Want to share our oil analysis program setup since it took us a couple years to get it dialed in.
Sampling location matters enormously and most people get it wrong. Don't sample from the drain plug — that gives you settled sediment and worst-case numbers. Don't sample from the fill port — that gives you the cleanest oil at the top. Install a dedicated sampling valve at mid-height on the return line while the system is running. That gives you a representative sample of what's actually circulating through the bearings.
Sampling frequency: we do quarterly on presses running single shift, monthly on 24/7 presses. But here's the key — always sample at the same operating condition. We sample after 2 hours of production, never cold start and never end of shift. Temperature affects viscosity readings and particle suspension, so inconsistent sampling conditions make your trend data useless.
Our action limits (developed over 5 years of data on our specific presses — yours may differ):
- Iron >50 ppm OR >15 ppm increase between samples = investigate
- Copper >25 ppm = bearing cage wear starting
- Silicon >20 ppm = contamination ingress, check seals
- Water >150 ppm = find the source immediately
- Viscosity change >10% from baseline = oil degradation or wrong oil added
- ISO cleanliness worse than 18/16/13 = filter maintenance needed
The copper one caught us off guard. We always watched iron for bearing wear, but copper spiked first on two of our presses because the bearing cages are bronze. By the time iron shows up, the bearing race is already damaged. Copper is your early warning.
Cost: about $35-45 per sample at most commercial labs. We use Polaris Laboratories — 3-day turnaround, results emailed as PDF with trend charts. For 6 presses sampled quarterly, that's under $1,000/year. Cheapest insurance in the shop.