Rusty Jones

SMS Experts
  • Content count

    276
  • Joined

  • Last visited

  • Days Won

    3

Rusty Jones last won the day on March 21 2011

Rusty Jones had the most liked content!

Community Reputation

6 Neutral

1 Follower

About Rusty Jones

  • Rank
    Senior Member

Contact Methods

  • ICQ
    0

Profile Information

  • Gender
    Male
  1. I think your best option would be to look at the RMA4 manual that is distributed with SMS. It contains the governing equations and overview of the model.
  2. Hi Vera, John isn't working with SMS anymore. Anyway, as stated in earlier posts RMA2 isn't a very stable model. If it is an option, you would be much better off investigating alternative engines. It will save you lots of time and headaches. In case this isn't an option, I'll try to explain where the error comes from and some ways to try to handle it. This error indicates that the model went unstable. One way to try to keep the model stable is to try changing the flowrate more slowly. Rather than going from 60 m3/s to 100 m3/s, try going from 60 m3/s to 80 m3/s. If you look at the convergence reports, it will tell you which nodes had the biggest changes. If you use the Node | Find option in SMS, you can find where these nodes are. Sometimes this will lead to see places that have poorly shaped elements or very rapid geometry changes that can increase the instabilities of the model. You may want to try using the steering module option in SMS. This tool runs the model over and over trying to get to a target boundary condition adjusting the model each run.
  3. Hi Matteo, What you should do is use the map module to create what we call a conceptual model to specify where the nodes should be placed. Then you interpolate the data from the DEM to the nodes. You can read the values in as a raster. A raster will keep all the original points (there is no filtering). However, it is managed in a way that allows more data and will display faster than the same data as a TIN. If you look at the tutorial labeled overview, it goes over a little bit about crating a mesh using a conceptual model. Try working through it to help. If you read the DEM as a raster, choose raster as the bathymetry source when doing the map->2d mesh.
  4. Hi, I'm sorry that this isn't working for you. This may have functioned this way at one time but it has always built all polygons for as long as I can remember. The way we store and work with polygon data it may have limited benefit. Part of the information stored is the topology of the polygons (who am I inside and who is inside me). We have what is called, the "universal polygon" which isn't a true polygon. It contains the information of all the polygons who aren't inside of any other polygons. One of the more time consuming processes is keeping that up to date correctly. Trying to keep that up to date while building a single polygon at a time, may not have saved much time and may have been tricky. In any case, somewhere along the line this functionality was discontinued. The good news is that a new polygon building algorithm was implemented that is far faster than the old algorithm. I can't remember exactly when this came along and it may be new to SMS 11.0. I would recommend getting a trial password to SMS 11.0 and seeing if working with polygons is less painful. I hope my explanation makes some sense. While I can't remember the functionality being discontinued, I can certainly appreciate some of the reasons it may have come about. As I say, because of some of the complications involved it may not have saved much time anyhow. I apologize for the error in the documentation. Cheers, Rusty
  5. I don't remember what changes have taken place since SMS 8.1. That is quite an old version. I imagine most of the information on the steering module on the wiki would apply to SMS 8.1.
  6. This message just indicates that SMS recognizes that the data has changed and SMS reflects the data from the new run.
  7. Hi, I'm sorry I don't think we have a tutorial on the steering module. How to operate it should be in our help. If you are using SMS 10.1, the help is on the wiki (xmswiki.com). If you are using SMS 11.0, the help is distributed with SMS and you can get to it from the menu command (in web menu?). RMA2 is likely stable enough to get the answer if you use the steering module or similar techniques. It is easier to get a solution to converge when running in steady state mode. However, there are more stable numeric engines available for SMS. My favorite numeric engine is TUFLOW. It is very stable even with rapid wetting/drying, runs in transient mode, has a very powerful model setup approach, very good error trapping and reporting, and is very versatile since you can use it in rivers, oceans, and even urban with storm drain systems. My second choice for this would be SRH2D which uses the generic model interface with SMS 10.1 (getting it setup to work with SMS 11.0 is currently in progress). SRH is a finite volume model that is very stable. You can learn more about these models by contacting our sales team. I hope that this helps.
  8. Hi, The problem that you are having is that RMA2 is sensitive to wetting/drying issues. You can disable the element and use the steering module to try to get your run to converge or use a more stable numeric engine.
  9. RMA2 boundary conditions are intended to be use to tell how water moves into and out from the mesh. It isn't possible to define a boundary condition that is zero velocity in an element that is part of the mesh. A couple alternatives are: 1. Change the material of the element to disable. This will force the element to be draw (unused) and all flow will move around the element. 2. Make the roughness in the material super high. This will cause a bunch of energy loss in the cell so that the velocity will be severely impeded. This option will show the element has having water in it but very little conveyance. I hope this helps.
  10. Hi Jose, I wish I knew how to avoid this and sorry for the hassle. Thanks for sharing your experiences for others to learn from.
  11. Hi, I'm sorry for the trouble this is causing. I have a workaround. Could you also get us a sample file that we can use to try to make SMS recognize this and read the file correctly in the future? Contact support@aquaveo.com for upload instructions. The workaround is to use the transform data option twice to flip the coordinates. To do this: 1. Go to Data | Transform. Choose the scale option. Use 1.0 for x and z but a -1.0 for y. Click OK. 2. Go to Data | Transform again. Choose the rotate option. Use 270.0 for the angle and leave the x and y at 0.0. Click OK. In the end you should end up with coordinates that are flipped from how they were read in. I hope this helps.
  12. You should still be able to do this in SMS 11.0 (it works fine for me). Here are the steps: 1. Create a CMS-WAVE coverage (right click on the item named "Map Data" and choose new coverage and choose CMS-WAVE type. 2. Use the create grid frame tool (looks like a box) and click out approximate locations for the gridframe. 3. Switch to the select grid frame tool (looks like the create grid frame tool with an added arrow to indicate selection). 4. Drag a box around the grid frame holding the left mouse button. When you let up, you should see the IJ indicators on the grid frame and highlighted points on the corners and midway along each side of the gridframe. 5. You should be able to drag any of these highlighted corners or mid-sides at this point. You can also right click on the grid frame to change the values of origin/size/etc. I hope this helps.
  13. Do you mean a "grid frame" rather than a grid? A grid frame is used to layout the physical location of the grid before actually creating it. What model are you trying to create a grid for?
  14. Hi Matt, In the same dialog, there is an option for a translate. Translate your nodes by 0.0 for x and y and 1.0 to raise your nodes 1.0 m. Cheers, Rusty
  15. Hi Matt, The value that you are specifying is the water level you want to use as 0.0 in the new system. So if you have an elevation of 5.0 and you want to get a depth of -4.0, you would specify 1.0. The value of 1.0 becomes 0.0 in the transformed coordinates. If you want to get 6.0, you would specify a -1.0. Does this make sense? If not, let me know and I'll take another stab at it.