Mohit,
You changed your muscle force output to output everything for the left side. From here it looks like what you need. Just remember to have everything consistent and it should work.
Regards,
Pavel
Yes sir
it works. Abaqus is showing now no error regarding anybody muscle forces during analysis.
It is showing now only one error as follows-
179596 elements have missing property definitions. The elements have been identified in element set ErrElemMissingSection.
Plz Guide me.
Mohit,
This is now an Abaqus question (this forum is for AnyBody) - it is further up to you to solve the task. Just read the error messages carefully. You are missing the material properties for your mesh to solve something. It might be possible to do from 3matic.
Regards,
Pavel
Dear sir
Everything is resolved now. Abaqus is giving output. But the results which we are getting seems to infeasible. Because bending due to the loads are too high which seems infeasible and very few stress points are showing. Im giving following material properties-
Modulus of Elasticity: 21 GPa
Density: 1900kg/m3
Poission Ratio- 0.3
Im attaching the output. Is anybody loads are too high that they are producing this results?
Thanks.
Mohit,
It looks like that all forces and moments were scaled when we converted everything to mm (I checked the files that you sent me).
A solution would be to search and replace the following string (shown by red):
*CLOAD, OP=NEW, AMPL=AMPL_Leg_Mus_AdductorBrevisProximal1_3
NS_Leg_Mus_AdductorBrevisProximal1_3, 3, 1.000000
by “, 0.001” for all values except the torques (those amplitudes that end with _4, _5, _6.
So the intention was to scale the moment from Nm to Nmm, but apparently this transformation is applied to everything. I think this bug in the converter was fixed, but apparently not released yet.
Regards,
Pavel
Dear Sir
i change the values as follows-
this is the case you mentioned that ends with _4,_5,_6
*CLOAD, OP=NEW, AMPL=AMPL_Leg_Mus_GastrocnemiusLateralis1_4
NS_Leg_Mus_GastrocnemiusLateralis1_4, 4, 1.000
And all other loads like follows-
*CLOAD, OP=NEW, AMPL=AMPL_Leg_Mus_GluteusMediusAnterior1_3
NS_Leg_Mus_GluteusMediusAnterior1_3, 3, 0.001
*CLOAD, OP=NEW, AMPL=AMPL_Leg_Mus_GluteusMediusAnterior3_1
NS_Leg_Mus_GluteusMediusAnterior3_1, 1, 0.001
*CLOAD, OP=NEW, AMPL=AMPL_Leg_Mus_GluteusMediusAnterior3_2
NS_Leg_Mus_GluteusMediusAnterior3_2, 2, 0.001
etc.
Sir is the changes are to be done in the amplitudes section which as follows-
*distributing, weighting method=UNIFORM
*AMPLITUDE, NAME=AMPL_Leg_Mus_Pectineus2_1
0.000000, 10.348173,
2.000000, 10.358976,
*AMPLITUDE, NAME=AMPL_Leg_Mus_Pectineus2_2
0.000000, 11.398773,
2.000000, 11.410673,
*AMPLITUDE, NAME=AMPL_Leg_Mus_Pectineus2_3
0.000000, 5.208563,
2.000000, 5.214001,
*AMPLITUDE, NAME=AMPL_MusclesSpineLeft_PsoasMajor_PML3I_TM_1
0.000000, 5160.580269,
2.000000, 5160.577883,
etc.
Thanks.
No, only the modifications that I mentioned - they will automatically fix the actual loading curves.
Regards,
Pavel
Dear sir
i had did the changes but the results are still same. im attaching the file.
The lines in red which i have not do any changes and all other by .001
Thanks.
Mohit,
It should be good enough. Does it bend too much again? Or is the problem somewhere else? I think the biggest loads will appear around hip and knee joints, which are a little inside - so you may not see the peaks on the surface.
Regards,
Pavel
Sir,
How can i confirm the units of loads are that in n/m or n/mm. Because i check the clavicle file in which there are 1.000 at all places and in that case the results are efficient.
Sir one question is more when i modify the material properties units or reducing the elastic modulus then there are less bending. But very few stresses are showing. Is you have an idea about the material properties that it should be accordance with the anybody loads?
Thanks.
All the forces that you see in the output XML file are in N and Nm. You can check against those values. Once you have done the conversion with -t “…” everything has increased by a factor of 1000 (all forces and moments) to be in mm, which is the units for your mesh. So the modification that we made (0.001) should return 1000*N to N and Nmm leave as Nmm.
The material properties, of course, have an effect on bending, but they should be something that you can find in the literature.
Regards,
Pavel
Dear sir
I tried everything in the coding replacing by .001 .But there is no changes in the results.
Then i did changes in the -t string by .001 instead of 1000
And now results seems to be ok. Plz check is it right approach.
-t" +strquote (".001 0 0 0 .001 0 0 0 .001 0 0 0");
Im attaching the results.
Thanks
Mohit,
At this point you should already know that this transformation scales the muscle attachment location by making X, Y, and Z components to be X0.001, Y0.001, Z*0.001. It means that all the force will be applied in the wrong place. What it will also do - it will scale all loads significantly down.
Basically scaling does not work well. And the correct scaling solution is the one where we replaced magnitudes. So if you want - you can run an FE simulation in meters. Please adjust all other units to be in meters and it should give you correct results.
Please run everything in meters (without -t). So at this point I will close this thread - there is enough information for you to solve the problem.
Regards,
Pavel
Thanks sir
For your support.
One last question which still creating a problem is that anybody works in meters and if a file that are in mm so we have to scale down by .001 in anybody which i had read in tutorials. But in the my problem i have a mesh which is in mm so where did i scaling down the file in anybody because 3-matic is unable to make a mesh of a file that are in meters.
Did you mean to say that by applying the t string (“1000 0 0 0 1000 0 0 0 1000 0 0 0”); mesh is scaling down to meters or body model femur is scaling large by 1000? Because if i open the anybody femur and mesh in 3-matic, mesh looks bigger than anybody femur.
Remain procedure i understand well.
Thanks for your valuable support for this project.
Thanks.
Mohit,
The AnyBody femur, if exported with a scaling factor of 1, will be in meters. I believe we used a factor of 1000 to get it in mm. And this 1000 times enlarged femur was used for your mesh.
When we specify -t “1000 0 0 0 1000 0 0 0 1000” - we convert the muscle attachments and all other force application locations to be in mm. We do not touch the mesh file itself. Such transformation is needed to ensure consistency of the force output with the mesh for FEA.
However, this transformation is also applied to the magnitudes of the loads (force conversions are wrong (they should not scale), moment conversions are correct). This is why we changed the amplitudes by a factor of 0.001 to convert the forces back to correct values, yet we kept the conversion of moments.
I hope this helps.
Regards,
Pavel
Dear Sir
You mentioned a new convertor which solve this problem earlier. When it is supposed to be launch?
Thanks.
Unfortunately I do not have a date yet. Most likely with the next version of the AMS.
Regards,
Pavel