Heel Drop Test: how reliable?Discussion requested 3

[IJR]

Pals

Are results from heel drop test always reliable?
Are there conditions when care should be excercised.

Your experience will be worth sharing w.r.t this apparently great yet simple tool.

respects
IJR

 

[JKStruct]

Here's a few statments that generalize my understanding, although I'm hardly a dynamics expert, and many people here probably know much more.

The heel drop is the age-old "gold standard" for vibration-inducing forcing functions. Thomas Murray has numerous publications about this, and vibration analysis in general.

For steel structures, AISC Design Guide 11 is a great resource. It poses alternative criteria to the standard heel drop test (I think chapter 4), that better represent the response of steel framed structures. This criteria is based on a lot of research and accounts for increased use of lightweight concrete and other common construction practices today. If you're analyzing a steel floor, use this guide.

We are just beginning a vibration analysis of a concrete framed structure, and we will likely revert back to the heel drop forcing functions for analysis.

That's my VERY rudimentry take on the matter.

 

[271828]

Reliable for what?

Heeldrops are a reliable excitation for determining the natural frequencies for low frequency floors. They're usually ok for high frequency floors also.

They actually come in two varieties: unmeasured (unreferenced) and measured (referenced).

Unreferenced means that someone heeldrops on the floor and measures accelerations on the slab. The amount of force and its exact frequency content are both unknown. The frequency content is fairly uniform for low frequencies, so the acceleration spectrum will indicate natural frequencies. That is ALL unmeasured ones are good for, period.

Referenced heeldrops are performed by heeldropping on a load-measuring device. These are relatively reliable ways to do modal testing. The output is the frequency response function (floor acceleration divided by force at each frquency). Natural frequencies, acceleration response, damping, and mode shapes can be determined from the set of FRFs measured using this method. This is less reliable than using a floor shaker, but is still an acceptable way to do modal tests on a floor. The FRFs can be used to predict floor vibration due to walking, but it's not easy.

 

[271828]

JK, I recommend you take a look at the newer British guidance for your concrete floors. I believe the Concrete Centre and the Steel Construction Institute both have prediction methods that you can use. The SCI one is very general, so should work for concrete also. Older heeldrop criteria are totally antiquated at this point. If you're looking for papers, look for stuff from Pavic, Reynolds, Willford, Fields, Young, etc. The UK is LOADED with floor vibe gurus and some of these guys have written extensively about concrete structures.

 

[IJR]

What if you have closely spaced joists with a thin slab supported.

You do heel drop on the slab and not on the beam

The slab is relatively flexible but the joist isnt.

What comes out of the test?

respects
IJR

 

[csd72]

RE: "The slab is relatively flexible but the joist isnt."

I find that hard to believe - check your span to depth ration I am sure you will find it is much lower for the slab - if not recheck the slab.

 

[271828]

csd72 is right. The slab is the high frequency part of this system unless the joists are extremely short and deep.

Heeldrops can be used on joist floors. It shouldn't matter if you heeldrop over a joist assuming they're closely spaced. The heeldrop can still be used to provide excitation for your tests. Like I typed before, if you're doing unreferenced tests, all you'll get are natural frequencies. You'll get no info on mode shape, FRF, damping, predicted response to walking, etc.

 

[IJR]

csd72
Dont find anything hard to believe my dear friend.

It is possible with long span structures supporting lightweight floor system, eg wood or OSB sheathing.

In that case where should a heel drop test be made, on top of sheathing or on top of long closely spaced joists. When I say joists here it could be a truss girder.

respects
IJR

 

[271828]

IJR, I believe CSD is referring to something like a slab on form deck supported by bar joists at a close spacing (say 2-3')--at least that's what I envisioned with your question.

If I had something VERY thing like wood, I'd heeldrop on a joist. Not sure it's totally necessary, but that's what I'd do.

 

[IJR]

Additional thanks 271828

That is exactly what I was after. Conventional floor has a deck and joists that talk one language. Mine is not conventional and there was no way a stiff slab could be made.

I should have mentioned this earliear in the post anyway.

thanks all pals
respects
IJR