odesolve problem #2 SOLVED

[MilchMachtMunte]

Hi

If you scroll (almost) all the way down you will find the variable "v.ho". Generally "h" stands for helicopter while "l" stands for load. -> "v.ho" is the inital velocity of the helicopter (with the slung load underneath). However if I change v.ho to a slightly higher value there's something strange going on. Can you tell me what it is and why it is?

If you have some suggestion for improvement of the file you're invited to tell me

many thanks in advance

[Werner_E]

Better turn the document in a discussion/question as people won't be able to attach files here (other than in your original post which can get quite confusing).

BTW, what do you mean with "the system is collapsing"? It seems to work fine for me with higher values of vh0.

I have just added a pdf with vh0 set to 12.8 for you to see how it looks here. It works with higher values, too.

In case it should be a matter of Mathcad version, I am using Mathcad 15 M020.

[MilchMachtMunte]

didn't get it to turn in into a discussion 😕
I use mathcad version 14.0.0.163

It seems that's the difference. I do not understand why there is this difference.

I uploaded my results.

[Werner_E]

I now see what you were talking about the system going crazy. Have no idea how to do a workaround (which I think should be possible).

I don't remember what service release (F001, M005, M010, etc.) that 14.0.0.163 was , but it seems to be rather early one, I guess.

If it was a version prior to M020 you should be able to update at least to this release even without being under maintainance, I guess.

No - correction: I think the version you use IS M020. I am not aware of any changes to odesolve since then. I remember that the wrong results of specific integrals had been addressed and from 14 to 15 the many invlaplace errors had been fixed. If I find the update history I'll try if I find a note to odesolve.

[MilchMachtMunte]

Alright, thanks for your effort. If I find a solution I'll publish here.

[Werner_E]

Here is the link to the release note for Mathcad 14 upto M035

http://www.ptc.com/WCMS/files/98347/en/Mathcad_14_M035_RELNOTES.HTM

I attach above the changes that were made in version 15.

At a quick glance I couldn't spot something related to a odesolve bug.

Have you tried different algorithms at odesolve? Your sheet is set to Radau, maybe one of the other three works better? I just tried with all of them and they all worked.

I can't edit the document anymore, so I cannot attach a file (maybe its a bug in the forum software that I was allowed to do so at the beginning). So here is a link to the Matchcad 15 version history - hope it works.

https://www.ptc.com/view?im_dbkey=151848

Just another idea: have you tried to increase the number of intervals (in case its an accuracy problem)?. That would be a number after the end value 20 in Odesolve.

[ValeryOchkov]

Welcome to the Group http://communities.ptc.com/groups/dynamic-models-in-mathcad

[MilchMachtMunte]

I do not have access to the files. However it's working fine now. It works with all solvers except radau. The problem remains if I change the number of intervals.
Thank you for your help!