OT- Pump Selection for fast flow

This is not PLC related at all but I recall there were a few folks well versed in pumping water so I'm hoping I can tap their brains.

I am trying to push a sample from a point some 800 feet upstream from an intake, and I need to get it to ravel that 800 feet as fast as possible.
This is where the hydraulics is fighting back. Without going to huge 50+ HP pump all I can get is about 2 1/2 minutes. If its the best I can do then its acceptable as the intake canal takes about 15 minutes to traverse the same distance. What seems odd to me that all I will get from this is less than walking speed...

15 foot vertical head, 750 foot 2 inch PCV pipe with 4 long sweep 90's.
If I did this right...

I figure this run as 22' friction and 15' static giving 37' of head.
A xylem KS2610 (for example) can pump that at 44 GPM giving me about 3 minutes of travel time.

I should add I will need to pre strain this pump as its raw water will contain a lot of algae. A chopper pump is not an option as I need the turbidity sample to be representative.

I'm not sure what else I need to know to optimize or improve this and I would appreciate any guidance you can offer.

I think your pump power is way off.

1) I don't know why you want to have such a large sample line. It seems like you are sampling surface water, and straining it, so solids plugging shouldn't be a big deal. Why not use 3/4" or 1"?

2) You can calculate head loss from this site: http://www.engineeringtoolbox.com/hazen-williams-water-d_797.html Each elbow is equal to about 30 diameters. Here is a chart: http://www.engineeringtoolbox.com/resistance-equivalent-length-d_192.html
and: http://www.engineeringtoolbox.com/hazen-williams-coefficients-d_798.html

3) Normal velocity is 3 to 5 ft/sec. You can push that higher if you are willing to take more pressure drop - delta p goes up as the velocity squared.

4) Pump power = gpm x fthead / 3960 : http://www.engineeringtoolbox.com/pumps-power-d_505.html

5) You should consider a suction switch so you don't cavitate your pump when algae clogs the suction strainer.

6) If there is an elevation difference between waster surface at pump suction and discharge include it in your calcs.

I've attached a spreadsheet that should help.

Hi Tom
Thanks for the assist

The reason I picked that large size was pressure.
There seems to be a curve where pipe diameter vs flow/velocity was optimal and a larger pipe was able to be used with a 2-3 HP pump whereas if I were to attempt to get the same flow/velocity in a smaller diameter it required much more pressure, even above the 150PSI of the pipe rating.

The spreadsheet you provided also reflects this, 228 psi if I used 3/4 pipe. 6 PSI with a 2 inch pipe, but of course slower flow.

The flow I was looking at is 45 GPM which allows for a smaller pump to drive the water at an acceptable speed. I can go faster but the pump requirement ramps up pretty fast.

For example, using that sheet...
2 inch Pipe
45gpm = 1.8 bhp(35PSI)
55 GPM = 3.14 bhp(48PSI)
100gpm = 16bhp (139PSI)

3/4" Pipe as you sent it.
10 GPM = 2.7 HP (228 PSI) !

That's too much pressure for PVC so it seems I have to use a bigger pipe.
Is there a flaw in my thinking?

Here is a great website that is easy to understand.
Well worth the time reading.

http://www.pumped101.com/

The pressure is essentially a function of velocity. Transit time is also a function of velocity. Velocity is a function of flow and diameter.

If you use a smaller pipe you can use a lower flow rate to get the same velocity, transit time, and pressure drop.For a given pressure drop lower flow means lower hp.

You can't cheat Mother Nature. To get the headloss down you need to use a lower velocity which means a longer transit time. Pick your priority!

(Neat site gas - thanks for posting)

Tom,
Right, so conversely if 150 PSI is my pipe limit, My only option is to make that pipe bigger if I want it faster.
It's starting to come together now. Thanks for the help.

Gas , I like that site, thanks.

Originally posted by TheWaterboy:

Right, so conversely if 150 PSI is my pipe limit, My only option is to make that pipe bigger if I want it faster.

Click to expand...

I don't think that is correct. Don't confuse flowrate with velocity. You can achieve the same flowrate with a small pipe at high velocity or a large pipe at low velocity. In your case you are trying to get a sample from one place to another in the shortest time possible. That infers operating at the highest material velocity possible. Friction head in a pipe is related to flow velocity not flowrate. The flow velocity that will produce 150 PSI will be roughly the same regardless of whether the pipe is 1/2" diameter or 2" diameter. And flow velocity (ft/sec) determines the time of sample transit, not the flowrate (gal/min). Also keep in mind that rated operating pressure of PVC pipe decreases as diameter increases.

So, unless you need a specific volume for your sample, I would go with a 1/2" diameter schedule 80 pipe. That stuff is rated for something like 500 PSI operating pressure. Talk about velocity.

Keith

Keith nailed it!

To get the pressure in a small pump you may need to look at positive displacement pumps instead of centrifugal pumps.

Ahh thats where I went off the rails, I see the PSI ratings of different diameters. Thanks guys, this changes my tactic. Considering my source water I am curious what positive displacement pump would handle water with stuff in it.

Go to some of the buyer's guides at www.wef.org or http://www.waterworld.com/index.html

There are sludge pumps out there that handle high solids content and high abrasion. Progressive cavity and piston pumps, for example.

How thick is the sample and how large a sample do you need.
If you are checking something like pH or some such then why not use a stainless steel tube and a smaller pump.
The velocity is really a function of cross sectional area.
3/8" has an area of .44 square inches X 12 inches = 5.29875 Cubic inches or 0.02293506 gallons per foot X 800 = 18.348048 gallons for the line.
I haven't checked the pressure and it might be too high, but you don't need to move much water.
The time to get the sample moved is the time it takes to flow the volume.
Again what are you sampling and can the sampler be moved to the testing point easier?

Not thick, and 2/3 gpm would be fine, but the sample can/will have filamentous algae that can be screened but not filtered.

I am testing for all manner of water quality including turbidity and this sample point is 1000' from the intake to give us a preview into whats coming down the channel. 1000' is where the power is and gives us about 12 minutes advance notice.

SO I'm trying to get the sample from that point to the existing measurement building as fast as I can. Some tests (copper sulfate for example), take 2 minutes to run leaving 10 minutes minus the transit time of the sample.

Ideally since I can't move the pump further down the channel I would extend the suction half a mile down the channel but I doubt that's practical.