Sizing New Sump Pump for Existing Drainage Netwrok

[GrimesFrank]

When sizing a new sump pump for a network does one size it to pump against the head of all other pumps running simultaneously regardless of their duty?

The other pumps are for sumps with infrequent duty (e.g. floor drains for areas never used, emergency flood pit, etc.) If I size my pump (with regular duty) against those it will run to the right (i.e. high flow not balanced with in-flow) all the time. This seems fine but with the pump I have we're talking pump out times of 45 sec., hold times of 1 hour, then pump out again. This doesn't seem good for the life of the pump. I have heard rule of thumb is one would like sump pumps to run for 2 minutes (min.) Any truth?

Today is gone. Today was fun.
Tomorrow is another one.
Every day, from here to there,
funny things are everywhere. ~'Dr.' Theodor Geisel

 

[Artisi]

If you have a wide fluctuation in flow rates look to installing 2 or even 3 pumps with float level control with alternating run sequence.

 

[PEDARRIN2]

The cycle time for the pump depends on the gallons per minute it is rated at. For lower flow rates - a shorter time of operation is acceptable. I generally size the duration between 30 seconds and 4 minutes, again depending on the pump flow rate and the volume of the sump. My pump vendor likes to see at least 1 minute of run time.

45 seconds would be at the short end of the range.

You did not say how large your sump was.

 

[BigInch]

You must size against the maximum head produced by any other pump that must be running at the same time.

It may be beneficial for you to carefully determine and set the running schedules of all pumps using a timer or other control system such that you always have only one pump running, or to schedule a group such that they always output the minimum possible head at any given time. As such you may be able to minimize the overall discharge head and system flow capacity at any given time.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[GrimesFrank]

You must size against the maximum head produced by any other pump that must be running at the same time.

Thats the key BigInch. My 'competitors' are an infrequent must;
1) FD(s) from unused areas
2) Dyke area in case something spills during drum changing
(Drum changing < 1 per month)
3) Pit drainage after emergency sump pump testing complete
(testing < 1 per 6 months)

My arguement has been;
"If another pump happens to be running then we hold (or they hold) pumping until its done."
My senior's arguement has been;
"You must be able to pump against the worst case, its just standard practice."

I'm trying to show him that my 2 NPS (80ft @ 20 GPM) pump is normally operating in a range not recommended. If my pump was infrequent as well I wouldn't mind, but it just so happens that my sump is a wash station (18"Dia / 30ft sump) so I'm thinking I'll be burning these pumps out.

Am I out in left field?

Today is gone. Today was fun.
Tomorrow is another one.
Every day, from here to there,
funny things are everywhere. ~'Dr.' Theodor Geisel

 

[stanier]

Check you sump's capacity to hold inflow whilst other pumps are operating. Presumably the other pumps will eventually empty their sumps and there will be a time that this sump pump can operate against a lower head.

In the design of low pressure sewer systems this application is commonly encountered. Dwellings all have a sump pumps reporting to a common sewer system network. The most challenged pumps have to wait until the system head is reduced so that they can operate. The sumps have to have sufficient free board so that they do not overflow whilst waiting the opportunity to run.

The pumps start and if the pressure is greater than set point they stop and try again after a fixed time period. If all your sumps levels are known by the SCADA then it is even easier.

 

[PEDARRIN2]

Does the discharge from the sumps connect to a common pressurized pipe or separately to the gravity drain pipe.

If you are following the plumbing code, then the discharges cannot connect to each other, just for this reason to avoid different pressures preventing a pump to operate.

Connections to the drainage pipe should occur on the top so that each pump is pumping against/to atmospheric pressure and avoid this problem.

 

[BigInch]

I can only say, if you want to pump when another pump is running, the discharge head MUST be at least equal. The "must" you mention is a scheduling constraint that has nothing to do with hydraulics, other than how the sump capacity might affect such a schedule.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[Artisi]

As pointed out by BigInch and others, if you are pumping to a common pipeleine then all pumps must be capable of similar discharge head otherwise 1 pump can overpower the others completely or force them to a highher head at a lower flow rate. You should also digest carefully what stainer has pointed out regarding pump queueing.

I guess you have now realised that sump design is not a hit and miss exercise but requires a careful study for operation under all conditions.

 

[GrimesFrank]

Thank-you all.
Yes Artisi, I realized unless you are a specialist, most designers treat sumps with the "Eh...whatever" attitude which made my job a whole lot difficult.

I think I've hit upon a compromise though.
Do you have any opinions about a lead/lag sump pump set-up?

I proposed to put to 2 smaller head pumps (~50 ft shutoff) put them in series and when the second one is needed (i.e. sump keeps rising due to insufficient head) it'll kick on.

I get what I want (operating at BEP and balanced in/out flow) and my supervisor gets what he wants (never have to stop operations due to inability to drain)

I have the cubicle to power another pump and my controls guy is using a depth sensor rather than float switches so its just and extra couple lines of code in the PLC.

Critiques please.

Today is gone. Today was fun.
Tomorrow is another one.
Every day, from here to there,
funny things are everywhere. ~'Dr.' Theodor Geisel

 

[GrimesFrank]

Sorry Artisi;
Credit to you for the idea.....

Today is gone. Today was fun.
Tomorrow is another one.
Every day, from here to there,
funny things are everywhere. ~'Dr.' Theodor Geisel

 

[BigInch]

One big one is better than two little ones.
<anonymous>

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[Artisi]

"I proposed to put to 2 smaller head pumps (~50 ft shutoff) put them in series and when the second one is needed (i.e. sump keeps rising due to insufficient head) it'll kick on."

I hope you mean in parallel - not series.

The first thing you should do is to produce a system head curve for the pipeline based on anticiated flow rates from all the pumps when operating, from there you can superimpose the individual pump performance curves over the system curve to check performance of each pump under maximun conditions - if you know the max. inflow and max. flow rate for each of the pumps you can then estimate the sump size.

For interest - do the other sumps have sufficient capacity if you are adding further flow to the pipeline?

 

[BigInch]

Artisi, I think he still wants to be able to pump (at the same rate) when the press in the pipeline goes higher than what one pump will do.

I'd consider putting a monster there .... <ready for this?> with a VFD! He'd be able to match the required head and pump more flowrate whenever he turned up the dial. Sounds like that's just what is needed here... for a change.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[Artisi]

BigInch - It's all a little confusing at the moment, seems he is worried about running a pump capable of pumping against all others plus friction losses if and when all are in operation and then running this pump on it's own at a lower head - it seems we are back to my first posting - install 2 or even 3 units based on a hydraulic study on min / max flows and heads.

 

[BigInch]

With high flow and head variations, 1 pump and one VFD often proves out well. Just have to see if head and flow match the pump with the VFD, otherwise its a wide open playing field.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[GrimesFrank]

BigInch has it.
I was proposing two smaller pumps in series to give me a constant flowrate (to pump out my inflow) but have a choice of heads depending on how many other pumps are running in the network.

BigInch I was hoping I didn't need a VFD as I've never dealt with them before as well as I only had two scenarios I was concerned about (1 pump running OR all pumps running). I guess this becomes a 'learning opportunity' now.
I was hoping to install a 2 NPS discharge submersible sump pump (~1-1.5hp). Is a VFD still suitable?

Today is gone. Today was fun.
Tomorrow is another one.
Every day, from here to there,
funny things are everywhere. ~'Dr.' Theodor Geisel

 

[BigInch]

Depends on what the instantaneous head required is in relation to the minimum discharge head needed. If static head is always high, a vfd may not get you much leverage. The required discharge head (system head) should be proportional to some factor times the square of the flow to get the best fit for a vfd.

I didn't know the maximum power requirement would be so small. Now I have to change my thoughts too. In this very low power consumption case I would rather use 1 constant speed pump x 3 HP and always have too much head rather than fooling with a vfd, as running 3 HP when you only need 1.5 just won't ever save enough power to make a vfd worth the effort of installing it.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[nationalpump]

Hello,

The problem you are facing may be able to be solved by an air operated submersible. The "Chicago Industrial Pump Company" produces the PITBULL pump line which should work in your situation. You would need to size the pump to operate slightly above the maximum header pressure which would allow the sump pit to always be drained. The output of the PITBULL would drop as the pressure in the line increased just like a centrifugal pump though you would not have to worry about "running off the BEP" and damaging the pump. There are other things to consider (air consumption, size and shape of available pumps, cost etc.) when looking into a PITBULL so this may not be right for your application. I would be interested to hear what you think. Thanks and good luck.

 

[BigInch]

With header pressure is increasing, a centrifugal pump naturally reduces flow and backs up on its curve, so there is no running out on the curve. How is your Pitbull different?

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein