The person accountable had been aware of the critical state of the gearing of a highly loaded rolling mill drive in a cement mill for a long time. The ordered replacement gear wheels, however, were only be assembled during the major revision some months later in order to avoid an operational stoppage. The gear was closely monitored by lubricant analyses. The attrition values determined were measured over weeks in the tolerable range and alarming deviations were not recorded. However, a filter change was then signalled on the basis of a changed differential pressure. The maintenance workers noticed a suspiciously high loading of the filter with metal abrasion. During an inspection made at short notice, a tear was discovered - along with many outbreaks (pittings) - which indicated an imminent tooth-breakage. The gear had to be exchanged as quickly as possible.
Had the lubricant analyses failed in this case? Why hadn‘t the laboratory flagged this up in time? Could something like this happen again?
The maintenance workers wanted to know precise details. They sent in an oil specimen again. This time too the investigation results were within the tolerated range. Both the PQ index for the total amount of magnetisable iron in the sample and the portion of iron particles determined with the ICP spectometry, which are smaller than 3 µm were not alarmingly high.
But quite different values were to be seen in the filter deposit investigation. For this purpose, the deposits of a 50mm x 50mm piece were extracted in the laboratory using solvent. Here, metallic wear particles with a diameter of up to 80 µm and large dirt particles with a diameter of up to 100 µm were manifested.
The filter deposits, and not the gear oil, were the deciding data media in this case.
However, it was not simply the proof of wear that was conveyed here. The large dirt particles had to have penetrated into the gear oil from the outside. During a further gear examination, it was ascertained they had got in through a defective ventilation filter. From this, the cause of wear was also found.
Also read Extract filter specimens correctly