as for the one to be calibrated (if it is an electric one) and a pressure source which can be a hand pump or for lower pressure a compressed air line with a pressure regulator.
The 2 gauges i.e. the reference one and the one to be calibrated are put on same manifold and loaded with same pressure. Results are read point by point for both and are analysed in order to determine the degree of uncertainty for the object to be calibrated.
For a good result (i.e. with a low degree of uncertainty) it is better to take at least 5 point over the FS (0 to FS). If there is time enough than up to 11 points are to be recorded. You could also use the Student approach for an estimation of the uncertainty band.
Right-click on the image. This (depending on your browser) should produce a pop-up menu with something like "view image information". Left-clicking that should expose the URL of the image.
__________________
"Love justice, you who rule the world" - Dante Alighieri
I got the link and saved, as per your guidance. While sopening and seeing/watching it is very small and not enlargeable. If it is enlarged I think it will be nice. Pl. advise me on this aspect.
This is more tricky. There are probably many ways of doing it, but the
only way I've tested is to save a local copy of the .gif (right click
> save image as ...), then open and resize it using appropriate image
manipulation software (I used jasc animation shop). You can now use it
within your own web publications etc. To include as an animation it in a
CR4 post, you'll have to save it somewhere on-line. This could be even
more tricky. I haven't yet found anywhere suitable (without using my own
webpage, which I'd rather not do). Good luck.
Note: copyright restrictions may apply.
__________________
"Love justice, you who rule the world" - Dante Alighieri
How is the 'uncertainty' expressed for the pressure gauge. And how is it translated into a meaningful value.?
All I have ever come across is (say) Full Scale ±1.0%. So I know
(believe) a reading of (say) 93psi on a 140psi FS gauge is somewhere
between 93±1.4. To me that is the 'certainty'. So how is uncertainty
expressed?
And when you consider the reading is by eye from an angle, and 93 is
an estimate (guess) of the position of the pointer in an ungraduated
space between 90 and 100, what value is the uncertainty then.
__________________
When arguing, remember mud-slinging = lost ground.
ha ha, thanks for your help. I already consider the uncertainty from the master, uncertainty due to resolution ,uncertainty due to hysteresis and uncertainty due to repeatability for uncertainty budgeting. I wonder whether I have to consider more constraint for my budgeting be more correct. That is why I asked this question.
Is it practical to find the uncertainty due to other parameters like humidity and temperature ? How we can find the same practically ?
__________________
If you can't find a way, make one........
Good Answers: