# Slow to converge, or failing?
**URL:** https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047
**Category:** General Questions
**Created:** [October 1, 2024, 12:57pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047 "2024-10-01T12:57:59Z")
**Posts on this page:** 20
**Page:** 2
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 3, 2024, 3:38pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/21 "2024-10-03T15:38:17Z")
Tried that setting - still got many errors on check part. Probably worse than before.
Non-positive jacobean
---
### Author:  [@synt](https://prepomax.discourse.group/u/synt)
#### Post date: [October 3, 2024, 4:23pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/22 "2024-10-03T16:23:58Z")
strange when the bolt has been simplified, geometry also not too complex enough for tetrahedral mesh. Can it be more specific for all mesh setting since previous error message may related to mesh in fit to geometry?
---
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 4, 2024, 6:52am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/23 "2024-10-04T06:52:07Z")
Went back to previous mesh settings. reverted to contacts instead of constraints for where the plates touch each other (have set these to Tied for now). That got rid of the non-positive jacobian errors.
Eventually it ‘fails with results’ - the results are all zero.
Biggest problem I have is that documentation is only a list of controls, it doesn’t give any detail on usage, or explain what control actually does. I have to guess everything.
Is any example of use of ‘Amplitude’ available? - the control makes no sense to me, and I see no way to link this to the loads.
---
### Author:  [@caque](https://prepomax.discourse.group/u/caque)
#### Post date: [October 4, 2024, 7:38am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/24 "2024-10-04T07:38:55Z")
I view it as a function f(time). You multiply the magnitude of your load (displacement…) by this function.
Timesteps are defined in the incrementation parameters of your step.
Check out FEAnalysts tutorials on youtube.
---
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 4, 2024, 8:01am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/25 "2024-10-04T08:01:12Z")
> [@caque](#):
>
> Check out FEAnalysts tutorials on youtube.
Any specific tutorial? The titles don’t give obvious hint as to which one might cover this…
---
### Author:  [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)
#### Post date: [October 4, 2024, 8:02am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/26 "2024-10-04T08:02:44Z")
Yes, it’s just a time-dependent multiplier for loads. It’s defined as a table (time vs amplitude value) first, then assigned to BC/load and its magnitude must be properly specified since it will be multiplied by the corresponding amplitude value in each time increment.
---
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 4, 2024, 8:15am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/27 "2024-10-04T08:15:22Z")
I tried removing all but one of the pressure loads to see if that would help. Still doesn’t seem to be converging.
I suspect I must be missing something more fundamental or basic with the set up.
Is there any tool or setting where the defaults might not be a good idea?
I can’t get any sort of plausible result from this - no clues from output as to what the issues might be.
To be frank, I don’t understand how to use most of the tips provided by others above.
My previous system would almost always solve, if it could mesh the model. Unfortunately it is no longer supported and won’t mesh this model.
---
### Author:  [@caque](https://prepomax.discourse.group/u/caque)
#### Post date: [October 4, 2024, 8:17am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/28 "2024-10-04T08:17:10Z")
Since you are doing a contact pb for instance this one
[https://youtu.be/IsSsUDkVlpM?si=UNXlLCWuO1RqY1kl](https://youtu.be/IsSsUDkVlpM?si=UNXlLCWuO1RqY1kl)
---
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 4, 2024, 8:39am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/29 "2024-10-04T08:39:46Z")
Thanks - what does ‘pb’ stand for in ‘contact pb’ ?
---
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 4, 2024, 9:14am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/30 "2024-10-04T09:14:07Z")
OK - I have an amplitude table set up, not sure yet if that is helping.
If results are still nonsense, could anyone take a look at my file to see if there are any glaring errors on my part?
IF I can get this to work at all, I can build on it by adding pre-tension perhaps, to be more realistic.
---
### Author:  [@synt](https://prepomax.discourse.group/u/synt)
#### Post date: [October 4, 2024, 9:38am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/31 "2024-10-04T09:38:33Z")
as i know, automatic tetrahedral mesh in PrePoMax have two mesher, default is Netgen (basic) and Gmsh as alternative. At some model is not always easy to run of the box by single preset, specifically in curved geometry and contact analysis. Maybe basic CAD files provided if possible, so another user can be trying to understand about the problem by several mesh setting.
---
### Author:  [@Trevor](https://prepomax.discourse.group/u/Trevor)
#### Post date: [October 4, 2024, 10:53am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/32 "2024-10-04T10:53:25Z")
have you shared the model ?
---
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 4, 2024, 11:48am UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/33 "2024-10-04T11:48:07Z")
Not yet - the .pmx file is about 155MB, so rather large to post to forum. I’ve no idea what is the accepted norm regarding upload size here…
---
### Author:  [@Trevor](https://prepomax.discourse.group/u/Trevor)
#### Post date: [October 4, 2024, 12:05pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/34 "2024-10-04T12:05:03Z")
people have used google drive or wetransfer in the past
---
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 4, 2024, 12:12pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/35 "2024-10-04T12:12:42Z")
Wasn’t until now aware of option to get link from WeTransfer - always learning!
> **[ModelB.pmx](https://wetransfer.com/downloads/6c738086521ca6f70af2b626e6440d7520241004120945/6e47e0?t_exp=1728302985&t_lsid=0cfd6d25-1901-4dfa-b094-5a78baba78c3&t_s=download_link&t_ts=1728043785)**
>
> 1 file sent via WeTransfer, the simplest way to send your files around the world
The model here is top half of of the actual assembly. It is symmetrical about the flat faces that surround the channels with Load3 applied.
I’ve omitted some minor parts whilst trying to get this to run at all.
---
### Author:  [@Trevor](https://prepomax.discourse.group/u/Trevor)
#### Post date: [October 4, 2024, 3:03pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/36 "2024-10-04T15:03:34Z")
Unconstrained parts ?
---
### Author:  [@DavidJ](https://prepomax.discourse.group/u/DavidJ)
#### Post date: [October 4, 2024, 3:36pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/37 "2024-10-04T15:36:24Z")
But only 2 parts are in the model - the ‘chokes’ were omitted.
I’ll delete them entirely to be sure and try again.
---
### Author:  [@Trevor](https://prepomax.discourse.group/u/Trevor)
#### Post date: [October 4, 2024, 3:39pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/38 "2024-10-04T15:39:22Z")
My mistake, I put them back in !
I’ll take another look.
---
### Author:  [@Trevor](https://prepomax.discourse.group/u/Trevor)
#### Post date: [October 4, 2024, 4:01pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/39 "2024-10-04T16:01:56Z")
Just the 2 parts. unconstrained I think.
---
### Author:  [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)
#### Post date: [October 4, 2024, 4:06pm UTC](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047/40 "2024-10-04T16:06:12Z")
Which PrePoMax version are you using ? I get this error when trying to open the file in v2.1.0 (seems to be created in this version but v2.1.11) shows the same problem:
[Previous page](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047.md?page=1)
[Next page](https://prepomax.discourse.group/t/slow-to-converge-or-failing/2047.md?page=3)