what is the difference between drop test and shock test?

Mintu

Paul Sharp_21301 said:

I'm not an expert in electronics test, but I would interpret the difference in definition as being whether you take into consideration the local effect of the impact itself. i.e. a shock test may just be an impulse on your model, applied on the whole structure (such as might apply to a device mounted in a vehicle or contained in packaging, or even just the shocks that maybe experienced by a phone while it's user is jogging), whereas a drop test is replicating an impact (contact with ground other object). So, a drop test will result in a shock (e.g. the corner of your phone hits the ground, the remainder of the phone experiences shock even if it doesn't contact the ground), but a shock test is not always necessarily an impact.

Hi sir,

I have performed my drop test simulation in radioss.

but how can i validate my results? while set up i have used rigid wall as infinite. can i model  rigid wall with elements? please help me out.

PaulAltair

Mintu said:

Hi sir,

I have performed my drop test simulation in radioss.

but how can i validate my results? while set up i have used rigid wall as infinite. can i model  rigid wall with elements? please help me out.

Validation: are you happy with the energy balance in your model? is the energy error low in the output? do you have physical test results to correlate against?

You can model the rigid wall (or non-rigid wall) with elements, you can simply mesh a plane (or any other shape) and include it in contact to your structure. If you want that plane to be rigid then you can assign a Rigid body (/RBODY) on all nodes of that wall/floor. You will need to constrain that rigid body in space by applying a boundary condition (/BCS) on all 6 DOF on the main node of the /RBODY.

 

Mintu

Paul Sharp_21301 said:

Validation: are you happy with the energy balance in your model? is the energy error low in the output? do you have physical test results to correlate against?

You can model the rigid wall (or non-rigid wall) with elements, you can simply mesh a plane (or any other shape) and include it in contact to your structure. If you want that plane to be rigid then you can assign a Rigid body (/RBODY) on all nodes of that wall/floor. You will need to constrain that rigid body in space by applying a boundary condition (/BCS) on all 6 DOF on the main node of the /RBODY.

 

I got peak at one point. i didn't understand energy graph for this simulation.after completion of simulation we will go for practical test.

Mintu

Paul Sharp_21301 said:

Validation: are you happy with the energy balance in your model? is the energy error low in the output? do you have physical test results to correlate against?

You can model the rigid wall (or non-rigid wall) with elements, you can simply mesh a plane (or any other shape) and include it in contact to your structure. If you want that plane to be rigid then you can assign a Rigid body (/RBODY) on all nodes of that wall/floor. You will need to constrain that rigid body in space by applying a boundary condition (/BCS) on all 6 DOF on the main node of the /RBODY.

 

Hi sir,

I have tried different cases 

rigid wall as infinite plane  and  Rigid wall as meshed component. The following is the energy plot for Infinite plane. can you help me by sharing any test case.

 

PaulAltair

Mintu said:

Hi sir,

I have tried different cases 

rigid wall as infinite plane  and  Rigid wall as meshed component. The following is the energy plot for Infinite plane. can you help me by sharing any test case.

 

That does not look atypical for a drop test energy curve, 'Total Energy' is not actually the Total (just a sum of KE and IE) you can also look at TTE and DTE which include more energies (such as rotational KE) I have attached at this link Phone Drop Example from the Radioss intro training

Mintu

Paul Sharp_21301 said:

You can't do it in OptiStruct 2019 the functionality wasn't in that version, you could do it in RADIOSS instead in 2019,

There is an example here:

https://support.altair.com/community?sys_kb_id=7b04d8b4db46ac10e8863978f496190c&id=kb_article_view&sysparm_rank=1&sysparm_tsqueryId=b7a903f91b0aec10c4dfdbd9dc4bcb6f

if you are more familiar with OptiStruct set up though, then it might be easier to install 2021 and do it there?

Regarding drop height/velocity in either OS or RADIOSS, you would calculate velocity after drop of 1.5m  (sqrt (2 * g * h)) (approx 5.42m/s) then set that as the initial velocity of your object, and position it just at the point of impact, you should add gravity loading too really, it will make a small difference.

Thank you very much for your guidance sir.

I have performed drop simulation from 1.5m of height but we can not simulate for 1.5 m of height.  to reduce solution time iam performing drop simulation from 1 mm of height. how much velocity need to define for 1 mm of height.

PaulAltair

Mintu said:

Thank you very much for your guidance sir.

I have performed drop simulation from 1.5m of height but we can not simulate for 1.5 m of height.  to reduce solution time iam performing drop simulation from 1 mm of height. how much velocity need to define for 1 mm of height.

I'm not sure I understand exactly. You don't need to model the drop phase (you don't need to model the object falling through space) The calculation in the original answer (sqrt (2 * g * h)) gives the velocity due to a fall from a height, e.g. 5.42m/s for 1.5m, you can give the object an initial velocity of 5.42m/s and model it 1mm above the floor (I thought this was what you were doing already!)

Mintu

Paul Sharp_21301 said:

I'm not sure I understand exactly. You don't need to model the drop phase (you don't need to model the object falling through space) The calculation in the original answer (sqrt (2 * g * h)) gives the velocity due to a fall from a height, e.g. 5.42m/s for 1.5m, you can give the object an initial velocity of 5.42m/s and model it 1mm above the floor (I thought this was what you were doing already!)

As you are said I have performed drop test for 5.42 m/sec. i got energy plots correctly. but in animation The 1d elements cannot drop along with the structure. i have used Type 7 contacts for self contact. is there any other contact should be used for drop test analysis?

Mintu

Mintu said:

how we can decide element size for drop test analysis?

Hi sir,

I have performed my drop test in radioss. Total weight of the electronic unit weight is 12.5 kg. dropped from 1.5 m of height.The box is made of Aluminium. Yield stress is 193 M Pa. I have got the results vonmises stress is 192 M Pa. Acceleration on PCB is 100g. how can i relate my results ?

PaulAltair

I'm not sure what you mean by your question. Are you questioning the confidence you can have in the results? If your energy balance is ok and you are confident in your material definitions and modelling practices then that is all you can do. Do you have any physical test results to compare to?

Mintu

Hi Mr. Paul. I have performed by drop test successfully and compared with practical testing by virtually dropping the Unit. The damaged locations are matched with Analysis results.My question is we can perform drop test repeated Drop test multiple times in analysis if Yes, what is the procedure is there any option in the software.

Thanks in advance,