Idea to reduce head height, alternative to 90° elbow

[inaspin]

A sweep 90 is basically what Im trying to accomplish, but larger. People recommend sweep 90s on return pumps because the 90 degree turn isn't as intense, and longer, consequently reducing pressure against the pump.

So by using Flex PVC, i am creating a large sweep to further the reduction in head loss.




Sorry, i don't exactly understand.

Sorry I was not clear. Bottom line of my thoughts is the 90 degree ell either a long or short sweep won't change enough to worry about if at all. The other part of your question being what schedule or type of pipe to use Its not a problem using what ever you want its not enough pressure for it to handle.

One idea I did have is you may want to take a look at PVC DWV pipe its made for drainage waste or vents , so the fittings are made so that what ever is going through them runs easy. ie- So you don't have blockages in your sewer. This should give you the easy flow you are looking for and would not have a problem with the small amount of pressure. Good luck again on your build.

 

[mdw]

Thank you all for your contributions - but my initial question has still gone unanswered.

Will replacing a standard 90° Elbow with an arc built with Flex PVC or a product alike in which a 90° turn is extended over a large radius (example: 20") reduce head loss?

Yes, your method will reduce head loss and back pressure on the pump, especially if more than one 90 is eliminated in your pipe run. I have used flex pvc on an Iwaki 100rlt (pressure rated, 2000gph) that pumps against significant head without any issues. The same flex pvc is commonly used in pool, pond, spa, and other applications which often use pumps with much higher output pressure than the Hammerhead. Hope this helps.

 

[mshaq]

I had the same goal as you so I bent PVC pipe. It was easy.

 

[DanK13]

Thank you all for your contributions - but my initial question has still gone unanswered.

Will replacing a standard 90° Elbow with an arc built with Flex PVC or a product alike in which a 90° turn is extended over a large radius (example: 20") reduce head loss?

The answer to your question is yes.
http://www.pondtrademag.com/articles/ar-71/

 

[Haffs09]

Here are a couple of excellent resources for plumbing info:

http://www.advancedaquarist.com/2003/1/aafeature2

http://www.advancedaquarist.com/2007/5/aafeature2

I modified the Pump Head Loss Calculator to include a Reeflo Hammerhead pump. With the following inputs:
Vertical height - 11'
Horizontal length - 17'
90° elbow - 4
45° elbow - 0
Gate valves - 0
Ball valves - 2 (either side of pump)
Unions - 4 (either side of valves above)
Swing checks - 0
Pipe exits - 2
Pipe entrance - 1

Results if calculated with all 1.5" pipe:
Head pressure (ft loss) = 18.76
Total flow rate (gph) = 3082

Same inputs but with 2" pipe:
Head pressure (ft loss) = 15.35
Total flow rate (gph) = 4018

If I eliminate the 4 90° elbows and add 8 45° elbows the results are:
For 1.5" pipe:
Head pressure (ft loss) = 18.2
Total flow rate (gph) = 3261

For 2" pipe:
Head pressure (ft loss) = 14.75
Total flow rate (gph) = 4144

Those calcs do not include any reductions in pipe diameter or losses due to the OM 4-way, but should be a good reference point

 

[EnderG60]

Will it reduce head loss? yes. Is it worth it? NO.

Advice from an actual plumbing engineer....Dont waste your time and effort.

If youre really that worried, use larger pipe and look at sanitary fittings, you can find long sweep 90's in 2" and up. 1.5" is around but hard to find.

Beyond that, honestly dont worry about it. End of the day your flow is probably half what you think it is anyway.

 

[TheFishTeen]

Yes, your method will reduce head loss and back pressure on the pump, especially if more than one 90 is eliminated in your pipe run. I have used flex pvc on an Iwaki 100rlt (pressure rated, 2000gph) that pumps against significant head without any issues. The same flex pvc is commonly used in pool, pond, spa, and other applications which often use pumps with much higher output pressure than the Hammerhead. Hope this helps.

Thank you for the clarification.

I had the same goal as you so I bent PVC pipe. It was easy.

Very cool. Nice job!

The answer to your question is yes.
http://www.pondtrademag.com/articles/ar-71/

Thanks for the link - that was a nice read.

Here are a couple of excellent resources for plumbing info:

http://www.advancedaquarist.com/2003/1/aafeature2

http://www.advancedaquarist.com/2007/5/aafeature2

I modified the Pump Head Loss Calculator to include a Reeflo Hammerhead pump. With the following inputs:
Vertical height - 11'
Horizontal length - 17'
90° elbow - 4
45° elbow - 0
Gate valves - 0
Ball valves - 2 (either side of pump)
Unions - 4 (either side of valves above)
Swing checks - 0
Pipe exits - 2
Pipe entrance - 1

Results if calculated with all 1.5" pipe:
Head pressure (ft loss) = 18.76
Total flow rate (gph) = 3082

Same inputs but with 2" pipe:
Head pressure (ft loss) = 15.35
Total flow rate (gph) = 4018

If I eliminate the 4 90° elbows and add 8 45° elbows the results are:
For 1.5" pipe:
Head pressure (ft loss) = 18.2
Total flow rate (gph) = 3261

For 2" pipe:
Head pressure (ft loss) = 14.75
Total flow rate (gph) = 4144

Those calcs do not include any reductions in pipe diameter or losses due to the OM 4-way, but should be a good reference point

Thank you so much! That was extremely helpful. I find it incredible though that only half a size up from 1.5" to 2" gives me 1000 more GPH!

You have been extremely helpful.

 

[TheFishTeen]

End of the day your flow is probably half what you think it is anyway.

Hmmm...let's not be pessemistic. ReeFlo is a prominent brand that stands behind their products, and would never agree with "half the flow".

 

[r-balljunkie]

Hmmm...let's not be pessemistic. ReeFlo is a prominent brand that stands behind their products, and would never agree with "half the flow".

Nope. the above is a 'true statement'.....coming from a mechanical engineer......you'll have other frictional losses your not accounting for as well. half...perhaps not, but much lower than you anticipate.

you should trust your local engineers

GL


C

 

[Haffs09]

I agree with r-balljunkie. Despite running the numbers for my system, I still have 2-4' higher head than calculated. I think you need to add a factor to account for the unknowns. Maybe add 20% to the calculated total dynamic head numbers and then size the pump for that value.

 

[EnderG60]

Hmmm...let's not be pessemistic. ReeFlo is a prominent brand that stands behind their products, and would never agree with "half the flow".

design conditions are a far cry from the real world.

I would bet money that your flow is no more then 75% of what you think it is.

Trust me on this, its not worth this much involvement and effort. Just plumb it up and dont worry about it. if you really need the extra flow that bad, get a bigger pump!

 

[yardboy]

Trying to maximize the flow from the pump you have; use biggest diameter pipe you can and sweeps, or bend the pipe. Spaflex pipe is smaller ID than rigid pipe so you'd be losing head with it versus sweeps or bending pipe or larger diameter pipe, but it'd be better than a hard 90.

 

[TheFishTeen]

Trying to maximize the flow from the pump you have; use biggest diameter pipe you can and sweeps, or bend the pipe. Spaflex pipe is smaller ID than rigid pipe so you'd be losing head with it versus sweeps or bending pipe or larger diameter pipe, but it'd be better than a hard 90.

Spaflex is smaller ID than standard rigid pipe? I don't think so - I'm holding two pieces of 1" PVC in my hand right now. One is rigid, one is spa flex. To the eye, they look exactly the same. If there is a difference, it is not even noticeable and will not alter the flow of the pump enough to notice.

 

[KeithP]

Yes, the wider curve will help reduce the overall friction loss of your piping. It reduces turbulence in the pipe.

All pipe systems suffer from "minor losses" at joints. They are called minor losses because they typically are minor but in some systems they can actually be the dominant energy loss. For the most part these involve high velocity. You can reduce velocity for a given flow rate by increasing the pipe size and you will reduce friction in the straight pipe as well as losses at the fittings. The fitting that increases the pipe size also has an energy loss associated with it so it only makes sense to increase pipe size if you have a lot of elbows or a long pipe. If you have to increase the pipe size, a long transition will have less energy loss than an instantaneous one. If you want to build the most efficient system possible, you can also gain a little at the pump intake through careful design. For external pumps a flared opening is more efficient. You can google "inlet control pipe" and try to emulate one of the more efficient inlet conditions. Outlets have a similar effect but in aquariums it is often more important to direct the flow than have the greatest possible flow rate.

Be careful using tables. The advanced aquarist article is pretty well thought out and it explains how to use the tables. It is easy for a lay person to misread the first table that was linked and there was no explanation on the page so I would advise against using that one.

For industrial piping projects, an engineer will typically develop a "system curve" which plots out the vertical head difference + friction + minor losses in head units vs. flow rate. This can be directly compared to the pump curve supplied by the manufacturer. The point at which the curves cross gives you the theoretical system flow rate using that pump. Developing a system curve allows you to quickly compare different pumps.

Be careful when bending PVC. You still want to put in only as many bends as is absolutely necessary. The earlier post showed some double curves that were probably done to save space but may have made the energy loss greater. Also if the pvc cross section area is reduced during bending, the energy loss will increase as the velocity increases offsetting some of the gains from wider bends. It does look cool though.


Having said all of this, I am actually inclined to agree with the other posters who state that it is not worth all of the calculations for most people. The calculations to determine the head loss at various velocities to come up with a curve are tedious without a computer program. The manufacturers of small pumps don't even provide pump curves. If you can do the sweeps without much cost, go for it but oversizing the pump a little won't hurt anything, adds only incremental cost, and is a whole lot easier. If you already have a pump and are trying to fix the plumbing to increase the flow, wide curves will help some and I have seen a number of posts where that was enough to make the system functional.

 

[TheFishTeen]

Yes, the wider curve will help reduce the overall friction loss of your piping. It reduces turbulence in the pipe.

All pipe systems suffer from "minor losses" at joints. They are called minor losses because they typically are minor but in some systems they can actually be the dominant energy loss. For the most part these involve high velocity. You can reduce velocity for a given flow rate by increasing the pipe size and you will reduce friction in the straight pipe as well as losses at the fittings. The fitting that increases the pipe size also has an energy loss associated with it so it only makes sense to increase pipe size if you have a lot of elbows or a long pipe. If you have to increase the pipe size, a long transition will have less energy loss than an instantaneous one. If you want to build the most efficient system possible, you can also gain a little at the pump intake through careful design. For external pumps a flared opening is more efficient. You can google "inlet control pipe" and try to emulate one of the more efficient inlet conditions. Outlets have a similar effect but in aquariums it is often more important to direct the flow than have the greatest possible flow rate.

Be careful using tables. The advanced aquarist article is pretty well thought out and it explains how to use the tables. It is easy for a lay person to misread the first table that was linked and there was no explanation on the page so I would advise against using that one.

For industrial piping projects, an engineer will typically develop a "system curve" which plots out the vertical head difference + friction + minor losses in head units vs. flow rate. This can be directly compared to the pump curve supplied by the manufacturer. The point at which the curves cross gives you the theoretical system flow rate using that pump. Developing a system curve allows you to quickly compare different pumps.

Be careful when bending PVC. You still want to put in only as many bends as is absolutely necessary. The earlier post showed some double curves that were probably done to save space but may have made the energy loss greater. Also if the pvc cross section area is reduced during bending, the energy loss will increase as the velocity increases offsetting some of the gains from wider bends. It does look cool though.


Having said all of this, I am actually inclined to agree with the other posters who state that it is not worth all of the calculations for most people. The calculations to determine the head loss at various velocities to come up with a curve are tedious without a computer program. The manufacturers of small pumps don't even provide pump curves. If you can do the sweeps without much cost, go for it but oversizing the pump a little won't hurt anything, adds only incremental cost, and is a whole lot easier. If you already have a pump and are trying to fix the plumbing to increase the flow, wide curves will help some and I have seen a number of posts where that was enough to make the system functional.

Thank you Keith, you have been much help.

I was looking into other pumps like the Wave II but honestly the cost to run does not fit well with me. The ReeFlo pumps are the most energy efficient pumps in their class.

I will do some more research.

 

[LouH]

I would just use more smaller pumps to accomplish the same thing. Odds are that you will get more flow/W, and your plumbing will be easier to execute.

I work in manufacturing where powder substances are transferred from silos to receiving vessels hundreds of feet away. Pneumatics are used to generate the flow, and the material flow is considered dense phase transfer. All of those systems are built with minimal changes in direction, and when one is required, the engineers use huge radius sweeps. When you use abrupt 90's, eventually the abrasion from the material slamming into the corner of the 90 eventually causes it to fail. Every component in a transfer line causes losses in laminar flow due to the turbulence created by the fluid in contact with the walls of the pipe hitting these in-line deviations. Overcoming that turbulence eats up lots of energy which corresponds to a loss in flow. Butterfly valves are probably the worst. I'm sure ball valves are very poor in this regard as well. I'm sure that there is a chart out there which provides comparable head losses for valve ccomponents. This alone helps make the case for running multiple small pumps and getting rid of a bunch of ball valves and 90s. The other advantage of multiple pumps is that one pump failure won't take out your entire flow system.

Lou

Lou

 

[LouH]

I would just use more smaller pumps to accomplish the same thing. Odds are that you will get more flow/W, and your plumbing will be easier to execute.

I work in manufacturing where powder substances are transferred from silos to receiving vessels hundreds of feet away. Pneumatics are used to generate the flow, and the material flow is considered dense phase transfer. All of those systems are built with minimal changes in direction, and when one is required, the engineers use huge radius sweeps. When you use abrupt 90's, eventually the abrasion from the material slamming into the corner of the 90 eventually causes it to fail. Every component in a transfer line causes losses in laminar flow due to the turbulence created by the fluid in contact with the walls of the pipe hitting these in-line deviations. Overcoming that turbulence eats up lots of energy which corresponds to a loss in flow. Butterfly valves are probably the worst. I'm sure ball valves are very poor in this regard as well. I'm sure that there is a chart out there which provides comparable head losses for valve ccomponents. This alone helps make the case for running multiple small pumps and getting rid of a bunch of ball valves and 90s. The other advantage of multiple pumps is that one pump failure won't take out your entire flow system.

Lou

Lou

 

[TheFishTeen]

Lou,

Im not sure what you consider a 'small pump', but having two pumps won't necessarily fix the problem. Smaller pumps usually can't handle 18'+ of head pressure, and even if they can, they're small! They don't push enough water to correctly implement an OM4.

I think from the calculations Haffs09 has provided me, plus removing ~500 GPH for unknown factors, I am still getting enough flow to use an OM4.

 

[KeithP]

Has anyone established a simple rule of thumb like assume an additional 2 or 3 feet of head for the piping on a typical tank? I think this would help a lot of people feel more comfortable ordering a pump.

 

[TheFishTeen]

Has anyone established a simple rule of thumb like assume an additional 2 or 3 feet of head for the piping on a typical tank? I think this would help a lot of people feel more comfortable ordering a pump.

I was wondering the same thing. Also what does putting a 90 on the intake of the pump do in terms of pressure/friction?