Jandy variable speed motor retrofitting - now with curves

[BenB]

Hey folks, happy Monday!

Apologies in advance that this will likely turn into a long one. I'll try to keep things short!

Summary
Having just installed solar heating on our garage roof, we're experiencing the usual challenge in that our existing 2.6 THP 2-speed pump isn't doing a good job. At low speed (and having had to move the VRV) the solar still works but likely at very reduced efficiency and with low flow overall, and at high speed it's a loud energy hog and probably pushes too much water through the system.

At the same time, we also have a sheer descent water feature that was plumbed with a single speed 2.21 THP pump with a set of valves that are choked almost all the way closed to hold back the flow, that seems to be massively overpowered the use case. We don't like running it (pH impact aside) because it's so loud and also because I know just how much power it's using up!

I'd like to explore the possibility of reconfiguring the pumps to:
1) Use the old 2-speed pump for the water feature and run it on low, with the valves opened back up to restore the flow as much as possible/needed; and
2) DIY retrofit a variable speed motor (V-Green is the obvious candidate but open to alternatives) onto the old single-speed pump and use that to run the pool. I would set up the speeds using the digital inputs vs. RS485 as laid out in this awesome post by @MSchutzer since it won't communicate directly with the Aqualink panel.

This would be a 'two birds with one stone' solution and minimize the expense involved, rather than replacing entire pumps which are otherwise perfectly functional.

List of Current Equipment
Pool pump - Jandy Stealth SHPF2.0-2 (2HP x 1.3 = 2.6 THP) [3450RPM 'high' / 1725RPM 'low']
Sheer descent pump - Jandy Stealth SHPF1.5 (1.5 x 1.47 = 2.21 THP) [3450RPM]
~20,000 gallons in-ground.
1 skimmer, 3 return jets.

Questions
1) Does this sound feasible overall?

2) Will the 2-speed on low likely provide 'good enough' flow to use the descent given the minimal distance and lift involved?

3) Given the starting point for the pool is a 2.6THP 2-speed, should I go straight to the 2.7HP V-Green motor, or could the 1.65 be sufficient? There is a large price difference ($400 vs $650) between these two options, so I don't want to just default to the larger one unless there is a good reason to do so; and at $700 that's a lot closer to just saying 'screw it' and getting an entirely new pump (maybe one that can talk to my Aqualink control panel) to avoid the fiddly labor of doing the motor swap.

A little more context here. While I see the general recommendation is to replace 'like with like' and match as close to the original THP as possible, I see plenty of folks around here with similar sized pools perfectly happily using lower rated pumps. If the PB oversized all the equipment on my pad, it feels silly to continue that trend when doing something like this unless it really needed it in the first place.

4) Am I right in saying that the practical difference between the two is that it the 1.65HP would need to run at higher RPM to achieve the same flow rate? How does the electricity usage compare between the two? At full speed the 2.7 would obviously use more overall but would the 2.7 running slower use more or less electricity than the 1.65 at the same flow? Presumably noise levels will also work similarly?

5) Would I need to plan making any changes to the impellers as well? I remember reading that water features do better with larger / oversized impellers? I could potentially also swap those over if needed, or just get a new one to go with the new motor vs. having to open them both up?

Thanks everybody!

Ben

 

[MSchutzer]

The Vgreen 2.7 HP would be the right choice given the 2.6 HP of your current pump. Unfortunately the 2.7HP version doesn’t directly support the four digital inputs like the 1.65 HP version. You have to buy their add on automation widget that adds another $180 dollars.

Now you are up to $830 dollars, you might just be better off getting a Pentair Intelliflow, or a Jandy VS pump, both of which Aqualink supports over the RS-485 datalink connection.

 

[BenB]

Oh interesting! Yeah that's disappointing that the bigger one is otherwise worse.

So I agree the conventional wisdom would point towards the 2.7. I guess my fundamental question is what if the original pump was oversized unnecessarily and all I ever needed was a 1.65HP to begin with? And how can I figure that out in the first place? Figure out the head of the whole system, desired flow rate and resulting RPM / power usage?

 

[MSchutzer]

While I don’t know Jandy pumps, with the Pentair Whisperflo that I have you can readily swap in an alternate impeller to change the SFHP of the pump. If the Jandy pump is similar you may be able to change to a 1.65 HP impeller. You may also need to change the volute depending if that part also changes with the pump HP.

I would suggest downloading the manual for the various different Jandy HP pumps and look at what changes there are for the parts on the wet end of the pump. You should also be able to look at the flow curves for a 1.65 HP version of the Jandy pump (assuming that they make one at that HP). Looking at the manuals should at least show you whether it’s feasible to use a smaller impeller.

If you are just filtering with that pump 2.7 HP is more than you will ever need. The nice part about a 2.7HP VS pump is that with the big impeller you can run that motor about 20% slower than you would need to run a 1.65 HP VS for the same amount of water flow.

I up sized my impeller one size to match the rating of the Vgreen 1.65. The fastest I ever run the pump is 2850 RPM when the pressure side cleaner is running for an hour. When my solar panels are active I run at 2600 RPM to have enough pressure to push the water 18 feet up in the air. When the solar is off I run at 2350 RPM and that moves 50 GPM at a filter pressure of 5 psi. I probably should drop filter only speed even lower, but I only run the pump 5 hours a day.

I used to run the original 1.25 SFHP single speed pump for 4 hours a day when the sun was out and that motor ran 1.55 kW or 6.2 kWh per day. At 60 gpm that moved 14.4K gallons per day. Now I run the larger impeller (1.65 SFHP) for 1 hr at 2850 RPM(cleaner speed) (.91 kWH) and 4 hrs at 2600 RPM (solar on) (4 x .66 kW = 2.64 kWh). So I now use 3.55 kWH a day and for both those speeds the water flow is 45 gpm, or 13.5k gallons per day. I’m essentially moving the same amount of water with a 42% savings in energy. The motor is also a lot quieter as it’s running slower.

I really could drop the speeds more and run longer for additional savings, but I need to run those speeds for the solar panels as that is my heating system.

The point of my example is to show that for most systems there’s really no need for a 2.6 SFHP pump as that more flow than you will ever need.

In my case I swapped to a Vgreen just because it made sense, it really doesn’t save me any real money as I also have Solar PV panels on the roof and I generate more power than I consume. I get paid about 2.3 cents for the excess power so it’s essentially free.

Mark

 

[BenB]

Thanks! I'll look up the spares for the smaller pumps to see how much changes. Am I crazy but I can't find any curves for the v-green? Is that because it significantly depends on the rest of the pump you're fitting it to?

Your post drew my attention because of the similarities in our setups. Basically same size of solar panels, both on single story roof. So on the face of it I'd have thought if it works well for you it might fit me also absent some other significantly different plumbing causing a huge difference in head. Also we're both Bay Area PG&E net solar electric producers and you like doing math (I have a spreadsheet doing the same calculations you just showed here!)

 

[MSchutzer]

Ben, the pump curves are based on the wet end of your pump, the motor just turns at a specific speed and the impeller and pump performance determine the pump curves. The Vgreen 1.65 power consumption is based on the motors speed and the size of the impeller as the impeller size determines the amount of work being done.

Here’s a chart from Century that lists the maximum power consumption for the Vgreen 1.65 when fitted with a 1.65 HP impeller.

This chart is close to what I measured on my motor. At 2850 rpm I measured 910 watts, at 2600 rpm 660 watts, and at 2350 rpm I measured 490 watts.


You can look at the curves for a Jandy 2.7 SFHP VS pump and get an idea of the curves for a 1.65 HP motor by using the 2850 rpm curve. That curve is about 20% down in speed for the 2.7 SFHP motor, but that would be roughly the water flow of the pump with a Vgreen 1.65 running at 3450 rpm when using a 1.65 SFHP impeller. The 2600 rpm curve would match the 3100 rpm curve for the Vgreen with the smaller impeller.

I hope this helps,

Mark

 

[BenB]

Thanks, Mark. That power table looks great and to hear that it's only a 20% difference between the 1.65 and 2.7 in terms of RPM is reassuring too.

At this point I'm just going to give it a shot. Worst case if it doesn't sufficiently work for the pool as a whole I can always use it as a cheap VSP for the sheer descent feature which will fix that issue regardless, or as a backup option if the 2-speed running on low doesn't give enough flow. So basically, I'll just get one and put it to the 'best use' I can find for it. For $380 + seals etc, it's significantly cheaper than any potential replacement pump as a whole, and won't require any plumbing.

Water temperature has gone from 58 last week to 69 yesterday. Postponed Monday's opening polar plunge for another week, by which time it should be nice and toasty in there. Looking forward to getting back in the water and enjoy the actual purpose of all the gadgets, testing and refilling that has been this winter!

Thanks again for your input!

 

[MSchutzer]

Make sure that a smaller 1.65 HP impeller is available for your pump if you go the Vgreen 1.65 route. The 2.6 HP impeller that you have in the pump now is too large for a 1.65 HP motor.

Mark

 

[BenB]

I'll make sure to have one on hand just in case it's not what I expect but since I'm swapping out from the 1.5 x 1.47 = 2.3 THP unit, and since Jandy's parts lists seem to show impellers for each model based on their face HP, I expect to find a "1.5HP" impeller in there which will pair nicely with the 1.65 HP motor. If they uprate the impellers too then yeah I'd need to get a 1.5HP one instead of the 2HP one I may find.

Edit: Spoke to Jandy. The impellers are actually based on the expected THP of the stock motor (equivalent to the uprated sister model). So I will definitely need the smaller impeller.

 

[BenB]

Can somebody please sense check my curves? (ooh matron!)



For the pump curves I used the published curve for the Jandy 1.5HP (assuming 1.5 vs 1.65HP is negligible difference) at 3450RPM and then used these two formulas to recalculate head and flow rates at different RPMs (thanks to @mas985 as always for the amazing hydraulics post):

GPM B = GPM A * (RPM B / RPM A)
Head B = Head A * (RPM B / RPM A) ^ 2

I compared the calculated curve for 1750RPM to the published Jandy curve for the same pump running at low speed. They matched up almost perfectly, so I think I did the math right. I'm happy with the pump curves.

For the system curve I used the Pentair calculator. This is based on 200' of 2" pipe, 25x90 degree turns, 2x45s, 1 check valve and NO vertical lift (based on what I've seen here that up/down even themselves out). I then add onto the curve that generates the published curves for my filter and solar panels. I have added a 15ft 'plug' for misc additional head loss e.g. SWG. Overall though the curve doesn't seem super sensitive to changes in lengths and fixtures - the main driver seems to be the size of the PVC.

As a sense check, I compared the calculated head to my filter pressure gauge readings that I have today with a 2.5HP motor (which I would therefore expect to be somewhat stronger). At low speed with the solar engaged, I see a pressure of 9PSI, or 21ft, which is not a million miles off what I see here. On high speed, however, I currently see 27psi or 62ft. Adding 10ft additional for suction side, that would imply an operating point at 3450RPM of about 70-80GPM, which seems very high compared with what I believe (but have no real way of knowing) I'm experiencing. With the solar disengaged entirely, my pressure is 16psi, or 47ft with the same 10ft suction estimate, which would suggest almost 100GPM.

Now I'll be honest, I don't know what 100GPM "feels" like in front of a return (there are 3 connected in a chain, so the furthest one is weak compared to the first, but I don't think that's what I'm getting really. So is there something obvious I've missed in my curve or plumbing calculations that is a source of additional head? My first guess would be that maybe what is buried in the ground (returns, suction) is 1.5" instead of 2" since that would have a large impact at higher flow rates but leave the lower end of the curve where it seems to line up fairly in tact. I know that the return jets are 1.5" at that point, but since everything at the pad is in 2" I have assumed that the buried PVC is 2" right up until the return outlet itself.

Either way - assuming the above is "accurate enough", then in planning for solar flow of 24-48gpm (4-8gpm per panel) would put me in the 2000-2500RPM range, which seems consistent with what I've seen folks around here posting as being their set points.

 

[mas985]

For the system curve I used the Pentair calculator. This is based on 200' of 2" pipe, 25x90 degree turns, 2x45s, 1 check valve and NO vertical lift (based on what I've seen here that up/down even themselves out). I then add onto the curve that generates the published curves for my filter and solar panels. I have added a 15ft 'plug' for misc additional head loss e.g. SWG. Overall though the curve doesn't seem super sensitive to changes in lengths and fixtures - the main driver seems to be the size of the PVC.

The return eyeballs can make a big difference if they are small. The spreadsheet in my signature has a full head calculation tab for ALL the elements in the pool plumbing so you can get a more accurate plumbing curve. Plus, it will give you predictions for any speed or any pump. BTW, SWGs can be ignored as their head loss is usually very low.

However, there is a much easier way to determine the correct flow rate. A normal panel setup will generate less than 1 PSI pressure rise when operating at the correct flow rate. If you add to that the head loss in the supply and return plumbing, the rise should be about 2-3 PSI. So basically, you only need a slight PSI rise to know you have sufficient flow through the panels.

However, the solar valve bypass setting will be different between full speed and half speed to get the same pressure rise so you need to take that into account as well.

 

[BenB]

However, the solar valve bypass setting will be different between full speed and half speed to get the same pressure rise so you need to take that into account as well.

Sorry I couldn't follow what you meant here? I use a Jandy no lube JVA triggered by my pool automation using the temperature sensors etc.

Also does my 9psi increase for both low and high speed indicate an issue? It's not a long run and while there are a lot of 90s involved unless the it's reading the static head at the filter that sounds high based on what you suggest.

Thanks!

 

[mas985]

My point is that pressure rise due to solar should be different for high vs low speed. So if you are targeting a 2-3 PSI rise due to solar, the solar bypass valve will have different settings for high vs low speeds.

But 9 PSI rise for solar is waaaaay too high. You need to bypass more of the water so you only get a 2-3 PSI rise. Of course total pressure needs to be high enough to keep the VRV closed.

 

[BenB]

Interesting... Ok I'll look into seeing how to do that with the JVA and of course see what I experience when I make the pump motor changes discussed above. Thanks for chipping in as always - we're very lucky to have you here!

 

[mas985]

I am a bit confused on your pressure measurements above. What are the following measured filter pressures:

Solar Off High Speed:
Solar On High Speed:
Solar Off Low Speed:
Solar On Low Speed:

Also, are you sure the VRV is closed on low speed? You can tell by running on low speed with solar for a while and then switch to high speed. If air comes out of the returns, the VRV is open on low speed.

 

[BenB]

With freshly cleaned filter:

High Speed
Solar ON - 27psi
Solar OFF - 16psi

Low speed
Solar ON - 7psi
Solar OFF "0"psi

11 vs 7, so not identical and the 0 is clearly a false reading so that might actually be right. High 11 could just be because the attempted flow rate is way too high?

To get it happy on low I have had to relocate the VRV from the top of the panels to the bottom. Single story garage roof. No bubbles anymore on low.

When I get the VSP set up I'll move it back up again and dial in the speed 100-200 above when the VRV closes properly.

 

[mas985]

11 vs 7, so not identical and the 0 is clearly a false reading so that might actually be right. High 11 could just be because the attempted flow rate is way too high?

The delta pressure on high vs low should not be the same. The filter pressure ratio high:low speed should be about 4:1 which it is with solar on so that makes sense. Solar off should still read about 4 PSI on low speed but the issue maybe that some gauges have a needle post stop location at around 3 PSI so it may not really be 0. Plus gauge accuracy is usually around 1-2 PSI. Aside from that, everything seems reasonable although the filter pressure with solar is quite high but that may be due to the solar plumbing setup. With a VS, you should be able to zero in on something that works ok and saves on energy cost.

One thing to note is that the lower (closer to the pump) you place the VRV, the lower RPM you will be able to use. However, priming the panels could be an issue so you might need to prime the panels on a higher RPM and step down to a lower RPM. But if you leave the VRV where it is, you won't have to do that but energy costs will be higher. A trade off.

 

[BenB]

Update!

The retrofit went well. Took my time going through all the steps and managed to swap everything out. The tidy-up is going to take longer as the inlet is in a different place compared to the old motor, so now my power cables and conduit are about 3" too short, so will have to replace the entire length! Argh. Then will be an afternoon to rewire the cabinet, remove unnecessary cabling, and hook up the digital inputs to new auxes on the Aqualink.

In dialing in the settings, I took filter pressure readings at each speed, with the solar both enabled and disabled. I also made notes as to the skimming performance and how strong the returns felt. For the suction side and to calculate total projected TDH I used the estimates from Mark's spreadsheets based on my plumbing. For the pressure side the spreadsheet was within 1 psi for most speeds, so I consider it a very good model. Where the filter reading was "zero", for the sake of precision I interpolated the result backwards from the next highest reading to get a truer reading to calculate the TDH with.

RPM	Filter Pressure - Solar Off	Filter Pressure - Solar On	TDH - Solar Off	TDH - Solar On
600	"0"	N/A	0	0
850	"0"	N/A	1	1
1100	"0" - SWG flow detected	N/A	1	1
1350	"0"	N/A - Cannot hold water	2	8
1600	"1"	4 - VRV drawing in air	6	12
1850	"3"	6 - Minor bubbling	10	17
2100	4	9	14	24
2350	6	10	19	29
2600	7	13	23	37
2850	9	16	28	45
3100	10	19	32	53
3450	12	21	38	59

First thing to point out is that the solar impact on pressure / head at reasonable working speeds is now reduced to more closely match the levels that were expected.

Next is that despite large the 'face' drop from 2.6HP to 1.65HP which some were concerned about (not just here - I asked InyoPools and they suggested getting the 2.7HP V-Green to keep the HPs roughly the same, though that assumes that the original motor was the 'right' size), the actual performance did not drop significantly at all and of course having room to select speeds in between 'high' (with higher TDH restricting flow) and 'low' (which didn't quite perform well enough for skimming and solar).

I plan to skim at 2350 RPM (75GPM, 540W), run the solar at 2100 RPM (40GPM, 395W), and general filtration at 1350 RPM (50GPM, 130W) each of which are '1 step' above the minimum levels I measured during my testing.

Based on the pump curves posted earlier, all of these choices will provide very good flow rates. Surprisingly good maybe? Can anybody sense check them and tell me if I'm crazy? When I measured with a hose with the old pump, I was only recording about 20GPM on low speed (1750RPM), but now I think I understand that the hose itself was a likely contributor to that reading - I only ever seem to get 6GPM through a 3/4" hose regardless of application - filling the pool, suction from sump pump - they always come out around 6GPM. Can I really expect to get 50GPM at 1350RPM? The math and graphs would seem to say yes, but... really? Sure there's clearly water flowing at that speed but to visualize 50 gallon bottles being filled up in a minute from 3 returns seems high - that's 1 bottle every 4 seconds at each return.

One thing to note is that the lower (closer to the pump) you place the VRV, the lower RPM you will be able to use. However, priming the panels could be an issue so you might need to prime the panels on a higher RPM and step down to a lower RPM. But if you leave the VRV where it is, you won't have to do that but energy costs will be higher. A trade off.

Could I ask you to elaborate on this point, please? I've seen this mentioned in passing on several threads but have not 100% grasped what is being said here. The obvious part I get - that the lower the VRV, the lower RPM the pump will work at without pulling in air. The second piece I think I get - if higher than higher RPM will be required and that will result in higher energy usage. The piece I'm not following is whether there is a separate difference between 'priming' the panels and presumably where the air gets trapped depending on where the VRV is located? Are you saying that by having it higher, the panels will automatically be more primed than if the VRV is lower and therefore little air pockets might remain? And this effect is independent of the RPM? Sorry if I've missed the obvious or am reading into it more than needed.

 

[mas985]

Next is that despite large the 'face' drop from 2.6HP to 1.65HP which some were concerned about (not just here - I asked InyoPools and they suggested getting the 2.7HP V-Green to keep the HPs roughly the same, though that assumes that the original motor was the 'right' size), the actual performance did not drop significantly at all and of course having room to select speeds in between 'high' (with higher TDH restricting flow) and 'low' (which didn't quite perform well enough for skimming and solar).

Did you downsize the impeller as well? If not, you may burn out your new motor. The motor must be sized to the impeller. You might be able to get away with it if you don't run at full speed but it looks like you were doing that for the above test.

The piece I'm not following is whether there is a separate difference between 'priming' the panels and presumably where the air gets trapped depending on where the VRV is located? Are you saying that by having it higher, the panels will automatically be more primed than if the VRV is lower and therefore little air pockets might remain? And this effect is independent of the RPM? Sorry if I've missed the obvious or am reading into it more than needed.

The minimum speed to keep the VRV closed could be less than the speed needed to prime the panel. Some controllers have the ability to bump up the speed when solar kicks on and then step it down after solar primes. If you don't have the ability, then you may need to run at the speed required for priming instead of the VRV so there is no point in lowering the height of the VRV.

 

[BenB]

Yes - I downsized the impeller too! That part is buried further up in the thread

What is the 'real world' implications of not fully priming? If water is flowing through the system, then is the issue just that there are air pockets and therefore some performance loss? Would having the VRV higher actually change how well the panel primes, or are you saying that a lower VRV provides a false sense of security? The Jandy does not have the functionality you describe, or at least, certainly not with my motor.