You lapped the seal face, checked it under the optical flat, and the light bands are wrong. Before you adjust your technique, stop. The most common mistake I see technicians make is assuming the problem is in their hands when the problem is under them — on the lapping plate itself.
Most lapping failures trace back to plate condition, not operator technique. A plate that has drifted 0.005 mm out of flat transfers that error directly to every seal face you lap on it. Yet plate verification is the step most technicians skip entirely. If you skip this step, you’ll be back in a week doing the same rework.
Check the Lapping Plate First
A plate drifts 0.005-0.01 mm for every 20 minutes of continuous lapping. That means a plate verified flat at the start of a shift can be out of spec before lunch. Plates in continual use need flatness checks at least twice a day — not once a week, not when results go bad.
How to Verify Plate Flatness
Three methods, in order of priority:
- Straight edge sweep. Slide a precision straight edge across the plate surface in multiple directions. The rocking point or gap location tells you whether the plate is convex or concave. Do this first — it takes 30 seconds and catches gross errors immediately.
- Dial indicator gauge. A flatness gauge calibrated to 0.001 mm, mounted on three support points and zeroed against a known flat reference, gives you a quantified reading. Positive reading means convex plate. The reading predicts the flatness you will transfer to every part lapped on that plate.
- Test block method. Lap a brass test block on the plate, then check the block under an optical flat. A convex test block surface means the plate is concave. This inverted relationship confuses operators constantly — they see a convex test piece and assume the plate is convex, then correct in the wrong direction.
The plate maintenance threshold is 0.004 mm. Anything beyond that, and you are lapping defects into your seal faces regardless of technique. With a properly maintained plate, achievable seal face flatness reaches one helium light band (0.0003 mm) for workpieces up to 400 mm diameter.
Correcting a Convex Plate
Remove one conditioning ring and position the remaining ring(s) toward the plate center. The outer edge of the ring should just sweep the plate perimeter — push it too far inward and you will create a step at the plate edge. Expect correction rates of 0.005-0.01 mm per 20 minutes on 15-24 inch plates.
Correcting a Concave Plate
Move conditioning rings outward. But here is the trap: concave correction runs several times faster than convex correction. Technicians who apply the same correction time they used for convex plates will overshoot into convex territory and create a new problem. Check the plate every five minutes during concave correction.
Replace any plate worn below 3 mm thickness. No amount of conditioning recovers a plate that thin.
Surface Defect Diagnosis
Scratches and scoring after lapping almost always point to contamination or abrasive mismatch, not hand pressure errors.
Scratching on Soft Materials (Carbon, Resin-Impregnated Carbon)
Carbon seal faces scratch when the abrasive is too coarse for the finishing stage or when previous-stage grit contaminates the plate. The most common mistake I see is skipping the plate cleaning step between grit changes. One particle of 15-micron diamond on a plate loaded with 3-micron slurry will score every carbon face you touch.
Target surface finish for carbon is 0.1 um Ra. If scratches are visible under 10x magnification after lapping, the plate surface is contaminated. Clean the plate completely, reload fresh abrasive, and restart. Do not try to “lap through” the scratches — you will embed the contaminating particle deeper into the plate.
Scratching on Hard Faces (SiC, Tungsten Carbide)
Hard face materials resist scratching from abrasive contamination but are vulnerable to microchipping when lapping pressure is too high. Tungsten carbide requires approximately 300 g/cm2 of lapping pressure; silicon carbide needs even less careful handling but targets 0.04 um Ra surface finish versus TC’s 0.01 um Ra.
If you see small pits or chipped edges on SiC or TC faces, reduce pressure and check that your diamond abrasive grade matches the material hardness. A wet plate — too much lubricant — causes the part to hydroplane, producing uneven contact and localized pressure spikes that chip brittle materials. The optimal plate condition is “just moist,” not wet.
Poor Surface Finish Despite Correct Abrasive
When the abrasive is correct but the finish will not improve, the plate itself is often the issue. A worn, glazed plate loses its ability to hold abrasive in the surface pores. For soft materials, switch to a polishing cloth overlay with 1-3 micron diamond paste for the final finish stage. This is not a workaround — it is the correct two-stage approach for carbon and similar materials.
Flatness and Geometry Problems
Concave parts, convex parts, and taper all trace directly to plate geometry. The defect on the part is the mirror opposite of the plate condition — and this inverse mapping is the single most useful diagnostic tool in lapping troubleshooting.
Concave Seal Faces
A concave seal face (dished center, high edges) means your lapping plate is convex. Verify with a straight edge: the plate will rock on the center high point. Correct the plate toward flat before relapping the seal face.
Convex Seal Faces
A convex seal face (high center, low edges) means your plate is concave. This is the more dangerous condition because concave plates also trap abrasive in the center, accelerating uneven wear and making the problem progressively worse with each lapping cycle.
Taper and Parallelism Problems
When seal faces show uneven thickness — one side higher than the other — the cause is usually uneven pressure distribution during lapping. Before you start, make sure your hand pressure weight is itself lapped flat. Use odd numbers of parts on the plate (three, five, seven) to prevent tracking patterns. Flip parts and lap a minimum of three full cycles to average out any asymmetry.
For high-pressure applications above 40 bar, the acceptable flatness tolerance drops to one helium light band maximum. Taper that passes inspection for general service will cause immediate leakage in high-pressure duty. Check your specifications before accepting borderline results.
Material-Specific Lapping Problems
Carbon faces, silicon carbide, and tungsten carbide each fail in different ways during lapping. Treating them identically is a reliable path to rejected parts.
Carbon Face Relaxation
Carbon is the material that fights back after lapping. A carbon seal face lapped to one light band may relax to two or three bands within hours as internal stresses redistribute. This is a material behavior, not a lapping defect. Allow carbon faces to rest for several hours after lapping, then reverify. If the face has relaxed beyond spec, a light re-lap with fine abrasive usually brings it back.
The most common mistake I see is technicians re-checking carbon faces immediately after lapping, seeing one band, signing off, and then wondering why the seal leaks a week later. The relaxation already happened — they just did not wait long enough to see it.
Silicon Carbide Grain Pullout
SiC faces can develop small voids during lapping when individual grains pull out of the matrix. This happens most often with sintered (pressureless) SiC when abrasive particle size approaches the grain boundary dimension. If you see tiny scattered pits on an SiC face that do not correlate with scratching patterns, reduce abrasive particle size and check your pressure. Target finish is 0.04 um Ra.
Tungsten Carbide Sensitivity
TC demands the finest surface finish of any common seal face material — 0.01 um Ra. It also requires the highest lapping pressure at roughly 300 g/cm2 compared to carbon’s 100 g/cm2. The combination of high pressure and ultra-fine finish means TC faces are unforgiving of plate condition errors. A plate that produces acceptable carbon faces can still produce rejected TC faces because the tighter tolerance amplifies plate geometry errors.
Lapping Mistakes Even Experienced Technicians Make
After ten years of reconditioning lapping plates and training maintenance crews, three mistakes stand out. They are not beginner errors — they are habits that experienced technicians develop precisely because they have been lapping for years without checking.
Skipping Plate Verification Between Jobs
The technician lapped a batch of carbon faces successfully, then immediately starts on a TC job without rechecking the plate. The carbon batch wore the plate unevenly, and now every TC face comes out with taper. Before you start, make sure the plate is verified for the job you are about to do, not the job you just finished.
Using the Wrong Lubricant
Petroleum-based lubricants, silicone oils, and PTFE sprays all contaminate lapping plates. The only correct lubricants are the water- or oil-based fluids designed for lapping — and even these should be used sparingly. Over-lubrication prevents conditioning rings from rotating freely and causes parts to hydroplane. The contamination from one application of the wrong lubricant can persist through multiple plate reconditioning cycles.
Handling Lapped Faces with Bare Hands
Fingerprints deposit oils onto a freshly lapped seal face. Those oils change the wetting behavior of the face during operation, creating localized dry spots that accelerate heat checking and wear. Use plastic drivers or a socket extension to handle finished parts. Never touch the lapped surface directly — the damage is invisible until the seal is installed and running.
Before You Lap Again
Every lapping defect I have traced in the field started with the same question: did anyone check the plate? Nine times out of ten, the answer is no. The seal face was telling the technician exactly what was wrong with the plate — concave face meant convex plate, scratches meant contamination, taper meant uneven pressure weight — but the technician skipped straight to re-lapping.
Start every lapping session with a 30-second straight edge check. Verify the plate with a dial gauge at least twice per shift. Match your abrasive, pressure, and lubricant to the specific material you are lapping. Let carbon faces rest before final verification. These four steps eliminate the majority of lapping rework and rejected seal faces.
When a seal face defect persists after correcting plate condition, technique, and abrasive selection, the face material itself may be damaged beyond what lapping can repair. Know when to stop lapping and replace the part.
