- Aug 13, 2010
- 74
Understood. Ok then, I have the option to... Um... Decrease the efficiency of my filter! Alright, I'll stick with the energy saving thing.bk406 said:
teapot said:
What's "glass media"?
Bernoulli vs. the pool guy
- Thread starter Simbilis
- Start date
What's "glass media"?
This is like saying that the resistance of transmission lines (and resulting heat losses) in the electric power grid shouldn't matter because it's going to end up powering simple light bulbs with lots of resistance at the end of the line. It simply does not work that way. There are energy losses in BOTH the transmission line AND in the light bulb just as there are frictional energy losses from water flowing through pipe AND coming out of a narrow return. The losses in the pipe would only be unimportant if the pipe run was short such that its losses were far smaller than from the return or when the flow rate is low such that the absolute savings is much smaller (though even this adds up when one considers 8 million pools in the U.S.).X-PertPool said:
A short but narrow constriction just increases the velocity of the water but does not hugely increase the head pressure or reduce the flow rate of the system unless, of course, the size of the restriction becomes extreme such that it dominates the overall resistance to flow. It definitely has an effect (as I describe below), but is not dominant unless pipe runs are short or flow rates are high (such as having only a single return).
My pool has directional eyeball fittings with a diameter of perhaps 3/4" and though this has a significant head loss as a result, this is around 7.4 feet of head (assuming K=1.0 for pipe exit, 0.75" diameter, and 30 GPM per each of the 3 returns and using this link for K factors of fittings and this link for a calculator of head loss for fittings). However, it is still only a part of the total system with the length of straight pipe still being the largest factor in the total dynamic head of the system. In my example it's around 7 feet for fittings vs. 15 feet for smaller pipe or 5 feet for larger pipe, though there are other losses I didn't count that make the total head loss much larger than that of the fittings. There is no question that the diameter of the eyeball fitting makes a difference. A 1" eyeball would have 2.3 feet of head while 1.5" would have only 0.5 feet of head (assuming the same 30 GPM flow rates -- in practice, the flow rates would be somewhat different if a fixed-speed pump were used since overall system head would be lower with narrower fittings, though head would be higher so energy usage would still be fairly high).
If one doesn't split to multiple returns and instead has a high flow rate of 90 GPM through one return, then a 3/4" opening is 66.5 feet of head, 1" is 21.0 feet of head, and 1.5" is 4.2 feet of head. I suspect that the actual K factor for an eyeball fitting is higher than a simple pipe exit, but I can't find what it is though sudden reducers are worst-case up to a K of 2 (so 2 times the feet of head I calculated above though most reducers have lower K), but the point is still the same that the piping is a significant, if not the dominant, factor in the overall resistance to flow.
This just underscores the importance of using larger pipe AND larger fittings especially when there aren't multiple returns. In fact, the use of narrow eyeballs is just a waste of energy. Circulation of water in the pool is due to the flow rate and the only thing that narrow fittings do is produce higher velocities of water in a narrow portion of the pool, but since the flow rate (GPM) is no higher this is very wasteful in terms of energy. Because water is an essentially incompressible fluid, there is no significant benefit to having a narrow high-velocity stream into the water vs. a wider slower entry at the same (or higher) flow rate since the same volume of water gets moved. By using wider fittings/returns and larger diameter pipe, the total dynamic head can be significantly reduced letting one use a smaller pump (or lower RPM in a variable speed pump) for substantial energy savings.
Simbilis said:
It's a recycled (imploded) glass made to the correct size for filtering water (course to cover the laterals and finer to make up the bulk of that filter). It has a lower weight density so requires approximately 15% less media and lower frictional loss.
I am not trying to sell it just commenting. There are stacks of information on the web with recycled glass or eco glass filter medium.
Now there is a proliferation of sellers but that is just crushed/imploded glass. I am using AFM from dryden aqua in Scotland, their glass is then re-treated "zeolitification" Drydens own treatment which improves the -negative charge on the glass which means it not only attracts small organic particles but actually has a high surface oxidisation which prevents bacterial growth within the filter. Ordinary crushed/imploded glass is better than sand but does not have the added plusses of AFM.
The bad side is the cost AFM costs a fair bit more than ordinary glass which costs more than sand and some people are not going to pay the extra.
Sand by the way DOES NOT wear out in filters, That is the truth it gets sticky with bacterial biofilm but can be washed clean. Whether it is worth the time to clean it compared with the relatively cheap replacement is down to pool owners but it does not need changing every few years it just needs cleaning.
Simbilis said:Heckpools said:
Yeah, I said that. He said "but that can never happen!"
I can show you a 3x4 foot patch in the concrete around my pool due to the pool builder using a Drain Pipe PVC elbow instead of a Pressure Pipe PVC elbow on the return line. On a positive note it did last 30 years before blowing out.
I understand the flowrate remains the same but is there any advantage to eyeballs in the sense of moving surface debris quicker to the skimmer?
X-PertPool said:
Absolutely, thats the point of using an eyeball to begin with...increased water speed at the surface.
I think what is being lost here in all these theoretical discussion of flowrates and water speeds is this is for a pool! It isn't the space shuttle, your not going to save hundreds of dollars a year in electricity going to a larger size plumbing setup. At best, if you remove the eyeballs...your going to turn the water over a little faster through the filter because the eyeballs are the dominant restriction in the system but the plumbing connector at the pool wall will be reduced? Who runs their pool without eyeballs and ruins the surface speed skim effect anyway? Plus we haven't even added in the restriction of the filter have we? I don't know about you folks but my pump is working at 10psi when the filter is clean and 16psi when its dirty (My gas heater pressure switch kicks out at 20psi), just for giggles I threw a pressure gauge on my return side where my slide outlet is today and it read zero on a 0-5psi gauge...no restriction there.
The analogy of electrical transmission lines doesn't apply to this situation because your aren't reaching the limits of the plumbing system where its capacity is taxed or the runs are miles long where line resistance becomes a factor. This is kind of silly because if you did the kilowatt/hrs calculation of how much harder a pump has to work with the smaller pipe it works out to pennies...not dollars per year.
The only analogy that definetly applies to this situation was already mentioned;
"I guess dudes will always talk about their pipe size and whats better.
"Sorry guys/gals, I'm a "bigger is always better" kinda guy but in this case it holds no water...figuratively speaking of course. :lol:
- Aug 13, 2010
- 74
4JawChuck said:
I don't agree that the electrical transmission line analogy is wrong. Neither power lines nor pool pipes are (or should be) run near the limit of their capacity. Chem Geek was just saying that in each case, line losses and load both need to be taken into account.
Chem Geek and mas985 have both posted with numbers showing that upsizing pipe diameter by 1/2" will reduce head loss due to flow resistance substantially, lessening some risks and yielding meaningful cost savings. I can follow and reproduce their calculations. I have to assume that they're framing the problem correctly, but their positions are consistent and seem well-supported.
If I read you correctly, You're arguing either that their interpretation is incorrect or their numbers are off. You believe line losses are such a small fraction of the total that they can be ignored. Could you please point out where they've gone wrong in disagreeing with that position?
Thanks,
Sim
The eyeballs are not the dominant restriction in the system in all pools. Yes, they are the narrowest piece, but for a very short distance. They do not contribute the largest part of resistance to flow rate compared to long pipe runs (unless the pipe diameter is large) though as I showed, they do contribute a significant, if not majority, portion. If you were to remove the eyeballs you would find an increase in flow rates with a fixed-speed pump (or a lower RPM and energy costs if you keep the flow rate the same with a variable-speed/flow pump), but if you were to instead put the pump right at the return with minimal piping (or if you would use large diameter pipe) you would find you would get an even larger increase in flow rate for fixed-speed pump (or an even lower RPM for a variable-speed/flow pump). Basically, if one significantly reduces resistance in piping and fixtures/returns, then one can achieve high flow rates at low cost (energy) and it is the high flow rates that should give better skimming action.4JawChuck said:
One does not have to reach the limits of the plumbing system, by which I'm guessing you mean the maximum velocity "limit" that is set to avoid noise and damaging wear and tear of pipes and fittings, nor does one have to reach the limit of pressure to prevent bursting of the pipe. Resistance to water flow is due to friction between the walls of the pipe and the water in the pipe, plus resulting side effects of turbulent (vs. laminar) flow in the pipe as a result and this doesn't just occur near "limits". I don't know why you think it cannot be a substantial part of head (pressure) loss in the system requiring larger (higher HP) pumps at higher energy cost (or higher RPM and higher energy cost in a variable speed/flow pump). As shown here for Schedule 40 PVC pipe, at 50 GPM 1.5" pipe loses 14.3 feet of head (6.2 psi) for every 100 feet of pipe while 2" pipe loses only 4.2 feet of head (1.8 psi) per 100 feet of pipe and 2.5" pipe loses only 1.8 feet of head (0.8 psi) per 100 feet of pipe.4JawChuck said:
Every pool is different, but in my own pool I would be saving a lot of money every year if larger plumbing for pipes and fixtures/returns had been used. I used to spend $1400 per year in electricity costs with the original equipment setup my PB had installed. By replacing the 1 HP (1.65 SF) Jandy HHP main pump and the 3/4-HP booster pump (for the pool sweep) with an IntelliFlo VF pump, I've cut my electric costs for the pool roughly in half to $700 per year. However, I still have long pipe runs to/from the pool and to/from the house and roof to the solar system and these contribute a substantial portion to the operating cost, especially when the solar is on since I have to run at 48 GPM for 4 GPM per panel in this case which results in around 1500 Watts (with the solar off I'm running at 26 GPM which is only around 275 Watts). Now, granted, electricity costs in my area are high with marginal rates at over 40 cents per kilowatt-hour, but even where rates are low it would be more responsible ecologically to use less energy. The savings I would get from larger piping and fixtures would not be pennies per year, but hundreds of dollars per year. I've added up the losses and the pipe sizes and fixtures/returns are a substantial part of the overall loss. If I wanted to, I would also be able to run with the solar off at higher flow rates for the same energy cost.4JawChuck said:
I have an oversized Jandy CL 340 square feet cartridge filter (a "4-cart monster" as waterbear used to put it) where the pressure doesn't show any rise even over a year of use so I clean it once a year which seems sufficient. Even at 90 GPM, the head loss is only around 5 feet of head (see page 13 in this link). I have a mostly opaque electric safety cover so the pool stays fairly clean and the idea of surface skimming isn't applicable except when the pool is open 1-2 hours every day during the week and longer on weekends. Even so, on windy days when leaves blow in when the pool is open, they do mostly end up in the skimmer even from the low flow rates. I am not at all convinced that having a larger diameter return would significantly lower skimming action. Leaves would just end up stagnant in the middle of the pool away from the returns if their movement was solely dependent on the high velocity stream from the eyeballs rather than the overall flow rate circulation pattern. In fact, that flow pattern away from the returns has very little to do with the eyeballs and has much more to do with the circular flow pattern from return and skimmer placement. I do think that eyeballs can be useful to direct flow to help eliminate dead spots or to point them up when wanting to aerate the water (such as for lowering the TA with acid/aeration). So in my situation, having returns with larger eyeballs would have been a reasonable compromise (along with larger piping).4JawChuck said:
I don't understand what you did here. Are you saying that you tapped the return line near where it goes into the pool with an eyeball outlet and it measured zero PSI? What is the GPM on this line and what is the diameter of the eyeball? Isn't your result inconsistent with saying that the eyeballs are the dominant restriction in the system?4JawChuck said:
Richard
Sigh.
There are two major restrictions in any pool plumbing setup, the eyeball returns and the filter. The filter represents 95% of the flow rate restriction in the plumbing while the eyeballs represent less than 5% since they are after the filter and free flowing to almost atmospheric pressure (about .5 psi of backpressure at 1ft below the water line). Frictional restrictions due to plumbing sizing is virtually non-existent unless you plumb the eyeball fittings with the same size plumbing as the exit port diameter I.E. 3/4", 1"...which nobody does.
Therefore if the plumbing to the eyeballs is 1.5" and your eyeballs are 3/4" the frictional pumping losses due to pipe restriction are fractions of what the eyeball restriction is. I can't make it any clearer than this. I understand where you are getting the plumbing calculations from, but they are being misinterpreted due to dominant restrictions in the system which make the plumbing frictional losses virtually non-existant in comparison. Essentially you are adding pipe frictional losses at peak flow without examining the entire system restriction as a whole which is dominant, this is hydraulics 101.
What you are proposing is that the pipe diameter flow restriction is dominant and is the limiting factor, it is not and hence can be negated...is this making any sense? This can easily be tested by removing the eyeball returns and noting the difference in pressure at the filter, removing mine causes no decrease in pressure at the filter which is the dominant restriction! I don't know how else to explain this in laymans terms that everyone can understand, plumbing is not complicated but pipe frictional losses only start becoming valid when they are the dominant restriction in a viscous fluids flow path...in a pool the filter and eyeballs are hugely dominant in the fluid flow path unless the runs are extremely long.
There are two major restrictions in any pool plumbing setup, the eyeball returns and the filter. The filter represents 95% of the flow rate restriction in the plumbing while the eyeballs represent less than 5% since they are after the filter and free flowing to almost atmospheric pressure (about .5 psi of backpressure at 1ft below the water line). Frictional restrictions due to plumbing sizing is virtually non-existent unless you plumb the eyeball fittings with the same size plumbing as the exit port diameter I.E. 3/4", 1"...which nobody does.
Therefore if the plumbing to the eyeballs is 1.5" and your eyeballs are 3/4" the frictional pumping losses due to pipe restriction are fractions of what the eyeball restriction is. I can't make it any clearer than this. I understand where you are getting the plumbing calculations from, but they are being misinterpreted due to dominant restrictions in the system which make the plumbing frictional losses virtually non-existant in comparison. Essentially you are adding pipe frictional losses at peak flow without examining the entire system restriction as a whole which is dominant, this is hydraulics 101.
What you are proposing is that the pipe diameter flow restriction is dominant and is the limiting factor, it is not and hence can be negated...is this making any sense? This can easily be tested by removing the eyeball returns and noting the difference in pressure at the filter, removing mine causes no decrease in pressure at the filter which is the dominant restriction! I don't know how else to explain this in laymans terms that everyone can understand, plumbing is not complicated but pipe frictional losses only start becoming valid when they are the dominant restriction in a viscous fluids flow path...in a pool the filter and eyeballs are hugely dominant in the fluid flow path unless the runs are extremely long.
- Aug 13, 2010
- 74
Thank you for patiently explaining. I understand your position better now, and I'll reward you by respectfully disagreeing 
You refer to "flow rate restriction" as if the the filter and the eyeballs have similar effects in the system. I don't believe they do. The filter causes loss of head due to surface friction. The eyeball increases the flow velocity by decreasing cross-sectional area, with almost no loss of head (much less than 5%). Head losses in the pipe cannot be disregarded - see mas985's and ChemGeek's responses for why.
Thanks,
Sim
You refer to "flow rate restriction" as if the the filter and the eyeballs have similar effects in the system. I don't believe they do. The filter causes loss of head due to surface friction. The eyeball increases the flow velocity by decreasing cross-sectional area, with almost no loss of head (much less than 5%). Head losses in the pipe cannot be disregarded - see mas985's and ChemGeek's responses for why.
Thanks,
Sim
Simbilis,
Your pipe runs are short so won't be the same issue as with my pool. You have a cartridge filter so if it is spec'd similar to mine, it will have a small effect on overall pressure -- as I showed earlier, 5 feet of head (about 2.2 PSI) at 90 GPM. This is not the same as a sand filter that 4JawChuck has in his system which does have a large part of the total head (though I disagree with it being 95%). Even so, sand filters vary as this one has only 9 feet of head at 90 GPM when clean while this one doesn't even get to 90 GPM and at 50 GPM has around 17 feet of head. So the specific filter that is used makes a huge difference in whether it is going to dominate overall head loss in a system.
As for the returns, smaller eyeballs most definitely have more resistance (head loss) resulting in lower flow rates, not just faster velocity coming out of them. You can see this for yourself by taking a hose and filling a bucket of water with no nozzle on it and then repeating with narrower nozzles (such as Jet) and see that the total flow rate is reduced (i.e. speed for the water to rise in the bucket) even though the velocity of the stream has increased (you can do this with the hose in water at the bottom of the bucket if you want to eliminate the water through air issue).
4JawChuck,
Are you saying that Jandy is lying when they published their head loss curves on page 13 of this link and that the head loss tables for pipe in this link are also wrong? You claim that the filter represents 95% of the head (pressure) loss, but I registered way more than 2-3 PSI (the filter's head loss) on my filter gauge at high flow rates (80-90 GPM) -- closer to 13 PSI (with solar off) which of course doesn't count suction head. Including suction head, where do you suppose the 18 PSI or so is coming from? It's from the piping and fixtures, not mostly from the filter (and my filter doesn't rise in pressure over a season before I clean it). Again, this is for a system with a cartridge filter though even one with sand isn't going to have 95% of the loss in the filter unless the pipe runs are very short and the fixtures not constricted. If the filter is 95% of the head loss in your system, then if you bypass the filter you should find that the pressure in the pipe near the pump is only 5% of what it was when the filter was in place (if you kept the same GPM) -- though it will certainly drop, I doubt very much that it will drop by that much; alternately, you could measure the pressure right after your filter with a gauge to see the pipe/fixture pressure (you already have a gauge on your filter measuring the pressure between the pump and filter) and have a suction gauge for the suction side of your pump.
Richard
Your pipe runs are short so won't be the same issue as with my pool. You have a cartridge filter so if it is spec'd similar to mine, it will have a small effect on overall pressure -- as I showed earlier, 5 feet of head (about 2.2 PSI) at 90 GPM. This is not the same as a sand filter that 4JawChuck has in his system which does have a large part of the total head (though I disagree with it being 95%). Even so, sand filters vary as this one has only 9 feet of head at 90 GPM when clean while this one doesn't even get to 90 GPM and at 50 GPM has around 17 feet of head. So the specific filter that is used makes a huge difference in whether it is going to dominate overall head loss in a system.
As for the returns, smaller eyeballs most definitely have more resistance (head loss) resulting in lower flow rates, not just faster velocity coming out of them. You can see this for yourself by taking a hose and filling a bucket of water with no nozzle on it and then repeating with narrower nozzles (such as Jet) and see that the total flow rate is reduced (i.e. speed for the water to rise in the bucket) even though the velocity of the stream has increased (you can do this with the hose in water at the bottom of the bucket if you want to eliminate the water through air issue).
4JawChuck,
Are you saying that Jandy is lying when they published their head loss curves on page 13 of this link and that the head loss tables for pipe in this link are also wrong? You claim that the filter represents 95% of the head (pressure) loss, but I registered way more than 2-3 PSI (the filter's head loss) on my filter gauge at high flow rates (80-90 GPM) -- closer to 13 PSI (with solar off) which of course doesn't count suction head. Including suction head, where do you suppose the 18 PSI or so is coming from? It's from the piping and fixtures, not mostly from the filter (and my filter doesn't rise in pressure over a season before I clean it). Again, this is for a system with a cartridge filter though even one with sand isn't going to have 95% of the loss in the filter unless the pipe runs are very short and the fixtures not constricted. If the filter is 95% of the head loss in your system, then if you bypass the filter you should find that the pressure in the pipe near the pump is only 5% of what it was when the filter was in place (if you kept the same GPM) -- though it will certainly drop, I doubt very much that it will drop by that much; alternately, you could measure the pressure right after your filter with a gauge to see the pipe/fixture pressure (you already have a gauge on your filter measuring the pressure between the pump and filter) and have a suction gauge for the suction side of your pump.
Richard
Lets see if I can explain this in a different way.
Flow restriction numbers are important in the design in plumbing systems, they are used to calculate the restrictions in the system for analysis to determine plumbing sizing.
The DOMINANT flow restriction is used in the design to determine pump capacity and power required to move the fluid through that restriction while maintaining the required flow rate.
Anything that restricts the flow is counted in the above calculations i.e. elbows, reducers couplings etc. These are typically listed as foot length equivalences in the literature.
If during the design phase, the dominant flow restriction is determined to be caused by the plumbing (by calculation) it is prudent to upsize the plumbing to reduce the restriction so it is not DOMINANT.
However in a pool design the dominant restrictions are the filter (any filter) and the return eyeball restrictions, common sense will tell you anything that reduces the plumbing diameter causes restriction and anything that interrupts flow (the filter) is a one also.
Upsizing the plumbing without reducing the inherent restrictions will gain you next to nothing in flow rate gains since piping wall friction is not dominant in the design. Upsizing the filter area or increasing the eyeball return diameter will be a much better return on investment by a large factor any way you want to calculate it. Just remember the eyeballs are after the dominant restriction so they add little to the total.
There are cases where Reynolds number of the fluid flow becomes important, if you are trying to create laminar flow through the plumbing such as near a SWG it is prudent to reduce the flow rate in that area by upsizing the plumbing IF IT IS REQUIRED. However, typical flow rates through pool plumbing does not require such designs but it is prudent to ensure a straight sections ahead of such devices to allow laminar flow to re-establish in the piping...as I mentioned in a another thread long radius elbows can be utilized to prevent mixed flow conditions ahead of such devices...or longer runs substituted.
In general plumbing designs are based on experimental data such as head loss graphs etc. for the various pieces of the design, however this design data is used to determine DOMINANT flow restrictions to ensure the design is optimal for the pump nomenclature. They are not to be used in an additive formula to determine theoretical flow rate gains such as has been proposed, this has to be determined through experimentation and actual field tests. I can tell you from experience, the scenario presented is not probable or possible due to the design of the common pool plumbing system.
Are there exceptions? Sure! Are your plumbing runs 100-300 ft from the pump to the splitter? Then a plumbing calculation for the run between the pump and the splitter would be prudent to ensure this long run of plumbing is not the DOMINANT restriction in the system. This is how this data is to be used, to determine losses from system components. Got a bunch of elbows in the system because you want the pump near the garage in the back to feed the pool in the front of the house? Better do a calculation to ensure all those elbows don't equal some huge length of straight pipe (kind of a dumb scenario since you would use smooth bore semi-flexible pipe for this setup right!).
In summary there are very knowledgable engineers in the profession whose job it is to ensure plumbing is accomplished properly and to spec, if they recommend upsizing the pipe for definite reasons it is wise to heed their recommendations. From my inspection of the common pool plumbing designs on the market these engineers have left huge margins for various possible designs so that 99% of the buying public (and the installers) don't have to think...just put it in as designed unless certain conditions apply such as long runs from pump to splitter.
Just as an example, using 2" lines from the splitter to the eyeball returns is plain silly unless you have more than one eyeball sharing the line or you have some ridiculously huge pool. Simple cross sectional area calculations are used in this scenario to ensure you have enough flow area for the return outlets cross sections. You don't use the head losses in the calculation since it is not the DOMINANT flow restriction...OK? I know it can be confusing because you will not find anywhere on the internet where it tells you this in plain language, it is assumed as common language/knowledge because thats how hydraulic engineering is taught.
I hate to beat a dead horse here but lets look at another common scenario that everyone has in their car sitting in the driveway. The typical fuel injection system.
Your vehicle has a certain size tube leading from the gas tank to the engine where injectors are used to meter fuel when the engine is running. Typically the maximum flow rate of this line is never approached because the pipes diameter is sized to feed the injectors at the engines maximum rpm under all possible conditions. Lets just say you decide Moms car is suitable to run in the Indy 500 and you are going to drive it at maximum rpm for long periods of time. Will there be anything gained by increasing the cross sectional area of the tube feeding the injectors? Will the fuel pump feel less backpressure (i.e use less energy) due to the less restrictive tube? Will you go faster because of the larger line? Will running a larger fuel filter allow more flow?
The answer to all of these questions is no, because the fuel injector is the DOMINANT restriction in the fuel line.
Now lets say you lost the Indy 500 big time because Moms car just wasn't fast enough, so you drop in a high HP engine with its larger double size injectors and drop in a huge pump capable of double the flow rate (to feed those new injectors) and you decide a tank and pump mounted in the back of a 20ft long trailer you are going to tow behind the vehicle is the hot ticket to winning this coveted race. Do you need to increase the size of the fuel line leading from the tank in the trailer?
Of course you do, because the fuel line has now become the dominant flow restriction in the line and judging from the doubling of the flow rate requirements at least a doubling of the tubes cross sectional area would be prudent. I have simplified this somewhat to make a point, car fuel injection systems are not pools and flow rates through fuel lines are much higher than a pools plumbing so Reynolds numbers become important...blah blah blah. :blah:
But there is an important lesson to be learned here, recirculatory plumbing is designed by calculating the restrictions to ensure the piping restrictions are not dominant to avoid pumping losses...going above this minimum size achieves next to nothing in savings or increases to flow rate. There are cases where this unavoidable such as municipal planning where you need to design for minimum acceptable pressure at the end of the line...but I digress.
To be honest I think the real discussion should be how often should I clean my filter to save energy (less pumping losses) and should I use a filter thats three times larger than recommended since that is the dominant restriction in the plumbing and pumps run for long periods with a dirty filter...thats where you should spend your money since it would represent the largest return on investment in daily use. I don't like to sound preachy on the subject but this is how its done in reality and real world experimentation has been done that backs this up since the 1800's when man devised the steam locomotive where accurate plumbing design became important to prevent danger to life and limb.
Can you hear me now?
Flow restriction numbers are important in the design in plumbing systems, they are used to calculate the restrictions in the system for analysis to determine plumbing sizing.
The DOMINANT flow restriction is used in the design to determine pump capacity and power required to move the fluid through that restriction while maintaining the required flow rate.
Anything that restricts the flow is counted in the above calculations i.e. elbows, reducers couplings etc. These are typically listed as foot length equivalences in the literature.
If during the design phase, the dominant flow restriction is determined to be caused by the plumbing (by calculation) it is prudent to upsize the plumbing to reduce the restriction so it is not DOMINANT.
However in a pool design the dominant restrictions are the filter (any filter) and the return eyeball restrictions, common sense will tell you anything that reduces the plumbing diameter causes restriction and anything that interrupts flow (the filter) is a one also.
Upsizing the plumbing without reducing the inherent restrictions will gain you next to nothing in flow rate gains since piping wall friction is not dominant in the design. Upsizing the filter area or increasing the eyeball return diameter will be a much better return on investment by a large factor any way you want to calculate it. Just remember the eyeballs are after the dominant restriction so they add little to the total.
There are cases where Reynolds number of the fluid flow becomes important, if you are trying to create laminar flow through the plumbing such as near a SWG it is prudent to reduce the flow rate in that area by upsizing the plumbing IF IT IS REQUIRED. However, typical flow rates through pool plumbing does not require such designs but it is prudent to ensure a straight sections ahead of such devices to allow laminar flow to re-establish in the piping...as I mentioned in a another thread long radius elbows can be utilized to prevent mixed flow conditions ahead of such devices...or longer runs substituted.
In general plumbing designs are based on experimental data such as head loss graphs etc. for the various pieces of the design, however this design data is used to determine DOMINANT flow restrictions to ensure the design is optimal for the pump nomenclature. They are not to be used in an additive formula to determine theoretical flow rate gains such as has been proposed, this has to be determined through experimentation and actual field tests. I can tell you from experience, the scenario presented is not probable or possible due to the design of the common pool plumbing system.
Are there exceptions? Sure! Are your plumbing runs 100-300 ft from the pump to the splitter? Then a plumbing calculation for the run between the pump and the splitter would be prudent to ensure this long run of plumbing is not the DOMINANT restriction in the system. This is how this data is to be used, to determine losses from system components. Got a bunch of elbows in the system because you want the pump near the garage in the back to feed the pool in the front of the house? Better do a calculation to ensure all those elbows don't equal some huge length of straight pipe (kind of a dumb scenario since you would use smooth bore semi-flexible pipe for this setup right!).
In summary there are very knowledgable engineers in the profession whose job it is to ensure plumbing is accomplished properly and to spec, if they recommend upsizing the pipe for definite reasons it is wise to heed their recommendations. From my inspection of the common pool plumbing designs on the market these engineers have left huge margins for various possible designs so that 99% of the buying public (and the installers) don't have to think...just put it in as designed unless certain conditions apply such as long runs from pump to splitter.
Just as an example, using 2" lines from the splitter to the eyeball returns is plain silly unless you have more than one eyeball sharing the line or you have some ridiculously huge pool. Simple cross sectional area calculations are used in this scenario to ensure you have enough flow area for the return outlets cross sections. You don't use the head losses in the calculation since it is not the DOMINANT flow restriction...OK? I know it can be confusing because you will not find anywhere on the internet where it tells you this in plain language, it is assumed as common language/knowledge because thats how hydraulic engineering is taught.
I hate to beat a dead horse here but lets look at another common scenario that everyone has in their car sitting in the driveway. The typical fuel injection system.
Your vehicle has a certain size tube leading from the gas tank to the engine where injectors are used to meter fuel when the engine is running. Typically the maximum flow rate of this line is never approached because the pipes diameter is sized to feed the injectors at the engines maximum rpm under all possible conditions. Lets just say you decide Moms car is suitable to run in the Indy 500 and you are going to drive it at maximum rpm for long periods of time. Will there be anything gained by increasing the cross sectional area of the tube feeding the injectors? Will the fuel pump feel less backpressure (i.e use less energy) due to the less restrictive tube? Will you go faster because of the larger line? Will running a larger fuel filter allow more flow?
The answer to all of these questions is no, because the fuel injector is the DOMINANT restriction in the fuel line.
Now lets say you lost the Indy 500 big time because Moms car just wasn't fast enough, so you drop in a high HP engine with its larger double size injectors and drop in a huge pump capable of double the flow rate (to feed those new injectors) and you decide a tank and pump mounted in the back of a 20ft long trailer you are going to tow behind the vehicle is the hot ticket to winning this coveted race. Do you need to increase the size of the fuel line leading from the tank in the trailer?
Of course you do, because the fuel line has now become the dominant flow restriction in the line and judging from the doubling of the flow rate requirements at least a doubling of the tubes cross sectional area would be prudent. I have simplified this somewhat to make a point, car fuel injection systems are not pools and flow rates through fuel lines are much higher than a pools plumbing so Reynolds numbers become important...blah blah blah. :blah:
But there is an important lesson to be learned here, recirculatory plumbing is designed by calculating the restrictions to ensure the piping restrictions are not dominant to avoid pumping losses...going above this minimum size achieves next to nothing in savings or increases to flow rate. There are cases where this unavoidable such as municipal planning where you need to design for minimum acceptable pressure at the end of the line...but I digress.
To be honest I think the real discussion should be how often should I clean my filter to save energy (less pumping losses) and should I use a filter thats three times larger than recommended since that is the dominant restriction in the plumbing and pumps run for long periods with a dirty filter...thats where you should spend your money since it would represent the largest return on investment in daily use. I don't like to sound preachy on the subject but this is how its done in reality and real world experimentation has been done that backs this up since the 1800's when man devised the steam locomotive where accurate plumbing design became important to prevent danger to life and limb.
Can you hear me now?
I repeat -- both Simbilis and I have CARTRIDGE FILTERS. I've shown you the head loss curves for such filters, especially if they are oversized as one would want if one does not want to clean them frequently. Do you not believe such curves that show only 5 feet of head (2.2 PSI) loss at 90 GPM for a 340 square foot cartridge filter?
You are right that the individual return lines even at 1.5" aren't a big deal so long as you have multiple returns (so splitting the flow rate between them) and I initially incorrectly multiplied the head loss by the number of lines which was incorrect because they are shared by each return (I've corrected that in the post). With Simbilis having 4 returns, 1.5" lines should be fine, especially since they aren't long. Nevertheless, 4 feet of head at 90 GPM from the filter is NOT the dominant head loss in my system as my pressure gauge on my filter shows 13 PSI (about 30 feet of head) at 80-90 GPM (with solar off with my older pump). In this post (which is conservative since the pipe runs are likely longer than I estimated) I show that the 1.5" piping to each of 3 returns at 30 GPM contributes 1.6 feet of head and the 2" piping from their combination at 90 GPM from the pump was another 5 feet of head and that the two 1.5" suction lines at 45 GPM contribute 8.2 feet of head for 14.8 feet (6.4 PSI) which is much more than that contributed by my filter and that had my PB used 2" for the returns and especially the suction lines and 2.5" for the combo line, I'd have less than ONE-THIRD of the loss from those lines. Even if I assume everything else stayed the same for the original 18 PSI total (i.e. keep 11.6 PSI the same), the better plumbing would result in at least a 4 PSI reduction so going from 18 to 14 (at the same GPM, so the RPM on the variable speed/flow pump would be reduced saving substantial energy -- around 30% reduction). If 1" eyeballs had been used instead of 3/4", there would have been at least another 2 PSI reduction.
As for the Reynolds number, the flow is turbulent flow for virtually everything in pool systems except for solar panels where 4 GPM per panel with their large area of small tubes results in laminar flow in the small tubes (it's still turbulent in the 2" headers). Even 5 GPM through 3" pipe is still turbulent since the Reynolds number is 6565 in this case so still > 4000.
Do you not believe that water flowing through pipe results in head loss according to these charts where even 30 GPM through 1.5" lines has a drop of 2.4 PSI for every 100 feet?
You are right that the individual return lines even at 1.5" aren't a big deal so long as you have multiple returns (so splitting the flow rate between them) and I initially incorrectly multiplied the head loss by the number of lines which was incorrect because they are shared by each return (I've corrected that in the post). With Simbilis having 4 returns, 1.5" lines should be fine, especially since they aren't long. Nevertheless, 4 feet of head at 90 GPM from the filter is NOT the dominant head loss in my system as my pressure gauge on my filter shows 13 PSI (about 30 feet of head) at 80-90 GPM (with solar off with my older pump). In this post (which is conservative since the pipe runs are likely longer than I estimated) I show that the 1.5" piping to each of 3 returns at 30 GPM contributes 1.6 feet of head and the 2" piping from their combination at 90 GPM from the pump was another 5 feet of head and that the two 1.5" suction lines at 45 GPM contribute 8.2 feet of head for 14.8 feet (6.4 PSI) which is much more than that contributed by my filter and that had my PB used 2" for the returns and especially the suction lines and 2.5" for the combo line, I'd have less than ONE-THIRD of the loss from those lines. Even if I assume everything else stayed the same for the original 18 PSI total (i.e. keep 11.6 PSI the same), the better plumbing would result in at least a 4 PSI reduction so going from 18 to 14 (at the same GPM, so the RPM on the variable speed/flow pump would be reduced saving substantial energy -- around 30% reduction). If 1" eyeballs had been used instead of 3/4", there would have been at least another 2 PSI reduction.
As for the Reynolds number, the flow is turbulent flow for virtually everything in pool systems except for solar panels where 4 GPM per panel with their large area of small tubes results in laminar flow in the small tubes (it's still turbulent in the 2" headers). Even 5 GPM through 3" pipe is still turbulent since the Reynolds number is 6565 in this case so still > 4000.
Do you not believe that water flowing through pipe results in head loss according to these charts where even 30 GPM through 1.5" lines has a drop of 2.4 PSI for every 100 feet?
- Aug 13, 2010
- 74
4JawChuck and ChemGeek,
Despite your best efforts, I clearly haven't learned enough to understand your disagreement and take an intelligent position. That's ok, though - I did learn enough to be able to ask questions and identify the most knowledgeable builder, and have an informed discussion about plumbing style with him. The result is that he agreed to do separate 1.5" runs back to the pad for my returns rather than teeing them underground, and to put a valve on each of them. He thinks that's a little bit of overkill, but not unreasonable.
Thanks again,
Sim
Despite your best efforts, I clearly haven't learned enough to understand your disagreement and take an intelligent position. That's ok, though - I did learn enough to be able to ask questions and identify the most knowledgeable builder, and have an informed discussion about plumbing style with him. The result is that he agreed to do separate 1.5" runs back to the pad for my returns rather than teeing them underground, and to put a valve on each of them. He thinks that's a little bit of overkill, but not unreasonable.
Thanks again,
Sim
Sim,
The separate 1.5" lines for each return is fine since you've got so many returns (so the flow is split between them so losses are small). If he did tee them together, then I would have suggested a 2.5" line back to the pump if that was a long run but that's a moot point now. So what about the suction lines? If there are only two (say, one for floor drains and one for skimmers), then that is where I would suggest using 2" lines instead of 1.5" (if there were 3 or more lines, then 1.5" is OK). The last efficiency would be the size of eyeballs (1" vs. 3/4"), but that's not as big a deal since, again, you have so many returns.
Do you know what cartridge filter he is proposing for the system? Do you have a spec sheet for it that shows the head loss vs. flow rate? Since your pool is about 30,000 gallons, I suspect it is a larger and probably oversized filter like mine so the head loss will be low. You won't need to clean the filter very often as a result of its large area and probably won't notice any pressure rise at all when it gets dirty. The manual instructions say to clean when it gets 10-12 PSI above original starting pressure, but I never see it get any noticeable rise over a 7-month swim season (so perhaps 1 PSI or so that I don't notice) yet it does get dirty mostly with suntan lotion (and used to get dirty with cedar needles before we removed the cedar tree).
As for the disagreement, 4JawChuck believes nearly all (95%) of the head loss for the system is in the filter with all filter types and that the remaining loss is almost all in the returns with hardly any loss in piping unless runs are (multiple) hundreds of feet. I disagree, especially with large cartridge filters and write about this later below.
This is not at all the case in my pool where the pump is closest to the shallow end and is still 50 feet away from that end (in a straight shot, which of course the pipe does not take since it goes into the ground and has elbows, valves, etc. at the pad) and 50 feet away from the house (straight shot) for the solar panels. So maybe my PB didn't follow what 4JawChuck is saying and didn't bother to calculate the head losses in the piping in my system, especially for the two 1.5" suction lines, combo (teed) 2" line and to/from 2" lines to the solar recognizing that with the low-head-loss cartridge filter the piping was indeed the dominant source of head loss in my system and he should have upsized accordingly.
4JawChuck,
As I showed above, an oversized cartridge filter has very low loss (and doesn't get a large rise in pressure under normal conditions either) so cannot be assumed to be dominant even with pipe runs of 50 feet. I have given a similar example (with somewhat longer pipe runs) before yet you have not explained why you believe the filter is still the dominant factor. For 50 feet:
Jandy CL-340 Cartridge Filter @ 90 GPM: 5 feet of head (2.2 PSI)
50 feet of piping:
1.5" at 45 GPM: 50*(11.7/100) = 5.9 feet of head (2.5 PSI) such as with two 1.5" suction lines
-OR-
2" at 90 GPM: 50*(12.4/100) = 6.2 feet of head (2.7 PSI) such as with one 2" teed line or to/from solar (so 25' each way), etc.
Please explain to me how the cartridge filter with 2.2 PSI is the dominant source of head loss in this system with only 50 feet of piping (not 100-300 feet) with 2.5-2.7 PSI for piping alone? Yes, the filter is assumed to be clean, but as my own experience and the experiences of others with large cartridge filters shows, they effectively do not rise in pressure before getting cleaned once or twice a year.
Richard
The separate 1.5" lines for each return is fine since you've got so many returns (so the flow is split between them so losses are small). If he did tee them together, then I would have suggested a 2.5" line back to the pump if that was a long run but that's a moot point now. So what about the suction lines? If there are only two (say, one for floor drains and one for skimmers), then that is where I would suggest using 2" lines instead of 1.5" (if there were 3 or more lines, then 1.5" is OK). The last efficiency would be the size of eyeballs (1" vs. 3/4"), but that's not as big a deal since, again, you have so many returns.
Do you know what cartridge filter he is proposing for the system? Do you have a spec sheet for it that shows the head loss vs. flow rate? Since your pool is about 30,000 gallons, I suspect it is a larger and probably oversized filter like mine so the head loss will be low. You won't need to clean the filter very often as a result of its large area and probably won't notice any pressure rise at all when it gets dirty. The manual instructions say to clean when it gets 10-12 PSI above original starting pressure, but I never see it get any noticeable rise over a 7-month swim season (so perhaps 1 PSI or so that I don't notice) yet it does get dirty mostly with suntan lotion (and used to get dirty with cedar needles before we removed the cedar tree).
As for the disagreement, 4JawChuck believes nearly all (95%) of the head loss for the system is in the filter with all filter types and that the remaining loss is almost all in the returns with hardly any loss in piping unless runs are (multiple) hundreds of feet. I disagree, especially with large cartridge filters and write about this later below.
This is not at all the case in my pool where the pump is closest to the shallow end and is still 50 feet away from that end (in a straight shot, which of course the pipe does not take since it goes into the ground and has elbows, valves, etc. at the pad) and 50 feet away from the house (straight shot) for the solar panels. So maybe my PB didn't follow what 4JawChuck is saying and didn't bother to calculate the head losses in the piping in my system, especially for the two 1.5" suction lines, combo (teed) 2" line and to/from 2" lines to the solar recognizing that with the low-head-loss cartridge filter the piping was indeed the dominant source of head loss in my system and he should have upsized accordingly.
4JawChuck,
I disagree. If you look at virtually any documentation for how to determine Total Dyanmic Head, it includes the piping as well as the equipment (i.e. filter) and does not simplify to assume certain dominant loss. For example, see this file from Hayward which says:4JawChuck said:
or this link and this link and this link and this link all of which include all components ADDED TOGETHER in calculating Total Dynamic Head (see where it sums main drain line loss, skimmer/gutter line loss, return line loss, filter loss when dirty, heater loss, other). This link is similar in summing up piping, filter, heater and other losses. You don't just take the largest one, especially when one isn't hugely above all the others combined. This link doesn't even include the filter in the TDH calculation and I would agree with you that this is wrong, especially if a sand filter is used. This link says the following about calculating TDH:
Though the above refers to friction from fittings and pipe, filters and heaters, it is incorrect with regard to elevation between the pump and the basin water level since that is static head to overcome when priming, but once primed any loss on the suction side is gained on the pressure side so cancels out (and isn't part of dynamic head).
As I showed above, an oversized cartridge filter has very low loss (and doesn't get a large rise in pressure under normal conditions either) so cannot be assumed to be dominant even with pipe runs of 50 feet. I have given a similar example (with somewhat longer pipe runs) before yet you have not explained why you believe the filter is still the dominant factor. For 50 feet:
Jandy CL-340 Cartridge Filter @ 90 GPM: 5 feet of head (2.2 PSI)
50 feet of piping:
1.5" at 45 GPM: 50*(11.7/100) = 5.9 feet of head (2.5 PSI) such as with two 1.5" suction lines
-OR-
2" at 90 GPM: 50*(12.4/100) = 6.2 feet of head (2.7 PSI) such as with one 2" teed line or to/from solar (so 25' each way), etc.
Please explain to me how the cartridge filter with 2.2 PSI is the dominant source of head loss in this system with only 50 feet of piping (not 100-300 feet) with 2.5-2.7 PSI for piping alone? Yes, the filter is assumed to be clean, but as my own experience and the experiences of others with large cartridge filters shows, they effectively do not rise in pressure before getting cleaned once or twice a year.
I completely agree with this. For energy efficiency and less frequent cleaning, one should definitely get an oversized filter and do not let the pressure rise very much before cleaning. For sand filters, they tend to clean better when a little dirty, but even so one shouldn't wait until the pressure rises so much that it significantly affects total head loss and therefore energy costs. For an oversized cartridge filter, one may see very little rise in pressure even when dirty (this is the case with my filter). I assumed that with Sim's 30,000 gallon pool that the cartridge filter would be large and likely have a very low head loss similar to my 340 square foot (4-cart) cartridge filter, hence the losses in piping could not automatically be ignored.4JawChuck said:
Richard
- Aug 13, 2010
- 74
There are three suction lines: Two skimmers and a linked pair of main drains, which I think is the norm. They're all supposed to be 2". I don't have the final revised bid yet, but I'll correct that if need be.
To his credit, the builder originally recommended oversizing the filter. For the Pentair CCP520, the spec sheet says that the head loss at 90 GPM is 3.0 PSI. At a more leisurely 42 GPM for a 12 hour turnover, head loss is about 0.5 PSI.
My equipment pad is only 7 ft. from the corner of the deep end, so the lines should be pretty short. Assuming 3 x 60' of 2" pipe suction side and 3 x 60' of 1.5" pipe return side, head loss at 90 GPM would be about 1.3 PSI suction side and 4.3 PSI return side using the Engineering Tables data.
Counting only pipes and filter, that puts me at 8.6 PSI or ~20 ft. of head at 90 GPM. Everyone seems to agree that bumping up to 2" return lines for the 3.0 PSI reduction isn't worth doing.
Thanks,
Sim
To his credit, the builder originally recommended oversizing the filter. For the Pentair CCP520, the spec sheet says that the head loss at 90 GPM is 3.0 PSI. At a more leisurely 42 GPM for a 12 hour turnover, head loss is about 0.5 PSI.
My equipment pad is only 7 ft. from the corner of the deep end, so the lines should be pretty short. Assuming 3 x 60' of 2" pipe suction side and 3 x 60' of 1.5" pipe return side, head loss at 90 GPM would be about 1.3 PSI suction side and 4.3 PSI return side using the Engineering Tables data.
Counting only pipes and filter, that puts me at 8.6 PSI or ~20 ft. of head at 90 GPM. Everyone seems to agree that bumping up to 2" return lines for the 3.0 PSI reduction isn't worth doing.
Thanks,
Sim
Sim,
Your situation is very well designed and the pipe runs are very short so quite frankly you didn't really need to worry much about the piping in your situation. With 3 suction lines, even 1.5" would be OK, especially with such short runs. Your filter manual also says to clean the filter when there is an 8-10 PSI rise, but I'd be surprised if you see that much rise even in one season given the 520 square foot area of your filter unless you have a lot of small debris that gets into your pool (you can also use skimmer socks to catch more before the filter, if necessary). 4JawChuck and I agree on this -- you should clean your filter long before there is a large PSI increase as in your system such a rise would clearly increase losses substantially though even then, your energy costs should be low until you get a solar system.
Your filter manual says that the head loss is the same for the filter regardless of size from 240 to 520 square feet but that's clearly not true and their graph is probably for the worst-case which would be 240 square feet (less area for water to flow through) and your losses through the 520 filter will probably be even lower.
When calculating head loss, you don't add together lines that are in parallel (I made that mistake initially, but have since corrected that post) -- you just split the flow rate and use one line for the head loss calculation assuming they are all the same. They are not added together as they are not in series. So your situation would be the following at 90 GPM:
3 x 60' 2" suction side (90/3 = 30 GPM each): 60*(1.6/100) = 1.0 feet of head (0.4 PSI)
3 x 60' 1.5" return side (90/3 = 30 GPM each): 60*(5.5/100) = 3.3 feet of head (1.4 PSI)
I thought you had 4 returns, not 3, but in any event as you can see the head loss is very low regardless and you could easily have 1.5" pipe for the suction side as well with no problem.
You are thinking of low flow rates of 42 GPM right now, but when you get a solar system you may find that you need higher flow rates depending on how many panels you get. I have 12 panels for my pool and each needs 4 GPM so in series this is 48 GPM and since your pool is larger, you might need more (depending on where you live and your target pool temperature). The energy cost when the solar is off is very low -- the dominant energy cost in my pool is when the solar is on and you will likely find that to be the case in your pool as well if you get a large area of solar panels. When you get your solar system in the future, it will likely be at a larger distance away and possibly on your roof in which case that is when you should get larger pipe to/from the solar, such as 2.5".
Richard
Your situation is very well designed and the pipe runs are very short so quite frankly you didn't really need to worry much about the piping in your situation. With 3 suction lines, even 1.5" would be OK, especially with such short runs. Your filter manual also says to clean the filter when there is an 8-10 PSI rise, but I'd be surprised if you see that much rise even in one season given the 520 square foot area of your filter unless you have a lot of small debris that gets into your pool (you can also use skimmer socks to catch more before the filter, if necessary). 4JawChuck and I agree on this -- you should clean your filter long before there is a large PSI increase as in your system such a rise would clearly increase losses substantially though even then, your energy costs should be low until you get a solar system.
Your filter manual says that the head loss is the same for the filter regardless of size from 240 to 520 square feet but that's clearly not true and their graph is probably for the worst-case which would be 240 square feet (less area for water to flow through) and your losses through the 520 filter will probably be even lower.
When calculating head loss, you don't add together lines that are in parallel (I made that mistake initially, but have since corrected that post) -- you just split the flow rate and use one line for the head loss calculation assuming they are all the same. They are not added together as they are not in series. So your situation would be the following at 90 GPM:
3 x 60' 2" suction side (90/3 = 30 GPM each): 60*(1.6/100) = 1.0 feet of head (0.4 PSI)
3 x 60' 1.5" return side (90/3 = 30 GPM each): 60*(5.5/100) = 3.3 feet of head (1.4 PSI)
I thought you had 4 returns, not 3, but in any event as you can see the head loss is very low regardless and you could easily have 1.5" pipe for the suction side as well with no problem.
You are thinking of low flow rates of 42 GPM right now, but when you get a solar system you may find that you need higher flow rates depending on how many panels you get. I have 12 panels for my pool and each needs 4 GPM so in series this is 48 GPM and since your pool is larger, you might need more (depending on where you live and your target pool temperature). The energy cost when the solar is off is very low -- the dominant energy cost in my pool is when the solar is on and you will likely find that to be the case in your pool as well if you get a large area of solar panels. When you get your solar system in the future, it will likely be at a larger distance away and possibly on your roof in which case that is when you should get larger pipe to/from the solar, such as 2.5".
Richard
I've done my best to illustrate my point but for reference the US Army Core of Engineers have an excellent manual for the design and construction of plumbing installations of all kinds available online, I would point everyone reading this thread to Appendix D and E, which gives excellent advice on pipe size selection and relates all those calculations to graphs for easy selection. A lot of useful information in that document.
Good luck on your project.
http://www.constructionknowledge.ne...hanical/Plumbing/Plumbing_Army_FM3-34-471.pdf
Good luck on your project.
http://www.constructionknowledge.ne...hanical/Plumbing/Plumbing_Army_FM3-34-471.pdf
Figure D-4 in your reference is a graph of friction loss for smooth pipe. If you look at the 1.5" line at 45 GPM, the Head is about 7 PSI per 100 feet. If you look at the 2" line at 90 GPM, the Head is about 6 PSI per 100 feet. These head losses are actually larger than found in the PVC pipe table I used that give 5.1 and 5.4 for the losses.
So the references you give actually demonstrate even higher pipe friction losses than I had used. You have yet to explain how you get your assertion that the filter (all kinds) has 95% of the losses in most pools with the fixtures (returns, etc.) having 5% and presumably the piping being negligible unless the pipes are long which you wrote was 100-300 feet. I gave an example using 50 feet of pipe. If I were to instead use Figure D-4 from your own reference, then 50 feet of smooth pipe would be about 3 PSI head loss for two 1.5" lines or one 2" line. The head loss at 90 GPM for the cartridge filter I have is 2.2 psi and the one that Sim will have is no more than 3 psi so please explain to me how the filter represents 95% of the loss and that the loss in piping is negligible because it is not dominant.
So the references you give actually demonstrate even higher pipe friction losses than I had used. You have yet to explain how you get your assertion that the filter (all kinds) has 95% of the losses in most pools with the fixtures (returns, etc.) having 5% and presumably the piping being negligible unless the pipes are long which you wrote was 100-300 feet. I gave an example using 50 feet of pipe. If I were to instead use Figure D-4 from your own reference, then 50 feet of smooth pipe would be about 3 PSI head loss for two 1.5" lines or one 2" line. The head loss at 90 GPM for the cartridge filter I have is 2.2 psi and the one that Sim will have is no more than 3 psi so please explain to me how the filter represents 95% of the loss and that the loss in piping is negligible because it is not dominant.
- May 3, 2007
- 17,029
- Pool Size
- 20000
- Surface
- Plaster
- Chlorine
- Salt Water Generator
- SWG Type
- Hayward Aqua Rite (T-15)
Simbilis said:
Maybe a detailed example using the actual design will help. Assuming 2" pipe is used on the pad, 1" eyeballs, no heater, and a typical number of 90s, 45s, valves, etc, here is what I get with my hydraulic model:
RPM: 2600
GPM: 91.7
Head: 33.4' (2.2' suction, 24.3' pad, 4.6' return, 2.4' eyeballs)
Filter PSI: 11
Eyeball Exit Velocity: 12.5 ft/sec (this is what matters for surface movement)
Gallons/Watt-Hr: 4.2
If you used 1.5" everywhere, here is what I get:
RPM: 2600
GPM: 70.4
Head: 42.4' (4' suction, 34.2' pad, 2.8' return, 1.4' eyeballs)
Eyeball Exit Velocity: 9.6 ft/sec
Filter PSI: 13
Gallons/Watt-Hr: 3.6
A loss of about 15% in efficiency and about 24% loss in eyeball exit velocity by using 1.5" plumbing on the suction and pad.
These numbers will change with pump RPM setting but I just wanted to give a detailed example to illustrate what changing the pipe size can really do. One thing to keep in mind is that head loss is proportional to the equivalent length of pipe but it is also proportional to the 5th power of the pipe diameter. So pipe diameter is much more important than the length of the pipe.
Note too that the filter is only about 30% of the pad head loss or around 20% of total head loss.