Cover Glass Deep Dive: The four-point bend test and why it lies
Apr 02, 2026
Every cover glass supplier gives you a strength number. Almost no one tests it the same way. That is not an accident. It is a negotiation tool.
The theory behind four-point bend
You place a glass strip on two lower support pins. Two upper loading pins press down from above. The glass bends until it breaks. A machine records the force at failure. Then you calculate flexural strength in megapascals (MPa).
Why four points instead of three? Four points creates a constant bending moment between the upper pins. That means the entire middle section sees the same stress. Three points concentrates stress directly under the single loading pin. Four points is theoretically cleaner.
The problem nobody talks about
Edge condition dominates the result. A perfect glass bar with polished edges breaks at 600 MPa. The same bar with as-cut edges breaks at 200 MPa. That is a 3x difference. And both are technically correct results for the same material.
Most suppliers send you polished edge samples for qualification testing. Then they ship you as-cut edge parts. Your real world drop performance drops by half. But the spec sheet says 600 MPa. That is the lie.
Other variables that change the number
Crosshead speed : Faster loading makes glass appear stronger (less time for subcritical crack growth). Some labs test at 0.5 mm/min. Some test at 5 mm/min. The faster one shows 15–20% higher strength.
Support span : Wider spans lower the stress for the same force. Narrower spans raise it. If you do not specify span length, your supplier will pick the one that makes their number look best.
Humidity : Glass is sensitive to water. Test at 30% relative humidity and you get 550 MPa. Test at 80% RH and the same glass gives 400 MPa. Moisture attacks the crack tips before they grow.
Sample size : Ten samples give a heroic number. Fifty samples reveal the truth. Weakest link theory applies here. The largest flaw determines the strength, not the average.
What the industry actually uses for quality control
Four-point bend is for R&D and spec sheets. For production, most cover glass makers use ring-on-ring or ball-on-ring testing. Why? No edge preparation needed. You test the as-used surface. That gives a realistic number, but it is usually lower. That is why they do not lead with it.
How to compare two suppliers honestly
Ask for three things in writing:
Edge condition : As-cut, ground, or polished. Only compare same condition to same condition.
Test environment : Temperature (23°C ± 2°C) and humidity (50% RH ± 10% RH) must match.
Weibull modulus : Ignore the average strength. Look at the Weibull modulus instead. It tells you how consistent the glass is. Higher is better. A supplier with m=10 is gambling. A supplier with m=20 has control over their process.
The real-world shortcut
Drop testing on actual product prototypes tells you more than any bend test. It is expensive and slow. That is why engineers keep using four-point bend anyway. Just remember: the number on the spec sheet is the best case. Your real strength is about 40% lower. Plan for that.
