Hydraflow Storm Sewers flow changes in closed system

[Anonymous]

I don't use this program much, but was trying to layout a small system and am getting what appears to be odd results.  I have some surface flow going into some drop inlets, then it goes to a Junction Box and the outflow.  My there isn't any flow added at the junction box and there isn't anywhere for the water to be leaving the system, but the flow drops 2 cfs from the inlet pipe to the outlet pipe.  How can the flow be reduced in a closed part of the system?

I've attached some screenshots of the programs output.  The line from from EX DI#1 to JB#2 to outfall.  Notice on the "Pipes" screenshot (capture2.jpg) it shows addition flow from a DA into the drop inlet, EX DI#1, but nothing being added at JB#2.  The results (capture.jpg) show a drop in flow 1.98 cfs.  I don't understand how this can happen.  The profile view doesn't show the water level above the pipe, so it shouldn't be leaving the system at JB#2.

Thanks for any help!

[tony1978]

Interesting. So just to make sure, the storm pipe is in partial flow not full flow? Bc the hgl jump is 1.43 ft at the junction box.  It's not hitting he top of your JB right? Can you show screenshots of the profile?

[MattAndersonPE]

This is a possibility based on the limited data you provided.

The flow in a pipe is a function of the CA of all inlets upstream, together with the overall system intensity at the upstream end of the pipe.  

So, the calculations appear to indicate that the systems time in the 395 feet is significant enough to change the overall system time for the next downstream pipe.  

I would suggest re-running the system with the enhanced method, which will perform a few more iterations to balance the estimates with the true EGLS.

[Anonymous]

Well, one of the nodes just lost all of it's data when I opened it this morning.  Let me see if I can get it to work again.  I've used Hydraflow Hydrographs for years and know it can be buggy at times.  Is it the same with this software?

[Anonymous]

Since the file was corruputed, I recreated it with only this line in the file.  Ran it with the 10yr storm analysis w/ design.  Capture.jpg shows the results as losing flow between the last two pipes, which is impossible as it's closed. Capture1.jpg shows the profile and HGL and it never goes above the manhole rim.

As suggested, I reran the design using the "Full Design" option and Capture2.jpg shows that result.  Notice that the flows stayed exactly the same, so at least that's consistant as it should be.  I screwed around with the pipe runs just to see if that had any effect on Q and of course it didn't.

This data is completely worthless to me as I can't turn in a design like this with flow vanishing.  Guess I'll have to go back and do it the old fashioned way, one pipe at a time.  

FYI, I am using Civil3D 2010, so behind the times, but looking around here some, it looks like this program is still exactly the same as it was when AutoDesk purchased it years ago.

Heck, I at least had reliable results a long time ago (when I did storm drain design more) using the design software in the Civil package of LDD.  It was clunky but at least I knew what was going on.  This makes no sense.

I will try it once more using "Known Q" instead of letting the software calculate it and see if that makes a difference.

[Anonymous]

And just so it's out there, here's the "Inlet" data from the program.  Key thing here is there the is zero flow being added at JB#2, which is correct.  Total flow into the system is 24.82 cfs, which of course should be equal to the total outflow unless somehow the Law of Conservation doesn't apply any longer. That number is about what's in the pipe prior to JB#2, but it disappears at that manhole.

[Anonymous]

Oh, Matt you did say run it in "Enhanced" mode, not Full, so I tried it again with that.  Oddly the flow in the last pipe went up some and I'm just missing 0.48 cfs now.  Still unacceptable as there is water vanishing inside the system (in doesn't equal out).

[Anonymous]

OK, using "Known Q" fixes my problem, well at least in my head.  Am I not grasping something here?  It appears the software is using Tc in the calcs to maybe account for some storage?  I've always designed storm drains using just peak flow, which is what every review agency does as well.  Seems like this is getting a bit too technical especially since we're just using the Rational Method to calculate flows.

[MattAndersonPE · Accepted solution]

Hydraflow is not a conservation of mass/moment tool.  Autodesk Storm and Sanitary Analysis is.

What is tripping you up is the pipe travel time.  Turn on the "Suppress Pipe Travel Time" on the Design Codes.

Hydraflow is using the Rational Equation to determine the peak flow pipe pipe based on the System CA, and System Tc.  If you look at DOT or FL-DOT Report, you will see the how this makes a difference.  Your last pipe is evaluating the system Tc at a much longer time than the prior pipe of 400 or so feet. It appears that the velocity assumption and the actual velocity are not in concert for that pipe.  Since you do not add flow at the last pipe, the system reevaluates the System Tc and adds the pipe travel time to that, resulting in a low System intensity, and hence lower Q.

[Anonymous]

That fixed it. Thanks Matt.