8fps max velocity on spa drains - I split the max flow rate in two (half per drain)

[venexiano]

Dear all,

I am installing a in-ground acrylic spillway spa (I already have a pool but I am not touching pool piping). I already have a Pentair 011018 IntelliFlo Variable Speed Pump VS+ 3.2kW 3HP pump (curve below) for the pool. I run the Head calculation and, given 60 ft of return and 60 ft of suction piping, given the heater loss (0.002*Q^2), the filter loss (1.4 feet), and 4 feet for miscellaneous losses, I have that with 10 jets (each with 8 GPM) my operational discharge when in SPA mode (no water to pool) is 80 GPM, with 32 ft of total losses, which gives a pump speed of 2350 rpm. The pump can provide at 3450 rpm a max discharge 140GPM with 57 feet of losses. All these numbers were computed with 3" piping in the suction line and 2" in the return line. 3" in the suction line was requested by the city when I pulled the Homeowner permit and I am using a licensed plumber as subcontractor. 2" in the return line were suggested by the plumber See screenshot below of the drawing I submitted to the city, given to me by the spa manufacturer, in which the city added the part in red about 3" drain pipes. Now the main point is: the spa comes with 2 drains that are pre-plumbed with 2" pipes up to a junction (see pic below). ANSI/APSP/ICC 3 7.6 requires 8 fps max on piping. My understanding (after speaking with my plumber) is that in my case 3" piping are needed only from where the two drain pipes are meeting (to give 6.3 fps<8fps at 140 GPM), and the pre-plumbed 2" are fine, because assuming the max flow rate (140 GPM) splits in two, it gives 70 GPM per drain, which gives a velocity 7.1 fps <8 fps at each drain. Is this correct? My plumber does not want the hassle of having to upgrade the pre-plumbed drain pipes to 3", and I think he is right.

 

[mas985]

I am installing a in-ground acrylic spillway spa (I already have a pool but I am not touching pool piping). I already have a Pentair 011018 IntelliFlo Variable Speed Pump VS+ 3.2kW 3HP pump (curve below) for the pool. I run the Head calculation and, given 60 ft of return and 60 ft of suction piping, given the heater loss (0.002*Q^2), the filter loss (1.4 feet), and 4 feet for miscellaneous losses, I have that with 10 jets (each with 8 GPM) my operational discharge when in SPA mode (no water to pool) is 80 GPM, with 32 ft of total losses, which gives a pump speed of 2350 rpm. The pump can provide at 3450 rpm a max discharge 140GPM with 57 feet of losses. All these numbers were computed with 3" piping in the suction line and 2" in the return line.

Are you sure you calculated the head loss correctly? Can you show the details of your calculation? Do you have a schematic of the return side as well?

140 GPM @ 57' of head has a plumbing curve of 0.00291 and if the heater is 0.002, that leaves 0.0009 for the rest of the plumbing. Each 3/8" jet has a plumbing curve of 0.19721 and 10 in parallel have a combined curve of 0.002. So something is not adding up.

Not only that but 140 GPM is very close to run out so the pump is likely to cavitate.

My understanding (after speaking with my plumber) is that in my case 3" piping are needed only from where the two drain pipes are meeting (to give 6.3 fps<8fps at 140 GPM), and the pre-plumbed 2" are fine, because assuming the max flow rate (140 GPM) splits in two, it gives 70 GPM per drain, which gives a velocity 7.1 fps <8 fps at each drain. Is this correct? My plumber does not want the hassle of having to upgrade the pre-plumbed drain pipes to 3", and I think he is right.

Yes that meets ANSI/APSP/ICC 3 7.6 standard. However, the 2" pipe would well exceed 8 ft/sec @140 GPM. Also, there are other regulations to take into account.

For entrapment safety (VGBA), each main drain line should be below 6 ft/sec when the other is plugged which means that each line of the MD pair needs to be below 3 fts/sec when operating together. However, after the MD TEE further up stream 8 ft/sec is allowed in the suction line although not recommended due to high head loss conditions under high flow.

https://www.poolspanews.com/how-to/the-whole-equation_o#:~:text=VGB%20says%20plumbing%20that%20connects,is%20limited%20to%2032%20gpm.

https://consensus.fsu.edu/FBC/Pool-Efficiency/APSP-15%20Pool%20Energy%20language%20(1-11-10)%20%20for%20Florida.pdf

https://duval.floridahealth.gov/programs-and-services/environmental-health/swimming-pools-spas-and-bathing-places/_documents/vgb-fact-sheet.pdf

 

[venexiano]

Thank you a lot. If you mean the schematic of the return 1" pipe, below are better pictures that go all around the spa (use the yellow measuring tape as reference for the pic angle). Basically, I will have 3" draining line from the equipment (starting at the junction of the two pre-plumbed 2" drain pipes), 2" pipe through the equipment and 2" back all the way to the junction where 1" pipes bring water to the 10 jets (see black circle in pic below).
As for the head, you are right I neglected the head loss on the 1" pipes and just shove it inside the 4' of miscellaneous head drop because I thought it was short. Also I am more conservative by neglecting it (like I am overestimating max velocity by neglecting it), since if for instance the head drop was 70 feet, then the flow rate would be 115 GPM and the drain velocity would be smaller. I can compute an (over)estimate of head loss on those pipes feeding the jets by considering two 1" pipes departing at the junction in the black circle below, each carrying half flow rate, and they are ~6' long. Using this calculator, for a Re=4000 (Darcy Friction factor for PVC of 0.04) I get 11' of head drop (34248 Pa). That would change little with different Re, I kept it simple. That seems highly overestimated, since in reality those pipes form loops with 10 pipes. Also, I shove in those miscellaneous losses also the exit losses of the jet. I have a 3/4" nozzle size, that gives an exit loss (using velocity/(2*gravity) and flow rate of 140/10=14 GPM for 1 out of 10 jets) of 1.6'. See pic of the drain specs, each drain is good up to 160 GPM. See at the end of this post below the list with all the losses I have considered.

My questions are:
1) Is it a problem if the 2" return pipe exceeds 8 fps? Would you do 3"? See screenshot below of ANSI/APSP/ICC 3 7.6. Why does it say " 8 ftp both suction AND pressure piping"? Does pressure piping mean "any pipe"? Why should the speed in the return pipe be limited? There is no entrapment issue there. Is that for energy efficiency?
2) The math above is done just to be safe when the max speed of the pump (3450 rpm) is turned on by mistake on the spa. That max speed is only used for the pool when using the robot vacuum (POOL mode, no water recirculation in the Spa). Instead, the spa operational rpm that I would set up on Omnilogic when in SPA mode would be 2350 rpm, or slightly above if I underestimated losses, and it would give a flow rate of 80 GPM, which gives just 4 fps velocity on each drain. I do think that spa has to be designed for max speed rpm, i.e considering that some one by mistake sets up the pump to his maximum rpm. But I think we are pretty safe about entrapment with ~2350 rpm. Do you agree?
3) You say "For entrapment safety (VGBA), each main drain line should be below 6 ft/sec when the other is plugged which means that each line of the MD pair needs to be below 3 fts/sec when operating together.". Can you point where it says it is enforced for residential? I did not see it in the ANSI
4) You said " after the MD TEE further up stream 8 ft/sec is allowed in the suction line although not recommended due to high head loss conditions under high flow." I guess you meant downstream, right?


Losses I Considered:
- 60 ft of drain line at 3" and 0.07 ft/ft gives 4.2 feet of loss
- 60 ft of return line at 2" and 0.07 ft/ft gives 9.5 feet of loss
- heater loss (0.002*Q_GPM^2): at 140 GPM gives 39.2
- the filter loss (1.4 feet),
- 4 feet for miscellaneous losses (bends, exit losses, 1" return loops etc, probably it is more than 4 ft)
Total loss: 57

 

[mas985]

Thank you a lot. If you mean the schematic of the return 1" pipe, below are better pictures that go all around the spa (use the yellow measuring tape as reference for the pic angle).

I meant for the entire plumbing system. All pipes, fittings, valves, equipment and distances between them.

Basically, I will have 3" draining line from the equipment (starting at the junction of the two pre-plumbed 2" drain pipes), 2" pipe through the equipment and 2" back all the way to the junction where 1" pipes bring water to the 10 jets (see black circle in pic below).

I would upsize those to at least 2.5". I am surprised that the city did not flag that because I believe Florida follows APSP-7 regulations for 3 ft/sec.

As for the head, you are right I neglected the head loss on the 1" pipes and just shove it inside the 4' of miscellaneous head drop because I thought it was short. Also I am more conservative by neglecting it (like I am overestimating max velocity by neglecting it), since if for instance the head drop was 70 feet, then the flow rate would be 115 GPM and the drain velocity would be smaller. I can compute an (over)estimate of head loss on those pipes feeding the jets by considering two 1" pipes departing at the junction in the black circle below, each carrying half flow rate, and they are ~6' long. Using this calculator, for a Re=4000 (Darcy Friction factor for PVC of 0.04) I get 11' of head drop (34248 Pa). That would change little with different Re, I kept it simple. That seems highly overestimated, since in reality those pipes form loops with 10 pipes. Also, I shove in those miscellaneous losses also the exit losses of the jet. I have a 3/4" nozzle size, that gives an exit loss (using velocity/(2*gravity) and flow rate of 140/10=14 GPM for 1 out of 10 jets) of 1.6'. See pic of the drain specs, each drain is good up to 160 GPM. See at the end of this post below the list with all the losses I have considered.

You need to make decisions based upon realistic numbers. What you are calculating is not realistic.

First, a Reynolds number of 4000 for 1" PVC is a flow rate of 1.4 GPM so I am not sure why you were using that.

Next, the 3/4" outlet you are seeing is the external eyeball and not the internal jet nozzle. Here is an illustration of a typical venturi jet:



The jet nozzle is embedded in the venturi tee. The schematic in the first post says 8-10 GPM per jet which is consistent with 5/16" jets. So you won't need more than about 100 GPM although it doesn't hurt to have the potential for a little more.

My questions are:
1) Is it a problem if the 2" return pipe exceeds 8 fps? Would you do 3"? See screenshot below of ANSI/APSP/ICC 3 7.6. Why does it say " 8 ftp both suction AND pressure piping"? Does pressure piping mean "any pipe"? Why should the speed in the return pipe be limited? There is no entrapment issue there. Is that for energy efficiency?

Yes, I would use 3" pipe. You don't want extra head loss for a spa or the jets will be very weak (see analysis below). The speed within a pipe is not limited by anything. The 8 ft/sec is a recommendation to keep head loss low for higher efficiency and the possibility of water hammer low to prevent plumbing failures.

2) The math above is done just to be safe when the max speed of the pump (3450 rpm) is turned on by mistake on the spa. That max speed is only used for the pool when using the robot vacuum (POOL mode, no water recirculation in the Spa). Instead, the spa operational rpm that I would set up on Omnilogic when in SPA mode would be 2350 rpm, or slightly above if I underestimated losses, and it would give a flow rate of 80 GPM, which gives just 4 fps velocity on each drain. I do think that spa has to be designed for max speed rpm, i.e considering that some one by mistake sets up the pump to his maximum rpm. But I think we are pretty safe about entrapment with ~2350 rpm. Do you agree?

Perhaps but again, the numbers you are calculating are not realistic. In most cities, to pass inspection, they base the entrapment flow rates on maximum speed because they do not know how you will use your system.

3) You say "For entrapment safety (VGBA), each main drain line should be below 6 ft/sec when the other is plugged which means that each line of the MD pair needs to be below 3 fts/sec when operating together.". Can you point where it says it is enforced for residential? I did not see it in the ANSI
4) You said " after the MD TEE further up stream 8 ft/sec is allowed in the suction line although not recommended due to high head loss conditions under high flow." I guess you meant downstream, right?

VGBA is a federal statute and applies to both residential and public pools. Each city has different rules and sometimes they can be more stringent than federal or state regulations so you should check with your local building department.

There is also APSP-7 which is often sited in federal, state and local regulations:

https://westfordma.gov/DocumentCenter/View/70/ANSI-Pool-Entrapment-Info-PDF

https://www.floridabuilding.org/fbc/commission/FBC_1015/commission_Education_POC/220/220-3-MATERIAL.pdf

https://www.cpsc.gov/s3fs-public/pdfs/foia_vgbincstd.pdf

https://www.poolsafely.gov/wp-content/uploads/2016/04/vgba.pdf

Losses I Considered:
- 60 ft of drain line at 3" and 0.07 ft/ft gives 4.2 feet of loss
- 60 ft of return line at 2" and 0.07 ft/ft gives 9.5 feet of loss
- heater loss (0.002*Q_GPM^2): at 140 GPM gives 39.2
- the filter loss (1.4 feet),
- 4 feet for miscellaneous losses (bends, exit losses, 1" return loops etc, probably it is more than 4 ft)
Total loss: 57

I don't think any of that is correct except for the heater. That is the same equation I use for my analyses. The fittings will account for much more head loss that what you are suggesting. Also, a cartridge filter is about half the head loss of the heater. Where did you get that number? I have some actual filter head loss numbers in this sticky:

Hydraulics 101 - Have you lost your head?

Edit - Before reading this article, you may want to first familiarize yourself with this Pool School article on Pump Basics Hydraulics 101 - Pump Head Curves, Plumbing Head Curves and Operating Points Hydraulic Head In fluid dynamics, head is a concept that relates the energy in an...

www.troublefreepool.com

Also, I have several spreadsheets that I make available to the forum specifically for this type of analysis. Based upon what you have told me so far, and quite a bit is still missing so I will need to guess, I have pre-populated the spreadsheet to come up with a more realistic view of your setup. You can review it here in Google Sheets:

Pool Pump Tools v027 - venexiano

docs.google.com

We can make adjustments with new information.

The operating point that I come up with for your current setup @ 3450 RPM is 91.41 GPM @ 85.4' of head.

Changing the return pipe to 3" and the pad pipe to 2.5" results in an operating point @ 3450 RPM of 97.5 GPM @ 83.2' of head.

Both designs actually fall short of the 100 GPM target. But if you can find the manufacture recommendations, we may be able to make some adjustments such as bypassing the heater. This will change the 3450 RPM operating point to: 108.1 GPM @ 79.2' of head.

Note for all of these cases, the velocity in the 3" pipe is well below 6'/sec and even in the 2" case, it is close to 8 ft/sec.

More spreadsheets:

Hydraulics 101 - Have you lost your head?

Edit - Before reading this article, you may want to first familiarize yourself with this Pool School article on Pump Basics Hydraulics 101 - Pump Head Curves, Plumbing Head Curves and Operating Points Hydraulic Head In fluid dynamics, head is a concept that relates the energy in an...

www.troublefreepool.com

 

[venexiano]

Thank you a lot. I really appreciate your help, you are very knowledgeable. Plumber plumbed 2" in the pad and 3" drain. After reading your message, I called the city, they first said I can go 2" on return, and 3" required only on drain side. But then I read to the city lady section ANSI/APSP/ICC 3 7.6 and she agreed I should keep 8 fps on return as well, so I need 3". So I am going for 3", this afternoon plumber started installing also 3" on the return, see pics below of pad. Thank you for the spreadsheet. I am not used to plumbing curve for exit losses, they should be proportional to velocity/(2*gravity)= Flow flow_rate/(cross_section_area_pipe*2*gravity). So they are proportional to flow rate squared, but they also function of pipe diameter. So head loss it is not a constant*GPM^2. But these are details, and I think your number are probably very close to be right, and mine are way underestimated because I was just trying to keep losses estimate at minimum for maximizing velocity to see I could kill someone in that spa due entrapment. I think we both agree I am fine with that. The problem here is that maybe I underestimated losses too much and I might be very tight with the flow rate as your math shows since I am around max rpm. I cannot change number in that spreadsheet, if you give me rights please i can change pad size to 2" and return to 3".

Other questions/observations:
1) When you say Section you probably mean suction right? And for plumbing curve you mean plumbing curve constant right?
2) I cannot see column B and it does not allow me to see it
3) I used Reynolds number = 4001 because it was the minimum allowed here and I was trying to be conservative to max velocity for entrapment. The friction factor in the Darcy-Weisbach equation decreases with Re (see screenshot), so dissipation is smaller with the actual Re that should be around 50000 (I used 0.003 mm as average roughness height for PVC from here).
4) I never used the plumbing rate curves, in my hydraulic class we computed directly losses with the Darcy-Weisbach equation, I know it is a pain because it has to be done iteratively because of the dependence of the friction factor on the Reynolds number, but head loss is not exactly proportional to the square of flow rate for all flow rates. But I believe that using a plumbing curve constant that fits the data is a very good simplified tool to get good estimate fast. I just do not have the experience with those constants to double check if your math is right, but I trust it is.
5) I got the loss per foot of pipe from the spreadsheet the city gave me, see screenshot below, I did not double checked the math with the Darcy Weisbach equation to be honest, so they could be wrong. I trust more your computation given your experience if you think the city gave me wrong numbers.
6) I have this filter: Hayward C150S SwimClear Cartridge Pool Filter, and head loss is below. It is for C200S but it is the closest I found.
7) From the equipment in the picture, plumber is gonna add only one 90 degree bend (or maybe two 45 deg) to get straight to the spa, so that is basically it (note that one drain in the spa picture above has two 90 deg elbows)

Thank you a lot, really appreciated

 

[mas985]

I am not used to plumbing curve for exit losses, they should be proportional to velocity/(2*gravity)= Flow flow_rate/(cross_section_area_pipe*2*gravity). So they are proportional to flow rate squared, but they also function of pipe diameter. So head loss it is not a constant*GPM^2.

The plumbing curve is dependent on both pipe length and diameter for straight pipes. You can see that in the equations in the spreadsheet. Currently I am using the equivalent length method for fittings which calculates the equivalent length of straight pipe for the same head loss but this too is dependent on pipe diameter.

But these are details, and I think your number are probably very close to be right, and mine are way underestimated because I was just trying to keep losses estimate at minimum for maximizing velocity to see I could kill someone in that spa due entrapment. I think we both agree I am fine with that. The problem here is that maybe I underestimated losses too much and I might be very tight with the flow rate as your math shows since I am around max rpm. I cannot change number in that spreadsheet, if you give me rights please i can change pad size to 2" and return to 3".

The issue with under estimating too much is that you may be making bad decisions upon that. Especially when it comes to pipe velocity. As a philosophy, I think it is better to make as good an estimate as possible and then add margin if necessary.

Other questions/observations:
1) When you say Section you probably mean suction right?

No, I actually mean section. The entire plumbing system is broken up into multiple sections. There are two sections for suction side and multiple sections for the return side. I just wanted to make it clear that the plumbing curve specified is only for that section of the plumbing.

And for plumbing curve you mean plumbing curve constant right?

Yes, or more accurately, the plumbing curve coefficient since it technically has many dependencies.

2) I cannot see column B and it does not allow me to see it

That is because it contains the constants necessary for the model to work and are not meant to be messed with. These numbers are the LoD constants for the fittings and the plumbing curve coefficients for the valves.

3) I used Reynolds number = 4001 because it was the minimum allowed here and I was trying to be conservative to max velocity for entrapment. The friction factor in the Darcy-Weisbach equation decreases with Re (see screenshot), so dissipation is smaller with the actual Re that should be around 50000 (I used 0.003 mm as average roughness height for PVC from here).

I understand your motivation but I think you took it too far.

4) I never used the plumbing rate curves, in my hydraulic class we computed directly losses with the Darcy-Weisbach equation, I know it is a pain because it has to be done iteratively because of the dependence of the friction factor on the Reynolds number, but head loss is not exactly proportional to the square of flow rate for all flow rates. But I believe that using a plumbing curve constant that fits the data is a very good simplified tool to get good estimate fast. I just do not have the experience with those constants to double check if your math is right, but I trust it is.

As you pointed out, DW is very cumbersome to use when trying to find the operating point of the pump as it would have required an iterative solution. I could have used Hazen-Williams as that is very similar to the plumbing curve but does not easily take into account surface roughness or water temperature. Also, given that many regulatory agencies (e.g. california energy commission) and standards bodies (e.g. ANSI, NSF, APSP) have adopted the plumbing curve methodology, I felt that using a PC was the best approach to pursue. I have seen some cities start to use this on their compliance forms so it is something they understand.

As for accuracy, I have used the tool on my own plumbing and probably several dozen other pools on the forum and for those cases where the OP has a flow meter installed or directly measured head loss (pressure/suction gauges), the calculations have come very close to the measurements. Well within the accuracy of flow meters, pressure gauges and suction gauges when the details of the plumbing are well known. Garbage-in-Garbage-out rule still applies.

5) I got the loss per foot of pipe from the spreadsheet the city gave me, see screenshot below, I did not double checked the math with the Darcy Weisbach equation to be honest, so they could be wrong. I trust more your computation given your experience if you think the city gave me wrong numbers.

You have experienced what I have with city building departments. They tend to use very broad approximates that are not all that accurate. The fact that they also missed the velocity in the return pipe and continue to miss the velocity in the MD cross pipe, indicates that they are not very thorough. That cross pipe velocity is probably the most critical for safety reasons.

6) I have this filter: Hayward C150S SwimClear Cartridge Pool Filter, and head loss is below. It is for C200S but it is the closest I found.

That one does have much lower head loss. I haven't seen many pools on the forum use that one. I updated the model for that.

7) From the equipment in the picture, plumber is gonna add only one 90 degree bend (or maybe two 45 deg) to get straight to the spa, so that is basically it (note that one drain in the spa picture above has two 90 deg elbows)

Thank you a lot, really appreciated

Adding one or two 90s/45s isn't going to change the results much. If you can give me a total for the suction and total for the return (pad and run) for each fitting, I can update the model for you.

 

[venexiano]

Ok I see, the coefficient in the plumbing curve varies with diameter, you just shove everything in that coefficient, one diameter one coefficient. It is a neat approach. I wonder how you determine the coefficient, do you use tables with list of coefficients for each diameter? Also, now I can see column B, what do you have in that?

You asked me the "total". I think you mean number of 90 elbows and 45s. So when in SPA mode I count:

- three 90s and one 45 in the drain line at 3"
- two 45s in the pad (upstream the pump) at 2"
- eight 90s in the return pad (downstream of the pump) at 2"
- three 90s and one 45 in the return line at 3"
- Then of course there are two Ts where the actuators are

Can you update spreadsheet with that please? The heater is Hayward HDF400 but I called Hayward but they dont have head loss specs. Important: I see on your spreadsheets ~2.5" for inner diameter of the 3" pipe, that gives higher losses, inner diameter should be 3.042" for 3" nominal.

Thank you a lot
A

 

[venexiano]

I just figured out I gave you the wrong filter. That was a filter cardridge I was offered as an upgrade but plumber said it is not 100% necesary. I currently have (see pic below) the Hayward Star Clear Plus C1200 filter with this cartridge, which says it has 120 sqf. The original Hayward cartridge for that filter is actually just slightly higher price and still 120 sqf, I could get that one to be safer. But please let me know if I should upgrade to the 150s (or even 200s) Swim Clear. From 120 to 150 sqf doesnt seem a big upgrade, beside square footage is there anything else that will improve?

 

[mas985]

Ok I see, the coefficient in the plumbing curve varies with diameter, you just shove everything in that coefficient, one diameter one coefficient. It is a neat approach. I wonder how you determine the coefficient, do you use tables with list of coefficients for each diameter?

Do you mean the coefficients 0.00065 and the exponential 5.23 in the plumbing curve equation? Those coefficients match the DW equation exactly at 6 ft/sec for all typical pool pipe diameters (1" to 3"). There is a small error off of 6 ft/sec but it isn't too bad.

Also, now I can see column B, what do you have in that?

As I mentioned in the last post, column B contains the LoD coefficients for the equivalent length method. If you are not familiar with that method, the equivalent length = Lod * Fitting Diameter.
For valves, those are just plumbing curves so treated separately.

You asked me the "total". I think you mean number of 90 elbows and 45s. So when in SPA mode I count:

- three 90s and one 45 in the drain line at 3"

There are no 90's or 45s between there and the pad? Normally there would be because a 60' run is not normally a perfectly straight shot. Don't forget about elevation changes as well. Usually, the run goes about 3' underground so there is a 90 right before the pad.

- two 45s in the pad (upstream the pump) at 2"
- eight 90s in the return pad (downstream of the pump) at 2"

I would really encourage you to change the pad to 2.5". At these flow rates, 2" is going to be very noisy and subject to water hammer.

- three 90s and one 45 in the return line at 3"
- Then of course there are two Ts where the actuators are

Can you update spreadsheet with that please? The heater is Hayward HDF400 but I called Hayward but they dont have head loss specs. Important: I see on your spreadsheets ~2.5" for inner diameter of the 3" pipe, that gives higher losses, inner

I am pretty sure I use 3" for a 3" pipe everywhere (3.042 won't make much of difference). However, I used 2.5" on the pad plumbing for the last two scenarios because using 3" on the pad would require a lot of bushings for the pump, filter and heater but should be able to accept 2.5" without bushings.

I have a Hayward heater and it has a head loss curve of 0.002*GPM^2 which is what I used in the model.

I just figured out I gave you the wrong filter. That was a filter cardridge I was offered as an upgrade but plumber said it is not 100% necesary. I currently have (see pic below) the Hayward Star Clear Plus C1200 filter with this cartridge, which says it has 120 sqf. The original Hayward cartridge for that filter is actually just slightly higher price and still 120 sqf, I could get that one to be safer. But please let me know if I should upgrade to the 150s (or even 200s) Swim Clear. From 120 to 150 sqf doesnt seem a big upgrade, beside square footage is there anything else that will improve?

Bigger is always better with cartridge filters. I would go with at least 300 sq-ft. I have 400 sq-ft and only clean it once per year and even then it doesn't really need it.

 

[venexiano]

Thank you a lot,
I am getting there. Yes there is a 90 between the pad and the spa, it was in the math above. A few things:
1) from your spreadsheet, when I put 2" instead of 2.5" in all pad lines I just lose 2 GPM (from 98 to 96 GMP) so I think that is negligible.
2) Could you elaborate a little be more on higher risk of water hammer with 2" versus 2.5" in the pad?
3) Plumber would not be happy with changing all pad to 2.5", so I would do it if really necessary. How do you suggest me to test for water hammer? by turning on and off the pump suddenly? by turning off all the 10 nozzles in the spa? The latter doesnt seem a good idea, but what if some guests does it as a stupid move? Would the pump just stop?
4) Spa seller's drawing says that each jets needs 8-10 GPM. With 10 jets and 96 GPM I should be fine as for "jet action", right?
5) Considering I would use the spa for 1-2 hours per week at most I do not think electric bill for the pump in SPA model is gonna add too much to the normal pool filtration and gas expenses. So, I do not think there is an economic advantage in spending $1200-1300 to upgrade the cartridge filter to 200-325 sqf, what do you think? 200 sqf would be for the C200S, 325 sqf would be for the Hayward W3-C3030, see specs and plumbing curve below from Hayward website and Hayward user manual. Max GPM are below and are around 120 GPM for all three filters. I am not gonna go above that value. Head loss at ~100 GPM is 14 feet for my 1200C, 3 feet for the c200s and 12 feet for the C3030. The c200s seems like a mistake, way too low. Which one do you have? Upgrading to the 325sqf C3030 would just decrease the loss by ~2 feet. Does it seem worth it to you?
6) I purchased this non-invasive flow meter. 5% accuracy. 2.3 feet head loss max. My plan is to measure the discharge downstream the filter (less debris) with the pump at 3450 RPM and get back to you. If we are around 96 GPM, jets intensity is satisfying, not water hammer I should be fine right?
7) Flow meter above requires drilling and I can return it. Alternatively, do you know any easy to use portable ultrasonic flowmeter at accessible price?


Thank you

 

[mas985]

Thank you a lot,
I am getting there. Yes there is a 90 between the pad and the spa, it was in the math above.

Based on the pictures and your post, I updated the model. I notice though you have two check valves, one after the heater and one after the 3-way in the return. Also, I noticed that there is already a heater bypass installed. You could actuate that so that the bypass opens automatically when the heater turns on. Do you have automation? I don't see any actuators, just valves.
Head loss has increased some based upon this new information and I also added a fourth scenario with the 3" Return and 2" Pad plumbing just for comparison:


All are above 8 GPM/jet but none are above 10 GPM/jet.

A few things:
1) from your spreadsheet, when I put 2" instead of 2.5" in all pad lines I just lose 2 GPM (from 98 to 96 GMP) so I think that is negligible.
2) Could you elaborate a little be more on higher risk of water hammer with 2" versus 2.5" in the pad?
3) Plumber would not be happy with changing all pad to 2.5", so I would do it if really necessary. How do you suggest me to test for water hammer? by turning on and off the pump suddenly? by turning off all the 10 nozzles in the spa? The latter doesnt seem a good idea, but what if some guests does it as a stupid move? Would the pump just stop?

2" on the pad was more about pipe velocity than head loss. Higher the velocity, the more noise there is and the greater the chance of water hammer with the check valves. However, given the pad is already in place, you can always change that latter if needed.

4) Spa seller's drawing says that each jets needs 8-10 GPM. With 10 jets and 96 GPM I should be fine as for "jet action", right?

With a VS pump, you can always reduce flow rate but never increase it beyond 3450 RPM. That range is most likely for moderate jets and should be satisfactory. However, have some margin is not a bad idea if possible.

5) Considering I would use the spa for 1-2 hours per week at most I do not think electric bill for the pump in SPA model is gonna add too much to the normal pool filtration and gas expenses. So, I do not think there is an economic advantage in spending $1200-1300 to upgrade the cartridge filter to 200-325 sqf, what do you think? 200 sqf would be for the C200S, 325 sqf would be for the Hayward W3-C3030, see specs and plumbing curve below from Hayward website and Hayward user manual. Max GPM are below and are around 120 GPM for all three filters. I am not gonna go above that value. Head loss at ~100 GPM is 14 feet for my 1200C, 3 feet for the c200s and 12 feet for the C3030. The c200s seems like a mistake, way too low. Which one do you have? Upgrading to the 325sqf C3030 would just decrease the loss by ~2 feet. Does it seem worth it to you?

Again, since the pad is already set up, I would just leave it. If you find the jets a little weak, then you can always redo the pad and change the filter and the pipe. I updated the model for your current filter.

6) I purchased this non-invasive flow meter. 5% accuracy. 2.3 feet head loss max. My plan is to measure the discharge downstream the filter (less debris) with the pump at 3450 RPM and get back to you. If we are around 96 GPM, jets intensity is satisfying, not water hammer I should be fine right?

Right. But for the jets, it isn't just about hitting the right flow rate, it is more about what feels good to you.

7) Flow meter above requires drilling and I can return it. Alternatively, do you know any easy to use portable ultrasonic flowmeter at accessible price?

You can get flow meters that replace check valves which I see you have one after the heater. If you are using the venturi type of flow meters (e.g. Blue-White) they are not very accurate. Flowvis meters are much better.

FlowVis flow meter — H2Flow Controls

FlowVis flow meters are a revolution in reliable flow measurement. With unrivaled average accuracy, greater installation flexibility, and more NSF 50 certified models than any other brand, it’s easy to see why FlowVis® is a brand of flow meter you can put your trust in.

www.h2flow.net

 

[venexiano]

Yes I have 2 Hayward GVA24.

1) Good point about the actuator on the heater bypass. I did not think about it. I could get another Hayward GVA24 on amazon by Saturday. That would decrease head pressure when in Spillway mode and pool mode. I see this cons in the forum: "I see more damage to heaters from an unintended stoppage in flow than the need for a bypass valve. I would also caution against running the heater with a bypass valve in any position other than fully open. The heater needs the full flow to cool the tubes and avoid warping or excessive heat of the plastic manifolds." This kinda scare me, what are your thoughts about it? What if automation messes up? Would automated bypass in Pool and Waterfall mode also extend heater life?

2) Good point about Flowvis! Should I get the FV-2, FV-2-U or FV-2-S? Not sure I understand difference between Union and Coupling.

3) Flowvis has 4 feet head loss at 90 GPM, so, since I am tight with max GPM for jets, it is only worth it if I used instead of an existing check valve (which has similar loss). The only option is that valve circled in the pic below. This way, though, I can only measure flow rate in spa mode and spillway mode, not in pool mode. Which I guess it is fine. The less accurate Blue-white would add two feet of loss but measure everywhere. I could also have flowvis for spa and add Blu-white on the pool side (need to return the 2" one and get the 1.5" model, no enough required space on that short 2" between pool drains) so I do not have extra loss for spa (before adding spa I was recirculating pool at 1200 rpm anyways 24/7 so 2' loss of the Blue-white no biggie).

4) Why is the Suction Dynamic Head Plumbing curve coefficient larger from 3" return and 2" pad than for 2" return 2" pad? See screenshot below

5) Pad size from 2" to 2.5" seems to change discharge very marginally anyways. I see no way to reduce head losses here beside changing the filter to a 200s. I got the following answer from Hayward. Their charts are correct. He said: " The Swim Clear C200S is a single element design with better hydraulic efficiency, but lower overall filtration area. Comparing the Swim Clear larger Capacity C2030, C3030 and larger are a 4 element design with a top and bottom collector manifold internally. The large capacity have a higher overall filtration area, but higher head loss due to the water pushing through 4 elements (cartridges). I hope this helps explain the differences between the head loss factors on these two designs." Unless you know of another 200-300-400 sqf filter with less head loss, the only way to go here to increase flow rate without changing pump would be buying a Swim Clear C200S, which would reduce my head loss by ~10 feet (currently ~12.5, will be ~2.5). What do you think? Not sure how you are computing head loss from the filter on that spreadsheet, so not sure how to change it and use the plumbing curve for C200S and 1200C so I can make a comparison on the actual flow rate gain (I guess it might be a gain of 7-8 GPM given the numbers)

 

[mas985]

Yes I have 2 Hayward GVA24.

1) Good point about the actuator on the heater bypass. I did not think about it. I could get another Hayward GVA24 on amazon by Saturday. That would decrease head pressure when in Spillway mode and pool mode. I see this cons in the forum: "I see more damage to heaters from an unintended stoppage in flow than the need for a bypass valve. I would also caution against running the heater with a bypass valve in any position other than fully open. The heater needs the full flow to cool the tubes and avoid warping or excessive heat of the plastic manifolds." This kinda scare me, what are your thoughts about it? What if automation messes up? Would automated bypass in Pool and Waterfall mode also extend heater life?

Not if you do it right. There are two ways to setup a bypass valve.

One way, which I do not recommend, is to set it up as either/OR so flow goes either to the heater OR the bypass. This is where you can run into issue where flow is cut off but the heater does have built in protection to prevent any damage.

However, you can set up the valve so it basically turns on/off the bypass but there is always flow through the heater. The handle needs to travel 180 degrees to do this. This actually has slightly lower head loss because you have flow through both the bypass and heater in parallel.

Just an additional thought that when using the spa, you will be at full flow rate anyway and the heater may still work fine even if most of the water is being bypassed around the heater. You could give it a shot and see. Otherwise, you may need to have the actuator. Do you know what the minimum flow rate is for the heater?

2) Good point about Flowvis! Should I get the FV-2, FV-2-U or FV-2-S? Not sure I understand difference between Union and Coupling.

3) Flowvis has 4 feet head loss at 90 GPM, so, since I am tight with max GPM for jets, it is only worth it if I used instead of an existing check valve (which has similar loss). The only option is that valve circled in the pic below. This way, though, I can only measure flow rate in spa mode and spillway mode, not in pool mode. Which I guess it is fine. The less accurate Blue-white would add two feet of loss but measure everywhere. I could also have flowvis for spa and add Blu-white on the pool side (need to return the 2" one and get the 1.5" model, no enough required space on that short 2" between pool drains) so I do not have extra loss for spa (before adding spa I was recirculating pool at 1200 rpm anyways 24/7 so 2' loss of the Blue-white no biggie).

What I was suggesting is not to get a full Flowvis check valve but to get a retrofit. That way it isn't adding to the head loss. It is replacing it head loss that is already there. They should be about the same.

Amazon.com

Do you have Jandy check valves?

Note too that you do not need a check valve after the heater because of the SWG. That is a bad spot for a flow meter as well if you have both the bypass and the heater open to flow.

4) Why is the Suction Dynamic Head Plumbing curve coefficient larger from 3" return and 2" pad than for 2" return 2" pad? See screenshot below

There was a wrong value in cell D30 so I fixed it.

5) Pad size from 2" to 2.5" seems to change discharge very marginally anyways. I see no way to reduce head losses here beside changing the filter to a 200s. I got the following answer from Hayward. Their charts are correct. He said: " The Swim Clear C200S is a single element design with better hydraulic efficiency, but lower overall filtration area. Comparing the Swim Clear larger Capacity C2030, C3030 and larger are a 4 element design with a top and bottom collector manifold internally. The large capacity have a higher overall filtration area, but higher head loss due to the water pushing through 4 elements (cartridges).

Actually the four cartridges are plumbed in parallel so they do not add much head loss at all. In fact, when I run by C4025 without the cartridges, head loss is about the same. The head loss is primarily due to the internal manifold and fittings.

The curve for the C4025 is slightly better than your current filter, (0.001 vs 0.0012).

I hope this helps explain the differences between the head loss factors on these two designs." Unless you know of another 200-300-400 sqf filter with less head loss, the only way to go here to increase flow rate without changing pump would be buying a Swim Clear C200S, which would reduce my head loss by ~10 feet (currently ~12.5, will be ~2.5). What do you think?

The issue I see with the C200s is that you will be cleaning it much more often and when a filter gets dirty, head loss can go way up with a small filter. So it may make maintenance a real pain.

Not sure how you are computing head loss from the filter on that spreadsheet, so not sure how to change it and use the plumbing curve for C200S and 1200C so I can make a comparison on the actual flow rate gain (I guess it might be a gain of 7-8 GPM given the numbers)

I am using a plumbing curve for the filter. The nominal plumbing curve for a filter is 0.001. There is a multiplier for that in line 50. So for your current filter, I used a multiplier of 1.2 instead of 1 so the resulting plumbing curve is 0.0012. I duplicated all the scenarios with a filter multiplier of 0.3 which is what the C200s curve shows.

 

[venexiano]

1) Minimum GPM for heater is 30 GPM. I feel I either I leave a manual bypass and I never bypass unless I need maintanane or I now I am not gonna use Spa for long time (i.e. summer), or I do the autmated bypass. Anything in between not worth the hassle I think.

2) I love the idea of that Flowvis replacement on the checkvalve. I do not have Jandy though, I have this Hayward Check Valve. Would Flowvis work on it?

3) Would you put the Flowvis on the check valve after the heater and one after the 3-way in the return? If I put it on the first one, it will measure flow rate both in pool mode and spa mode, but not measure it when I manually bypass the heater in the summer. If I put in the second one, I get only the spa flow rate (not the pool rate) but I have it both with heater and without heater, so I can estimate the heater losses at different flow rates by varying speed of pump, so estimate how much loss I have when in pool mode. Is it easy to screw it from one location to the other or it takes time?

3) As for your new computation, for 3" pipes and 2" pad (my configuration), I see that the new operation point by using the C200s instead of my original 1200C has just 1' difference in total head loss. Upgrading from 1200C to C200s should reduce the head loss by 10 feet around 90 GPM. So I would expect the new operation point to have about 8-9 feet in reduction of head loss, not only 1 foot. Something is wrong there.

Thank you a lot again, and merry Christmas!

 

[mas985]

1) Minimum GPM for heater is 30 GPM. I feel I either I leave a manual bypass and I never bypass unless I need maintanane or I now I am not gonna use Spa for long time (i.e. summer), or I do the autmated bypass. Anything in between not worth the hassle I think.

Keep in mind that for the spa, the heater will only remain on during the initial heat up and occasionally to top off. So it isn't on all the time. An automated bypass is going to be much more convenient.

2) I love the idea of that Flowvis replacement on the checkvalve. I do not have Jandy though, I have this Hayward Check Valve. Would Flowvis work on it?

They didn't use to but it looks like they might.

Amazon.com

You will just need to find on in stock.

3) Would you put the Flowvis on the check valve after the heater and one after the 3-way in the return? If I put it on the first one, it will measure flow rate both in pool mode and spa mode, but not measure it when I manually bypass the heater in the summer. If I put in the second one, I get only the spa flow rate (not the pool rate) but I have it both with heater and without heater, so I can estimate the heater losses at different flow rates by varying speed of pump, so estimate how much loss I have when in pool mode. Is it easy to screw it from one location to the other or it takes time?

If it were me, I probably wouldn't bother. I don't have one on my system. I just go by the alarming of the heater and flow rate estimates. To me, that is good enough.

3) As for your new computation, for 3" pipes and 2" pad (my configuration), I see that the new operation point by using the C200s instead of my original 1200C has just 1' difference in total head loss. Upgrading from 1200C to C200s should reduce the head loss by 10 feet around 90 GPM. So I would expect the new operation point to have about 8-9 feet in reduction of head loss, not only 1 foot. Something is wrong there.

You can't assume the same flow because those two scenarios do not have the same flow rate. Reducing the head loss in the filter, increases the flow rate in that scenario but it also increases the head loss in the rest of the plumbing making up for some of that delta. But you are also getting higher flow rates.

So if you compare column D with Column I, here are the operating points:

Column D: 89.18 GPM @ 86.11'; 0.01083 PC; Filter Head Loss = 9.54'
Column I: 92.52 GPM @ 84.99'; 0.00993 PC; Filter Head Loss = 2.57'

Yes, there is only 1' of TOTAL head difference but there is also 3.34 GPM difference which increases the TOTAL head loss of column I but the filter head loss alone has a difference of about 7'.

To illustrate this further, I created another scenario in column M that matches the flow rate in column D by changing RPM. The change in head loss for these two scenarios is now 7.15' which is exactly the difference in filter head loss alone.