When the condition of a machine is judged on site, the pressure gauge and the vacuum gauge are the two instruments used most often. Yet few people put the instrument itself first on the list of suspects when a reading looks wrong. A misreading gauge leads to a wrong diagnosis more often than the actual machine fault does, because it turns the whole search in the wrong direction from the first step, and everything done afterwards rests on a reading that was never right.
How an inaccurate gauge misleads the diagnosis
When the pressure gauge reads low, the site concludes the filtration section is not blocked yet and keeps running; the actual pressure drop has already reached or passed the alarm value of 0.35 MPa for the 150-500-5 fine filter element, and the problems of a vacuum level that will not come up and of incomplete dehydration follow. A gauge reading high works the other way round: the pressure drop is perfectly normal, but the operator concludes the element is due for replacement and throws away an element that was still usable.
The vacuum gauge is similar and in some ways more deceptive. The working vacuum of the ZYD-100 is -0.08 ~ -0.0999 MPa and the limit vacuum is ≤5 Pa. If the pointer of the vacuum gauge has drifted away from zero, the dial may show -0.09 MPa while the machine is actually only at -0.06 MPa, and then the driving force for the vacuum degassing is short, the treatment of gas content and water content falls short of what it should be, and the number on the dial looks perfectly normal all the same.
What suggests the instrument itself is at fault
An instrument fault rarely announces itself with a sudden jump in the reading; it is mostly gradual, and without attention it is easily taken for a change in the conditions. The signs below are worth watching for, and as soon as one of them appears the instrument belongs on the list of suspects.
| How the reading behaves | Possible cause | How to verify it |
|---|---|---|
| The pointer does not return to zero and sits clearly away from the zero mark | Zero drift, a change in the force on the pointer mechanism or the hairspring | Compare it with a gauge of known condition at the same tapping point |
| Under the same conditions it moves in the opposite direction to the other instruments | The sensing line is blocked or holds air, so the damping is too high | Tap the case lightly and watch whether the pointer jumps, and clear the sensing line |
| Water vapour or oil mist has entered the instrument and the pointer responds sluggishly | Media residue in the sensing line and in the case of the instrument | Remove the instrument and check the diaphragm and the cleanliness of the case |
| Gauges fitted at different points on the same machine read very differently | Differences in mounting position, an unsuitable tapping point | Swap them to the same tapping point and compare |
| The reading changes a great deal after a new gauge is fitted | The range or the accuracy class of the new gauge does not match the conditions | Check the range, the accuracy class and the direction of mounting |
Simple checks the site can do
- Keep a spare gauge of a similar range in known condition and fit it at the same tapping point in turn to compare.
- Close the line and release the pressure, then watch whether the pointer comes back to zero; if the zero offset is obvious, do not keep using the gauge.
- Check that the sensing line is clear, because liquid or a kink in it makes the pointer respond sluggishly.
- Check that the range of the gauge matches the pressure range being measured; reading consistently outside 10% to 90% of the range makes the error noticeably larger.
- Record the date and the result of every comparison or calibration and keep them in the equipment file so that the history can be traced.
- Once a deviation has been found, correct the operating conclusions that were drawn from that gauge as well, so that a wrong judgement is not carried forward.
Calibration interval and records
Calibration does not have to wait until something is suspected. Build it into the routine inspection: at every shutdown for maintenance, make one comparison with the spare gauge, and send the instrument for calibration or replace it if the deviation is beyond the limit. The interval does not need to be tight, but it needs a named owner and a record. The record should cover at least the instrument number, the range, the mounting position, the result of the comparison and the date, so that a set of operating records can still be explained when it is reviewed later.
There is one more point that is easily overlooked: after an instrument is replaced, the operating data taken before the change has to be compared again. A pressure drop curve that suddenly drops by a large amount does not always mean the conditions improved; sometimes it means a gauge that reads low has been fitted. Data does not become trustworthy on its own because the instrument was changed.
When a reading cannot be trusted, suspect the gauge first, the oil second and the pump last.
