# \*ERROR: increment size smaller than minimum

**URL:** <https://calculix.discourse.group/t/error-increment-size-smaller-than-minimum/199>\
**Category:** Uncategorized\
**Created:** [September 22, 2020, 8:37am UTC](https://calculix.discourse.group/t/error-increment-size-smaller-than-minimum/199 "2020-09-22T08:37:48Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![konstantin](https://avatars.discourse-cdn.com/v4/letter/k/858c86/32.png) [@konstantin](https://calculix.discourse.group/u/konstantin)\
**Post date:** [September 22, 2020, 8:37am UTC](https://calculix.discourse.group/t/error-increment-size-smaller-than-minimum/199/1 "2020-09-22T08:37:48Z")

</div>

Hello, colleagues.

I try to calculate the perfect shell without imperfections with non-linear geometry.

 ![shell](https://global.discourse-cdn.com/free1/uploads/calculix/original/1X/b23e60649556401ed99a453ff79db0b2762c6a8c.png)

**My model**

> **[Meet Google Drive – One place for all your files](https://accounts.google.com/ServiceLogin?service=wise&passive=1209600&continue=https:%2F%2Fdrive.google.com%2Fdrive%2Ffolders%2F1v855NBFnOJFBJSbgy6K6KI9nlihSXAqb?usp%3Dsharing&followup=https:%2F%2Fdrive.google.com%2Fdrive%2Ffolders%2F1v855NBFnOJFBJSbgy6K6KI9nlihSXAqb?usp%3Dsharing)**

**Expected**

Finalize without errors for full time line.

**Problem**

If I draw graph force-reaction dependency of time, then error happen at the top.

 ![graph](https://global.discourse-cdn.com/free1/uploads/calculix/original/1X/d59d50c3c272460da120a16a2f1585fc0a0859b9.png)

```auto
average force= 0.023996
time avg. forc= 0.019977
largest residual force= 0.000198 in node 333 and dof 1
largest increment of disp= 8.813149e-05
largest correction to disp= 2.800649e-05 in node 247 and dof 2

divergence; the increment size is decreased to 2.574921e-07
the increment is reattempted

*ERROR: increment size smaller than minimum
best solution and residuals are in the frd file

exit status 201

```

---

<div class="post-metadata">

**Author:** ![konstantin](https://avatars.discourse-cdn.com/v4/letter/k/858c86/32.png) [@konstantin](https://calculix.discourse.group/u/konstantin)\
**Post date:** [September 22, 2020, 8:03pm UTC](https://calculix.discourse.group/t/error-increment-size-smaller-than-minimum/199/2 "2020-09-22T20:03:48Z")

</div>

As I understood now, it is happening because **Calculix** implement **newton raphson method** , but for that case need **Riks method**

---

<div class="post-metadata">

**Author:** ![konstantin](https://avatars.discourse-cdn.com/v4/letter/k/858c86/32.png) [@konstantin](https://calculix.discourse.group/u/konstantin)\
**Post date:** [September 25, 2020, 7:38am UTC](https://calculix.discourse.group/t/error-increment-size-smaller-than-minimum/199/3 "2020-09-25T07:38:38Z")

</div>

After comparing by:

```auto
Nonlinear Analysis of Structures
The Arc Length Method: Formulation, Implementation and Applications

```

```auto
*NODE
1, 0.0, 5.0, 0.0
2, 5.0, 0.0, 0.0
*ELEMENT, TYPE=B21, ELSET=PROP
3, 1, 2
*ELSET, ELSET=PROP
3

*NSET, NSET=load
1
*NSET, NSET=fix
2

*MATERIAL, NAME=STEEL
*ELASTIC
2.05e+5, 0.3

*BEAM SECTION,MATERIAL=STEEL,ELSET=PROP,SECTION=RECT
0.01,0.01
1.,0.,0.

*BOUNDARY
1, 1, 1
1, 3, 3
1, 4, 4
1, 5, 5
*BOUNDARY
2, 1, 1
2, 2, 2
2, 3, 3
2, 4, 4
2, 5, 5

*STEP , NLGEOM, INC=1000
*STATIC
5e-6, 1.0

*BOUNDARY
load, 2, 2, -15.0

*NODE PRINT, NSET=load
U
*NODE PRINT, NSET=fix
RF

*END STEP

```

Result is good.