XPSWMM experience/input 1

[Nole954]

Hi all new to the forum and relevantly new to stormwater design. The firm work at just purchased the full version of XPSWMM due to municipalities in the area using it(North Florida). I have learned the basics of the software for pond design, conveyance and hydrology. I have submitted projects to NWFWMD and local municipalities, however It seems the data output files and results are extremely long and filled with useless information. I have used the XP tables to show results and that seems the best way without handing in a 200 page report. Anyone have experience with XPSWMM and if so any ideas on how to submit results. I would rather use ICPR or PONDs, but my boss wants to use XPSWMM so I have to learn it. If you know of any forums or blogs on XPSWMM that be helpful please share.

 

[beej67]

I worked with it for about a decade, and quit around 2010. It's reports were terrible, it's graphing was terrible, and basically everything about its output was terrible. What I used to do was to tell it to include a detailed export of the routing data at important nodes, whereupon it would put a wall of tab delimited text in the output file. Then I would copy/paste that into Excel, use the "text to columns" function to break it out, and then graph what I wanted to graph in Excel directly. Then save that Excel file as a template to do the same process later, with new projects. It was a real headache, but it was the only way I found to get output I felt was responsible to present to anyone.

I spoke with XP customer support about that issue once, and they told me they had no control over it because the company had outsourced all their graphing module to another software company. Go figure.

One of the more hilarious bugs I ran into, was that if I was trying to print a hydrograph to a PDF, and XP-SWMM was up on my right hand (dual) monitor, then it would print in color, but if it was up on my left hand monitor it would print in black and white, despite being shown as color on the screen. That was in 2007, they may have fixed it by now, I just thought it was very funny.

XP-SWMM is a very powerful, very capable hydrologic and hydraulic modeling software. But boy oh boy you better know how it works or you can get yourself into some trouble. In particular, learn about how to use Configuration Parameters. They can vastly change your output, and without knowing how to use them, the software might be doing something totally different than you expect it to do.



Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[Nole954]

My head hurt just reading your first paragraph and completely agree....you hit it spot on, the configuration parameters are key but idk where/who configured the software bc the parameters are not practical whatsoever. And as far as printing goes, it is the only software that will NOT print to PDF si you must scan every page after printing. It is a very powerful piece of software with the 2D capabilities but imo it was pieced together throughout the years with no master plan. Anyways just wanted to get someone else opinion on it and you hit it on the money...Just curious, in your opinion what is the best software for site development/flood analysis??? Need to talk my boss out of using this shizz

 

[beej67]

I switched to HydroCAD for site hydrology applications after discovering it on the eng-tips forums, and am a big fan. Peter is a very responsive guy when it comes to technical questions, and the license is very affordable. Your modeling challenges down on the Panhandle are different than up here in Atlanta, though. I've done half a dozen panhandle projects, half a dozen SWFMD projects, and two or three in other WMDs in Florida so I have a good sense of what you're up against.

I know Peter will say HydroCAD does variable tailwater and interconnected ponds, and it sorta does, but the models have a tendency to explode once you connect them up in that way. Then you have to spend time adjusting time increments and calculation methods and those sorts of parameters in HydroCAD to get any output that's meaningful. So while I would consider it capable, it's touchy. XP-SWMM is much more robust at that particular task, from a modeling perspective, despite the headaches it has producing deliverables for a report. It might be nice (looking at you Peter) if HydroCAD had a button that would re-run a model over and over with different "finer routing" settings until it found a setting that didn't oscillate. One thing I really liked about XP-SWMM is at the very end of the output file, it would tell you what your continuity error was, so you could see if model fluctuations were causing an additive error that either removed or added water into your routing analysis. I wish all software had that. XP-SWMM was definitely very powerful.

I've never used it, but supposedly ICPR is the bees knees in central and southern Florida, for site development work. I've even seen guys up here in Atlanta using that, although most of our site stuff doesn't need variable tailwater analysis. I get the sense that anyone using that up here is a FL transplant. I've often thought that if I had to get back in to XP-SWMM style modeling, I'd just learn EPA-SWMM, since it's free. I've only used it once, to tweak a flood model put together by another firm. I was able to do what I wanted to do in it, but I don't know how much of a headache it would be to model something from scratch.

For flood modeling, HEC-RAS is still the thing to use, and probably always will be.


Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[psmart]

Modeling tailwater is challenging, especially as the ponds equalize and the difference in WSE approaches zero. In the past my first recommendation has been to use the "finer routing" parameter in HydroCAD, as beej67 suggests. This causes the model to repeat the calculations at each time step several times and use the average result to minimize the appearance of oscillations.

Although finer routing increases the run-time, it doesn't use any more storage, making it a reasonable solution on computers with limited RAM. But with the vast increase in RAM size over the last 10+ years, it's now possible to use a smaller time increment than was previously feasible. This allows the model to remain stable by updating the results more frequently, rather than trying to "average out" the oscillations.

So I'm now tending to recommend a smaller time step as the first recourse to resolve oscillations, rather than the finer routing. In fact, several of the warning messages have been changed accordingly in HydroCAD-10. Normally the smallest time step you can use is 0.01 hours, but it's easy to customize the "TimeInc" parameter in order to add another decimal place and go down to 0.001 hours. In many cases, especially with small storage volumes, this completely resolves the instability.

As you compare the different programs, have a close look to see what they're using for a default time step. While some programs may seem to handle tailwater better, this may be primarily a function of using a smaller time step.


Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[beej67]

Normally the smallest time step you can use is 0.01 hours, but it's easy to customize the "TimeInc" parameter in order to add another decimal place and go down to 0.001 hours.

Oh man, I never thought to check if that was an option. I might try that next time I have a variable tailwater HydroCAD model giving me problems.

While we're on topic, I've had XP-SWMM models develop oscillations because oscillations were actually occurring naturally in the model as I had it built. I'd crank down on the time step and then discover that the oscillations didn't go away, but became much more clearly defined, as water was physically sloshing back and forth through chains of conveyance nodes in the model. In order to get rid of these, I'd have to add a tiny bit of storage to each node, which would round out the oscillation as it sloshed around the system.



Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[HydrologyStudio]

As for oscillations, I agree with beej, they're occurring naturally and we shouldn't be over alarmed. After all, we're typically stuffing a large amount of water through an orifice or something similar and it's going to create an immediate backwater, regardless of the Time Interval used. We have done a lot of research on this and have determined that a smoothing algorithm employed on the receding side of the outflow hydrograph is the best solution.

As for the hydrology software in general, most in use today are decades-old, legacy software that has been the victim of acquisitions or years of add-ons, add-ons and more add-ons, resulting in bloated, unfocused, hard-to-use software. Making easy-to-read printouts and easy to use software is not easy but it has been done. It's 2014. It's time to move forward.


Terry Stringer
Better Hydraulics & Hydrology Software

http://www.HydrologyStudio.com