Are two small pumps equal to one big pump?

[Sapo]

For a fountain I am planning, I was recommended to use one 5HP pump. I could special order it, but all my local suppliers have 3HP pumps and everyone knows how to service them and has the parts at hand. Special ordering one might mean down time in the future waiting for parts.

Can I use two 3HP instead? How do you hook up two pumps to work in concert? An expert plumber will do the actual installation, btw. I just need to have a rough idea of what can be done and how so I am not taken for a fool by such plumber.

For an alternative setup I was recommended two 5HP. Would that then be four 3HP pumps? Is there a practical limit to how many pumps you can hook up? Thanks in advance.

[Harmonious Discord]

You need to look at the gallons per hour rating, not the horsepower.

[beowulff]

You realize that a 5HP motor is going to cost almost $9/day or $260/month to run, if electricity costs 10¢/KWH, don't you?

[Sapo]

You need to look at the gallons per hour rating, not the horsepower.
That was my original impression from my first internet research but everyone I have contacted, both suppliers over the internet and local vendors speak of HP all the time.

In any case, can you hook up two smaller pumps to do the job of a large one? Practical limits and considerations to keep in mind if it is indeed possible.

[Sapo]

You realize that a 5HP motor is going to cost almost $9/day or $260/month to run, if electricity costs 10¢/KWH, don't you?
That's about right for my location and although I hadn't done the math, it is not too far from what I had guesstimated. This is not for my backyard, though (and if it were it wouldn't be on 24/7). This is for a commercial setup and the city will be paying for the electric bill (woot).

[Squink]

Centrifugal pumps are limited in the amount of backpressure they'll work against.
For that reason, connecting two small pumps to a single pipe will not give twice the flowrate of a single large pump, unless it's a big pipe.

[Harmonious Discord]

There is going to be a way to use two pumps, but I'm not going to be able to tell you how. It would certainly require a y junction before and after the pumps to run them in parallel and a back flow valve for each pump.

[Sapo]

So, yes, there is a way. I presume that it is too complicated to explain to the layman over the internet but my plumber should know how to do. But it comes at a cost in efficiency and the smaller pumps won't just add up to the sum of their powers. Right?

You are not my plumber and all that. Considering that the 5HP pump costs about 6 times what the 3HP pump costs (and factoring in the issue of not being the daily bread of my local suppliers), is this something I should consider or is it so much of a hassle that I am always better off buying the big pump?

[Bill Door]

You need to look at the pump curves, it's not just a matter of horsepower. A pump's power requirements can be calculated by multiplying the flow in gpm by the head in feet by the specific gravity of the liquid then dividing by (3960*efficiency). So a pump pumping 80 gpm of water with 100' total developed head (TDH) at 70% efficiency needs around 2.9 horsepower.

The same pump with a different impeller might pump 40 gpm with 200' TDH and at the same efficiency take the same horsepower to pump half the volume but at twice the pressure. The head requirements can be estimated by adding together the static head plus frictional losses in pipes and fittings. The efficiency comes from the pump curve and will vary with the flow rate. A well designed pump will usually be in the 70 to 80% efficiency range most of the time.

So what you'll need are two pumps, each designed to pump half the volume of the 5 HP pump, but at the same TDH. Pumps usually don't do well on a shared pump suction, which gets into net positive suction head (NPSH) issues, but if there's enough static head between the fountain and the pumps to provide NPSH you should be all right.

[YamatoTwinkie]

You're really going to have to figure out what flowrate(GPM) and pressure is required for your fountain. Those are your two main factors you should be looking at first. Not all 5hp pumps are created equal.

Yes, you can configure two pumps in series (doubling the pressure), and parallel (doubling the flowrate), if your system is properly designed. But you really need more information first.

[gotpasswords]

How about splitting your grid of nozzles into two halves and running each half on its own pump? This would eliminate all of the mysteries of how to plumb two pumps together, while still letting you enjoy the benefits of cheaper and readily serviced smaller pumps.

You'll have two runs of pipe (four pipes total - suction and output times two) between the fountain and the pumphouse, but that would be about the only complexity vs one large pump.

[boytyperanma]

I am a professional in this field so far their is not enough information to answer your questions. Moving water is and can be much more complicated then your trying to simplify it down to. A motor and a pump are two different things. A motor is what powers a pump. A pump moves the water. Just in the realm of 5HP motors I can pick up books from several manufacturers and each will have 5 pages of related flow charts for the different pumps that can be powered by a 5HP motor. To determine the correct pump/motor combination you need a lot of data.

submersible pump or an above ground variery
Available gpm from your source
Desired gpm at the point of use
friction loss calculations-size and type of piping
elevation changes
available power- amperage voltage and phases available

If you can describe what this project is I'm sure we can help. Likely your going to have to talk to the 'plumber' anyhow to get the information so he might be the best person to start this line of questioning on.

[Bill Door]

You've already got enough data to point you in the right direction, but without trying to hurt your feelings, it's like you said, "I'm thinking about sailing solo across the Atlantic, the pointy end of the boat is the front, right?" You probably want to practice in some shallow water first.

That being said, if you want to try, buy this book; Cameron Hydraulic Data (http://www.amazon.com/Cameron-Hydraulic-Data-Reference-Hydraulics/dp/9990620393). It's detailed enough to design your system and still accessible to someone without a PhD in fluid dynamics. For the size of your project, $90 is pretty cheap.

[Sapo]

You've already got enough data to point you in the right direction, but without trying to hurt your feelings, it's like you said, "I'm thinking about sailing solo across the Atlantic, the pointy end of the boat is the front, right?" You probably want to practice in some shallow water first.

Point well taken. It is a bit more like "I'm thinking about sailing solo across the Atlantic and I hired this captain to sail the boat. Help me not look like a total idiot when I talk to him". (or maybe "... and I hired this dude with an anchor tattooed on his arm. How do I know if he knows what the pointy end of the boat is?"). Special note of the fact that I call him a "plumber" because that's what he is, not a hydraulic engineer or anything of the sort. He has done pools and fountains before, just maybe not at the scale we are thinking now.


I did consider the idea of just splitting the system into smaller parts that can be managed by each smaller pump. Any caveats at this other than the multiple pipe sets running to and from the installation?


The project is a Fountain Maze. It will be a zero grade fountain (aka pop up jets. Jets that come out from the floor with no standing water) with a grid of water jets that will define the maze "walls". We are talking about a 100'x100' grid with a 6'x6' grid. I started a thread about it a while ago that didn't go too far.

Thanks for all the responses so far. I appreciate the fact that I am asking for stuff well over my head and that I am not providing enough detail. I will be contracting professionals for all the work, I just want some general info on what I am dealing with so I can talk to these professionals and not be bullshitted too badly.

[YamatoTwinkie]

When active, are all of the maze "walls" constantly running, or are you talking about popping up individual walls as a user goes through?

[Sapo]

When active, are all of the maze "walls" constantly running, or are you talking about popping up individual walls as a user goes through?
Constantly running. The maze is configurable by turning walls on and off but that happens off line, not when people is in there. Doing that would be almost trivial from a technical point of view but well out of our budget right now. v2.0 maybe.

[Snnipe 70E]

So, yes, there is a way. I presume that it is too complicated to explain to the layman over the internet but my plumber should know how to do. But it comes at a cost in efficiency and the smaller pumps won't just add up to the sum of their powers. Right?

You are not my plumber and all that. Considering that the 5HP pump costs about 6 times what the 3HP pump costs (and factoring in the issue of not being the daily bread of my local suppliers), is this something I should consider or is it so much of a hassle that I am always better off buying the big pump?

Yes you can run two pumps in plaace of one larger pump. But the system will have to be engineered. And you will need to get two pumps that are designed right. It is done all the time. when I worked in SF the buldings central plant had three chilled water pumps and three tower water pumps. Depending on the cooling load the plant was designed to run on one, two, or three pumps.

[Spezza]

For a fountain I am planning, I was recommended to use one 5HP pump. I could special order it, but all my local suppliers have 3HP pumps and everyone knows how to service them and has the parts at hand. Special ordering one might mean down time in the future waiting for parts.

Nothing to add except: while one 5HP pump might require longer downtime, two 3HP pumps double the odds of a failure.

[Snnipe 70E]

Nothing to add except: while one 5HP pump might require longer downtime, two 3HP pumps double the odds of a failure.

Consider 3 pumps. No down time.

[Sapo]

Consider 3 pumps. No down time.
This is where I was going.

[CutterJohn]

Considering its a backyard fountain, I think downtime isn't a grave concern. Certainly not enough of one to engineer in redundancy.

[billfish678]

If its a swedish made pump you are only gonna need one baabbieee.

[Sapo]

Considering its a backyard fountain, I think downtime isn't a grave concern. Certainly not enough of one to engineer in redundancy.
Post #5

[HongKongFooey]

FWIW

Considering its a backyard fountain, I think downtime isn't a grave concern. Certainly not enough of one to engineer in redundancy.

This is not for my backyard, though (and if it were it wouldn't be on 24/7). This is for a commercial setup and the city will be paying for the electric bill (woot).

[HongKongFooey]

Just to clarify Sapo, my understanding is you're not looking for design help on this just to kick around some of the potential issues that could arise so that you'll have the right information on hand when you talk to your plumber is that correct? Things such as whether you need check valves on the discharge and that type of thing?

[Sapo]

Just to clarify Sapo, my understanding is you're not looking for design help on this just to kick around some of the potential issues that could arise so that you'll have the right information on hand when you talk to your plumber is that correct? Things such as whether you need check valves on the discharge and that type of thing?
Right, I just don't want to look completely clueless when I talk to the hired professionals. Also, since I am just the idea guy on this project and have very little say in hiring these professionals, I want some way to gauge if they know what they are talking about. It might not be my money, but I do want to see the project successfully completed.

[HongKongFooey]

Will you have an engineer involved? Undoubtedly some plumbers could handle a project like that but there an awful lot who can't (but would try anyway).

[Sapo]

Will you have an engineer involved? Undoubtedly some plumbers could handle a project like that but there an awful lot who can't (but would try anyway).
I have a pool guy who has done big[ish] projects with fountains and waterfalls and all that. I am now getting the feeling that this might be more than he can handle as he is not an engineer. Someone suggested I contact my nearest university and get an engineering student, which I though was a brilliant idea for getting some brainpower on the cheap.

The current incarnation of the idea is actually rather simple as all the jets will be static, most of them are very low (1-2') and the whole project is not really that big and there is plenty of room for pumps and all that right next to where we want the installation. Also, it is a new build so we are not constrained by existing structures and all that.

ETA: The tricky part is that there will be a lot of jets and I want them all to look uniform. Here the fact that they are small jets might actually be a difficulty since they are then easier to compare.

I want to believe that our guy is capable of handling it, but I would like some way to gauge his competence over his confidence. He has so far said that this is all piece of cake. My problem is that his answer to everything is "yes, that can be done" with no further explanations of exactly what and how he is planning to do. Granted, that's what he is paid to do and as my auto mechanic says all the time, I am paying enough for a car fix, not for an education. Still, I wish I knew enough about this to know if what he wants to do is ok before the whole thing explodes.

[billfish678]

A thought regarding the two pump approach. Lets say its something really simple like a ring of jets spewing upwards along the edge of birdbath.

If you didnt think about it at all, you'd make one half the circumference one run on one pump and the other half on the other pump. And that would be a bad idea.

If the pump pressures were not close to equal, jets on half the side would be higher than the other, looking like crap. And if one pump is temporarily down, again it looks like crap.

Now, if you get a bit smarter, you run two rings around the perimeter, with each ring running off a different pump. If pressures arent equal, its still good. And if one pump is down it still looks decent.

Is this kinda distribution something that can be done for this project?

[Sapo]

Is this kinda distribution something that can be done for this project?

Theoretically, I guess yes. It would mean a royal nightmare of multiple piping nets underneath, though. Plus, the walls tolerate only a maximum of space between jets before they stop being read as a continuous wall. And with one pump down among more than two, then the missing jets start reading as gaps on the wall.

This is why I would like multiple pumps running in parallel. If one is down, all jets go down a little bit but the effect holds. Also why I would like say 4 pumps where 3 are needed so you always have a spare ready to kick in when one goes down.

[HongKongFooey]

How will you control the pumps? Is someone going to manually turn on the fourth pump or will it be done automatically?

[Sapo]

How will you control the pumps? Is someone going to manually turn on the fourth pump or will it be done automatically?
I was thinking manually. This is either a planned event (for maintenance) or a rare event (for failures) and this is not a life-critical installation. 10 minutes of the thing looking wonky while someone gets up and flips a breaker switch is acceptable.

[ZenBeam]

ETA: The tricky part is that there will be a lot of jets and I want them all to look uniform. Here the fact that they are small jets might actually be a difficulty since they are then easier to compare.Making them all uniform sounds to me like the hardest part of the project, and the part where you'd need an engineer and not just a plumber who's done fountains a few times before. Do you have an idea what your requirement is for uniformity? Does your plumber, or whoever you hire, have a way of showing that he will satisfy that requirement prior to the work being performed?

[astro]

White a university student many have the "brains" this really, really sounds like a project where you need someone with hands on practical experience in designing fountain systems.

[Bill Door]

If they're all identical nozzles the flow will be the same through each as long as the pressure on the upstream side is the same. The pressure on the upstream side will be affected by friction, and since every nozzle will be a different distance from the original pump, the pressure can't be the same.

Let me illustrate with a simplified example. We have a straight piece of pipe. Ten feet from the pump we have a nozzle, and 10 feet from that nozzle we have a second nozzle, and then the pipe continues on.

We send 20 gallons per minute with a pressure of 34' of head from the pump through a 1" pipe. The pressure loss at that flow through new schedule 40 pipe is 25.2' per 100', so we lose 2.52' of head getting to the first nozzle. At the first nozzle the pressure is therefore 31.7'. We get a certain flow through the nozzle based on P1 and P2. Say we engineer the nozzle for a flow of 5 gpm at an inlet pressure of 31.7'.

Now the flow through the pipe is only 15 gpm. The friction loss for that flow through the same pipe is 14.6' per 100', so we lose 1.46' of head getting to the second nozzle. We started at 31.7, so the pressure at the second nozzle is only 30.2 feet. The flow will be slightly different.

A third nozzle further along will be even less, and so on. Feeding the system with two pumps on either side would help, and the larger the pipe diameter the less the frictional effects will be, but if you want all the spray heights to be the same you'll need to engineer every nozzle separately, and then it will only work within a specific flow range, so you'll have to add either flow or pressure controllers to keep it.

A 1/4" nozzle at 35' inlet pressure gives a flow of 7.24 gpm but by the time we get to the fourth or fifth nozzle at 23.1' inlet pressure the flow would be 5.91 gpm. You could try a 5/16" nozzle, and the next one might be 11/32" but they'll all have to be as individual as a snowflake.

In an industrial setting we'd use individually engineered orifices and control valves to maintain flow. With a pump that small I might try a variable frequency drive and a pressure control loop, I don't know. The last pump I bought was a 250 HP pump delivering 4000 gpm at 160' total delivered head, so I don't have a lot of experience with small systems.

[Sapo]

Making them all uniform sounds to me like the hardest part of the project, and the part where you'd need an engineer and not just a plumber who's done fountains a few times before. Do you have an idea what your requirement is for uniformity? Does your plumber, or whoever you hire, have a way of showing that he will satisfy that requirement prior to the work being performed?

Since this is purely for show, the requirement is that it looks uniform to the eye. As four our plumber, he was "chosen" because he is the guy that does this kind of work for the city. The city officials seem pretty relaxed about the whole deal which is what's panicking me.

If they're all identical ...

Thanks for the detailed example. It does help a lot and it has confirmed some of my hunches (I may suck at math but I am very proud of my physics intuition).

Long story short. We need an engineer. I just came back from making an appointment with the big guys. This will be my own point for this meeting. No engineer, no pretty fountain.