# Result Comparison With Abaqus **URL:** \ **Category:** General Questions\ **Tags:** results\ **Created:** [February 1, 2024, 12:47pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474 "2024-02-01T12:47:14Z")\ **Posts on this page:** 20\ **Page:** 1
**Author:** ![ajinkyagote](https://avatars.discourse-cdn.com/v4/letter/a/3ab097/32.png) [@ajinkyagote](https://prepomax.discourse.group/u/ajinkyagote)\ **Post date:** [February 1, 2024, 12:47pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/1 "2024-02-01T12:47:14Z")
This post is referred to model in my previous post “[Kinematic Coupling / Rigid Body with Moment load having higher magnitude](https://prepomax.discourse.group/t/kinematic-coupling-rigid-body-with-moment-load-having-higher-magnitude/1464)” After successfully running the simulation, I made comparative study of results with abaqus and I have found following results. Please refer to images below: ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/3/3d38474c4dee285733cbfaebca031e1856ec1ee8.jpeg) ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/c/c266c409333038fe15bb9772e4b99156f4e08252.jpeg) ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/2/2906fc8b6e1e5f69f417ccc0c519e4aa0899c619.jpeg) ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/8/8744edb0d1a26e35a03ec7e8b72ce2cd3a69c71b.jpeg) ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/c/cb5b8507e329d27d67d652bd3e973f6e9c31e189.jpeg) 1. Displacement - Max displacement is observed in both model at same node with small deviation in magnitude 2. von Mises Stress / Max Principal (Abs) - Again the location for maximum value is same but there is deviation in max value 3. For S11 component of stress, as you can see the location of maximum value and the magnitude differs a lot. As I have understood, in PrePoMax all values are nodal values ( but I don’t know what kind of avaraging is done here ) , and in abaqus I am using default averaging as shown in picture below: ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/e/e18e2becd3a51694aa77b89e8534c344227b02c8.png) So can anyone guide me in case I want to compare such results what is the good way or settings which I should be aware of? Thanks in advance! ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [February 1, 2024, 12:54pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/2 "2024-02-01T12:54:38Z")
First of all - are the models identical ? Which basically means that you either used the same mesh and manually redefined everything in Abaqus or (the best way) used the same input file, only with the necessary modifications due to small syntax differences. > [@ajinkyagote](#): > > As I have understood, in PrePoMax all values are nodal values ( but I don’t know what kind of avaraging is done here ) Check this thread (and the other thread referenced there): [Result Intepretation](https://prepomax.discourse.group/t/result-intepretation/150) ---
**Author:** ![ajinkyagote](https://avatars.discourse-cdn.com/v4/letter/a/3ab097/32.png) [@ajinkyagote](https://prepomax.discourse.group/u/ajinkyagote)\ **Post date:** [February 1, 2024, 12:58pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/3 "2024-02-01T12:58:09Z")
> [@FEAnalyst](#): > > First of all - are the models identical ? Yes, both models are identical in all aspects, with node numbering, element numbering, material, BCs and load. Those are the same models which I have shared in my previous post by google drive. ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [February 1, 2024, 12:58pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/4 "2024-02-01T12:58:58Z")
And there are still those membrane elements ? How do you account for them in postprocessing ? ---
**Author:** ![ajinkyagote](https://avatars.discourse-cdn.com/v4/letter/a/3ab097/32.png) [@ajinkyagote](https://prepomax.discourse.group/u/ajinkyagote)\ **Post date:** [February 1, 2024, 1:08pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/5 "2024-02-01T13:08:46Z")
The results in abaqus are only displayed Solid elements. No membrane elements are considered here. Although I can again confirm by having purely solid elements, by running new job. ---
**Author:** ![ajinkyagote](https://avatars.discourse-cdn.com/v4/letter/a/3ab097/32.png) [@ajinkyagote](https://prepomax.discourse.group/u/ajinkyagote)\ **Post date:** [February 1, 2024, 1:11pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/6 "2024-02-01T13:11:29Z")
> [@ajinkyagote](#): > > by running new job. The results stays same by running job with only solid elements. ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [February 1, 2024, 2:02pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/7 "2024-02-01T14:02:09Z")
I used the same input file (with only solid elements to avoid differences in membrane formulation/visualization) and I can confirm those differences in results. You could try using different meshes to see how close you can get. ---
**Author:** ![Matej](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@Matej](https://prepomax.discourse.group/u/Matej)\ **Post date:** [February 1, 2024, 4:14pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/8 "2024-02-01T16:14:30Z")
In this case the highest stress is at the rigid connection. Maybe there is some averaging error at the constrained nodes. To check it you can try inverting the load and supports. ---
**Author:** ![SergioP1975](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/sergiop1975/32/37_2.png) [@SergioP1975](https://prepomax.discourse.group/u/SergioP1975)\ **Post date:** [February 1, 2024, 6:01pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/9 "2024-02-01T18:01:59Z")
If I rememeber well, you need to adjust the stress averaging to 100% in Abaqus to have the same results in CCX ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [February 1, 2024, 6:04pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/10 "2024-02-01T18:04:06Z")
In general yes but that’s not the culprit here. ---
**Author:** ![Gunnar](https://avatars.discourse-cdn.com/v4/letter/g/ed655f/32.png) [@Gunnar](https://prepomax.discourse.group/u/Gunnar)\ **Post date:** [February 1, 2024, 8:51pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/11 "2024-02-01T20:51:43Z")
Calculix’s results are very close to those of Code\_aster. 🤔 ![cxcaaq](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/f/f9e24925849d9d512e2c3a392d0fcb96bbd6f92c.jpeg) ---
**Author:** ![Gunnar](https://avatars.discourse-cdn.com/v4/letter/g/ed655f/32.png) [@Gunnar](https://prepomax.discourse.group/u/Gunnar)\ **Post date:** [February 2, 2024, 6:24am UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/12 "2024-02-02T06:24:40Z")
If you could kindly show a plot from abaqus, with the stresses at the gauss points, can maybe help to brings light into the darkness. ![gauss](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/9/9f79d5a8c0ee55b3a33d99f44085b351c1dfffd8.jpeg) ---
**Author:** ![ajinkyagote](https://avatars.discourse-cdn.com/v4/letter/a/3ab097/32.png) [@ajinkyagote](https://prepomax.discourse.group/u/ajinkyagote)\ **Post date:** [February 2, 2024, 6:54am UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/13 "2024-02-02T06:54:04Z")
> [@Matej](#): > > To check it you can try inverting the load and supports. I would give it a try!! ---
**Author:** ![ajinkyagote](https://avatars.discourse-cdn.com/v4/letter/a/3ab097/32.png) [@ajinkyagote](https://prepomax.discourse.group/u/ajinkyagote)\ **Post date:** [February 2, 2024, 6:55am UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/14 "2024-02-02T06:55:56Z")
> [@Gunnar](#): > > can maybe help to brings light into the darkness. That can be a good approach also. I will have a look into it and come to some conclusion and perhaps a solution. ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [February 2, 2024, 8:17am UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/15 "2024-02-02T08:17:42Z")
> [@Gunnar](#): > > If you could kindly show a plot from abaqus, with the stresses at the gauss points, can maybe help to brings light into the darkness. Abaqus doesn’t have this form of plots. I can only disable averaging and create so-called quilt plot. Or query the stresses at particular integration points. ---
**Author:** ![Gunnar](https://avatars.discourse-cdn.com/v4/letter/g/ed655f/32.png) [@Gunnar](https://prepomax.discourse.group/u/Gunnar)\ **Post date:** [February 2, 2024, 9:15am UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/16 "2024-02-02T09:15:22Z")
Alternative the extrapolated stress at nodes per element? ![elno2](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/5/5f7290da042a4502921b2bfe8756401a37bdaabf.png) > [@FEAnalyst](#): > > I can only disable averaging and create so-called quilt plot. I think this must be the same ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [February 2, 2024, 3:56pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/17 "2024-02-02T15:56:55Z")
Quilt plots (no averaging): ![quilt x](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/0/018261c456f3693cab8832ea98c2ce426d87d14b.png) ![quilt y](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/4/44cc66de4c8faebae4de49756024856bba13205f.png) ---
**Author:** ![Gunnar](https://avatars.discourse-cdn.com/v4/letter/g/ed655f/32.png) [@Gunnar](https://prepomax.discourse.group/u/Gunnar)\ **Post date:** [February 2, 2024, 7:05pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/18 "2024-02-02T19:05:18Z")
There is a big difference in Abaqus between the averaged and unaveraged stress for component S11: averaged = -0,884 / 0,884 unaveraged = -1,515 / 1,515 while the values for S22 are very close: averaged = -2,603 / 2,603 unaveraged = -2,611 / 2,611 in code aster i have for S11: averaged = -3,013 / 3,013 unaveraged = -3,096 / 3,096 for S22 averaged = -3,915 / 3,915 unaveraged = -4,071 / 4,071 For a closer look attached the aster values for just one node ID 7410 and the gauss values for Element ID 5134. ![7](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/3/34c5471d618435fe712a3238f316803cf0a3b328.png) ---
**Author:** ![ANYS](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/anys/32/7737_2.png) [@ANYS](https://prepomax.discourse.group/u/ANYS)\ **Post date:** [February 3, 2024, 9:39pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/19 "2024-02-03T21:39:25Z")
HI, I have request SOF and SOM at the fixed support and overall My moment is over the requested 10.000Nmm: ```auto *SECTION PRINT,SURFACE=Disk,NAME=Reaction_forces SOF,SOM ``` Expected: 10.000 Nmm Result: 11.882 Nmm ```auto statistics for surface set DISK and time 0.1000000E+01 total surface force (fx,fy,fz) and moment about the origin(mx,my,mz) 4.503173E-04 1.890045E-11 -7.493543E+01 9.126258E-05 -1.188226E+04 3.964118E-08 ``` ---
**Author:** ![Gunnar](https://avatars.discourse-cdn.com/v4/letter/g/ed655f/32.png) [@Gunnar](https://prepomax.discourse.group/u/Gunnar)\ **Post date:** [February 3, 2024, 10:24pm UTC](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474/20 "2024-02-03T22:24:13Z")
True, but when you fix in addition the reference point and leave only UR2 free, the moment is 10.901 Nmm. If you then fix only U1 of the surface (U2+U3 free), you get 9.664 Nmm. I don’t think there is something wrong with the calculix results because all values i compared with code aster are nearly exact the same. For example the stresses for element 5134 at the gauss points: stresses (elem, integ.pnt.,sxx,syy,szz) for set ELEMENT\_5134 calculix → code aster 5134 5 -0.10 0.91 0.69 → -0.10 0.91 0.69 5134 6 -0.27 0.90 0.71 → -0.27 0.90 0.70 5134 1 0.66 2.35 1.75 → 0.65 2.34 1.75 5134 2 0.58 2.50 1.91 → 0.58 2.49 1.91 5134 3 1.07 2.45 1.64 → 1.07 2.45 1.64 5134 4 1.18 2.70 1.91 → 1.18 2.70 1.90 5134 7 0.40 1.08 0.76 → 0.40 1.08 0.76 5134 8 0.41 1.16 0.86 → 0.40 1.16 0.86 [Next page](https://prepomax.discourse.group/t/result-comparison-with-abaqus/1474.md?page=2)