FM discharge is lower than high point of FM 3

[douginpc]

howdy all.

i have a small sewer lift station where the discharge point is lower than the high point in the force main and also lower than the submersible pump.

it's understood the pump must initially overcome the high point in the FM. once the FM is full, and the system is operating steady state, can i count on a reduced static head due to a siphon effect from the lower downstream portion? a siphon would seem logical at low or zero velocities but i don't know if the same principle applies to a higher velocities of 2-3 fps. do the dynamics of moving water overcome any siphon effect?

simply put, does static head decline and the dominating force become the friction head?

thanks,
doug

 

[BRIS]

I have a problem answering your initial question because I cannot see the logic of your argument


1) the maximum siphon vacuum pressure is the vapour pressure - you cannot siphon more than about 9.0m head.

2) once you have primed the siphon - i.e pumped through the hump and evacuated the air you can theoretically switch off your pump and the flow will continue under gravity flow.

3) However you have to vent the air from the crest of the siphon and for siphon action to continue after priming you need to have a vacuum at the crest. Most air valves vent out and in and out and don't allow siphon action. i.e they also act as vacuum breakers.

3) You have a negative static head on the pump. i.e you are pumping down hill

4) your concern is the rate of flow. The head loss flow calculations are the same as normal.

5) The concern is that the siphon de primes and has to be re-primed after each pump cycle. this usually creates odour problems !!


6) -Providing pressure all along your system remain above vapour pressure, and you don't allow air into the system then you can ignore the siphon.

7) To determine whether you have a siphon under pumped conditions you will need to draw the hydraulic gradient (HGL) on the pipe long profile - if the HGL remains above the top of the pipe you have pressure flow.
If it falls below the pipe you have a vacuum (sub atmospheric pressure)
If it falls 9.0m below the pipe you have vapour.

 

[BRIS]

PS once you have determined whether you have (or don't have) sub atmospheric pressure (siphon conditions) under pumped flow the next step will be to determine whether it is sensible to operate with a siphon or to provide a vacuum breaker.

Then you need to consider whether or not you have a water hammer problem

 

[douginpc]

Thanks for the response. Let me be more descriptive:

The high point is about 17’ above the pump. The FM discharge is 6’ below the pump. The FM length is about 2770 ft. It pumps up and then back down to the discharge.

It's understandable that the pump will need to overcome the high point of 17’ at start up. In this case, 17’ would be the starting point for the system curve.

Questions:
1) Once steady state flow is achieved, will the pump always feel the 17’ of static head plus friction head?
2) Does the net drop in elevation assist the pump in some manner due to a siphon effect?
3) Will the system curve drop down on the pump curve thereby resulting in increased flow through the system?

Thanks

 

[bimr]

In addition to the comments by Bris, you may consider the use of valve on the dicharge end of the force main to prevent siphoning.

If you have significant elevation difference, the negative pressure from the siphon may collapse the pipe or drain the pipe.

In addition, it is probably not good practice to allow the force main to drain since you will get gases and two phase flow when you start pumping again.

These problems can be eliminated if you use a full ported diagphram valve or a pinch valve on the discharge end of the force main. The valve would open and close with the pump operation.

 

[BRIS]

I agree with bimr's point - I have provided a valve on the end of long FM's to prevent draining and odour problems. Also the fall from the crest to the outlet is 23 feet so you could have significant vacuum and you need to design the pipe not to collapse.

If you really want to save energy then there is no reason not to have a siphon. On the other hand if the energy saved is not significant you would be better breaking pressure at the crest and designing the downstream leg as a gravity sewer. You have 23 feet drop.

In answer to the questions

1) No once the system is primed the pump sees a static head of minus 6 feet plus friction head.

2) The drop in elevation assists the pump - same answer as 1) above

3) Once the pipe is full the system curve is the same as any system curve. You work out friction losses at various flow rates and add them to the static head which is - 6 feet. That gives you the system curve.

what you need to do is ensure your pump is operating within its duty range.

I guess for a velocity of 3 to 4 ft/sec. you have a friction loss of about 20 feet and therefore a pump head of 20 - 6 = 14 feet at the duty point. (I don’t work in US units)

This could work.


You realise that without a valve you can't stop it: once the pump stops the siphon will continue to flow until your wet well is empty, the pipe draws in air and breaks the siphon.

 

[douginpc]

Ok, so the pump does see a reduction in static head. That aside, I still need enough to get over the hump when I lose the prime. I'm not counting on a power savings. I just wanted to make sure my logic is correct so I could calculate some operating points on the curve.

There is an anti-siphon valve at the top of the system to prevent siphon once the pump stops. I shouldn't develop any negative pressures in the pipe until the pump shuts off.

Unfortunately I can't gravity the sewer from the top to the bottom. Too many undulations between the pump station and discharge point. The net elevation difference is 23 feet but it's not straight line.

Thanks

 

[redbridge]

You need an air/vacuum valve at the high point to prevent a siphon. The pump will only have to overcome the high point unless the headloss through the entire line is greater than the static head (head between high point and pump). If static head is greater than headloss through entire line than the tdh will be static head plus headloss through pipe to the high point, from the high point to the discharge the pipe will not be full flow. I personally would not design for a siphon. I have had problems in the past.

 

[BRIS]

If you have a vacuum breaker at the high point then you don't have a siphon.

The anti siphon will prevent a siphon when the pump runs.

Until you plot the HGL and determine whether it is above or below the pipe at the hump you don't know whether you have potential for a siphon.

You need to plot the hydraulic grade line from the outlet. Whenever the line goes below the top of the pipe you have sub atmospheric prssure in theory. The pipe gradient at this point is steeper than the HGL. Follow the pipe gradient until you reach a point where the HGL is steeper than the pipe grade and carrying on ploting the HGL until it next intersects the pipe and so forth until you get to your pump. The height of the HGL above the pump is the pump head.

I am not sure I understand Lbrgs hydraulics but I do agree with Lbrg and Bmir; at the end of the day a siphon is not your best solution.

 

[BRIS]

Lburg

I think you are saying if the HGL from the outlet to the pump is at all points above the pipe (i.e positive pressure) then the pump head is static head plus friction loss; as for any pipeline.

If the HGL is below the hump (siphon crest) and if there is a vacuum breaker at the top of the hump so that there is no siphon then the pump head is the static head up to the hump plus the friction head between the pump and the hump. - I agree. (The pipe downstream undulates. If it is ventilated - there will be part full flow, hydraulic jumps and full flow sections).

If there is a vacuum breaker the pump always has to overcome at least the high point plus friction between the pump and the high point!.

 

[douginpc]

thanks for the responses.

i'm not counting on a siphon to make this thing work. i was mainly curious if i could expect higher flows since the discharge is lower than the pump.

 

[rconner]

I think the subject of forcemains that do not flow full, or are in effect at some point "pumping downhill" have been discussed multiple times on these forums, and many more thoughts and (perhaps some varying) opinions could be derived by "advanced search". It would appear that in some cases the hydraulic situation could be most predictable if the pumped pipelines are somehow/dependably designed to flow full (sort of the definition of a forcemain?), and additionally it might be helpful if the liberation/release of the bulk of sewer gases e.g. hydrogen sul(ph)fide is most isolated/controlled to be at the end of the pipeline (and not at other locations, e.g. in neighborhoods etc. along the pipeline). If the pipeline does not flow full, it likewise appears there could be more uncertainties/potential issues to deal with (as a result of what could happen under some conditions in the non-full extants). When sewer forcemains do not flow full and are allowed to contain an "atmosphere", references indicate that in some cases and under some conditions this can contribute to safety, odor, hydraulic/surge, and/or corrosion problems along the pipeline. I guess as an alternative or supplement to specific control valving (as previously mentioned in this thread) to keep such sections charged or full, I have also noticed some folks have reportedly employed in some cases where they can (an arguably some simpler principle?) vertical “goosenecks” or “barometric loops” at the end of the pipeline (e.g. see the posts near the end of thread at

http://www.eng-tips.com/viewthread.cfm?qid=24447&page=8).

As far as siphons or inverted siphons, while it is true they are widely employed as a hydraulic device I have also noticed the first bullet/guidance you often see when this is discussed is "Minimize the length of siphon."

 

[Yellowtetra]

Doug,

I agree with the other posters on plotting your hydraulic profile to determine if a siphon is even possible.

If it is, you can almost never count on a dependable siphon since any air or gases that get into the high spot will destroy the siphon. Your system would perfrom erratically because of this. Install an automatic air/vac valve at the high spot and design the system with a discharge elevation at the high spot.

Dan Barr, PE
Burgess and Niple