Sunday, January 24, 2010

Drop Test on PVC Profiles

Skippy - hey Buzz - here is another typical extrusion scenerio -

“Using ASTM D4495 to test rigid PVC profiles from extrusion. The problem I’m having is that very few of the profiles are failing . . .

The standard calls for room temperature conditioning, and when a 10 lb weight strikes the profiles, even from a height of 30 inches, no damage is done as far as failure, denting yes, no cracking . . .

I’ve even increased the weight to 14lbs and nothing. Any ideas”

Buzz - As the standard points out


results obtained by use of this test method can be used in two ways:

a) As the basis for establishing impact-test requirements in product standards, and or

b) To measure the effect of changes in materials or processing.

Skippy - isn't drop dart impact testing by profile extrusion houses done in foot pounds per mil (so two different wall thickness samples from the same materials would be subjected to different pass/fail drop heights/weights but the same overall criteria?)

Buzz - yes and is used to determine the limits of acceptable ductile or brittle failure modes - mention made above is that "only a few are failing". Actually at the maximum for acceptable failure mode - none should fail - perhaps the testing is at too high a standard to be appropriate (just beyond what would be "acceptable") or the testing isn't destructive enough if you are hoping for failure more often or every time?

Skippy - Additional questions requiring study come to mind given the presented information

- Are you testing for ductile or brittle failure
- is your calculation for inches of height per pound correct
- how long is the product resting between production time and testing
- would the application benefit or call for an annealing process involved to relax stress
- how thick is the product in mils
- what tup shape are you using
- etc

Buzz - Perhaps just as important, after the drop dart testing, are you giving it a more "real life" type 'whammo' testing - the kind of test when a worker punches through it with a nail, or drops it from the truck, or off a roof, in hot and more importantly cold weather; etc?

Also, we are assuming that dimensions and aesthetics are acceptable and accounted for in the testing at current (measured and reproducable) run conditions - aka you should be doing this testing to failure on parts that are in all other ways acceptable - if not, when or as you make additional changes, your test results may be suspect or worse, value-less.

Just our two cents -
Skippy and Buzz