# Thread: Estimating GPM from pump speed

1. ## Estimating GPM from pump speed

Hello all, if there is already a post answering this question please redirect me; my searches brought back too many results...

I am trying to figure out a rough pump speed to get approx 45 gpm GPM with a vs3050

Where do I start?

2. ## Re: Estimating GPM from pump speed

the spreadsheet referenced here helped me.
http://www.troublefreepool.com/rpm-t...rs-t19602.html

3. ## Re: Estimating GPM from pump speed

Perfect thanks!

4. ## Re: Estimating GPM from pump speed

So according to the excel file (which is awesome!):

I am running my pump at 900 RPM to run my cleaner (a Hayward Navigator) and the controller is showing that I am using 106 watts so this calculates to 7.3 gpm. At this speed the navigator is actually moving pretty quickly so I'm finding it hard to believe it's that low a flow. With my old 1HP pump, to get the cleaner moving this quickly I would have to set the suction valve to about 75% open for the cleaner port (and 25% to the skimmer) - that has to have been more that 7.3 gpm right?

Likewise with filtration set to 600 rpm, I'm pulling 93 watts which calculates to 25 gpm and will therefore take a little over 10.5 hours to turn over 16k gallons, and that's about 1kWH per day. Is that good? 103 watts would be 50 gpm...so 10 watts more but half the time so 515 kWh. Is there a larger error factor at the lower and upper ends of the speed spectrum?

This is kind of an interesting problem to setup a desired gpm vs. efficiency

5. ## Re: Estimating GPM from pump speed

Yes, the error in the model tends to increase with decreasing watts and RPM. The model basically fits the published Pentair power/RPM curves to an equation which is then solved for other values and the worst error is about 15% at 20 GPM but you usually can't get much better with a flow meter either as those are rated at full scale. Also, a pressure side cleaner runs at high pressure and low GPM so it is not surprising that the GPM is low.

6. ## Re: Estimating GPM from pump speed

It's a suction side cleaner so is that still normal?

I thought about getting a flow meter but I haven't seen one that reads below 20 gpm which it seems like we could easily be in this range at low pump speeds

7. ## Re: Estimating GPM from pump speed

So I spent some time last night running the pump at different speeds and recording the watts, then running them through the calculator and here's what I found:

RPM Watts GPM
400 132 299
500 118 135
600 94 28
700 89 4
800 93 1
900 112 14
1000 119 9.5
1100 134 11
1200 154 13.5
1300 186 20
1400 222 25.5
1500 271 33.5
1600 318 37.5
1700 370 41
1800 424 43
1900 496 48
2000 584 54.5
2100 652 55
2200 735 57
2300 832 60
2400 826 46
2500 1035 63.5
2600 1136 63.5
2700 1259 66.5
2800 1400 68.5
2900 1543 70.5
3000 1695 72.5

Clearly some unusable data for the lower speeds, but it does start to follow an even progression from about 1,000 to 2,800 RPMs. Anything above 2,800 flattens out so that seems to be the point at which my 1.5" piping maxes out.

Here are some turn over times for various flow rates:
Gals 16000 16000 16000 16000 16000 16000 16000
Hours 4 6 7 8 9 10 12
GPM 66.67 44.44 38.10 33.33 29.63 26.67 22.22

- The problem I'm running into is that to get a turn over in 12 hours I need to run at 1,300-1,400 RPMs which is about 185-225 Watts....am I just limited by my piping such that I will never be able to get approx 100 Watt power usage with a high enough flow? How are others on this board running sub 1K rpms and seeing sub 100 watts? It would appear that longer run times with slower flow aren't really going to save me any (or at least not much) money In fact, I'd save 600 kWh by running for 6 hours at 2X the speed - this just isn't adding up

I took these measurements running only the skimmer and returning to three returns (how it wuld be setup for filtration). All piping on the pad and to the solar is 2" but the suction cleaner port, skimmer and returns are all 1.5" - am I choking the system with ony a single suction line (the skimmer) open? Should I scrap trying to separate cleaning from filtration and always have both open to some degree?

8. ## Re: Estimating GPM from pump speed

How are you determining the Watts? They should be increasing with increasing RPM. So something is not right below 1000 RPM.

However, your most efficient setting will probably be around 1000 RPM. Although for the data you posted seems to be aroung 1500 RPM which is about 7.5 gallons/watt-hr.

Code:
I took thsemeasurements running only the skimmer and returnng to three returns (how it wuld be setup for filtration). All piping on the pad and to the solar is 2" but the suction cleaner port, skimmer and returns are all 1.5" - am I choking the system with ony a single suction line (the skimmer) open? Should I scrap trying to separate cleaning from filtration and always have both open to some degree?
Yes, this will greatly reduce your efficiency. For normal operation, you will want all lines open except the cleaner if it is not attached but if it is port to the pool, you might just leave that open as well unless someone swimming.

9. ## Re: Estimating GPM from pump speed

I'm reading the watts off of the suntouch controller - pumps menu, status sub menu shows rpm and watts consumed. Yeha the data below 1K is just wacky...

The cleaner is on it's own suction port, besides this port I only have the skimmer. I currently have the cleaner automated for 2-3 hours per day at a correct speed for it, and then run filtration for the rest of the day (still figuring out the run times), with solar being able to kick on when available. The reason I want the separation is that when the solar kicks in the cleaner is either running too fast or struggling to move depending how I tweak the suction valve. I was expecting that with a low flow rate that only havign the single suction line would not be a problem...

10. ## Re: Estimating GPM from pump speed

Do you happen to have any check valves in your plumbing? That could explain the strange results at lower RPM.

Also, looking at the numbers, I think that the wattage is way off at lower RPM. The RPM to GPM relationship remains linear down to about 1500 RPM and below that this start to deviate. I have seen this with other VS systems that have check valves.

I forgot to mention that with a measurement at the higher speeds, you can scale the flow rate by the RPM level and get a reasonable estimate to flow rates at lower speed. This works well for plumbing systems without spring loaded check valves and may be more accurate than using the wattage which seems to be incorrect. Your flow rate should follow:

GPM = 72.5 * RPM / 3000

11. ## Re: Estimating GPM from pump speed

Yes there is a check valve! A black valve-like body with a clear round cover and a spring-backed rubber gasket inside of it (I think it's a pentair check valve) between the pump and the filter. What's the theory on how this affects the #s? Is the pump really drawing that much power or not? Should I remove the valve?

Thanks for the equation; I'll go plug that in and plot it out.

Also: It looks to me that the curve is pretty linear from 1,100-2,000 ... why are you picking 1,500 - to have a bias from the higher-end bias of the curve? - EDIT: Strike that, took a look at it in my graph and I agree it's 1,500 up.

12. ## Re: Estimating GPM from pump speed

I noticed that when my pump went to low speed, the check valve that I have for solar, similar to yours, closed about 3/4 of the way. The flow rate was not sufficient to keep the valve 100% open. Reducing the orifice that water can flow through at lower flow rates would suggest that the check valve represents more equivalent length of pipe at low flow rates than at high flow rates. Normally, fittings and pipe itself does not change in equivalent length no matter what the flow rate so the head loss is somewhat predictable over any flow rate and the head loss would follow a GPM^2 curve. Also, GPM and RPM would remain linear over a larger range of values.

It would be interesting to see if in your case, the check valve starts to close below 1500 RPM. This might explain some of the decrepencies.

13. ## Re: Estimating GPM from pump speed

First off: Mark - thanks so much for your help! You are clearly a Jedi Master in this subject

Back to the issue: The check valve is definitely partially closed at lower speeds i.e. sub 1,000 RPM. If memory serves it was up a ways before it really opened so 1,500 may be pretty close. I won't be able to play with them again for a few days but I'll definitely check that when I get back.

So what does this mean for the power consumption? Is it what the controller is showing then? Because those power number lead to funny GPM numbers being calculated...

14. ## Re: Estimating GPM from pump speed

I still suspect that that controller is not displaying the correct power at some points. Part of the problem is precision. At 400 RPM, the range of Watts and GPM is pretty small. For 690 RPM, Pentair only publishes a range of 0-30 GPM but the watts range only goes from 87-103 watts. So a small error in watts can result in a very large error in GPM.

Your test results look reasonable down to about 800 RPM with 1500-800 RPM accounted for by the check valve non-linearities. Below 800 though it could be the check valve is completely closed off and not allowings any or not very much water through. Did you check the returns to see if there is water comming out for the lower RPMs?

BTW, I removed the spring from my check valve and the flapper now stays fully open on low speed. This is an option to consider but will only work if the valve is installed horizontally.

15. ## Re: Estimating GPM from pump speed

I'll have to check - I know the cleaner (suction) was moving slowly at 700 so I assume water was making it through the whole system. The pump was making a lot of noise at 400 and there was a lot of water thrashing in the check valve so I'm pretty sure nothing was flowing that low. I don't think I'd want to run that low anyway. I'm looking for a balance between efficiency and a reasonable turnover and I'm ok with a level of error I would just like to get closer than a ballpark measurement. I realize I'm over thinking this but its in my nature

16. ## Re: Estimating GPM from pump speed

Originally Posted by mas985
BTW, I removed the spring from my check valve and the flapper now stays fully open on low speed. This is an option to consider but will only work if the valve is installed horizontally.
Update: I have the Pentair 2" check valve. I bought a replacement check valve kit with the intent to break off the spring but I would still have the original with the spring if I wanted to go back. Turns out there was no need to spend \$45. The cover and valve door are one assembly - the door, a swing arm that the door is connected to and the spring are held to the cover by a pin that makes a hinge with the valve door and cover. If you look at the cross pin that the swing arm swings on one end is beveled - I tapped on that side with a punch and the pin slid out - the other end does not taper and is actually roughed up to bite into the hole. The pin holds the spring and arm/door together so as you slide it out the assembly comes apart. Super easy to reassemble without the spring - just push the pin back in.

I did find that the door did not swing smoothly though. So I sanded the parts of the faces of the hinge mechanism that were on the door that rubbed on mating surfaces. The plastic is pretty soft so 150 grit sand paper made quick work of it. After a few tries I got it to where hte door swings back and forth freely wihtout binding up at all.

I swapped out the covers and fired the pump on high - the door swung up to the top as expected. It swung back down and sealed when the pump was off too - my initial fear was that it would not swing back down; I'm pretty sure that sanding down the friction points helped with this. Then I tried low speed - the door opened almost all of the way at 1K RPM, and closed just fine when the pump was turned off.

I haven't had a chance to re-measure power usage vs. speeds yet. I'll do that soon and update this thread.

Interestin design detail: the valve door is actually dual hinged. There's the main hinge described above, and there is a second hinge point where the arm and door attach to each other. So the door actually has a little bit of swing movement at this point. I think this may have been Pentaire's solution for low flow rates that can't push the valve fully open due to the spring weight. The door pivots on this point allowing for a slightly larger opening than if the door was fixed solidly to the arm.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•