How Often Should I Calibrate?
Frequency of Calibration
The calibration interval of any instrument should depend upon its use. Re-calibrations must be carried out at appropriate intervals so that confidence in the traceability chain can be maintained. The length of the interval will depend on a number of variables, such as the importance of the measurements, the stability of the instrument, the frequency of use, the manner of use, the uncertainty required, etc.
So what is an ‘appropriate interval’ for density measurement equipment?
Few, if any, of the international standards lay down fixed recalibration intervals.
Deciding when to recalibrate therefore rests with the user. Only you know the importance of the results, or conversely, what effect a systematic density error will have, either from a safety, production or financial point of view.
If a hydrometer is used and stored at normal room temperature, tests have shown that in a well made glass hydrometer, any change in volume of the instrument is likely to occur in the first year after manufacture. Note, however, that hydrometers which are stored at room temperature but used at elevated temperatures (e.g. at 80°C when testing heavy fuel oils), will change in volume. These will therefore need shorter recalibration intervals.
A review of about 400 hydrometers which had been submitted for calibration two or more times over a 10 year period showed that many tended to indicate higher density at second and subsequent calibrations. As part of the calibration procedure, the hydrometers had been weighed in air and it was noticed that there was a tendency for the hydrometers to lose mass at each calibration. Examination of the base of the bulbs showed that they were losing small chips of glass, presumably due to contact with rough surfaces.
A small number of hydrometers re-presented for calibration were found to have slightly increased masses. In most instances, this was due to a thin chemical coating, deposited on the glass from the samples they were normally used in. However, in several cases microscopic analysis showed that it was due to very small defects in the glass wall, allowing small quantifies of sample to pass slowly into the hydrometer, so increasing the mass.
It is recommended that hydrometers should be calibrated when new. If they are in regular use at ambient temperature, recalibrate after one year, and then every two to three years.
Hydrometers used above or below ambient temperature should be calibrated annually.
Liquid density meters are an extremely simple way to make measurements to relatively high accuracy. The ease of use and repeatability of such instruments can, and often does, lull the user into a false sense of security with them being treated as a black box that is never or seldom recalibrated. It is essential that they should be recalibrated using reference liquids that are supplied by suitably accredited laboratories.
Some manufacturers and suppliers offer a service where an engineer will take accredited reference liquids to an industrial site, calibrate and if appropriate adjust the meter.
Like all instruments, density meters will drift with time. The degree of drift will depend on both the instrument and its design, and the application.
This drift may be due to:
- Instrumental changes due to physical changes in the U tube itself, such as mass, volume and coefficient of elasticity.
- Changes in the way the electronics of the instrument function. This is less of a problem with newer instruments using modern electronics.
- Damage to the instrument by incorrect handling, dropping it, getting liquid inside.
- Moving the instrument, especially if it is at a different angle to the horizontal, or was knocked during the move.
- The effect of the liquid in the tube on the surface of the tube. This can be due to either deposition of material on the inside of the tube, or erosion of the tube by the sample or the cleaning method.
Monitor the oscillation period of the empty cell, and set acceptance criteria. If the period goes outside these limits, clean the cell and then recalibrate.
For routine recalibration, you should consider:
- How important are the results from the meter? This might be in terms of keeping the production lines running, financial losses which the company might incur if a faulty batch of product is sent out, meeting Customs & Excise requirements, keeping auditors happy. What level of uncertainty is required from your measurements?What effect is the sample likely to have on the U tube?How effective is your cleaning regime?Is the cell subject to temperature cycling?
- Laboratories who use 5 place instruments and who are trying to get the very best traceable measurements for fiscal purposes should calibrate at least once a month. Others, who never change cell temperature, are perhaps working with samples which are easily flushed from the U tube after each reading, and who do not need a high level of precision, might only calibrate every 3 months.
It is recommended that the longest acceptable recalibration interval for a density meter in routine use is six months.
A new pyknometer must be calibrated before it can be used. The recalibration interval then depends on:
- how it is used.
- how it is stored.
- the importance of the density results to your laboratory.
For a new pyknometer in frequent use, It is recommended that it should be calibrated when new, again after 6 months, and then annually, for several years. Once you are confident that the results are consistent, the recalibration interval can be increased.