Views: 0 Author: Site Editor Publish Time: 2026-10-09 Origin: Site
A steel ball inspection report is the only proof you get that the balls in the carton match the grade on your purchase order. That works fine until a batch causes noise or vibration on the assembly line, and someone has to work out what the report actually said.
Steel balls end up in bearings, gearboxes, valves, medical devices and a long list of other equipment, and their size and finish decide how smoothly that equipment runs and how long it lasts. A report is where those micron-level requirements become numbers you can check.
The items are much the same for chrome steel (AISI 52100 / GCr15), stainless steel balls (304, 316L, 440C), carbon steel, alloy steel and ceramic balls. What changes is the limits.
| Item | What it measures | G10 limit | G100 limit |
Vdws: ball diameter variatio | Largest minus smallest diameter measured on a single ball | 0.25 µm | 0.25 µm |
SPH (ΔSph): deviation from spherical form | How far a single ball departs from a true sphere, i.e. roundness | 0.25 µm | 0.25 µm |
Vdwl: lot diameter variation | Largest minus smallest mean diameter among the balls in one lot | 0.5 µm | 0.5 µm |
| Ball gauge / mean diameter deviation | How far the lot's mean diameter sits from nominal | As ordered | As ordered |
For the full grade table, see our buyer's guide to steel ball tolerance grades.
The thing to notice is that Vdws and SPH describe individual balls, while Vdwl describes the lot. A batch can be made of perfectly round balls and still fail Vdwl if some balls are slightly bigger than others. In a bearing, that matters: balls of different sizes don't share the load evenly.
Precision balls are often sorted into gauges, sub-lots whose mean diameter sits a fixed step above or below nominal.
Surface roughness (Ra) is the average height of the surface profile, in µm. The limit depends on the grade: 0.020 µm for G10 and 0.10 µm for G100 under ISO 3290-1. It is measured with a stylus profilometer or an optical instrument such as a white-light interferometer. On very smooth balls, the instrument and settings affect the reading, so for G10 to G28 orders it's worth asking which method was used. How lapping produces that finish is covered in our lapping process guide.
Visual inspection checks for cracks, pits, rust, scratches and soft spots under defined lighting. On most reports this is a single line reading "no defects" or "OK". For critical applications, ask what lighting and magnification were used and how many balls were checked.
A hardness line usually looks something like "HRC 62–64 (n = 5)". Check three things.
The scale. Hardened steel balls are tested in HRC. Annealed or soft balls are tested in HRB or HV, and very small balls are often tested in HV. A value without a scale isn't a result.
Where and how it was tested. A Rockwell test on a curved surface reads lower than on a flat one. Some reports show corrected values, others test a small flat ground onto the ball. If the report looks low against your specification, check this first. Our article on Rockwell corrected values for spherical hardness explains why.
For case-hardened carbon steel balls, surface hardness alone doesn't describe the ball. Case depth matters just as much. See case-hardened carbon steel balls: hardness and case depth.
These items usually don't appear on every batch report. They come from separate documents, from periodic testing, or are done on request.
Chemical composition normally comes from the steel mill certificate, traced by heat number. Check that the heat number on the certificate matches the one on the report.
Microstructure For bearing steel balls, metallographic checks cover carbide distribution, banding and non-metallic inclusions, rated against the steel standard. This is usually done per heat-treatment lot.
Crushing load is the force needed to break a ball. The required value depends on diameter and material, so a crushing load figure without a ball size means nothing. The test is destructive and is done on samples.
Residual magnetism matters for bearings and instruments. It is measured with a gaussmeter and is only reported if specified.
Non-destructive testing Eddy current and magnetic particle testing find surface cracks; ultrasonic testing finds internal defects in larger balls. These are normally done on request.
Many factories sort 100% of balls for diameter and surface defects on automatic sorting machines. The measured values on the report, though, come from a sample. The sample size should be stated, and for attribute checks it is usually drawn to a sampling plan such as ISO 2859-1 (GB/T 2828.1 in China).
No lot number, or a lot number that doesn't match the cartons
Measured values with no limits beside them
Every value identical, or suspiciously round
Hardness with no scale, or no number of balls tested
Ra with no method on a G10–G28 order
A report dated before the balls were produced
A generic certificate of conformity instead of measured data
A standard or grade that doesn't match your purchase order
Every batch from RISHENG ships with a material certificate and a dimensional inspection report. Our standard report covers diameter, ball diameter variation (Vdws), deviation from spherical form (SPH), lot diameter variation (Vdwl), hardness and visual inspection, plus surface roughness (Ra) for precision grades.
The measurements are taken in our own QC lab, using roundness testers, surface roughness testers, Rockwell hardness testers and three-point diameter gauges. If your application needs more, we can also arrange crushing load tests, residual magnetism measurement and non-destructive testing on request.
If you’d like to see what our reports look like before ordering, ask us for a sample report.
Vdwl is lot diameter variation: the difference between the largest and smallest mean diameters among the balls in one lot. It shows how consistent the batch is, which matters in bearings because balls of different sizes share the load unevenly.
Check the scale and the test method first. A Rockwell test on a curved ball surface reads lower than on a flat surface, so the value may need a curvature correction, or the test may have been done on a ground flat.
Usually not. Crushing load is a destructive test, so it is done on samples, typically per heat-treatment lot, periodically, or when the customer asks for it. The required value depends on ball diameter and material.
Diameter and surface defects are often 100% sorted by machine, but the measured values on the report come from a sample. A good report states the sample size for each item.
