Skimmer Suction Air Leak - Hypothetical?

[Texas Splash]

The test is if the water level doesnt change in the basket and the pump runs low speed for 24 hours constantly then it is unlikely pulling air. Turning the pump off may result in an unexpected outcome as there could be small amounts of air come back into the pump from the discharge side and that may cause unexpected outcomes.

To confirm, once the water level drops in the pot, it basically stays there 24/7. Also, when I turn the pump off there is no change in water level or shifting of water in the system as would be expected with a typical equipment pad air leak.

 

[mas985]

What seems to happen is the pump doesnt pull a full basket. I am making an assumption the centrifugal action creates a vacuum of say -100 (whatever units you like, just for explanation) when immersed in 3 inches of water. The pipe restrictions and losses head are perhaps -20 (again whatever units). The difference shows up as 'air' which I am assuming is (compressed) in the pipe normally. When the pump runs at full speed the pressures change perhaps the vacuum from the pump becomes -150 and this is more than enough to pull X volume to, compress the air and fill the pump as this vacuum is enough to pull a full pipe.

Any trapped air in the suction line would not remain there during the priming phase of the pump. The flow rate at high speed is more than sufficient (well above the critical flow rate) to flush out any air. That is exactly how a pump primes. So that may happen the first time after the pool is built or after pool opening but after that, where does the air come from? Once a suction line has all the air purged, it will not fill with air again unless there is a leak.

Also, the suction line is well below the water line and when the pump is running is below atmospheric pressure so not under compression.

The test is if the water level doesnt change in the basket and the pump runs low speed for 24 hours constantly then it is unlikely pulling air. Turning the pump off may result in an unexpected outcome as there could be small amounts of air come back into the pump from the discharge side and that may cause unexpected outcomes.

Any air coming from the discharge side of the pump would fill the pump basket and you see that after the pump shuts off. But even if that happened, during the priming mode, that would be purged out of the plumbing system

Hope that clarifies what I mean.

Sorry but no. None of it makes any sense.

Oh and by the way different pumps even with the same published pump curve may act differently as the domestic pump curves do not include the vacuum side, and vacuum can change from one pump to another depending on internal set ups.

The head curve most certainly does include the vacuum side of the pump.

Total head = suction head + return head

How to Quickly Calculate a Centrifugal Pump’s Total Dynamic Head

With centrifugal pumps, these calculations typically deal with calculating a pump’s total head to select the optimally sized pump required.

www.pumpsandsystems.com

Total Head for the Pump & System

Total head is also commonly referred to as total dynamic head (TDH); however, the Hydraulic Institute (HI) uses the term total head and it will be used throughout this article.

www.pumpsandsystems.com

 

[DavidLast]

The head curve most certainly does include the vacuum side of the pump.

Total head = suction head + return head

How to Quickly Calculate a Centrifugal Pump’s Total Dynamic Head

With centrifugal pumps, these calculations typically deal with calculating a pump’s total head to select the optimally sized pump required.

www.pumpsandsystems.com

Total Head for the Pump & System

Total head is also commonly referred to as total dynamic head (TDH); however, the Hydraulic Institute (HI) uses the term total head and it will be used throughout this article.

www.pumpsandsystems.com

Sorry this is not quite correct in the way I am meaning.

In swimming pool pumps the following could occur:
Vacuum - 20
Head 80

And

Vacuum - 30
Head 70

Both of these will be seen and recorded as 100 as you drop the negative and add the two numbers together.

In practical terms, on most pools, these two pumps would act nearly identically. We are dealing with buckets of water, not ounces, so the difference between the two would be marginal most of the time and not important. Even commercially it is acceptable to work off this type of pump curve for pool pumps.

Yet one will 'suck' better than the other.

Here is a simplified industrial pump curve.
Pump Curve

This is what I am referring to. I have asked the pump guys in the office for a better explanation and will see if they come up with something. As I said I have seen this a lot. It is not overly unusual.

 

[DavidLast]

To confirm, once the water level drops in the pot, it basically stays there 24/7. Also, when I turn the pump off there is no change in water level or shifting of water in the system as would be expected with a typical equipment pad air leak.

And when it starts again the water level stays the same?

Final test. Stop pump. Alter to full speed. Start does the basket fill with water? All of this is saying the speed is the issue not an air leak.

 

[mas985]

Sorry this is not quite correct in the way I am meaning.

In swimming pool pumps the following could occur:
Vacuum - 20
Head 80

And

Vacuum - 30
Head 70

Both of these will be seen and recorded as 100 as you drop the negative and add the two numbers together.

Assuming you are referencing units of head, I agree. That represents total head which is what is referenced on a pump's head curve.

In practical terms, on most pools, these two pumps would act nearly identically. We are dealing with buckets of water, not ounces, so the difference between the two would be marginal most of the time and not important. Even commercially it is acceptable to work off this type of pump curve for pool pumps.

Yet one will 'suck' better than the other.

The flow rates for both of these are identical. So one does not technical "suck" more than another. The pressure in the pump basket is different but that is all.

Technically, pump's do not suck at all. They create low pressure at the impeller and the atmospheric pressure is what pushes the water toward the pump.

PUMPS DON'T SUCK! And Other Centrifugal Pump Basics

This post gets back to the basics of the centrifugal pump. Learn how a centrifugal pump functions, and the best and worst ways to apply them.

blog.craneengineering.net

Here is a simplified industrial pump curve.
Pump Curve

This is what I am referring to. I have asked the pump guys in the office for a better explanation and will see if they come up with something. As I said I have seen this a lot. It is not overly unusual.

Yes, I know what a head curve is. That represents TOTAL HEAD which is the combination of suction and return head (static + dynamic) as was shown earlier.

 

[JamesW]

The main requirement for the suction is that you meet the manufacturer's NPSHr or the pump can begin to underperform.

The other thing to know is where the maximum efficiency is on the pump curve.

Ideally, the operating point will be in the range where the wire to water efficiency is the highest, which is more important than the WEF number.

Sell Sheet | IntelliFlo3 VSF Pool Pump

online.flippingbook.com

IntelliFlo3® Variable Speed & Flow Pool Pump

The award-winning, energy-efficient IntelliFlo3 VSF Pool Pump has a quiet design with sensorless flow control, app connectivity, and automation capability.

www.pentair.com

https://www.pentair.com/content/dam/extranet/nam/pentair-pool/residential/pumps/intelliflo3-vsf/manuals/install-guide/intelliflo3-pro3-vsf-install-guide.pdf

https://www.pentair.com/content/dam/extranet/nam/pentair-pool/residential/pumps/intelliflo3-vsf/manuals/user's-guide/intelliflo3-pro3-vsf-users%20guide.pdf

 

[JamesW]

To get above about 53% efficiency, you would want the flow to be between about 65 and 134 gpm.

The maximum efficiency happens at about 100 GPM and about 58% efficiency.

A System curve of about 0.00715F² would get abouit 58% efficiency at 100 gpm at full speed.

y = 0.00715(x)^2, x from 0 to 120 - Wolfram|Alpha

Wolfram|Alpha brings expert-level knowledge and capabilities to the broadest possible range of people—spanning all professions and education levels.

www.wolframalpha.com

 

[mas985]

Where the pump's head curve crosses the NPSHr curve is where cavitation begins and the reason why the performance drops. This is often called the run out point where the pump runs out of suction head.

 

[Texas Splash]

Appreciate everyone's input. Most of it goes over my head. But today was a gorgeous day in San Antonio, so back to pressure testing one more time. Results as follows:

When inserting the blow-thru plug directly into the 2" pipe below the skimmer, I am able to hold a psi of 12-14 pretty solid. I stopped testing after about 30 min.

However, like before, if I insert an extension that screws into the threaded portion of the skimmer hole to get a better grip on the blow-thru plug, with several wraps of Teflon tape, it does not hold well at all.

One could assume it's just leaking at the threads under pressure, but I also worry there is a defect in that threaded area that does not show up with a static dye test. That would also explain why I never saw a leak when I had the whole area dug out last August.

Once I put everything back together, there were the tiny bubbles again flowing in through the inlet port into the pump pot. Wife said she could see them flowing through the new check valve as well. Not sure if any of you recall from my other threads and pics, but that lower half of the skimmer is tilted. That, along with air in the pump, is what started my relentless journey.

I'm considering trying another Frankenplug to seal that area into the skimmer port to see if that's the issue.

 

[JamesW]

This example uses a 2.5" PVC pipe at 100 feet long as the model for the suction.

The model is about H = 0.0006F^2 for the suction line.

NPSHa = Net Positive Suction Head Available.

NPSHr = Net Positive Suction Head Required.

The point where the NPSHa curve intersects the NPSHr curve is the maximum flow possible without cavitation.

y = 34 - (0.00062284(X)^2), x from 0 to 180 - Wolfram|Alpha

Wolfram|Alpha brings expert-level knowledge and capabilities to the broadest possible range of people—spanning all professions and education levels.

www.wolframalpha.com

Hazen-Williams Water Flow Formula: Head Loss, Data, Charts & Calculator

Friction head loss (<i>ft_H2O per 100 ft pipe</i>) in water pipes can be estimated with the empirical Hazen-Williams equation.

www.engineeringtoolbox.com

 

[mas985]

Appreciate everyone's input. Most of it goes over my head. But today was a gorgeous day in San Antonio, so back to pressure testing one more time. Results as follows:

When inserting the blow-thru plug directly into the 2" pipe below the skimmer, I am able to hold a psi of 12-14 pretty solid. I stopped testing after about 30 min.

However, like before, if I insert an extension that screws into the threaded portion of the skimmer hole to get a better grip on the blow-thru plug, with several wraps of Teflon tape, it does not hold well at all.

One could assume it's just leaking at the threads under pressure, but I also worry there is a defect in that threaded area that does not show up with a static dye test. That would also explain why I never saw a leak when I had the whole area dug out last August.

Once I put everything back together, there were the tiny bubbles again flowing in through the inlet port into the pump pot. Wife said she could see them flowing through the new check valve as well. Not sure if any of you recall from my other threads and pics, but that lower half of the skimmer is tilted. That, along with air in the pump, is what started my relentless journey.

I'm considering trying another Frankenplug to seal that area into the skimmer port to see if that's the issue.

There could possibly be something in the plumbing that does not leak under pressure but does under suction.

 

[JamesW]

The NPSHa should always be above the NPSHr.

NPSHa = 34 - suction side head loss in feet.

So, you need to know the maximum flow that you want so that the suction side head loss System component curve can be designed to always keep the NPSHa above the NPSHr.

For example, if the maximum flow is going to be 170 gpm, the NPSHr is about 16 feet.

So, the maximum suction side head loss in feet has to be limited to about 18 feet at 170 GPM.

18 feet at 170 GPM is a curve of about 0.00062284F^2.

For a 100 foot PVC pipe at 2.5" inside diameter, the head loss would be just under 18 feet.

If you follow the 6 feet per second water velocity rule for suction plumbing, the head loss should never be an issue unless the pump was elevated or the run of pipe was very long.

 

[JamesW]

There could possibly be something in the plumbing that does not leak under pressure but does under suction.

Maybe you can do a vacuum test rather than a pressure test.

You can buy or rent a vacuum pump and the process is basically the same as a pressure test.

Put the system under vacuum and watch the gauge to see if it holds steady or drops.

 

[mas985]

The problem is that the actual pressure along the suction line varies from very low near the impeller to zero at the skimmer and not a constant. Also, when the pump is on high speed, the pressure along the pipe run is lower than at low speed so one should think that there would be more of a leak but it does not appear to be the case. So the leak seems to appear only in a narrow band of pressures. You may need to test at many different suction pressures to get confirmation.

 

[DavidLast]

Guys this is not a leak.

The problem is summarised according to the pump guys in our workshop as:

It is complicated and I am unsure how to explain it.
I will think about a simple explanation .

It is a combination of the pump being off its BEP (best efficiency point) so its BEP would be between 70% and 120%
When we spin below 70% we fail to meet the specific speed required for the impeller design and the flow rate becomes too low to meet the suction needs.
IE it is not meeting the NPSHr requirements of the pump.
Net positive suction head rotation

Hence I don’t like VSD driven pool pumps , buy a smaller pump.

VSD is good for trimming , soft start ramp down water hammer BUT running a pump designed to operate at 3000 rpm below 2100 rpm is bad .

To make matters more complicated we need to look at the Affinity laws ( sort of what you were saying right track )

This is from a guy who works with pumps all the time and has for over 30 years.

After discussion with him our conclusion is the air is simply in the system. It is not being 'sucked' in, it is more likely compressed somewhere in the system either as I suspect somewhere in the suction line or perhaps on the discharge side for instance at the top of the filter. Neither of us believe the pump is fully clearing all the air from any pool system.

If anyone is interested enough and has the equipment, we think you could connect a submersible pump to the skimmer box and blow all the air from the system if you could open a bleed on the filter and then see if at the lowest speed the problem is still there.

As I mentioned different impellors have different charateristics so one will lift better than another. In fact when I was in product development for pool pumps the design the engineers had was in theory perfect, however could not lift at all, we drilled a hole in the diffuser and it enabled that pump to lift.

Classic example for me was we were making vacuum trolleys. These are used to vacuum 50 metre pools. They were fitted with a Sta-Rite Max-E-Pro of abut 2.0 hp from memory. I got approached by a local manufacturer with an alternative pump, on paper same curve, same horsepower, same amp draw. We set up a vacuum trolley, and attempted to use it for 1 day. Called up and ordered a new Sta-Rite as the other pump would not lift. Gauges showed it was struggling to pull 2.0 m vacuum whilst at the time the Sta-Rite was testing about 3.0 m from memory.

 

[JamesW]

Guys this is not a leak.

The problem is summarised according to the pump guys in our workshop as:

Pretty much everything you said makes zero sense.

It is a combination of the pump being off its BEP (best efficiency point) so its BEP would be between 70% and 120%

This makes no sense.

How are you going to get 120%?

120% of what?

When we spin below 70% we fail to meet the specific speed required for the impeller design and the flow rate becomes too low to meet the suction needs.

This makes zero sense.

IE it is not meeting the NPSHr requirements of the pump.
Net positive suction head rotation

NPSHr is Net Positive Suction Head Required.

You have no basis to conclude that the NPSHr is not being met.

Hence I don’t like VSD driven pool pumps , buy a smaller pump.

This makes zero sense.

VSD is good for trimming , soft start ramp down water hammer BUT running a pump designed to operate at 3000 rpm below 2100 rpm is bad .

This makes no sense.

To make matters more complicated we need to look at the Affinity laws ( sort of what you were saying right track )

What do the affinity laws have to do with anything?

In my opinion, you have no idea what you are talking about.

 

[DavidLast]

With VSD the speed can be increased to about 120% of original motor speed. Sorry if this isnt clear.

Affinity laws are what is used to design a pump.

Affinity Laws for Pumps: Principles, Formulas & Calculator

Turbo machines affinity laws can be used to calculate volume capacity, head or power consumption in centrifugal pumps when changing speed or wheel diameters.

www.engineeringtoolbox.com

 

[mas985]

Guys this is not a leak.

The problem is summarised according to the pump guys in our workshop as:

It is complicated and I am unsure how to explain it.
I will think about a simple explanation .

It is a combination of the pump being off its BEP (best efficiency point) so its BEP would be between 70% and 120%
When we spin below 70% we fail to meet the specific speed required for the impeller design and the flow rate becomes too low to meet the suction needs.
IE it is not meeting the NPSHr requirements of the pump.
Net positive suction head rotation

Hence I don’t like VSD driven pool pumps , buy a smaller pump.
VSD is good for trimming , soft start ramp down water hammer BUT running a pump designed to operate at 3000 rpm below 2100 rpm is bad .

To make matters more complicated we need to look at the Affinity laws ( sort of what you were saying right track )

First, the pump is a two speed, not a VSD pump so I don't know where he got that info but it is wrong.

Second, I did a complete head calculation for the plumbing and came up with an operating point shown here:

Pool Pump Tools v027 - Texas Splash

ReadMe Version:,v027 Disclaimer,These spreadsheets are provided as is without any guarantees or warranty. Every effort has been made to produce reasonable estimates with these modeling tools. However, unknown measurement errors in the CEC/APSP/EnergyStar data may significantly affect the results...

docs.google.com

The operating point is only 20% off of BEP and that is fine (even your "expert" says it's fine) and fairly typical of pool pumps so yes it actually is meeting NPSHr criterial. I think your so called "expert" believes the pump is in cavitation. It is not. That is NOT what is causing the air in the pump basket. If the pump cavitates, you would first hear it but more importantly, when a pump cavitates, it occurs in the impeller inlet and you would never see it in the pump basket. The vapor bubbles would collapses before reaching either the outer edge of the impeller or the pump basket because both are above the vapor pressure of water.

This is from a guy who works with pumps all the time and has for over 30 years.

But he is obviously working with incorrect information so his conclusions well be incorrect.

After discussion with him our conclusion is the air is simply in the system. It is not being 'sucked' in, it is more likely compressed somewhere in the system either as I suspect somewhere in the suction line or perhaps on the discharge side for instance at the top of the filter. Neither of us believe the pump is fully clearing all the air from any pool system.

Are you suggesting the air moves from the filter backwards against the flow of water into the pump basket? That would violate many laws of physics. Air travels with the flow of water, not against it.

Plus, the OP has already stated that this does not happen so again, you are ignoring evidence.

If anyone is interested enough and has the equipment, we think you could connect a submersible pump to the skimmer box and blow all the air from the system if you could open a bleed on the filter and then see if at the lowest speed the problem is still there.

Are you suggesting that a submersible pump is more powerful than a pool pump? If so, you are mistaken. The OPs pump has a much higher head curve than any consumer products.

As I mentioned different impellors have different charateristics so one will lift better than another. In fact when I was in product development for pool pumps the design the engineers had was in theory perfect, however could not lift at all, we drilled a hole in the diffuser and it enabled that pump to lift.

Absolutely true. The later pump you are describing is a not a self-priming pump so you needed to add a recirculation port. But both pumps the OP is using ARE self-priming. Again, ignoring evidence already stated.

Classic example for me was we were making vacuum trolleys. These are used to vacuum 50 metre pools. They were fitted with a Sta-Rite Max-E-Pro of abut 2.0 hp from memory. I got approached by a local manufacturer with an alternative pump, on paper same curve, same horsepower, same amp draw. We set up a vacuum trolley, and attempted to use it for 1 day. Called up and ordered a new Sta-Rite as the other pump would not lift. Gauges showed it was struggling to pull 2.0 m vacuum whilst at the time the Sta-Rite was testing about 3.0 m from memory.

Irrelevant to this situation as the Matrix IS a self-priming pump and the required lift in less than a foot.

With VSD the speed can be increased to about 120% of original motor speed. Sorry if this isnt clear.

No it can't. Look at any VS pump specificationon the market today and the maximum speed is 3450 RPM. The controller does not allow speeds above that.

 

[Texas Splash]

Below is a nice cutaway pic. Since my pressure testing seems to hold well when plugged below the threads with the plug in the pipe, I'm going to try a couple more tests at deck level before contemplating opening up the deck to get down/around the skimmer from below. My area of concern is the threaded area before the pipe into the ground. That's about all I can think of as suspect at this point.

 

[mas985]

TS, can your confirm my understanding. I just want to make sure I am not misunderstanding something.

1) On high speed, you checked the filter for air and there was none. But I wasn't clear, is that when you saw the small bubbles in the pump basket and in the check valve?

2) On low speed, the pump basket fills with air to a certain point and then stops. Do you also see small bubbles in the check valve? How long has it remained at this level with the pump running? Does the level change when the pump is off?

3) Do you run your pump 24/7 or on a time clock? If the later, does a controller prime first or just start on low speed?