Level Finish · Technical Test Report

Alpha P800 vs 3M Cubitron II 737U: fine refinement, measured

Level Finish Alpha Ceramic P800 disc vs 3M™ Cubitron™ II Hookit™ Clean Sanding Abrasive Disc 737U P800. P800 is a finishing grit: its job is to refine the scratch profile left by the previous stage, not to remove stock, so this test measured surface roughness, refinement from the documented P600 stage, disc condition, and cleaned surface appearance. The test ran on the same surfaces carried forward from the P600 comparison, in two phases: a controlled 40-second trial, then an additional 1 minute 40 seconds of endurance sanding with the same discs. Alpha recorded the lower average Ra in both phases and the larger P600-to-P800 refinement. Cubitron II recorded two narrow consistency edges in the initial sample. Disc condition was a tie. All of it is published here.

The P800 test panel, its strips coordinate-marked for matched Ra sampling, carried forward from the documented P600 surfaces.

What happened, in short

• Initial finish (40 seconds): Alpha averaged 0.238 µm Ra vs 0.304 µm for Cubitron II, 21.70% lower, with the lower median as well (0.237 vs 0.315 µm).
• Refinement from P600: from the documented P600 averages (Alpha 0.409 µm, Cubitron II 0.395 µm), P800 reduced average Ra by 41.76% for Alpha and 22.99% for Cubitron II. Alpha started this stage slightly rougher and finished materially finer.
• Endurance (additional 1:40): Alpha averaged 0.226 µm vs 0.261 µm, 13.37% lower, with the tighter distribution.
• Where Cubitron II led: the initial sample's absolute range was slightly narrower (0.087 vs 0.092 µm) and its coefficient of variation lower (10.49% vs 12.80%). Both are published in the tables below.
• Disc condition: both discs passed pre-test inspection, stayed intact through both phases, and are scored a tie. No loading winner is assigned, because photographs do not quantify loading.

Method and context

Initial P800 testing was performed on the surfaces carried forward from the previous P600 comparison (see the P600 test results linked below), so each product refined the surface its own P600 stage had produced. Each product sanded for 40 seconds, and 10 Ra readings per product were recorded. The endurance phase added 1 minute 40 seconds per product with the same discs (2 minutes 20 seconds total use), with 7 Ra readings per product, retained in the order supplied. Discs and panels were photographed before testing, after the 40-second trial, and after endurance, with dust present and after clean-off. No material-removal metric was taken: P800 was evaluated as the refinement step it is.

For application context: 3M's blend-panel sanding process specifies grade 800 with a DA sander and soft interface pad in the color-blend area, while other systems specify finer preparation for entire blend panels (Axalta documents P1000 to P1200 in one current process, and PPG directs users to the selected product's data sheet). P800 is not a universal final grit. The right choice is the product that performs best inside the coating manufacturer and OEM procedure for the repair.

Results: surface finish, 40-second trial

Ra, 10 readings each	3M Cubitron II P800	Level Finish Alpha P800
Average Ra	0.304 µm	0.238 µm
Median Ra	0.315 µm	0.237 µm
Minimum reading	0.256 µm	0.196 µm
Maximum reading	0.343 µm	0.288 µm
Range	0.087 µm	0.092 µm
Sample standard deviation	0.032 µm	0.030 µm
Coefficient of variation	10.49%	12.80%

Alpha's average was 0.066 µm lower, and its median was lower too, so the result does not rest on one unusually fine reading. Cubitron II's readings spanned a slightly narrower absolute range and varied less relative to their own mean. All 20 readings, exactly as recorded: Cubitron II 0.343, 0.329, 0.329, 0.323, 0.303, 0.256, 0.325, 0.264, 0.307, 0.263. Alpha 0.288, 0.235, 0.223, 0.257, 0.196, 0.274, 0.238, 0.200, 0.255, 0.216.

Average Ra after the 40-second trial (µm). Lower is finer.

Results: refinement from the P600 stage

P600 to P800	3M Cubitron II	Level Finish Alpha
Documented P600 average Ra	0.395 µm	0.409 µm
P800 40-second average Ra	0.304 µm	0.238 µm
Absolute reduction	0.091 µm	0.171 µm
Percent reduction	22.99%	41.76%

Alpha produced about 1.88 times the absolute average-Ra reduction measured for Cubitron II across the documented P600-to-P800 progression. This describes the tested surfaces and timing; it is not a universal cutting-rate ratio for other substrates, tools, or operators.

Results: endurance, additional 1 minute 40 seconds

Endurance Ra, 7 readings each	3M Cubitron II P800	Level Finish Alpha P800
Average Ra	0.261 µm	0.226 µm
Median Ra	0.256 µm	0.219 µm
Minimum reading	0.199 µm	0.195 µm
Maximum reading	0.307 µm	0.274 µm
Range	0.108 µm	0.079 µm
Sample standard deviation	0.033 µm	0.028 µm
Coefficient of variation	12.75%	12.61%

All 14 endurance readings, exactly as recorded: Cubitron II 0.254, 0.254, 0.256, 0.279, 0.199, 0.276, 0.307. Alpha 0.231, 0.195, 0.219, 0.211, 0.200, 0.274, 0.251. One measured comparison worth stating plainly: Alpha's average after only the initial 40 seconds (0.238 µm) was already lower than Cubitron II's endurance average (0.261 µm). Also stated plainly: Cubitron II's average fell further between its own phases (14.30% vs 5.18%) because it began the endurance phase from the rougher initial average, so the report does not score that drop as an effectiveness win for either side.

Disc condition and surface appearance

Before testing, both discs passed visual inspection. Both showed localized discoloration around portions of the perimeter and several extraction holes, reported by the operator as consistent with the laser-cutting process for both products. It was recorded as a pre-existing cosmetic condition, not a defect, and it is not counted as sanding damage anywhere in this report.

After the 40-second trial, Cubitron II showed light, broadly distributed residue with holes predominantly open; Alpha showed light-to-moderate distributed residue with localized accumulation around some openings. After endurance, both showed moderate distributed residue, predominantly open extraction patterns, and intact perimeters, center openings, and backings. No severe clogging, major tearing, delamination, or gross edge failure was established for either product at any point. Photographed disc condition is scored a tie.

On the panels, cleaned photographs of both products retained some overlapping curved and orbital pass structure. Cubitron II's structure appeared more conspicuous and more broadly distributed in both phases; Alpha's appeared comparatively more uniform, with residual structure more localized. These are photographic observations, not scratch-depth measurements, and no mark in this P800 test is classified as a confirmed pigtail defect for either product.

The endurance panel after clean-off. Pattern structure that persists after dust removal is photographed pattern, documented as such; photographs do not establish scratch depth.

Category results and the report's overall finding

Category	Leader	Basis
40-second average Ra	Alpha	0.238 vs 0.304 µm, 21.70% lower
40-second median Ra	Alpha	0.237 vs 0.315 µm
P600-to-P800 Ra reduction	Alpha	41.76% vs 22.99%
40-second standard deviation	Alpha, slight	0.030 vs 0.032 µm
40-second range	Cubitron II, slight	0.087 vs 0.092 µm
40-second coefficient of variation	Cubitron II	10.49% vs 12.80%
Endurance average and median Ra	Alpha	0.226 vs 0.261 µm; 0.219 vs 0.256 µm
Endurance distribution	Alpha	SD 0.028 vs 0.033 µm; range 0.079 vs 0.108 µm
Cleaned visual uniformity, both phases	Alpha, photographic	Less broadly persistent pass structure in the supplied images
Disc integrity and condition	Tie	Both intact; no supported durability winner

The test report's overall finding: for fine P800 refinement with low average Ra, a controlled distribution, and a comparatively uniform cleaned surface, Level Finish Alpha is the stronger choice under these documented conditions. Cubitron II remained mechanically serviceable, continued to refine during extended use, and recorded the two narrow consistency edges noted above. That finding reflects this test's documented conditions, not third-party certification.

How to read this test

At P800 the contest is not removal, it is control: how fine the surface gets, how repeatable the readings are, and what the cleaned panel looks like before paint. Alpha led the measured finish in both phases and the refinement from P600. Cubitron II's strengths here are real but narrow: a marginally tighter initial range and relative variation, and sound physical condition throughout, which Alpha matched. If the priority is the finest, most uniform refinement surface this test could measure, the record points to Alpha. We publish every category either way, so you can decide for yourself.

More test results

Alpha P600 vs 3M Cubitron II, the stage these surfaces came from: see the P600 test results. Alpha P180 vs 3M Cubitron II, two controlled tests: see the P180 test results. Alpha P80 vs 3M Cubitron II, two controlled tests: see the P80 test results. Omega P1500 vs 3M Purple 260L: see the P1500 test results. Omega P2000 vs 3M Purple P2000, filmed: see the filmed test results. More reports are filed here as testing is completed.

Notes and limitations

This is a single controlled comparison under its documented conditions, not third-party certification, and the results describe this test. Ra is an arithmetic average and does not by itself characterize individual deep scratches or the complete scratch field. Photographs are qualitative evidence; lighting, angle, abrasive color, and residue contrast limit image-to-image comparison, which is why no loading or durability winner is assigned from them. Endurance readings were supplied without coordinate labels and are published in supplied order. P600 averages are imported from the previous P600 report solely as the documented preceding-stage reference. Paint-system and OEM sanding requirements differ; always follow the procedure and technical data for the coating system being applied. Results from one grit or substrate do not transfer to another.

Frequently asked questions

Which P800 disc finished finer?

Alpha, in both phases: average Ra 0.238 µm vs 0.304 µm after 40 seconds, and 0.226 µm vs 0.261 µm after the endurance phase.

How much did P800 refine the P600 surface?

Alpha reduced average Ra by 41.76% from its documented P600 average (0.409 to 0.238 µm). Cubitron II reduced its average by 22.99% (0.395 to 0.304 µm).

Did either disc fail or clog?

No. Both discs remained structurally intact through 2 minutes 20 seconds of total use, with distributed residue and predominantly open extraction holes. Condition is scored a tie, and no loading winner is assigned because photographs cannot quantify loading.

Were pigtails found in this test?

No mark in this P800 test is classified as a confirmed pigtail defect for either product. Both cleaned panels showed curved and orbital pass structure; Cubitron II's appeared more broadly distributed. Photographs document pattern, not scratch depth.

Did Cubitron II lead any category?

Yes. In the initial sample its absolute Ra range was slightly narrower (0.087 vs 0.092 µm) and its coefficient of variation lower (10.49% vs 12.80%). It also continued to lower its average Ra during extended use and matched Alpha on disc integrity.

Level Finish · St. George, Utah · level-finish.com · Source: Level Finish P800 Abrasive Comparison Test Report (September 2026), 3M Cubitron II 737U vs Level Finish Alpha Ceramic. Page prepared November 13, 2026. All numbers transcribed exactly from that report; no values added or altered.