DEATH TO THE LASER PARTICLE COUNTER! What do you call 1,000 particle counters in a trash compactor? A good start!
I've only once in my career seen a request for particle count that made sense to me. The customer had purchased a new filtration system and wanted to submit samples from just upstream & just downstream of the filter to see what effect it was having. Those results are only useful as relative values.
1 - The repeatability (multiple tests of the same sample, same unit, same operator) looks really bad, until you get a look at the reproducibility (multiple tests of the same sample by different labs), which is a joke. The only way those numbers give the illusion of precision is by abandoning accuracy. Anyone who has read ISO 4406 knows that the range of actual counts that those two-digit classifications represent is huge, especially when you get into the mid-teens. Do you really want to get those counts down? Send samples around to a coupla-three labs. At least one of them will give you the numbers you're looking for. It won't change what's actually occurring in your system, but it will give you a pleasantly false sense of security.
2 - OK, I'm counting particles, but I'm not characterizing particles. I can get a warm fuzzy if I have a nice low count, even if they are primarily hard, abrasive particles. I can run around with my hair on fire if I have a high count, even if they are mostly soft & phantom particles.
3 - Yeah, those phantom particles - silicone antifoam droplets (that do not dissolve but form a micro-suspension, when they're good), in high detergent formulations you can get micelles that form just because the polar ends of the molecules are looking for something to do. Any aspect of the lube that prevents the complete transmission of the light can be "read" as a particle, water droplet, air bubble, bit of lint. Yeah, we're doing real useful science here!
4 - I know I have particles, but I'm really not certain where they came from. They may be from the reservoir, they may be from a sample line that wasn't flushed enough, they may be from the sample container, they may be from the pipet you used to pull the sample from the dirty bottle, they may be from the lab atmosphere if the container was left open while the tech had coffee.
Where did your "preferred" level come from, actual experience or someone's lower torso? (Or from a filter supplier who wants to sell you an upgrade?) Are you plugging filters or jamming servos at 16/14/12? Do problems start at 17/15/13? There is one analysis that is as least as important as all of the viscosities & ICPs you can run, but people typically neglect to value it because it's free and it's not provided by some geek in a dirty lab coat: how is your system running?
That's not to say you can't get an early warning from testing, but you also have to guard against alarmism. Is life good at 16/14/12? Then just enjoy life and don't worry until the numbers start trending up.
If you want a useful particle count, you need to take a good sample in an ultraclean container and find someone who does the old-school visual count with a microscope on a gridded filter pad. If you want to know what the particles are, find a lab that will look at the filtrate with a scanning electron microscope with an energy disperse spectrometer (SEM/EDS). That will give you a good idea of the elemental composition of the particles. Compare that to the elemental compostion of the oil & you can start to puzzle out what is contamination & what is degradation.
Thank you for your kind indulgence.
