Variable speed pump efficiency

[HeavyD328]

So there should be a minimal drop in pressure going from the pool pump to filter is that correct?

The pump doesn't sound like it's pumping gravel at full speed, and the skimmer and pump baskets are clean.

I posted a couple pics of the check valve at full speed but it's probably hard to discern in the pictures. The flap comes up more as I increase up to full speed but it's not completely horizontal but I don't have enough experience to know if this is normal.

My gut has been telling me that something seems wrong even though the PB and Pentiar rep both day everyday is working in perfect order so not sure what I should do

 

[mas985]

So there should be a minimal drop in pressure going from the pool pump to filter is that correct?

Should be. The only thing we know for sure is the head loss from the filter to the pool and the total head loss. The difference is the head loss in the suction side plumbing and the plumbing between the pump and the filter. It could be either place but given the check valve looks ok, it is most likely in the suction side plumbing or the backwash valve. One thing you might try is to cycle the back wash valve with the pump off (push plunger down and pull back up) and see if that changes anything. If not, then we can start to focus on the suction side plumbing.

Also, can you get a full speed set of readings when in spa mode?

 

[HeavyD328]

Should be. The only thing we know for sure is the head loss from the filter to the pool and the total head loss. The difference is the head loss in the suction side plumbing and the plumbing between the pump and the filter. It could be either place but given the check valve looks ok, it is most likely in the suction side plumbing or the backwash valve. One thing you might try is to cycle the back wash valve with the pump off (push plunger down and pull back up) and see if that changes anything. If not, then we can start to focus on the suction side plumbing.

Also, can you get a full speed set of readings when in spa mode?

I cycled the back wash valve. At 3450 RPM, pump pressure is still 40 PSI, filter pressure 23 PSI, 51 GPM

In SPA mode at 3450 RPM, pump pressure is 40 PSI, filter pressure 30 PSI, 50 GPM

 

[JamesW]

Assuming that the pressure and flow readings from the pump are correct, it definitely looks like there is excessive restriction somewhere, but nothing is obvious.

At this point I would open the suction three-way valve and see if there is anything stuck in the valve.

Make sure that the pump basket is installed correctly with the hole lined up with the hole in the basket.

I would remove the slide valve. It's useless because the filter cannot be backwashed effectively.

I would also take the filter apart and clean the cartridges.

It might be worthwhile to get a 1/4" npt vacuum gauge to screw into the suction drain plug to get a suction reading.

 

[HeavyD328]

I can try to work on some of these things. In your opinion are the pressure and flow readings I'm getting way out of the norms of a brand new pool design and build? Or is what I'm seeing considered within the acceptable standards of pool construction?

Just wondering if I should be pursuing this potential issue with the pool builder more aggressively as something they should diagnose and fix if needed since everything is under warranty. They've actually been tremendous throughout the whole construction phase, it's just been during the startup that the service has been lacking. The service tech assigned to the startup of our pool does not seem to have a great knowledge base regarding the function and dynamics of the equipment, hence I think that's why he keeps telling me everything is functioning perfectly. Same goes for the Pentair rep saying that the posted flow rates should not be believed (basically saying the product he sells doesn't work). The project manager during construction was very good, but when I ask questions above how everything functions, at this point he has been referring me back to our startup service tech.

 

[JamesW]

I think that the restriction is excessive.

The whole system has about the same amount of resistance as about 580 feet of 1.5" pvc pipe.

I would start with the suggestions that I posted earlier.

The pump measures the difference between the suction and discharge pressure of the pump and that's the total head loss.

The pump's software then calculates flow based on the curve at the running rpm and the total head loss.

It should be relatively accurate assuming that the pressure reading is correct.

You could add a flowmeter to independently measure the flow.

If we can get a vacuum reading and an independent flow reading, that might tell us something.

You could also screw a regular pressure gauge into the back drain plug to get a pump discharge pressure.

Overall, nothing jumps out as actionable that you could require the builder to address.

In my opinion, there's definitely something off if you need to run a 3.95 hp variable speed pump at 2,900 rpm to generate the required 40 gpm for a heater.

Here's a similar situation. I don't think that we really found anything.

Pump Efficiency and Head Loss... I’m at a Loss :(

 

[HeavyD328]

The pump measures the difference between the suction and discharge pressure of the pump and that's the total head loss.

Can you help me clarify some of the aspects on the science of pool circulation. So in my pool the 40 psi @ 3450 RPM is the total pressure or the sum of the pressure on the suction side and return side?

With a filter pressure of 24 psi, is that the pressure from the plumbing past filter until it reaches the returns in the pool?

And the difference of the two values is 16 psi, or essentially the suction side pressure. But from what I understand this is not all exclusively suction pressure because this includes pressure changes between the pump and the filter itself, which is technically part of the return side of things?

So if there were not excessive flow restriction present, what would I then expect under more ideal flow conditions? Assuming 40 psi at the pump was the correct pressure reading, would the suction pressure and return pressure be about equal, with 20 psi on suction side and 20 psi on from the filter (pressure side)?

The pump's software then calculates flow based on the curve at the running rpm and the total head loss.

How is the total head loss calculated? Is it extrapolated from the pump pressure reading (40 PSI) along with the pump speed (3450 RPM) to determine that the TDH is ~90 to obtain a flow of 50 GPM according to the performance curve of the pump itself?


In my opinion, there's definitely something off if you need to run a 3.95 hp variable speed pump at 2,900 rpm to generate the required 40 gpm for a heater.

The Pentair rep seems to be pretty convinced still that the flows displayed are incorrect and they've taken this issue to some of the engineers at Pentair. I will try to see if they can go ahead and verify whether the flow is correct or not with a flow meter as well as measure suction and return side pressures and will report back once I get more information

 

[mas985]

The pump measures the difference between the suction and discharge pressure of the pump and that's the total head loss.

The pump's software then calculates flow based on the curve at the running rpm and the total head loss.

I believe the way that it actually works is that the motor wattage is measured (and displayed) and combined with the RPM setting, both flow rate and total head loss can be calculated. The pump does not measure pressure directly nor does it need to. There is only one flow rate and total head loss possible for each RPM and wattage pair.

 

[JamesW]

The pump has a pressure sensor that measures the difference between the suction and discharge. It displays the pressure in the display in PSI.

To convert to feet of head, you multiply by 2.31. So, at 40 psi, you have 92.4 feet of head. You then use that number and the pump curve as published by Pentair to determine the GPM. See curve in the manual.

https://www.pentair.com/content/dam...peed_Installation_and_Users_Guide_English.pdf

Using the 3450 rpm curve in the manual, I get 60 GPM.

The filter pressure is the pressure in the filter before the water goes through the DE filter. So, the filter pressure includes the DE and cartridges and any equipment and plumbing past the filter.

The pump discharge pressure should be close to the filter pressure with the difference being everything between the pump and the filter including the check valve and slide valve. The check valve and slide valve should not add more than maybe 4 psi at 50 gpm.

You might be able to find published curves for the check valve, slide valve, heater, filter etc which show the head loss as a function of flow on a graph.

The suction can’t be more than 14.7 PSI because that’s a perfect vacuum. So, if the total is 40 psi, the discharge has to be at least 25.3 psi assuming that the suction is maxed out. Water turns to gas at about -14.2 psi at about 82 degrees.

Using a filter pressure of 24 psi and 4 psi between the pump and filter, we get a pump discharge pressure of 28 psi and a suction of 12 psi. At 12 psi suction, I would expect cavitation, which is the water converting to gas at the eye of the impeller and collapsing after the impeller.

If we assume 50 gpm and 12 psi on the suction side, we can roughly approximate the suction plumbing to about 180 feet of 1.5” PVC pipe. Since this does not match your actual pipe, it indicates a blockage of some sort.

At 50 gpm, the normal filter pressure will be about 15 psi. At 28 psi, you’re about 13 psi more than normal.

Overall, I would say that there is some sort of issue somewhere, but it’s not obvious where it is.

If we can get the suction reading and pump discharge reading and a flow rate, we might be able to get a better idea of what the issue is.

 

[mas985]

The pump has a pressure sensor that measures the difference between the suction and discharge. It displays the pressure in the display in PSI.

Sorry I have to disagree. Every VS that I have seen has no such sensor in it and I have taken a few of them apart. As I said before a sensor is not necessary, it can be calculated from the wattage and RPM.

But think about what you are saying, the sensor would need to be inside the wet end on both sides of the impeller. Since all nearly all VS pumps use wet ends from standard pumps and no modification is necessary. People on this forum have used whisperflo wet ends and just replaced the motor and drive so they did not need to replace the entire pump and it still displays the pressure. PLUS, there are no wires going into the wet end which would have to be the case if it was measuring pressure directly. Just doesn't make sense.

 

[JamesW]

OK, you're correct. I thought that it had a sensor because it gave a readout. I found this in the manual.

Sensorless Flow and Pressure Control Technology

Maybe the pressure and flow readings are not super accurate?

 

[mas985]

One of issues that I found using this method, my spreadsheet does the same calculation, is that the wattage vs flow rate relationship is very flat so a tiny error in the wattage measurement results in a very large error in flow rate. Then you have the head curve lookup to get headloss. So the calibration of those curves is crucial.

However, having said that, I dont think that the calculation is that bad given the op's observations.

Pressure and suction measurements at the pump would confirm.

 

[jonpcar]

Mark/James in your experiences, are pressure and suction pump measurements the most accurate for gpm determination? I tried it on my system and it didn’t seem that close to my flowvis measurements, but I could have done something wrong.

 

[mas985]

Every measurement has error. It is just matter on magnitude. Flovis is advertised as being pretty accurate but other methods can be close enough if done properly. Pressure and suction gauges have some error as well but if the measurements are taken at full speed, they should also be fairly accurate in the flow rate determination. However, if the measurements are taken at low speeds, error climbs and the measurements can be pretty far off.

What was the RPM, pressure/suction measurements and Flovis flow rate?

 

[kadavis]

Why does the builder go to all the trouble of plumbing in 2.5" and 3" plumbing, then use a XF pump and then throttle it all down to a 2" slide valve. He would have done you favor if he used the Full Flow backwash valve.FullFloXF High Performance Backwash Valve

 

[jonpcar]

Every measurement has error. It is just matter on magnitude. Flovis is advertised as being pretty accurate but other methods can be close enough if done properly. Pressure and suction gauges have some error as well but if the measurements are taken at full speed, they should also be fairly accurate in the flow rate determination. However, if the measurements are taken at low speeds, error climbs and the measurements can be pretty far off.

What was the RPM, pressure/suction measurements and Flovis flow rate?

Mark, I hesitate to post these...it's peripherally related but probably hijacking this thread...moderators feel free to sack this post...

I have learned a few things already in this this thread: a perfect vacuum is -14.7, water vaporizes at -14.3...SO, obviously the pump psi numbers I have read from my vacuum gauge are NOT accurate, probably one source of my problems. Also Mark, it sounds from your reply that the PSI method of calculating gpm is only good for high rpms (maybe only max speed?). This means that it probably cannot be used to determine gpm dynamically at lower speeds , one of the things that interested me.

So anyway, thanks for the discussion so far in this topic as I am learning quite a bit. Here are my numbers, but like I said, not sure we should have that conversation here. For these numbers, I effectively have 2" piping on both suction and return sides, short runs. Also, I haven't posted any tables yet on the new site, and these are definitely not a table.

These numbers should give the OP a comparison of numbers to a system that is similar to his with two exceptions: I 100% agree with kadavis about the slide valve (which I don't have in my system). Also, personally I am always suspicious of the heater (which I also don't have) in these threads where systems do not seem to perform up to their potential. I know they usually have an internal bypass, but I've seen a couple (maybe more) threads where the heaters was clearly a limiter in the system gpm throughput.

FlowVis-gpm Suction-PSI Filter-PSI Watts RPM
31 0 0 165 1207
50 -2.5 2 331 1725
82 -10 10.5 1198 2760
103 -15.5 17.5 2368 3450

EDIT: I should have mentioned these numbers are for 100% return through the wall returns and NONE back through my In-Floor-Cleaning-System popups.

 

[JamesW]

That valve definitely has a better curve than the slide valve.

At 50 gpm, the slide valve should be about 3.5 feet of head. It's not good or horrible.

So, it shouldn't be relevant to the excessive restriction that seems to be present.

Since it's a quad de filter, it really should not have a valve (in my opinion).

Note that the vacuum gauges might not be in psi. Inches of mercury are common units. So, be careful about the readings and the units of measure.

 

[JamesW]

I think that if the majority of the head loss is on the pressure side, the pump's calculated flow and head loss might be fairly accurate.

However, if there is significant head loss on the suction side, then the calculated head loss might be significantly off.

For example, if the suction was almost completely clogged, the flow would slow way down and the pump would overestimate the head loss.

 

[mas985]

FlowVis-gpm Suction-PSI Filter-PSI Watts RPM
31 0 0 165 1207
50 -2.5 2 331 1725
82 -10 10.5 1198 2760
103 -15.5 17.5 2368 3450

The suction PSIs are obviously wrong unless those are actually in units of in-hg.

But assuming the wattage is correct, the Watts2GPM spreadsheet gives me a flow rate of 106 GPM which is pretty close to the 103 GPM. 3% error is actually very good for that model so the pump itself should be just as good at the estimation. BTW, what does the pump estimate for GPM?

But considering the filter PSI, wattage, RPM and GPM as correct, the suction side head should be about 29.4' of head or a suction gauge should read about 26.5 in-hg. This is extremely high head loss (close to cavitation) for the suction side and is also consistent with the very low filter pressure for such a high RPM. So I believe you have a suction side issue as well unless the pressure gauge is off.

If we assume that the suction is in in-hg and the filter pressure is wrong, then the correct filter pressure would be 22.8 PSI which is much more reasonable.

If I were to put your pump on my pluming (2" return and 2.5" suction), I would get 98 GPM with return pressure of 25.8 PSI and suction of 10 in-hg.

 

[jonpcar]

Mark, once again you are “right on the money” about my gauge...it IS in in-hg, not in PSI. I had bought that for a different purpose and hadn’t really used it until I decided to put in on my pump’s suction side. I blindly looked at the numbers assuming they were the same as my other gauge. Thanks for finding at least one of my errors. My ecostar unfortunately doesn’t have estimates for the gpm, one of the initial reasons I installed the flowvis.

I took these numbers a couple months ago and I think there is one more parameter...I vaguely remember that my vacuum gauge read -15.5 in-hg when 100% suction was from the main drain...I “think” I remember it being about -10 in-hg when 100% suction was from the skimmer. I will have to recheck that at a later date.

Thanks again.