Karagozian & Case LCRATE

[Nava Farhadzadeh]

Hi Len/James

I have used Karagozian & Case material model to simulate high velocity impact between concrete samples, as far as I remember in the recent releases of LS-DYNA, LCRATE can be put equal to -1 and the system automaticaly takes strain rate effects into account. I'm using Version : smp d R10.1.0 and for some unknown reasons, I get an error message saying I haven't defined a load curve for the LCRATE -1.  For my other models (lower impact velocities) LCRATE = -1 works without defining a load curve and as far as I know LCRATE -1 doesn't require a load curve, have you seen this error message before? Could you advice why it doesn't work for one model and works fine for the others (I have checked the model and can't seem to find any other issue with it)?

Best regards

Nav  

[Nava Farhadzadeh]

Also in completion of my question in my previous email, I have realized that when I use units of mm, g, ms, MPa the LCRATE -1 doesn't work, however, when the same file is set on m, N, Pa, S, the LCRATE -1 works. Unfortunately this doesn't resolve my problem because my model is set up in mm, g, ms, MPa. Is there a unit conversion for LCRATE -1?

Kind regards

Nava

[James M. Kennedy]

Dear Nava,

 

-----------------

 

A reply I found to a slightly different question:

 

Because your model is in microseconds, the abscissa values of LCRATE should be input

in  1/microseconds. For your model it would not be correct to use the LCRATE =-1 option,

because the DIF equations in the paper implemented when using this option are in units

of seconds.

 

----------------

 

From the User’s Manual:

 

LCRATE - Define (load) curve number for strain-rate effects; effective strain rate on

abscissa (negative = tension) and strength enhancement on ordinate. If LCRATE is set to

-1, strain rate effects are automatically included, based on equations provided in Wu,

Crawford, Lan, and Magallanes [2014]. LCRATE = -1 is applicable to models which use time

units of seconds, for other time units, the strain-rate effects should be input by means

of a curve.

 

----------------

 

A note previously shared by Len Schwer:

 

Let me preface my remarks by saying I am firmly in the camp of those that think NO strain

rate enhancements should be added to concrete constitutive models. In compression, the

laboratory observed strain rate enhancement is in large part due to the frictional (sloping)

nature of the shear failure surface. In tension, the dominant physics is fracture mechanics

and not continuum strain rate effects. If you are interested, you can read my paper and

especially the reference works cited.

 

https://www.dynalook.com/conferences/european-conf-2009/M-I-04.pdf/@@download/file/M-I-04.pdf

 

I think you can best answer your MAT072R3 questions by exercising a simple single element

model subject to a prescribed strain rate. Your complicated model illustrates a problem with

MAT072R3, but provides no hope of isolating what might be the problem.

 

Some responses to your questions:

 

1. Is my assumption of the curve being horizontal after reaching the max strain rate correct?

 

Read Remark 1 in the User Manual under the description of *DEFINE_CURVE. Personally, I never

trust what LS-DYNA does in terms of load curve extrapolation – I always make a simple test

case.

 

2. Do you know what happens in MAT72R3 concrete response when we put all the DIF values

as 1 (I have kept the same strain rate values and ranges for all the models)? I have

attached a photo for you which shows the comparison of different displacement-time

histories obtained from different load-curves.

 

See my preface remarks about single element tests of strain rate effects.

 

3. Do I need to include a steel re-bar DIF curve as well for my shaking table frame model?

How can I input the load curve in the steel material model if I am using (*MAT_PLASTIC_

KINEMATIC) material model?

 

You should use MAT024 to include strain rates for the steel in your model.

 

                --len

 

----------------

 

Sincerely,

James M. Kennedy

KBS2 Inc.

August 18, 2021

--
You received this message because you are subscribed to the Google Groups "LS-DYNA2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ls-dyna2+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ls-dyna2/4db824ba-8f95-44d3-b298-b8d16a3a5888n%40googlegroups.com.

[l...@schwer.net]

From the User Manual Volume II description of MAT072R3

 

LCRATE -- Define (load) curve number for strain-rate effects; effective strain

rate on abscissa (negative = tension) and strength enhancement

on ordinate. If LCRATE is set to -1, strain rate effects are

automatically included, based on equations provided in Wu,

Crawford, Lan, and Magallanes [2014]. LCRATE = -1 is

applicable to models which use time units of seconds, for other

time units, the strain-rate effects should be input by means of a

curve.

 

I am not sure when this feature was implemented in MAT072R3, so it may work differently with older version of LS-DYNA – currently at R13

 

                --len

 

From: ls-d...@googlegroups.com <ls-d...@googlegroups.com> On Behalf Of Nava Farhadzadeh
Sent: Tuesday, August 17, 2021 4:46 PM
To: LS-DYNA2 <ls-d...@googlegroups.com>

--

You received this message because you are subscribed to the Google Groups "LS-DYNA2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ls-dyna2+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/ls-dyna2/8ac3e807-0b27-4fee-97bc-2500c9ac8a49n%40googlegroups.com.