Craters and Fish Eyes: How to Trace Silicone Contamination on Your Line
Craters and fish eyes in a paint film are the signature of a contaminant that breaks surface tension, and the source is usually compressor oil, silicone-based maintenance products or welding anti-spatter sprays. This article shows how to trace the contamination station by station, how to tell it apart from pinholing, and how to fix the compressed air line permanently.
Enlarge
Craters and fish eyes in a paint film are almost always caused by silicone contamination or oil residue. A contaminant film far thinner than a micron lowers the surface tension of the molten coating at that spot; the paint retreats from it and leaves a round depression with an open centre and a raised rim. A few parts per million of silicone is enough to produce the defect, which is why the answer is never "apply a thicker film" — it is to find the source and cut it off. On the lines we look at, the source list comes down to the same four headings: compressor oil in the compressed air line, welding and anti-spatter sprays, machine maintenance lubricants, and silicone-bearing products the operators use. Pushing pretreatment towards the upper end of the 3–7 stage band removes part of this load, but if oily air reaches the gun no washing stage will close the defect, because the contaminant travels to the surface together with the coating.
The defect looks like this: a circular depression between roughly one and a few millimetres across, sometimes opening all the way down to bare metal, with a slightly raised rim. The name fish eye comes from the contaminant particle often visible at the centre of the crater. The defect never distributes at random — its distribution is what identifies the source.
Crater or pinhole? Name the defect correctly first
The most expensive mistake on the shop floor is mistaking a crater for a pinhole and then spending weeks adjusting the oven. They are different physics: a crater is an external contaminant disturbing surface tension, while a pinhole is gas escaping from the substrate or the coating and punching through the film while it is still fluid. Confuse the two and you will chase oven temperature for weeks without ever opening the compressor room.
| Distinguishing feature | Crater / fish eye | Pinhole |
|---|---|---|
| Root cause | Oil, silicone, dirty compressed air — surface tension gradient | Gas escape: casting porosity, galvanising, trapped moisture |
| Appearance | Round depression, raised rim, open centre | Needle-point hole, no raised rim |
| Size | Millimetre scale, clearly visible | Usually very small, visible under raking light |
| Distribution | Clustered: one zone, one hanger position, one shift | Around welds and cast areas, worse in thick film |
| Relation to the oven | Persists even after the oven regime is corrected | Drops noticeably when ramp and pre-drying change |
| Correct action | Trace and cut the contaminant source | Fix pre-drying and the heat ramp |
We covered that distinction and the separate diagnostic sequence for gas-driven defects in the article on pinholing in powder coating. The practical rule: if the defect is clustered, think crater; if it follows the material, think pinhole.
Where does the contaminant come from? Four main sources
1. The compressed air line. The most common source and the one most often overlooked. Oil vapour carried over from a screw or piston compressor condenses into a thin film inside the pipework and is sprayed straight onto the part through the gun. The symptom is typical: the defect concentrates in the middle of flat surfaces along the gun's sweep direction, and it peaks at the start of the shift, after the line has stood idle longest.
2. Welding and anti-spatter sprays. Silicone-bearing anti-spatter agents are the classic contamination source. On parts arriving from the welding station, the defect arranges itself in a ring around the weld seam.
3. Maintenance and lubrication products. Conveyor chain oil, hanger lubricant, penetrating sprays and pneumatic line lubricators. Oil dripping from the conveyor leaves an unmistakable signature: the defect appears on the upper surface of the part, always along the same line.
4. Operators and environment. Hand cream, silicone polish, aerosol sprays, silicone sealing work carried out on a neighbouring line. The mark of environmental contamination is a scattered, irregular distribution that cannot be tied to any single station.
Tracing the contamination station by station
You cannot find the source by guessing; you split the line and eliminate. The sequence below runs from the cheapest test to the most expensive, and usually gives an answer within the first three steps.
| Step | Test | What you see | Conclusion |
|---|---|---|---|
| 1 | Take a clean panel, blow gun air only on it, then coat and cure | Craters appear on that panel too | Contaminant is in the compressed air |
| 2 | Drop water onto a dry part leaving pretreatment | Water beads up and runs off instead of spreading | An oil film remains on the surface |
| 3 | Cure a welded and a non-welded part on the same hanger | Craters only on the welded one | Anti-spatter spray is the source |
| 4 | Match hanger position to defect for a full shift | Always the same hanger position | Conveyor oil is dripping |
| 5 | Run a small batch with virgin powder and a separate gun | Defect disappears | Contaminant is in the reclaimed powder |
| 6 | Measure booth air velocity, inspect filter outlet | Velocity outside the band, oily filter surface | Problem on the booth / filter side |
Step five matters most: once a contaminant enters the recovery circuit, the whole powder inventory is contaminated and the defect keeps appearing for days after the source has been cut. In a filtered booth, recovery runs at 95%+ and cartridge filter efficiency at 99.9%+, so the system collects powder very efficiently — which is a virtue, but in a contamination event it recirculates the dirt just as efficiently. Once contamination is confirmed, you do not go back into production before the fluidising hopper, hoses, sieve and cyclone have been emptied and cleaned.
The permanent fix: rebuild the compressed air line
Most crater cases come from an air preparation chain that is either incomplete or unmaintained. The air feeding the coating line must be separated from the general workshop supply and given its own preparation set. The correct order is: compressor outlet → aftercooler → receiver → dryer → prefilter → coalescing (oil vapour) filter → activated carbon filter → pressure regulator → gun. There must never be a lubricator upstream of the gun; if pneumatic equipment needs lubrication, that branch is fed separately from the coating line.
Points to watch when specifying the air line:
- Dryer capacity is selected for summer conditions; the dew point must stay below the temperature of the coldest point in the building, otherwise condensation forms in the pipework.
- The carbon filter is the only stage that captures oil vapour and it has a finite life; replacement goes on a calendar, because "I change it when it clogs" does not work here.
- Pipe material must be non-rusting; rust and scale forming in galvanised or black pipe generate contamination downstream of the filters.
- Drainage must be automatic; oily water left to manual draining in a receiver will reach the line sooner or later.
- Line capacity is calculated from gun count; in the booths we build with 2 to 12 guns, air flow and filter sizing are engineered together with the number of guns.
Once this chain is built correctly, the large majority of crater cases close permanently. Compare the investment against the cost of a single scrapped batch — the air preparation set is normally the cheapest item in that comparison.
Lasting measures on the pretreatment and booth side
The degreasing stage is the first line of defence against craters. A dip surface treatment line runs degreasing, phosphating, passivation and drying stages in sequence; if bath temperature, chemical concentration and rinse water conductivity are not tracked, degreasing efficiency drops silently and the defect returns without warning. How many stages you actually need, and when adding one is justified by the soil profile, is set out in how many pretreatment stages you need.
Two points matter on the booth side. The first is internal air velocity: the 0.5–0.8 m/s band in a filtered powder booth was chosen to be fast enough to hold powder inside the booth and slow enough not to strip fresh powder off the part; a booth running outside that band also carries ambient contamination onto the part. The second is the position of the openings — if the booth mouth faces an oily machine or a welding station, contamination is being invited into the line. Booth selection criteria are collected in choosing a powder coating booth, and particle-driven surface defects in dust and fibre contamination on the coated surface. On the curing side, running samples in a box type electric oven working between 80 and 250 °C with ±3 °C uniformity is the most practical way to hold the variables constant while you diagnose.
Frequently asked questions
How do I tell a crater from a pinhole?
Look at the rim and the distribution. A crater is a round depression with a raised rim and an open centre, and it appears in clusters; a pinhole is a needle-point hole and it follows the material. To be certain, cure two parts from the same batch on the same hanger, one welded and one not: if the defect appears only on the welded part it is almost certainly a contaminant, and if it appears on both in thick film areas it is gas escape.
How do I confirm there is oil in the compressed air?
The fastest test is this: blow nothing but gun air onto a clean, uncoated panel for a few minutes, then coat and cure it. If craters appear on that panel too, the contaminant is in the air. The second check is the drain valve on the filter bowls; if what comes out looks milky or smells of oil, the coalescing and carbon filters are long since saturated.
How do I clean a line that has been contaminated with silicone?
Silicone does not come off with water and it tends to spread. Every affected surface — hangers, conveyor, booth interior, hoses, fluidising hopper — is wiped with a suitable solvent, using disposable cloths, and the same cloth is never used on a second surface. Contaminated powder is discarded rather than rescued by sieving. The first batch after cleaning is always cured as a sample and inspected.
Can hangers and fixtures cause craters?
Yes, and often. Hangers carrying built-up coating lose conductivity and stop earthing the part properly, and residues of welding oil and lubricant accumulate exactly at the hanger contact points. Hangers must be cleaned regularly by burn-off or stripping, and a cleaned hanger must never be set down on an oily surface.
Can cratered parts be sanded and recoated?
Not before the contaminant is removed. If you apply a second coat over a crater, the defect usually reopens in the same place because the contaminant is still there. The correct sequence is: solvent-clean the surface, abrade it mechanically, degrease and dry it again, then coat. Any rework done before the source is cut simply produces a second scrap part.
Summary
Craters and fish eyes are not a settings defect; they are a cleanliness and contamination control defect. The diagnostic order is clear: separate the defect from pinholing, test the compressed air line, eliminate welding, maintenance and operator sources, and finally check whether the reclaimed powder itself has been contaminated. The permanent fix is nearly always in the same place — a compressed air preparation chain dedicated to the coating line, with a dryer, coalescing and carbon filters and no lubricator anywhere upstream of the gun. Replacing equipment before that chain exists is money spent in the wrong place. Every line we build is engineered around the customer's largest part and delivered with a 2 year manufacturer warranty.
Send us a close-up photograph of the defective part, the air preparation schematic of your line and your pretreatment stage list — we will work out together which station the contamination is coming from.