Polycarbonate Plate, Window Grade

Window-grade polycarbonate is thick transparent plate selected for optical quality: polished rather than textured faces and a low count of black specks and inclusions. It is the grade for sight glasses, viewports and machine windows - anywhere somebody has to see through a thick section clearly.

Here is what window grade is NOT, because trade copy gets this wrong constantly: it is not annealed, and it is not lower-stress. No manufacturer datasheet supports either claim. The grade that is stress-relieved is machine grade. Window grade and machine grade are the same resin in the same thickness range, and every published mechanical and thermal property is identical - tensile 9,500 psi ultimate and 9,000 at yield on a 340,000 psi modulus, flexural 13,500 psi on 345,000, compressive 12,500 psi, Rockwell M70 and R118, heat deflection 275F at 264 psi, specific gravity 1.20. The difference is surface and optical selection, nothing else - and note that published haze is the same for both at 0.5 to 2.0%, so haze is not actually the differentiator either. What you are buying is the polished face and the low inclusion count.

Temperature, stated honestly. Heat deflection temperature for plate and rod is published at 275F at 264 psi and 280F at 66 psi by ASTM D648 - the extruded sheet figure is 270F at 264 psi, so do not mix them - but that is a short-duration softening test and must not be used as a service limit. Published maximum continuous service temperature in air is 240 to 250F depending on the publisher, with intermittent exposure to 275F published. Parts under sustained load should be derated well below that, and anything load-bearing near the top of the range needs creep data.

So specify window grade when the part is a window and machine grade when it is a bracket. Paying for window grade buys you appearance, not strength.

Machining and annealing. Polycarbonate machines cleanly with sharp tools and good chip clearance, and it is notch-sensitive, so avoid sharp internal corners, scribed lines and deep tool marks - that sensitivity is the same effect that makes published notched impact figures scatter so widely. Machine-grade plate and rod are supplied stress-relieved and most machined parts need no annealing. Consider annealing when you remove a large proportion of the stock, when you remove it mostly from one face, or when the part will see chemical exposure, because residual machining stress plus a marginal chemical is the main cause of stress crazing. The cycle we work to for unfilled polycarbonate is a ramp to 275F over about four hours, a hold of 30 minutes per 1/4 inch of thickness and a cool-down at 50F per hour, fixtured to prevent distortion, with critical dimensions finish-machined afterwards. Glass-filled grades anneal at 290F.

Chemical compatibility and cleaning, which is where most polycarbonate failures start. Clean it with mild diluted dish soap and a soft microfibre cloth, or an ammonia-free glass cleaner. Do NOT use ammoniated glass cleaner - the common blue kind - because ammonia attacks polycarbonate, and the resin maker's own testing shows 0.1% ammonia is enough to do damage in the presence of water. Also avoid abrasive or high-alkaline cleaners, scrapers and squeegees, dry rubbing, and solvent in direct sun. Isopropyl alcohol diluted 50/50 with water handles oil and tar. Acetone, MEK, methylene chloride, chloroform, toluene, xylene and aromatic fuels all attack it, and so do amines and the plasticisers that migrate out of flexible PVC - which is why PVC gasket, tape and cable tie contact is a real failure mode. The important subtlety is that chemical resistance is not a fixed property: stress cracking only develops where the part is in tension, so a cold-bent, over-torqued or tightly framed part will fail with chemicals a compatibility chart calls safe.