I am grappling with a gravity fed water system and have questions

[Jeff Thompson]

How did you find PF?: Google

I have an off grid home that uses a spring fed gravity flow water system. The spring is about 200 feet up a steep mountain- and very low flow. The pipe run starts at 2” pvc and eventually is 1.5” pvc. Once down the hill the water flows about 1 mile down an old ditch - pipes exposed at times and buried most of the way. The elevation is slightly lower at house below the initial drop of 200 ft. The pipes go slightly up and down over the long run.

I’d say the end is around 250 ft below the spring. Though the initial water supply is quite low - maybe 3gpm - it is continuous and ends up around 20gpm at house. A 1000 tank fills in about 5 hours - but it goes uphill from ditch and pressure is reduced way back.

There is a small pond where water can flow in order to always have an outlet- which is a bit above the end of the line at house. The pond water spout gets the maximum volume of water - at times gushing out. 100 feet lower is the house where water pressure is very low - less than 20psi though it fluctuates depending on spring flow and time of year. Now to the hard stuff.

Due to storms, snowy winters and bears breaking pipes- I regularly have to hike up ditch to locate problems and repair. Often it’s pipe breaks. But even with optimal flow from spring I’ve never been able to get adequate pressure to house.

So. I found a much better spring and tapped it - this flows about 20gpm right out of ground. The elevation of second spring is at base of mountain - thus 200 feet below main spring. I connected the spring pipes about 100 feet down the ditch - using a 90. The lower spring flows by 90 into original spring. With a valve.

The lower spring has ar least 3x water volume. In connecting the 2- it seems the original spring pipe that delivers water down the whole run - is sputtering air and not happy with the second source of water. Eventually the flow returns but it is stifled and uneven. Air sputters out at bottom (pond). If I turn off new soring source - the flow is normal but not very good. The second source has a lot more water volume to add - and is also a backup If the first line breaks above. But obviously connecting the two lines caused air locks? They are connected in a spot where the pipes travel slightly upward before going steadily flat and down the long ditch run.

Questions:
How do I utilize new spring source and optimize flow? Air lock release fittings?
Not possible due to elevation differences from 2 sources? The water has to flow somewhere once it enters pipe- I don’t believe it can go back the way it came. It’s possible it goes back up the original pipe - but eventually that goes straight up a steep hill so the pressure would force water back the other way if enough time goes by, the pipe would backfill with water until it couldn’t go any higher and that water would be pushed back down the line?

I don’t understand the physics enough to improve on the system but my suspicion is I need air lock releases somewhere?
Any thoughts?

 

[DaveE]

How did you find PF?: Google

Though the initial water supply is quite low - maybe 3gpm - it is continuous and ends up around 20gpm at house.

This is a typo right? You do know you can't make 17gpm by magic inside a pipe?

I found your description really hard to follow, although I wasn't motivated enough to reread it and do some puzzle solving. Maybe you could make a hand sketch of the complete system and post a photo of it, like a side view elevation vs. distance. In particular, we really need to know where the piping is continuous and at which locations air can enter the system.

I know it may not be practical, what with bears and such, but the best system will have continuous piping without any chance of air entry to allow a big pressure difference between the ends of the pipes.

 

[Borg]

They are connected in a spot where the pipes travel slightly upward before going steadily flat and down the long ditch run.

It’s possible it goes back up the original pipe - but eventually that goes straight up a steep hill so the pressure would force water back the other way if enough time goes by, the pipe would backfill with water until it couldn’t go any higher and that water would be pushed back down the line?

These two lead me to believe that you have a backflow issue related to the higher leaky pipe. The fact that it frequently leaks leads me to believe that the water is backflowing up that pipe when you turn off it's source and then leaking out. This would be a possible cause for the new pipe's pressure to drop when you do this. I would test it by closing off the upper source at the junction instead of at the top and see if that helps.

 

[Jeff Thompson]

This is a typo right? You do know you can't make 17gpm by magic inside a pipe?

I found your description really hard to follow, although I wasn't motivated enough to reread it and do some puzzle solving. Maybe you could make a hand sketch of the complete system and post a photo of it, like a side view elevation vs. distance. In particular, we really need to know where the piping is continuous and at which locations air can enter the system.

I know it may not be practical, what with bears and such, but the best system will have continuous piping without any chance of air entry to allow a big pressure difference between the ends of the pipes.

Yes. Sorry. I meant the flow at source is quite low but once down hill it has about 20-30psi.
That pressure is pretty steady at bottom of first steep hill. But friction and a mile of pipe reduces pressure so that it’s about 10-15 psi at house. I’ll draw diagram when I get home. The spring inlet is the only opening besides the bottom where the pond is. The spring barrel is buried tight. There are no other openings - with the exception of the second spring pipe which is also sourced from a buried spring tap. My diagram will with luck make this easier to visualize. Thanks.

 

[Averagesupernova]

I'm confused with your elevations. Where is the house, the 1000 gallon tank and both springs relative to each other elevation wise?

 

[Jeff Thompson]

The main spring is around 4400 ft elevation. The house is around 4150 ft. The tank is slightly above house but is not part of the continuous run. It fills if I open valve off main line. The second spring (that I recently tapped) is at base of hill around 4200ft. More volume and pressure at source. But has to travel slightly uphill for a bit after it joins main spring pipe- which runs a mile to house.

 

[berkeman]

The main spring is around 4400 ft elevation. The house is around 4150 ft. The tank is slightly above house but is not part of the continuous run. It fills if I open valve off main line. The second spring (that I recently tapped) is at base of hill around 4200ft. More volume and pressure at source. But has to travel slightly uphill for a bit after it joins main spring pipe- which runs a mile to house.

Maybe you could make a hand sketch of the complete system and post a photo of it, like a side view elevation vs. distance. In particular, we really need to know where the piping is continuous and at which locations air can enter the system.

This ^^^^

Please use the "Attach files" link below the Edit window to upload a JPEG or PDF copy of your sketch. Thank you.

 

[Lnewqban]

... More volume and pressure at source. But has to travel slightly uphill for a bit after it joins main spring pipe- which runs a mile to house.

Do you believe that you may have an inverted P-trap situation, in which a big pocket of air that does not have a way to escape is restricting the flow at the highest point of the layout?

Please, see:
https://en.wikipedia.org/wiki/Air_lock

Welcome!

 

[Jeff Thompson]

This ^^^^

Please use the "Attach files" link below the Edit window to upload a JPEG or PDF copy of your sketch. Thank you.

This ^^^^

Please use the "Attach files" link below the Edit window to upload a JPEG or PDF copy of your sketch. Thank you.

Do you believe that you may have an inverted P-trap situation, in which a big pocket of air that does not have a way to escape is restricting the flow at the highest point of the layout?

Please, see:
https://en.wikipedia.org/wiki/Air_lock

Welcome!

I think there is a negative or counter active pressure- causing restricted flow. Air can escape on both ends of run. But my thought is that the difference in elevation of the second spring is causing problems because it enters main line and goes uphill a bit. The water wants to travel the easiest direction which is backwards - against the flow. But the line still flows eventually. It’s just sputtering a lot.

 

[Lnewqban]

I think there is a negative or counter active pressure- causing restricted flow. Air can escape on both ends of run. But my thought is that the difference in elevation of the second spring is causing problems because it enters main line and goes uphill a bit.

If the pipe from second spring is connecting to top of pipe from first spring with a 90, and downstream from that point of connection goes uphill a bit, air does not have a way out.

When the valve at the pond closes, water from the higher spring is flowing into the lower spring.
Perhaps a check valve could prevent that from happening.

When the valve at the pond opens, water from both springs converge at the point of connection, but as flow rates are different, some air may be sucked from the higher spring into the main pipe, therefore the sputtering.
Perhaps lowering the intake of the pipe within spring 1 could prevent that from happening.

 

[Jeff Thompson]

That’s a good response. I think you’re right that the air from line two can’t totally escape.
I can’t change the input sources as the springs are located where they are at. But do you think installing a vertical air lock relief valve near the highest point after the pipes join would help? Keeping in mind it’s arduous work and takes a few hours up and back to do even slight changes. And I’m only up there once in a while. So I don’t always have time to see if things work. The main spring normal flow once restored always beats me back to house. But changes to the second line take a few days it seems to reveal.

 

[Tom.G]

You mentioned a 1000 gallon tank but it is not on your drawing. Where is it connected, what is the elevation of both the empty and full levels.

Assuming gravity feed without pumps, somehow the numbers you gave do not make sense.

The static pressure of water is about 0.43psi per foot of elevation difference. Since you have an open pond feeding the house with an elevation difference of 20ft., the maximum static pressure at the house would be 8.6psi. Due to friction, this of course will decrease as soon as the water is flowing.

The same holds for the plumbing from the springs. With just the higher spring (4400ft.) connected to the pond (4120ft.) the static pessure at the pond is 120psi. The lower spring (4200ft.) will yield about 35psi at the pond (that's about the typical water pressure (40psi) in a house connected to a city water supply).

Static pressures below these show the level of the water in the pipe. Again, reduced pressure with water flowing could be from any or all of:

1. friction losses in the pipe
2. not enough incoming water to keep the pipe full
3. an air lock in one of the high points of the plumbing, partially blocking water flow

With the valves open to both springs, expect all pressures to the pond inlet to be between the values for the two springs (120psi and 35psi).

Cheers,
Tom

p.s.

1. static pressures are measured with water in the plumbing and no water flow
2. all altitude differences are calculated between the nearest point that the water is exposed to the free atmosphere, and the point where the pressure measurement is taken.

 

[Jeff Thompson]

You mentioned a 1000 gallon tank but it is not on your drawing. Where is it connected, what is the elevation of both the empty and full levels.

Assuming gravity feed without pumps, somehow the numbers you gave do not make sense.

The static pressure of water is about 0.43psi per foot of elevation difference. Since you have an open pond feeding the house with an elevation difference of 20ft., the maximum static pressure at the house would be 8.6psi. Due to friction, this of course will decrease as soon as the water is flowing.

The same holds for the plumbing from the springs. With just the higher spring (4400ft.) connected to the pond (4120ft.) the static pessure at the pond is 120psi. The lower spring (4200ft.) will yield about 35psi at the pond (that's about the typical water pressure (40psi) in a house connected to a city water supply).

Static pressures below these show the level of the water in the pipe. Again, reduced pressure with water flowing could be from any or all of:

1. friction losses in the pipe
2. not enough incoming water to keep the pipe full
3. an air lock in one of the high points of the plumbing, partially blocking water flow

With the valves open to both springs, expect all pressures to the pond inlet to be between the values for the two springs (120psi and 35psi).

Cheers,
Tom

p.s.

1. static pressures are measured with water in the plumbing and no water flow
2. all altitude differences are calculated between the nearest point that the water is exposed to the free atmosphere, and the point where the pressure measurement is taken.

Thanks for thorough response. The tank is an offshoot. And the pond doesn’t supply water to house. It is also an offshoot of mainline that allows water to flow out once the house lines are full and not being used.
In other words - the pond is a place to always allow water to flow out - and I usually have that valve slightly open to allow the flow to continue to house.
The psi at house is still very low - as in less than 20. Despite the long run and elevation drop - that’s what I get at the house. Bear in mind the main spring is just a slow trickle entering a pipe. The pipe is not filled at any point along the route - other than maybe the very last part going to house where the pipe diameter is smaller. So the water is flowing but fills less than half the pipe. So air is present regardless -
There are potentially any number of issues along the pipe run impacting flow and pressure. Trees fallen. Underground breaks or cracks. I only know that I fix the obvious breaks and water flows when I do. There’s a lot of guesswork involved because I didn’t design and install the system and I’ve never been able to get info on it’s proper functioning. If the original owner were to tell me he used to get 60psi at house from that little trickle of spring water - then at least I would know there is a fundamental issue preventing the maximum flow. Either air locks. Or a damaged section or partial break that I’ve yet to detect. My hope was that connecting the second spring line would increase water flow and pressure at bottom. As well as giving me backup source to at least keep water flowing if main line stops working above. But so far. It’s just creating more uncertainty of water flow. And I’m hoping that installing a few air release valves at the high points along the run will help the flow.

 

[Lnewqban]

That’s a good response. I think you’re right that the air from line two can’t totally escape.
I can’t change the input sources as the springs are located where they are at. But do you think installing a vertical air lock relief valve near the highest point after the pipes join would help? Keeping in mind it’s arduous work and takes a few hours up and back to do even slight changes. And I’m only up there once in a while. So I don’t always have time to see if things work. The main spring normal flow once restored always beats me back to house. But changes to the second line take a few days it seems to reveal.

It is hard to properly evaluate the installation without being there.
I would not give you any advice with enough confidence, especially if hard work would follow.

I would install an automatic air relief valve at the points where air could accumulate and a vacuum breaking automatic valve inmediatelly upstream the connection point, if at all feasible.

Let me ask you: is there a reason for not just using the lower spring, installing a valve upstream the connection point, and leaving the higher spring as a back up?

 

[Jeff Thompson]

It is hard to properly evaluate the installation without being there.
I would not give you any advice with enough confidence, especially if hard work would follow.

I would install an automatic air relief valve at the points where air could accumulate and a vacuum breaking automatic valve inmediatelly upstream the connection point, if at all feasible.

Let me ask you: is there a reason for not just using the lower spring, installing a valve upstream the connection point, and leaving the higher spring as a back up?

My thought is the lower spring might not travel on its own all the way because it has to go uphill a bit. The upper spring travels straight down before reaching the long level run. It forces the water through - I think after a few days worth of buildup in line. I have a few “see through” backflow preventers I could install. My thought was to put one a bit upstream of connection and view what is happening (to see if water is knotted up from backflow). Downside is these devices could “stick” closed and cause blockage. But it might be worth trying. Thanks

 

[256bits]

With the valves open to both springs, expect all pressures to the pond inlet to be between the values for the two springs (120psi and 35psi)

With the valves open to both springs, expect all pressures to the pond inlet to be between the values for the two springs (120psi and 35psi).

shouldn't it just be hovering around 35 psi, with both valves open
Springs are pressure sources, not flow sources,
The higher spring would be feeding into the lower spring.
That would be for a completely filled pipe from the connection to the house.

With only the higher spring connected, the pressure at the connection point would be the head difference of 200 ( 4400 - 4200 ) if the pipe is full to the top.
With the lower spring connected and operating, one basically has lost all that 200 feet of head to send water to the house, leaving just the remaining 100 feet of elevation from the connection point to the house.
But you said that the pipe is not filled from the higher spring due to its low output.

The pipe is not filled at any point along the route - other than maybe the very last part going to house

The psi at house is still very low - as in less than 20

If the pipe is not filled all the way to the top, you do not get the benefit of the 4400 elevation of head, but only that to where the pipe is filled. How do you know it is not filled? It would fill to the 4400 level if you shut off the exit to the pond, and spring 2 is also shut off.
Perhaps you are getting head only from where the tank is connected.

My thought is the lower spring might not travel on its own all the way because it has to go uphill a bit.

Not on its own.
But if the pipe is full it would depends upon the velocity and pressure head at the connection.
If not full then atmospheric pressure.

It is hard to properly evaluate the installation without being there.

Spring 1, spring 2, pond, house, tank...
It is really quite difficult to evaluate since the information is hard to follow.
And some confusing statements

More volume and pressure at source.

for the lower spring?? The head is less than the 4400 if the pipe is full, which it isn't, or is it?

Though the initial water supply is quite low - maybe 3gpm - it is continuous and ends up around 20gpm at house

for how long? The pipe would empty out if 20gpm is going out and only 3gpm entering.

 

[Jeff Thompson]

shouldn't it just be hovering around 35 psi, with both valves open
Springs are pressure sources, not flow sources,
The higher spring would be feeding into the lower spring.
That would be for a completely filled pipe from the connection to the house.

With only the higher spring connected, the pressure at the connection point would be the head difference of 200 ( 4400 - 4200 ) if the pipe is full to the top.
With the lower spring connected and operating, one basically has lost all that 200 feet of head to send water to the house, leaving just the remaining 100 feet of elevation from the connection point to the house.
But you said that the pipe is not filled from the higher spring due to its low output.
If the pipe is not filled all the way to the top, you do not get the benefit of the 4400 elevation of head, but only that to where the pipe is filled. How do you know it is not filled? It would fill to the 4400 level if you shut off the exit to the pond, and spring 2 is also shut off.
Perhaps you are getting head only from where the tank is connected.Not on its own.
But if the pipe is full it would depends upon the velocity and pressure head at the connection.
If not full then atmospheric pressure.

Spring 1, spring 2, pond, house, tank...
It is really quite difficult to evaluate since the information is hard to follow.
And some confusing statements

for the lower spring?? The head is less than the 4400 if the pipe is full, which it isn't, or is it?

for how long? The pipe would empty out if 20gpm is going out and only 3gpm entering.

Yes. Last first. The spring produces about 3 gpm. 20 psi at house is what I meant to write. At no point along the route are either pipes full. The system provides continuous water - about 4000 gallons a day for the upper spring alone. So the water is slowly entering pipe from barrel but the pipe is not submerged in water completely at spring barrel. The water immediately travels down a steep hill about 200 ft. Where it then travels relatively flat for one mile to house. It goes directly to house. The pond and tank are valved and stem off line. I turn pond valve off to maximize house pressure occasionally but otherwise leave it slightly open to allow water to have exit point as it is closed system otherwise.
Your comment that lower spring might negate the head prsssure of upper spring is possibly what is causing the problem. I’m trying to increase the volume of water (and hopefully the pressure at house) but its possible I’m just messing up the flow and would be better turning off the lower spring.

 

[Jeff Thompson]

Wow. I’ve never considered just shutting off everything to let the line fill up. Maybe that’s the answer? All of that pipe - a mile - with 4000 gallons a day trickling in - would eventually fill the pipe and the water could escape out the source or out an air spout. I was told the back pressure would blow pipe connections apart and to always leave pond valve open a bit. But I’ve never shut off pond for more than a day or so. How long would it take to fill pipe - if there Is appx 5000 feet of 1 1/2 inch pipe with 3 gallons minute at entry?
And can I assume all of that weight and volume of water filling all that pipe would increase pressure at bottom once I opened a valve again? I wonder how long it would take for the excess water to exit and how long that pressure would hold if I began using more water at once?

 

[Tom.G]

How long would it take to fill pipe - if there Is appx 5000 feet of 1 1/2 inch pipe with 3 gallons minute at entry?

I calculate the pipe holds 459 gallons. At 3 gallons per minute, that works out to 153 minutes, or about 2.5 hours to fill the pipe.

Is the pipe rated for 120psi, as calculated in post #12 above?

Cheers,
Tom

 

[Jeff Thompson]

Yes. It’s rated for 120 psi. And yes I already did the calculations and came up with similar numbers. Thx. Now I wonder if this is a key to increasing pressure?

 

[Hutch178]

I have a similar (kinda) situation.
I have a 1 1/2" pipe at the top of my property that draws off from a small brook. It travels down about 50' then switches to a 3/4" pipe that travels down 1800' to my 1/2 acre fish pond. It to seems to have trapped air pockets. In my situation I prefer the air bubbles! The reason is oxygenation for my fish. The issue I have is: If I submerge my pipe into the pond with an aerator at the end to create small bubbles will this drastically reduce water flow? The pond is spring fed under ground and has no above exit, water comes in from springs and goes out I believe where it comes in.
What I wonder is: will the water draining out of the pond through the springs cause a vacuum?

 

[BillTre]

If I submerge my pipe into the pond with an aerator at the end to create small bubbles will this drastically reduce water flow?

Probably if the air bubbles can't get out.

In my situation I prefer the air bubbles! The reason is oxygenation for my fish.

You could make something like a trickle tower where water splashes over a bunch of "solid media" but with a lot of air present.

The large surface of air-water contact and water turbulence promote gas exchange.
Aquacultural engineers have a rule of thumb that water falling 5 feet through a good trickle tower puts the water's dissolved gasses at equilibrium with the atmosphere. This will also desaturate the water or remove unwanted gasses, if those are problems.

Trickle towers can be made using a great variety of media in a great variety of containers (professional or amateur, expensive or cheap).

They are often used as a form of biological filter, but that issue is a different from oxygenation.