turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- Navigation
- Simcenter
- Forums
- Blogs
- Knowledge Bases

- Siemens PLM Community
- Simcenter
- 3D Simulation - NX Nastran Forum
- negative tangent modulus in TableS1 (sol601)

Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

N/A

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-21-2009 05:50 AM

Hi,

i am performing an advanced non linear analysis using contact. i am considering

small strain and displacement so i am using the true engineering strain Vs stress

data.

but it does issue an error when reading nastran data complaining that the tangent

modulus is negative in TableS1 for the material.

could you please shed any light on this?

Many thanks,

Engrequest

7 REPLIES

N/A

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-21-2009 07:08 AM

"engrequest"

>

>Hi,

>

>i am performing an advanced non linear analysis using contact. i am considering

>small strain and displacement so i am using the true engineering strain Vs

>stress

>data.

>but it does issue an error when reading nastran data complaining that the

>tangent

>modulus is negative in TableS1 for the material.

>could you please shed any light on this?

>

>Many thanks,

>Engrequest

>

>

>

>

Hi,

probably you filled the table with some wrong values. TABLES1 uses linear estrapolation

outside the table, maybe it is the extrapolation. Could you post the TABLES1 from

your bulk (bdf or dat) file?

Anyway if you consider small displacement and small strain (hence the only non linearity

is the contact) you should not need any TABLES1.

Hope this helps

Francesco

N/A

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-21-2009 08:45 AM

"francesco pelizza"

>

>"engrequest"

>>

>>Hi,

>>

>>i am performing an advanced non linear analysis using contact. i am considering

>>small strain and displacement so i am using the true engineering strain Vs

>>stress

>>data.

>>but it does issue an error when reading nastran data complaining that the

>>tangent

>>modulus is negative in TableS1 for the material.

>>could you please shed any light on this?

>>

>>Many thanks,

>>Engrequest

>>

>

>Hi,

>probably you filled the table with some wrong values. TABLES1 uses linear

>estrapolation

>outside the table, maybe it is the extrapolation. Could you post the TABLES1

>from

>your bulk (bdf or dat) file?

>Anyway if you consider small displacement and small strain (hence the only

>non linearity

>is the contact) you should not need any TABLES1.

>

>Hope this helps

>Francesco

The error message simply means that one (or more) of the segments of your stress-strain

curve has a negative slope. As noted by remark 9 for MATS1 in the QRG:

If TYPE = ¡°PLASTIC¡±, all tangent modulii ETi must satisfy the following:

HR = 1: 0 ¡Ü ETi < E

HR = 2: 0.0001 * E < ETi < E

HR = 3: 0.0001 * E < ET < E

Note that in the upcoming NX Nastran 7 release, this restriction will be removed

and a negative slope will be allowed, if that is what you truly need.

Also note that this is not an extrapolation issue. Linear extrapolation simply extends

the last segment as needed. If the last segment has positive slope, so will the

extrapolated data.

Regards,

Jim

N/A

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-21-2009 09:29 AM

"Jim Bernard"

>

>"francesco pelizza"

>>

>>"engrequest"

>>>

>>>Hi,

>>>

>>>i am performing an advanced non linear analysis using contact. i am considering

>>>small strain and displacement so i am using the true engineering strain

>Vs

>>>stress

>>>data.

>>>but it does issue an error when reading nastran data complaining that the

>>>tangent

>>>modulus is negative in TableS1 for the material.

>>>could you please shed any light on this?

>>>

>>>Many thanks,

>>>Engrequest

>>>

>>

>>Hi,

>>probably you filled the table with some wrong values. TABLES1 uses linear

>>estrapolation

>>outside the table, maybe it is the extrapolation. Could you post the TABLES1

>>from

>>your bulk (bdf or dat) file?

>>Anyway if you consider small displacement and small strain (hence the only

>>non linearity

>>is the contact) you should not need any TABLES1.

>>

>>Hope this helps

>>Francesco

>

>The error message simply means that one (or more) of the segments of your

>stress-strain

>curve has a negative slope. As noted by remark 9 for MATS1 in the QRG:

>

> If TYPE = ¡°PLASTIC¡±, all tangent modulii ETi must satisfy the following:

>

>

> HR = 1: 0 ¡Ü ETi < E

> HR = 2: 0.0001 * E < ETi < E

> HR = 3: 0.0001 * E < ET < E

>

>Note that in the upcoming NX Nastran 7 release, this restriction will be removed

>and a negative slope will be allowed, if that is what you truly need.

>

>Also note that this is not an extrapolation issue. Linear extrapolation simply

>extends

>the last segment as needed. If the last segment has positive slope, so will

>the

>extrapolated data.

>

>Regards,

>Jim

Dear all,

of course the error message means that there is a negative slope on the stress strain

curve (it is in the question Engrequest posted). I still think that if Engrequest

would post the TABLES1 we could help him better than reporting the QRG, I also think,

as stated in the preceding post, that if he wants to perform an analysis with small

strains and displacements in which the non linearity is only the contacts the real

problem is that he should not use MATS1 (and corresponding TABLES1) that are used

for material non linearity.

I'm happy than in the NX Nastran 7 release this limitation there is not anymore

(as it is already in MSC NASTRAN), but I really don't think it is what Engrequest

need, or meant with this post.

Regards

Francesco

N/A

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-21-2009 09:48 AM

"francesco pelizza"

>

>Dear all,

>of course the error message means that there is a negative slope on the stress

>strain

>curve (it is in the question Engrequest posted). I still think that if Engrequest

>would post the TABLES1 we could help him better than reporting the QRG, I

>also think,

>as stated in the preceding post, that if he wants to perform an analysis with

>small

>strains and displacements in which the non linearity is only the contacts

>the real

>problem is that he should not use MATS1 (and corresponding TABLES1) that are

>used

>for material non linearity.

>

>I'm happy than in the NX Nastran 7 release this limitation there is not anymore

>(as it is already in MSC NASTRAN), but I really don't think it is what Engrequest

>need, or meant with this post.

>

>Regards

>Francesco

>

>

Actually, if the model is truly linear in all respects except for contact, he could

use SOL 101 (which MSC does not have) rather than SOL 601.

N/A

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-21-2009 09:52 AM

Dear All,

Iam applying the Sol601 for the contact surfaces and as well to consider the non

linearity of the material ( ( when perfoming a sol101 i had a lof of stress on the

contact surfaces (high compressive load applied) )

when i tried the basic non linear, the contact surfaces were not modelled proferly

with gap elements so the soltuino was not converging so i decided to go for the

sol601.

i used the same stress strain curve i have used for sol106 but for the sol601 i

had the error or negative tangent modulus.

> francesco, this is the TableS1 on the bulk data :

$* NX Material: HIGH-STRENGTH_STEEL

MAT1 11.8600+8 0.3100008.0000-6 25.0000

MATS1 1 1 PLASTIC 1 11.5500+6

TABLES1 1 +

+ 0.0000 0.00000.0083331.5500+60.010000 1600.000.020000 1700.00+

+ 0.040000 1800.000.050000 1900.00 ENDT

$*

Thanks,

"francesco pelizza"

>

>"Jim Bernard"

>>

>>"francesco pelizza"

>>>

>>>"engrequest"

>>>>

>>>>Hi,

>>>>

>>>>i am performing an advanced non linear analysis using contact. i am considering

>>>>small strain and displacement so i am using the true engineering strain

>>Vs

>>>>stress

>>>>data.

>>>>but it does issue an error when reading nastran data complaining that the

>>>>tangent

>>>>modulus is negative in TableS1 for the material.

>>>>could you please shed any light on this?

>>>>

>>>>Many thanks,

>>>>Engrequest

>>>>

>>>

>>>Hi,

>>>probably you filled the table with some wrong values. TABLES1 uses linear

>>>estrapolation

>>>outside the table, maybe it is the extrapolation. Could you post the TABLES1

>>>from

>>>your bulk (bdf or dat) file?

>>>Anyway if you consider small displacement and small strain (hence the only

>>>non linearity

>>>is the contact) you should not need any TABLES1.

>>>

>>>Hope this helps

>>>Francesco

>>

>>The error message simply means that one (or more) of the segments of your

>>stress-strain

>>curve has a negative slope. As noted by remark 9 for MATS1 in the QRG:

>>

>> If TYPE = ¡°PLASTIC¡±, all tangent modulii ETi must satisfy the following:

>>

>>

>> HR = 1: 0 ¡Ü ETi < E

>> HR = 2: 0.0001 * E < ETi < E

>> HR = 3: 0.0001 * E < ET < E

>>

>>Note that in the upcoming NX Nastran 7 release, this restriction will be

>removed

>>and a negative slope will be allowed, if that is what you truly need.

>>

>>Also note that this is not an extrapolation issue. Linear extrapolation simply

>>extends

>>the last segment as needed. If the last segment has positive slope, so will

>>the

>>extrapolated data.

>>

>>Regards,

>>Jim

>

>Dear all,

>of course the error message means that there is a negative slope on the stress

>strain

>curve (it is in the question Engrequest posted). I still think that if Engrequest

>would post the TABLES1 we could help him better than reporting the QRG, I

>also think,

>as stated in the preceding post, that if he wants to perform an analysis with

>small

>strains and displacements in which the non linearity is only the contacts

>the real

>problem is that he should not use MATS1 (and corresponding TABLES1) that are

>used

>for material non linearity.

>

>I'm happy than in the NX Nastran 7 release this limitation there is not anymore

>(as it is already in MSC NASTRAN), but I really don't think it is what Engrequest

>need, or meant with this post.

>

>Regards

>Francesco

>

>

N/A

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-21-2009 10:11 AM

"engrequest"

>

>Dear All,

>

>Iam applying the Sol601 for the contact surfaces and as well to consider the

>non

>linearity of the material ( ( when perfoming a sol101 i had a lof of stress

>on the

>contact surfaces (high compressive load applied) )

>when i tried the basic non linear, the contact surfaces were not modelled

>proferly

>with gap elements so the soltuino was not converging so i decided to go for

>the

>sol601.

>i used the same stress strain curve i have used for sol106 but for the sol601

>i

>had the error or negative tangent modulus.

>

>> francesco, this is the TableS1 on the bulk data :

>$* NX Material: HIGH-STRENGTH_STEEL

>MAT1 11.8600+8 0.3100008.0000-6 25.0000

>MATS1 1 1 PLASTIC 1 11.5500+6

>TABLES1 1 +

>+ 0.0000 0.00000.0083331.5500+60.010000 1600.000.020000 1700.00+

>+ 0.040000 1800.000.050000 1900.00 ENDT

>$*

>Thanks,

>

With that data, the second segment of the stress-strain curve has a negative modulus:

Strain Stress Slope

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

0 0

0.008333 1.55E+06 186007440.3

0.01 1600 -928854229.2

0.02 1700 10000

0.04 1800 5000

0.05 1900 10000

If you plot this data, the chart certainly does not look correct. Are you sure that

the last 4 stress values are not off by a factor of 1000? The following data would

be valid and "looks" more like a typical stress-strain curve:

Strain Stress Slope

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

0 0

0.008333 1.55E+06 186007440.3

0.01 1600000 29994001.2

0.02 1700000 10000000

0.04 1800000 5000000

0.05 1900000 10000000

Regards,

Jim

N/A

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-21-2009 10:21 AM

Thanks Jim,

normally i write the stress strain cruve data on the NX interface in the material

properties (values are in N)

it seems that in writing the imput solver file, there is an issue with the units,

because the first value of stress (yield ) is written in mN and the next ones in

N that's why i am having a factor of 1000 missing.

Regards,

"Jim Bernard"

>

>"engrequest"

>>

>>Dear All,

>>

>>Iam applying the Sol601 for the contact surfaces and as well to consider

>the

>>non

>>linearity of the material ( ( when perfoming a sol101 i had a lof of stress

>>on the

>>contact surfaces (high compressive load applied) )

>>when i tried the basic non linear, the contact surfaces were not modelled

>>proferly

>>with gap elements so the soltuino was not converging so i decided to go for

>>the

>>sol601.

>>i used the same stress strain curve i have used for sol106 but for the sol601

>>i

>>had the error or negative tangent modulus.

>>

>>> francesco, this is the TableS1 on the bulk data :

>>$* NX Material: HIGH-STRENGTH_STEEL

>>MAT1 11.8600+8 0.3100008.0000-6 25.0000

>>MATS1 1 1 PLASTIC 1 11.5500+6

>>TABLES1 1 +

>>+ 0.0000 0.00000.0083331.5500+60.010000 1600.000.020000 1700.00+

>>+ 0.040000 1800.000.050000 1900.00 ENDT

>>$*

>>Thanks,

>>

>

>With that data, the second segment of the stress-strain curve has a negative

>modulus:

>

> Strain Stress Slope

> -------- -------- ------------

> 0 0

> 0.008333 1.55E+06 186007440.3

> 0.01 1600 -928854229.2

> 0.02 1700 10000

> 0.04 1800 5000

> 0.05 1900 10000

>

>

>If you plot this data, the chart certainly does not look correct. Are you

>sure that

>the last 4 stress values are not off by a factor of 1000? The following data

>would

>be valid and "looks" more like a typical stress-strain curve:

>

> Strain Stress Slope

> -------- -------- ------------

> 0 0

> 0.008333 1.55E+06 186007440.3

> 0.01 1600000 29994001.2

> 0.02 1700000 10000000

> 0.04 1800000 5000000

> 0.05 1900000 10000000

>

>Regards,

>Jim

Follow Siemens PLM Software

© 2017 Siemens Product Lifecycle Management Software Inc