Impact force (testing an adhesive bond) 1

[bronx68]

We recently got a requirement from a customer that a bond needed to be able to with stand an impact force of about 15 lbs. We built up a simple test rig to basically drop an 8 lbs. weight a distance of 6 inches onto the bond area. Both objects are hard materials (i.e. one is steel and the other is delrin). I calculated the velocity of the object just before collision to be 5.67 feet/sec. The change in velocity (deltaV) is the difference between the velocity at impact and 0 velocity. I then solved for the average impact force. Aver Force Impact = (m * deltaV/time). Easy so far but what should I use for time?? I used .1 seconds because it seems reasonable and I got 14 lbs force. Any better ideas for testing an adhesive bond for impact force?

 

[btrueblood]

Without any more definition, I would default to a "step" loading - i.e. a 15 lb. weight dropped from essentially "zero" height. Your estimate of the time duration of an impact between a steel part and a Delrin part seems high (I'd make a WAG at less than .01 sec), but the real answer depends on the flexibility and damping characteristics of both parts.

 

[Compositepro]

Force is not a measure of impact. Impact is almost always measured in units of energy. Even energy, by itself is not a good measure unless the test method is defined. Better use an ASTM or other establihed test method.

 

[zekeman]

How did you ever equate 15 lbs of impact to dropping the 8 lbs at 1/2 foot? What is the application so that we may assist you in making an adequate test setup?

 

[rb1957]

as the OP notes v = sqrt(2gh),
impulse = mv
force = impulse/time (assuming an average force over the assumed time interval) ...
but be careful that you're using MASS units for m (wt/g),
then if time = 0.1sec
force = 10*impulse = 10*(wt/g)*sqrt(2gh)
wt = 8lbs, g = 32ft/sec2, h = 0.5ft ...
force = 10*(8/32)*sqrt(2*32*0.5) = 2.5*sqrt(32) = 14 lbs.

you might want to check with your customer if your test satisifies what he was thinking of.

a "peel" test is another test of adhesive bonding.

good luck

 

[bronx68]

rb1957 your math is the same as mine and I also got the 14 pounds of force as listed in my original email. So we agree on the basic math but the question is whether you agree with the 0.1 seconds I used for time of impact. I choose this number but can't back it up with any math or other means.

 

[zekeman]

I can't believe what I am reading here. First off, how do you make two dubious assumptions, namely the impact time and the average force when in fact you don't have a clue about the former and your interest should not be in the average, but the peak value of force. The facts are, the author of this thread hasn't given enough information on the requirements to make any case for a solution and his model is questionable.
Finally, it is well known that the peak force on impact is the most important force, not the average, and the shape of the impact response would have to be known in order to get it.
So, the only thing that is true is that
Fdt=d(mv) and the integral of this is the impulse or change of momentum; impulse is in momentum units, not energy units.
Since F is a function of its shape which is unknown during the impact time interval we don't have a handle on its peak value.

 

[arto]

How about MIL-S-901 shock test?

 

[btrueblood]

Whack it with a 3-lb sledge. If it don't break, whack it with a 15-lb sledge. Still good? Passed 15 lb. impact test.

 

[Astroboy]

Hi Bronx68
I used to build and do this kind of testing using instrumented impact load cells etc. For steel and delrin an impact time of 3-5 milliseconds would be reasonable depending on how dented the delrin gets.

 

[bronx68]

Astroboy,
How did you arrive at the 3-5 milliseconds? Is this from the data you took with the load cells using similar materials? Is this the complete duration of the impact cycle?

 

[Astroboy]

Hi bronx68
The test lab I worked at used instrumented drop testing for measuring impact energy of various materials and structures. The load cells would give a load-time curve which was integrated for energy. A typical rigid impact say, with no penetration or some other complication would produce a load time spike or stretched simple bell curve. A steel on steel impact could be less than 1 ms while steel on aluminum would be a bit longer. I'm guessing at the time for your case based on similar tests on similar materials. The equipment was called "Dynatup" and they were bought by Instron who still makes it. I'm sure you could have your sample tested by them as well. Good Luck

 

[MintJulep]

Clearly the requirement from your customer is unclear.

You need to go back and either ask for a clarification of what they really want you to demonstrate, or propose something and see if they accept what you propose.

There are many recognized test standards for bond strength under various conditions. As a starting point I suggest you get a copy of ASTM D6465.