Optimize Your In-Floor Cleaning Effectiveness and Efficiency

[jonpcar]

So, my re-plumbing job is finished and it went really well. I probably missed jonpcar off for his cruise, but will quickly report that the main goal of my re-plumb was achieved. That is, the pressure loss from filter to Caretaker manifold was drastically reduced. I am really happy about that. Instead of losing 7-9psi between these 2 points, I now only lose 2-3.5psi. These readings were taken at the likely cleaning speed of 3000rpm.

I need to take lots more data, but it means that I can now achieve the minimum Caretaker psi of 14-15psi to see how much it helps clean up the dead spots, both on the floor and in the corners. I'm not sure if I will have enough pressure and flow to contribute testing of jonpcar's theory, sending some return amt to the wall, but I will try.

Note that I did NOT install a flow meter. By far the biggest change is the heater bypass, but I also halved the run of 2" pvc and took out more than half of the elbows.

Hey Dodger...back from my GREAT cruise vacation! Now I need time to recuperate, haha. Nice job on your pipe rework. I am going to do mine as well but first I am going to run some experiments on my IFCS suction side that may take some time. Eventually, I want to try to use the bottom drain as exclusively as possible when running the popups since I am convinced that is one key to making these IFCS as effective as possible.

No dead spots in my pool over the 12 days that I was gone...the "swirl" is definitely working in my system to clean up those areas where debris continually falls back down the wall and collects in the same spots.

Any experiments on your side...do you have automation on your suction side (particularly skimmer/drain select)? If not, I would suggest 1) try running with a 75%+ drain/25%- skimmer mix (while using the popups) or 2) ensure that the skimmers "just" operate (with some margin for basket collection blockage) when your VSP is at low speed...this setting should allow increased suction from the main drain when the VSP is at high speed for floor cleaning (using the popups). Not sure if these two different methods are mutually incompatible. That is an area where I am planning to run some experiments.

 

[jonpcar]

I am not just 100% convinced a parallel path will reduce my TDH. Anyway since I do have type 2 heads I can upgrade them to venturi heads which would further reduce the need for increased pressure at the manifold:
Gamma 4 Venturi Power A Manufacturing

The venturi heads seem like a good idea...if I were you, I would just buy a couple to put in your "problem areas" and see how they perform. From your activity on these boards, you are making quite a few changes to your pool...automation of some of your valves would seem to be in the cards? I really like OpenSprinkler as a pool/landscape controller (especially its interface) but for the fact that they only allow 1 or possibly 2 outputs (of upto unlimited outputs) to be active at any time...bummer. It would be an awesome control system.

 

[Dodger]

Any experiments on your side...do you have automation on your suction side (particularly skimmer/drain select)? If not, I would suggest 1) try running with a 75%+ drain/25%- skimmer mix (while using the popups) or 2) ensure that the skimmers "just" operate (with some margin for basket collection blockage) when your VSP is at low speed...this setting should allow increased suction from the main drain when the VSP is at high speed for floor cleaning (using the popups). Not sure if these two different methods are mutually incompatible. That is an area where I am planning to run some experiments.

Welcome back jonpcar. I do not have very usable skimmer/drain valve options. There is no MD/Skimmer Jandy valve. Both skimmers have dual ports in the bottom and the space-ship cover on top. They join underground somewhere and come up as one pipe into the Intake valve.

Skimmer #1 includes a suction line from the main drain. This is the skimmer with much better skimming action, perhaps because it is close to the equipment and/or perhaps because it receives more of the prevailing breezes. I will play with the flap underneath the space-ship to try to shift preference to MD suction, but it's very imprecise and I don't expect much controllability.

Skimmer #2 includes suction from an equalizer port in the wall just below the skimmer. This skimmer barely has enough action as it is, so I've decided against closing down the space-ship flap at all on that side. The equalizer port could theoretically act as wall suction for suspended debris on that side of the pool, though I've never actually seen any visible debris go in there, only dye. Maybe I'll drop some twigs and do a more specific dirt test on that port. (Don't tell Dirk )

I do plan to try a larger nozzle size in at least one pop-up at some point in the off-season. This one problem pop-up contributes to at least 2 dead spots because it doesn't spray far enough. It's a unique situation though - a relic of the replaster we did last summer, such that the pop-up sits in a recessed bowl and the nozzle path may be partially impeded.

 

[jonpcar]

Welcome back jonpcar. I do not have very usable skimmer/drain valve options. There is no MD/Skimmer Jandy valve. Both skimmers have dual ports in the bottom and the space-ship cover on top. They join underground somewhere and come up as one pipe into the Intake valve.

Skimmer #1 includes a suction line from the main drain. This is the skimmer with much better skimming action, perhaps because it is close to the equipment and/or perhaps because it receives more of the prevailing breezes. I will play with the flap underneath the space-ship to try to shift preference to MD suction, but it's very imprecise and I don't expect much controllability.

Skimmer #2 includes suction from an equalizer port in the wall just below the skimmer. This skimmer barely has enough action as it is, so I've decided against closing down the space-ship flap at all on that side. The equalizer port could theoretically act as wall suction for suspended debris on that side of the pool, though I've never actually seen any visible debris go in there, only dye. Maybe I'll drop some twigs and do a more specific dirt test on that port. (Don't tell Dirk

)

I do plan to try a larger nozzle size in at least one pop-up at some point in the off-season. This one problem pop-up contributes to at least 2 dead spots because it doesn't spray far enough. It's a unique situation though - a relic of the replaster we did last summer, such that the pop-up sits in a recessed bowl and the nozzle path may be partially impeded.

Dodger, good to know you have the spaceships...I am actually going to characterize mine in my next set of experiments (if I can get to it in the next couple weeks). I have ordered a suction gauge (negative psi) and am planning to use it to determine gpm flow rate through the skimmer vs the drain using the diverter flap. There is one other change that I am contemplating based on this characterization. I would suspect that the characterization done on my system would be somewhat applicable to others with this Sta-rite/Pentair mechanism, maybe to yours.

But, I’m still not exactly clear how your skimmers/drains all connect. It seems you only have one pipe coming back to your pump (but you called it an Intake valve, is that just the input to the pump or is there actually a valve there, possibly for a spa?). A Is there a dedicated line from the pump to one of the skimmers (possible, but not probable, because then the drain, skimmers and equalizers would all have to be connected via that skimmers 2^nd port)? B Is there line dedicated between one of the skimmers and the drain (doesn’t seem probable, seems like it would still share the equalizer port for that skimmer at minimum)? C Is there one line going from the pump that branches to both skimmers connected underground…possible, still would have the issue with B? D Is there one line going from the pump that branches underground to the drain and both skimmers…possible as well, doesn’t have issue B.

I had to look up what the equalizer ports were…if I were you I would cap both of those ports off (leave a tiny little hole in it so that a small amount of circulation continues to avoid algae growth). Why take away from your skimming? Seems to me their only purpose is so that your motor doesn’t dry run if your water level goes below the skimmer and your float closes off the skimmer drawing air. You probably have a water leveler like I do…I’ve never had that happen in the 25 years I’ve had my pool. In any case you still have the drain feed (I think your description indicates that) to at least one of your skimmers to avoid that issue.

The reason I am asking the questions is because I have a plan for my system, I don’t know if it would apply to yours. My single skimmer has dedicated connections to the pump and to the drain, simple but probably pretty common.

Finally, when they did your replastering, did they replace your drain lid? The new requirements for drains (2008) obsoleted my old one that used to do a decent job collecting larger debris. I got a decent one from the plastering company, but it doesn’t deal with larger debris. I am looking at the PDR2 Retro from A&A (not a single match in TFP forums for PDR2).

 

[jonpcar]

So, I finally had some time this last weekend and ended up re-plumbing my pad so that I could add the "parallel path" to my cleaning system. Here are before and after pictures of what I added.



With the change...



The new In-Floor Cleaning System (IFCS) path allows me to "tune" the IFCS. It actually works very well. For any given pump RPM I can turn the new ball valve to set the pressure at my valve manifold. Water through the valve goes to the wall returns which contributes to the cleaning "swirl" in my pool. This water also increases the effectiveness of the IFCS by running more water through the filter in the same amount of time (at about the same energy efficiency per gallon as discussed throughout this thread). The goal is to find what I believe are the optimal settings for my pool’s cleaning system (rpm, psi, cleaning gpm) and then use that setting. Once I choose the “optimal”, the new ball valve won’t be changed.

I completed this yesterday and so have just started testing. To start with, when cleaning, I am running the valve manifold at 14PSI and a pump RPM of 2760 (pump 80%). This sends about 30 gallons to the IFCS popups and 30 gallons to the wall returns. Currently I am still running my popups for 3 hours a day and this particular case runs at just over 1000w of power usage resulting in 3kWh/day; I will be looking to reduce this (either RPM and/or cleaning time). This will be TBD by how my pool cleans over the testing period.

I made another major change to my system in addition to this. I was unsuccessful in my attempts to characterize the flow of water through the "spaceship" diverter in my skimmer. The PSI was simply too low below the "spaceship" to characterize (I needed a negative PSI valve that went from 0 to -3, pressures were that low). The goal was to determine how much water was being “drawn” from the drain vs the skimmer based on the flap setting on the bottom of the spaceship...mission failed.

So instead, I jettisoned the skimmer "spaceship" and designed a diverter to take its place that (at higher cleaning system gpm volumes) automatically suctions most of the water from the drain when running the IFCS popups, and suctions most of the water through the skimmer when I am skimming/filtering at low gpm. This device allows me to suction from the drain when using the pop ups for cleaning and from the skimmer when skimming/filtering, helping to overcome the fact that my drain and skimmer are not piped back to the pump where I could use a 3-way valve to automatically select between them.

Over the next couple weeks I'll be testing how this new diverter and the pump/valve settings work out.

 

[Dodger]

Can you remind us how this is different than just adjusting your "IFCS Select" actuator to a partially open position and sending some past the waterfall valve and to the wall returns? Didn't that also allow you to tune the pressure at a given rpm?

 

[jonpcar]

Can you remind us how this is different than just adjusting your "IFCS Select" actuator to a partially open position and sending some past the waterfall valve and to the wall returns? Didn't that also allow you to tune the pressure at a given rpm?

The difference is that this change allows the IFCS to be tuneable at ANY given RPM (within reason...the pump RPM has to be at a minimum to drive the popups effectively). The valve added in the picture can be adjusted "infinitely". Remember that I initially started out this thread with my experiments in which I adjusted the actuator on the valve labelled IFCS Select. The actuator had only discrete "stops". With one "click" (I called that the 85/15 case) of that actuator, too little water was being sent through to the wall returns. With two "clicks" of that actuator (I called that the 50/50 case) I needed to raise my pump RPM all the way up to 85% to get my IFCS popups to work effectively...also, there were 86 gallons going through the cleaning system in that case, too much and at the max of my pump recommendation for 2" pipes. Three clicks of that actuator in that original experiment opened the valve way too much...effectively all the water flow went to the wall returns and the popups would not function at all.

I now can set the pump RPM anywhere at 65% or above and then adjust the new valve so that my popups operate at "exactly" 14PSI which is currently the spot where I believe my popups operate effectively. This in turn allows me to set the total flow through the cleaning system (suction flow through the filter using pump RPM and valve adjustments) to anywhere between roughly 30 GPM to 80 GPM, all points which operate near optimum energy efficiency (GPM/kWh). So, total cleaning GPM is what I am experimenting with now...starting at 60GPM...30 to the walls, 30 to the popups.

 

[Dodger]

Okay, so now you have an analog adjustment if I understand. I never did relate to the "click" or discrete adjustment you described. I guess your actuator is different than my JVA. I don't think it has finite, clickable stops (the finite points would only be if a SmartJVA was in place.)


Next question: Are you saying that you can adjust the GPM without the IFCS pressure changing from 14psi? I thought the pressure and GPM at the IFCS valve had a proportional relationship (inverse).

 

[jonpcar]

Okay, so now you have an analog adjustment if I understand. I never did relate to the "click" or discrete adjustment you described. I guess your actuator is different than my JVA. I don't think it has finite, clickable stops (the finite points would only be if a SmartJVA was in place.)

Exactly! I looked up SmartJVA and you are right...with the actuator on my Jandy valve, I only have a finite number of set points, like your SmartJVA.

Next question: Are you saying that you can adjust the GPM without the IFCS pressure changing from 14psi? I thought the pressure and GPM at the IFCS valve had a proportional relationship (inverse).

Not quite...with my new valve closed, this is the same as my previous 100% case where at 65% max pump RPM, I had approx 14 PSI at the manifold and was sending 31 gallons through the popups. See this post for the numbers:

Optimize Your In-Floor Cleaning Effectiveness and Efficiency - Page 6

In order to increase the cleaning GPM (by sending water through the wall returns as well), I must "UP" the pump RPM. My pump RPM is controlled by my Hayward Ecommand 4 and can only be adjusted in 5% increments. So, if I increased it (from 65% in the prior paragraph) to 70%, I then open the new valve slightly to bring the pressure at the manifold back down to 14psi which results in flow through the valve and an increase in total cleaning GPM. At 80% (where I am running currently), I had to open up the valve even more to bring the manifold pressure back down to 14psi, total flow through the system is currently 60gpm (as I said previously, approximately 30gpm to the popups, 30gpm to the wall returns).

 

[Dodger]

Got it. I hope the ball valve works for you. I have one on each of our 3 sheer descents and they get very hard to move, let alone to move with any precision. I think it gets worse as they age in the heat, even if coated to protect from sunlight.

 

[jonpcar]

Dodger, my claim that I can adjust the total cleaning gpm between 30-80 gpm might have confused you a bit. With my current pump control, it is somewhat granular (stipulated by the 5% pump increments). I am actually thinking about changing this for a different reason (there are not enough time controls for my pump with my Ecommand 4). But even with the current system, I can adjust the gpm by targeting slightly above/below the 14psi manifold level (say 13.5 to 15) to reduce/increase flow through the wall returns (and thus, total cleaning gpm) at the expense of possibly losing some cleaning effectiveness or energy efficiency.

- - - Updated - - -

Got it. I hope the ball valve works for you. I have one on each of our 3 sheer descents and they get very hard to move, let alone to move with any precision. I think it gets worse as they age in the heat, even if coated to protect from sunlight.

I debated a Jandy valve here but thought it was overkill. I am going to find the mythical “optimum” setting for my pool and then leave it, I only need the valve to make adjustments to find what works best for my pool. My pump equipment has a solar shade over it so I'll hope for the best. I also installed these ball valves on my three waterfalls...once again as a “set and forget” fashion.

 

[Dodger]

Got it and absolutely agree that as long as you don't have to change it often, the ball valve is fine.

Your pump is different than mine in that I can control to the single digit rpm, as opposed to percentage increments of the max. So your granularity has some limits I don't have. I just wish I had more granularity on the IFCS pressure gauge!

 

[jonpcar]

Dodger…when you talk about granularity of the IFCS manifold gauge, do you mean that you wish you had more control over it, or do you wish the PSI was simply higher so that you can play with it?

Backtracking a bit. For energy analysis, I used one starting back in the 2^nd post of this thread that was based on Gallons/kWh. Unknown to me at the time was that there is already a standard, called Energy Factor (EF), for pool pumps that is essentially the exact same number I calculated, but divided by 1000. I will use EF from now on. The higher the EF, the more energy efficiently you are operating your pump at. Generally, lower pump RPMs result in higher EFs, until a limit is reached. For a given pump operating state, EF is calculated as follows:

EF = (Pump GPM)* 60 / (pump watts)

EX: If your pump is moving 70gpm and using 762 watts, the EF is calculated as follows:
EX: EF= (70 * 60) / 762 = 5.70

To determine the number of gallons pumped by your pool pump per kWh (this is what the electrical company charges for), simply multiply the EF by 1000. In our example: 5.70 * 1000= 5700 gallons/kWh.

I ran some experiments today and am going to post some charts of my system’s EF vs other numbers. My In-Floor Cleaning System (IFCS) was one of the early ones built in 1993. As a result, there may be some disadvantages that my system has that are fixed in more modern IFCSs, but I am not sure because I haven’t seen any other pool data. Disadvantages in my system are that I have older popup heads which can’t be upgraded (A&A type 1), 1.5 inch return lines to the popups (45-50 max gpm recommended for that size pipe). Advantages I have in my system: relatively short 2” suction pipe from skimmer/drain to the pump with few fittings, large filter to reduce pressure loss, simple system with no heater, and low loss 2” return path via wall returns.

Here are two of the graphs and the experimental data tables used to create them. Currently I am operating under the assumption that my IFCS needs 14psi at the manifold to properly drive the popups and clean my pool effectively; 14psi at the manifold sends about 32gpm back through my popups. Point A on the charts is where I am currently operating my pump when running the popups (14psi @ manifold, plus sends 30 gpm back through the wall returns via my new IFCS bypass valve shown in the picture a few posts back). Point B on the charts is where I am currently operating my pump when skimming/filtering, all through the wall returns. Point C on the charts is where I would most efficiently operate my pump IF I ran 100% of the return through my popups (14psi @ manifold, 0 gpm back through the wall returns). Point D on the charts is the approximate location that I ran my single speed 1.5hp pump for 21 years, and lost all that money because of the inefficiency...because I could have been running at point A by simply opening the wall return valve slightly.

There are some interesting conclusions for my system, and possibly others, that can be gained by looking at those curves, but I don’t have time to comment on that this afternoon. I believe that the 100% wall return curve (blue) and the 100% popup return curves (green) would be very useful to graph for those with Variable Speed Pumps (VSPs) in their IFCS. I also have some other plots that I hope to post later.

Pool Return r100% through Wall
RPM	GPM	Watts		EF
3450	108	2385		2.72
3277	104	2046		3.05
3105	101	1727		3.51
2932	97	1448		4.02
2760	91	1202		4.54
2587	86	989		5.22
2415	79	807		5.87
2242	73	648		6.76
2070	67	520		7.73
1897	60	415		8.67
1725	53	331		9.61
1552	46	264		10.45
1380	41	212		11.60
B 1207	35	161		13.04
1035	28	136		12.35
862	21	136		9.26
690	12	131		5.50

Pool Return r100% through Popups
RPM	GPM	Watts	mPSI	EF
3450	48	1602	30.5	1.80
3277	44	1377	27	1.92
D 3105	41	1179	24	2.09
2932	39	989	21	2.37
2760	37	836	18.5	2.66
2587	34	691	16	2.95
C 2415	31	573	13.75	3.25
2242	29	469	12	3.71
2070	27	384	9.75	4.22
1897	24	315	7.75	4.57
1725	21	257	6.5	4.90
1552	19	212	5	5.38
1380	15	169	4	5.33
1207	11	140	3	4.71
1035	9	136	2	3.97
862	8	131	2	3.66
690	7	131	2	3.21

The way I generated the next table was to adjust my new IFCS bypass valve so that at a pump RPM of 2760, the manifold PSI=14. That bypass valve position was not changed (it was left in its partially open position) as I then ran the pump through its RPM range (from 3450 down to 690 in 5% increments).

TEST POINT #1, 14PSI @ 2760RPM
RPM	GPM	Watts	mPSI	EF
3450	79	2018	23	2.35
3277	76	1743	20.5	2.62
3105	72	1487	18	2.91
2932	68	1225	16	3.33
A 2760	64	1035	14	3.71
2587	59	851	12	4.16
2415	54	695	10	4.66
2242	50	565	8.5	5.31
2070	46	462	7	5.97
1897	41	367	5.5	6.70
1725	36	296	4.76	7.30
1552	31	241	4	7.72
1380	28	194	3	8.66
1207	24	152	2	9.47
1035	20	136	2	8.82
862	15	131	2	6.87
690	9	131	2	4.12

 

[jonpcar]

Keep in mind that I am documenting these experiments because I am having fun with it and hopefully some can gain more insight into how their IFCSs may be improved. In-floor cleaning systems are not actually that complex, but it may seem that way with all this analysis. I have learned much about my system and despite its limitations, I am extremely happy with its performance and cost effectiveness.

All of the following is discussing the two charts in the previous post...

My IFCS (and probably many others) was poorly designed. My green curve (r100% popups) shows how restricted the flows are through my ICFS. At my current “optimum” point (manifold @ 14psi), I only get 31gpm through my filter. Even at a higher, inefficient manifold psi, my IFCS popups max out at less than 50gpm. I attribute this to 3 things: (a) small 1.5" manifold valves (b) 1.5” piping to popups between the manifold and popups (c) restrictive popup heads: older technology AND only two popups per zone (larger-hole/flow versions). Newer (than mine) IFCSs are probably implemented using 2"+ piping/valves throughout the popup system and should significantly change their characteristic curves. I would love to see the data from someone with those systems.

Here are some generalizations for those with curves that may resemble mine:

One speed pump IFCS systems

One speed pumps IFCS systems won't have curves associated with their 100% Wall Ret (blue curve) and 100% Popup Ret (green curve) operation ...there is only a single green point (a C/D point) and a single blue point (a B point) for their operation. There are multiple A Points possible for them.

Point D on the green curve (100% popups) was “approximately” the spot that I was limited to for 21 years with my one speed, 1.5hp pump. If your Point D looks as bad as mine, you are probably not happy with your IFCS unless you run it for many hours a day and simply accept the added cost to your electric bill. I did that for many years and spent minimal time taking care of my pool. However, it is fairly obvious that Point D is not an energy efficient operating point; nor does it clean the pool effectively.

Your system would benefit if you can successfully "tune it" by finding a Point A. This is done by slightly opening wall returns/jets appropriately. This would result in a more efficient EF, AND your pool would clean faster.

Two speed pump IFCS systems

Two speed pumps IFCS systems won't have curves associated with their 100% Wall Ret (blue curve) and 100% Popup Ret (green curve) operation ...there are only two green points (C/D points, high rpm and low rpm ) and two blue points (B point, high rpm and low rpm) for their operation. There are multiple A Points possible for them.

At high speed, two speed pump systems have the same problem as one-speed-pump IFCS systems when 100% of the return is through the popups: that operating point is probably not optimized for your system. Also, unless your system has automation to switch returns from popups to walls/jets when operating at low speed, it is doubtful that running at low speed pump mode will result in significant increases to your EF (at the low RPM green point). The IFCS return path is simply too restrictive.

Once again, your system would benefit by "tuning" to a Point A when using the popups at high pump speed. In addition, at low pump speed, your resulting red curve (actually only a point in this case: the low rpm A point) may allow significant increase in EF when operating at that point (even without automation), but this is probably a stretch. Without seeing a system that successfully accomplishes an increase in EF at low speed, I wonder if it is really possible...my personal opinion is that a 2-speed pump will provide little benefit to an IFCS unless valve automation can be used at low speed to switch over to 100% wall return (a blue point).

Variable Speed Pump IFCS systems

These systems can easily generate the green (100% popup) and blue curves (100% wall return) by varying pump rpm and recording the gpm/wattage. By creating these two curves it is easy to see how much disparity there is between them..my system has a huge disparity. For systems with less disparity between the two curves, potential "gains" may be correspondingly diminished. In my system, there is no way to take full advantage of the low rpm of my VSP if I am limited to operating along the green curve. The highest EF I can attain running through the popups is about 5.5.

Automation is a requirement in order to use my VSP at low RPMs and obtain the highest possible EF (Point B) for skimming/filtering. This is accomplished by using a 3-way valve to switch over to the blue curve, 100% Wall Return. For those systems without automation that successfully can find a Point A (partial popup return AND partial wall/jet return), higher EFs can be obtained when operating the VSP at low RPM (in my system about 9.5 at the 1200 rpm point, red curve vs green curve)…but still not as high as can be obtained with automation (blue curve vs green curve).

My opinion is that there are limited gains to using a VSP for ICFSs which have a large disparity between the 100% popup and 100% wall return curves (green and blue curves), UNLESS automation is involved and/or a Point A can be found.

 

[jonpcar]

I am going to post a couple more charts and it’s going to get more complicated, haha. It seems like a lot of data, but there is a lot to explain. Taking this data was quick (about 45 minutes) because I simply had to set my new bypass valve, set my pump gpm, and then record my Flowvis GPM and pump watt usage. Interpreting and graphing the data is what took more time…

How do I choose/determine my Point A? For me, this was a combination of things.

(1) Through these experiments, I realized that my system (pump, filter, pipes) could support up to about 100gpm. However, my system is implemented using 2” piping. On the suction side, I have all 2”. On the return side I have 2” returns when using my wall returns. When using the popups, I effectively have 2”+ returns only if I open up my wall returns in parallel to running my 1.5” popups. For more modern IFCSs than mine, I would hope that most plumbing (including the popup path and manifold valves) is implemented using 2”+.

My Ecostar pump recommends an 80gpm max for 2” pipes. Also, the widely recommended/accepted limit for 2” pool pvc pipes is an 80gpm flow rate. Even though my system’s pipe lengths are relatively short, I don’t want my system to exceed those recommendations. Note that all my gpm measurements have been made when ONE particular popup zone is fully “active”. When the zones “transition”, the gpm increases by about 10-12, so I need some margin to compensate for that. Thus, I have decided to run my IFCS in the 60-70 gpm range.

(2) Over the course of the past month and half, a manifold pressure of 14psi seems to be adequately cleaning my pool. It may not be the optimum, but that is my starting point.

(3) The most important factor is testing that Point A actually cleans my system effectively (which I am still currently doing).

 

[jonpcar]

Cleaning GPM (cGPM) is a number I defined that is the total gpm being filtered while I am operating my IFCS popups. For me (because I am operating at Point A), this number includes return flow through my popups AND through my wall returns. I have charted a few different IFCS configurations in the diagram below that show how adjusting my new IFCS bypass valve and pump rpm affects my cGPM.

Below, the green curve is derived from the same data as the green curve in previous charts…it shows the gpm through the popups when 100% of the return is sent back through the popups. For my system near Point C (pump RPM=2415, mPSI=14), about 31gpm is pushed through the popups. At p80 of the green curve (pump RPM=27260, mPSI=18.3), about 38gpm is pushed through the popups. This is the starting point from which I determined my Point A, the same Point A as in the previous charts.

With my new “tuning valve” closed, I started at point p80 (pRPM=2760, mPSI=18.3). [Now follow along to the upward left along the black curve] I then opened the valve slightly until mPSI is reduced to 18. This increased cGPM to 40gpm, about 37gpm to the popups and 3gpm through the wall returns. Next, I opened the “tuning valve” more until mPSI decreased to 17, which set cGPM=47. I kept opening up the “tuning valve” and recording the results. At Point A (mPSI=14, cGPM=62), about 32gpm is being sent to the popups and the remaining 30gpm is sent through the wall returns.

This spot hits my first two measurable targets for Point A: (1) cGPM=60-70 (2) mPSI=14. I continued measuring to complete the process for manifold pressures between 10-20psi. This process completed the black curve. As can be seen, I also went through this same process for other pump rpms.

If I decide that 14psi is NOT my optimum manifold pressure, I can choose a new Point A on one of the other curves to hit my goal of cGPM=60-70. For instance, if I decide I need more pressure for my popups, I could move to the p85 point (pRPM=2923, mPSI=16) or possibly the p90 point. If I decide I need less pressure (say mPSI=12), I might move to the p80 point or possibly the p75 point.

Systems don’t have to be tuned this finely to take advantage of a Point A, but this level of granularity (gpm, psi, watts measurement) does instill some confidence in the result.

Keep in mind, poorly designed IFCSs like my pre-automation version don’t need to install a special tuning valve like I did for my current system to take advantage of this method…if there is a return valve that can add a parallel return path to your popups (wall/jet returns, venturi skimmer, water feature but that effects PH), it simply needs to be appropriately “opened”.

 

[jonpcar]

Last post on these latest experiments…

Previously in this thread, there was some concern that increasing the cGPM (by running flow through both the popups and the wall returns) might decrease the energy efficiency of cleaning. As it turns out, the data we collected previously indicated that it was essentially a wash in my system. But I took more data to make sure that was the case.

My system may be a bit unique in this case because it is simpler than many. I don’t have pressure loss from a heater, small filter, extra pvc fittings, etc. But, until there are more systems with data, we can’t be sure.

The following graph shows the Energy Factor (EF) vs the manifold PSI (mPSI). It is populated by data points that are from the same charts as all those curves on the last graph (all the color curves match). There are five different configurations of the IFCS shown (various pump rpms, various settings of my bypass valve, etc). What we see is that for all those curves, there is relatively little difference in the EF…all probably within the noise of the data accuracy.

Look at the mPSI=15 values, they are shown in the box. Despite a wide range of gpm differences (33-85gpm), the EF for all the curves is close enough to be a wash (3.1-3.5). Other mPSI points are similar. The 100% popup case (green curve) seems to have the lowest EF for all curves but I think that is an artifact of the FlowVis itself. Mark has pointed out that devices like that (including check valves) have a larger impact on lower gpm flows than higher gpm flows. As we know, the green curve (100% popup return) has the lowest gpms of all the curves displayed.

Based on the results of my system, I suspect there are many systems whose EF would not be impacted by increasing the cleaning GPM in the fashion that has been described in the last few posts.

That’s it for the data on my new “bypass” valve. I am planning to use these charts/data to continue my search for an optimum cleaning point (Point A) over the next 6-8 months. I’ll update on that as I find out more. I may take out my FlowVis at this point to see how much it actually impacted some of my data, but probably won't rush to do it. I am also continuing experiments on the suction side of my system (with my new skimmer diverter) and looking into the possibility of replacing my drain cover to one that is more "debris friendly".

 

[Dodger]

First of all, just want to let you know that someone is reading along.

Dodger…when you talk about granularity of the IFCS manifold gauge, do you mean that you wish you had more control over it, or do you wish the PSI was simply higher so that you can play with it?

I was simply referring to the resolution on the manifold pressure gauge. I don't like that it's analog and I don't like that the notches are every 2psi. Makes the pressure readings somewhat subjective and the margin of [human] error is unsatisfactory.

Note that all my gpm measurements have been made when ONE particular popup zone is fully “active”. When the zones “transition”, the gpm increases by about 10-12, so I need some margin to compensate for that.

Since water is flowing through 2 zones during the transition, doesn't that remove your 80 GPM limit? I don't think you have the possibility of 80 GPM through either zone.


Edit: Sorry, when I wrote above, I was thinking of the pipes, but realize that you are probably talking about a pump limitation. Is cavitation the concern when pushing too much GPM?

 

[KDpoolguy]

To start with, when cleaning, I am running the valve manifold at 14PSI and a pump RPM of 2760 (pump 80%).

I made another major change to my system in addition to this. I was unsuccessful in my attempts to characterize the flow of water through the "spaceship" diverter in my skimmer.

So instead, I jettisoned the skimmer "spaceship" and designed a diverter to take its place that (at higher cleaning system gpm volumes) automatically suctions most of the water from the drain when running the IFCS popups, and suctions most of the water through the skimmer when I am skimming/filtering at low gpm. This device allows me to suction from the drain when using the pop ups for cleaning and from the skimmer when skimming/filtering, helping to overcome the fact that my drain and skimmer are not piped back to the pump where I could use a 3-way valve to automatically select between them.

Jon,
We appear to have very similar setups as far as A&A manifold & Hayward VS and “spaceship” diverter (mine is Pentair 08650-0079 Float Assembly for Sta-Rite U3 Skimmer). My IFCS system was gutted after years of poor management, and didn’t even know about vital need for float diverter until maybe a year ago. Now my in-floor works much better by forcing more intake from main drain. I don’t know which adjustments of weir on float diverter are ideal, but my skimmer takes in a good amount about 2/3 in favor of main drain.

I’d love to see your new 3 way valve for this, as I’m assuming it’s via a motorized Jandy-style valve? Mine is buried and I don’t have ability to adjust any suction away from main drain, it’s all plumbed into skimmer (hence the float). If your access is above ground that explains it, but perhaps you’re thinking of something I’m not.

I also run my Hayward VS for 3-4 hours at 2750rpm and get excellent floor cleaning. Then run at 1100rpm for skimming/circulation for another 7-9 hours
-Kevin

 

[jonpcar]

I was simply referring to the resolution on the manifold pressure gauge. I don't like that it's analog and I don't like that the notches are every 2psi. Makes the pressure readings somewhat subjective and the margin of [human] error is unsatisfactory.

Is it easily replaceable? They are pretty cheap on amazon and the one I put in, 0-30PSI, can be read in roughly .25PSI increments (of course not for absolute accuracy, but for comparison purposes). It’s pretty easy to tell when it is the middle of the 1 PSI marks (.5), or “not quite” in the middle to one side (.25, .75).

One advantage I have possibly is that I stop my valves from rotating when I am doing these manifold psi readings so it is pretty rock solid. I believe you mentioned one time that your valve manifold does not have that capability?

Since water is flowing through 2 zones during the transition, doesn't that remove your 80 GPM limit? I don't think you have the possibility of 80 GPM through either zone.

Edit: Sorry, when I wrote above, I was thinking of the pipes, but realize that you are probably talking about a pump limitation. Is cavitation the concern when pushing too much GPM?

Yes, you are right…I am concerned mostly about the pump recommendation but also the suction pipe (from the skimmer/drain). Not sure why my pump documentation has that 80gpm recommendation for 2” pipes; I don’t think it’s a cavitation concern but not positive; I think it’s more that it is pushing the pump to extremes. A general google search on pool pipe flows shows that there is pretty much consensus that a 2” pipe should only carry about 80gpm. That pipe flow recommendation probably is based on “typical pipe lengths” for a pool install. There is also a concern about the flow rates at the main drain and the dangers for entrapment/entanglement, etc.

Over the last month or so, I have developed a 2nd regret in the remodel of my pool...1st was that I didn't try to route the main drain/skimmer pipes separately to my pump. The new one is that I didn't get a channel drain put in when I had my pool surface refinished. Much better for IFCS, and safer too.

Related to the 2-zone manifold transition time and flow during that time…I am still planning a special “manifold cam” to increase the flow during the transitions that I mentioned in a previous post, though I don’t know when I will get to it. I have realized through this latest round of experiments that there is not quite as much to gain as I previously thought. Eventually, I am going to have my son 3d print a special manifold cam to test this theory out…just for fun, haha.

Dodger, I know you didn’t put in a FlowVis, have you done any gpm calculation using Marks tables to look at your EF ratings? Also, do you run at lower speeds (for filtering/skimming) through your wall returns, or do you do that through your popups? You did see that there was a huge difference in my system for the EF between those two cases...yours is probably better but just wondering if you checked it.

Kevin (KDPoolGuy), the “gadget” I have designed as my skimmer diverter (to replace the spaceship) is pretty specific for my pool. Don’t know that anyone would want to try and duplicate it but I’ll take a couple pictures and post tonight. We have the grandkids today.