 Rank: Newbie
Groups: Registered
Joined: 20/02/2023(UTC) Posts: 6 
|
I have attached a mathcad file.. and the moment diagram at the end doesn't match with expcted output.. it should have gone smoothly to end with x coordinate 5m. but at certain point it has jumped to postive rigion. Please help fix it. Thanks   00Two Overhang Beam with UDL.xmcd (645kb) downloaded 8 time(s).
|
|
|
|
|
|
 Rank: Advanced Member Groups: Registered
Joined: 23/07/2013(UTC) Posts: 1,198  Was thanked: 548 time(s) in 368 post(s)
|
I have no idea why mathcad plots that segment faulty. But it is possible to plot correct graphic in SMath. However SMath integration is not very good and fast. It will take much more time than mathcad, and zig-zagy. You can improve plot by increasing Integrate: Accuracy. It is in Tools>Options>Calculation. Regards PS: I increased accuracy to 500 to get a decent plot. It took 50 seconds to plot in the picture!! Integration should really be improved. UDL.sm (19kb) downloaded 21 time(s).
|
|
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 15/04/2012(UTC) Posts: 2,048  Was thanked: 1177 time(s) in 752 post(s)
|
In the interactive handbook you can find examples for beam analysis. They use symbolic integration and equation solving with Maxima. This is very fast, except for the Maxima restart at the beginning of the sheet (can be removed if all works well).
example maxima beam.sm (42kb) downloaded 14 time(s). |
|
|
|
|
|
|
Rank: Newbie
Groups: Registered
Joined: 20/02/2023(UTC) Posts: 6
|
Sorry but I have problem again
with M(5m)=-0.2083 kN m
it should have been
M(5m)=0
|
|
|
|
|
|
Rank: Newbie
Groups: Registered
Joined: 20/02/2023(UTC) Posts: 6
|
Originally Posted by: yoswos  Sorry but I have problem again
with M(5m)=-0.2083 kN m
it should have been
M(5m)=0 My sorry it is correct.. since it is overhang it should have moment at the end;;
|
|
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 23/07/2013(UTC) Posts: 1,198 Was thanked: 548 time(s) in 368 post(s)
|
Though it should be smaller. Probably zero by the way. Result can be reduced with integral accuracy set to 10000. SMath integral algorithms are not good enough. Regards 
|
|
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 28/08/2014(UTC) Posts: 1,533  Was thanked: 990 time(s) in 618 post(s)
|
Hi. Remembering that solving an ODE is also told to integrate it: UDL.sm (23kb) downloaded 17 time(s). The domain can also be restricted to M(x) by multiplying it by the corresponding inequalities. Best regards. Alvaro. Edited by user 2 years ago
| Reason: Update the ode solver.
|
 2 users thanked Razonar for this useful post.
|
on 21/02/2023(UTC), on 21/02/2023(UTC)
|
|
|
 Rank: Guest
Groups: Registered
Joined: 04/07/2015(UTC) Posts: 6,866  Was thanked: 983 time(s) in 811 post(s)
|
Originally Posted by: yoswos My sorry it is correct.. since it is overhang it should have moment at the end;; Finite Differences [FD] is a very good yardstick insensitive to noise. It that what you mean moment at the end ?Cheers ... Jean. Page2 FD.sm (58kb) downloaded 4 time(s).
|
|
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 23/07/2013(UTC) Posts: 1,198 Was thanked: 548 time(s) in 368 post(s)
|
Originally Posted by: Razonar Hi. Remembering that solving an ODE is also told to integrate it:
The domain can also be restricted to M(x) by multiplying it by the corresponding inequalities.
Best regards.
Alvaro. This is a marvelous solution. I learned a new thing today. Thanks Razonar. Maybe I am wrong but I think formula should be like this. Huge kN*m values appear if calculated as in the picture above. Regards 
|
1 user thanked overlord for this useful post.
|
|
|
|
Rank: Guest
Groups: Registered
Joined: 04/07/2015(UTC) Posts: 6,866 Was thanked: 983 time(s) in 811 post(s)
|
ODE integrator(s) range from exact [rarely] approx. 4 decimals ... to glorious freak ! 64 bits Smath does not have 32 bits Matlab ODE. Recently checked on my PC2 this damned bug. Cheers ... Jean. ODE 0 TEST [17].sm (46kb) downloaded 3 time(s).Revisit from few months ago.
|
|
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 28/08/2014(UTC) Posts: 1,533 Was thanked: 990 time(s) in 618 post(s)
|
Originally Posted by: overlord Maybe I am wrong but I think formula should be like this.
Huge kN*m values appear if calculated as in the picture above ... Hi overlord. You're absolutely right. Thanks for catching that error. Only 3 orders of magnitude. Who could worry about that? The Mathcad block can also be used, although I don't know if this is the best way to deal with the units, using an intermediate function Mo.  UDL.sm (27kb) downloaded 19 time(s).Best regards. Alvaro.
|
|
|
|
|
|
Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.
