Roman Aqueduct w/Siphon lock?

leonz

Hello Rick,

This really sounds like a job for two three inch Gorman Rupp diaphragm pumps
and a very small screw small screw compressor to power them.
Does the sump have a mud floor?

RickDelta

leonz said:

Hello Rick,

This really sounds like a job for two three inch Gorman Rupp diaphragm pumps
and a very small screw small screw compressor to power them.
Does the sump have a mud floor?

Brick pavers over crushed stone

leonz

Ah, good you can either use a single suction line with a
basket strainer with the pump out of the water or leaving
the pump in the sump with the discharge line connected
to a flap check valve connected to the drainage pipe.

RickDelta

"@leonz"

Selling the water for pools and selling it by the truck load for household use would make him a great deal of money with 1,000 gallon lots at $45.00 per thousand gallons which is what they charge up here for water.

think about it! .... were already piping all this spring water across that highway, that's a big WAWA gas and convenience store right there!

Put "mountain fresh spring drinking water" dispensers right next to the Tesla charger stations and sell for $2 a bottle! : )

( I would have to find a mountain in the area though) : (


We do have the New Falls land fill mountain about a mile down the road here, its now like the size of Mount Everest! ..... but that may not be very appealing to the discerning bottled water drinkers. : (

pecmsg

What’s the operating costs for that?

RickDelta

pecmsg said:

What’s the operating costs for that?

sorry! ..... what is your question directed at?

Pumpguy

My first visit to this thread. Referring to @RickDelta's first post and illustration of the syphon, this is a classic vacuum priming application.

At the highest point in this piping, a vacuum priming valve (see attached example) would be added, the air outlet of which would be piped to the inlet of a vacuum pump.

Once the vacuum pump establishes the lift and water starts flowing downhill, gravity takes over and flow continues until all the water is drained and air enters the lift pipe.

The vacuum pump would run start/stop, controlled by a vacuum switch. So long as the float chamber of the priming valve is flooded, the air outlet to the vacuum pump would be sealed. If any air might come out of solution or otherwise enter the lift pipe, it would bubble into the float chamber of the priming valve causing the float to drop and open the valve. The vacuum on the air outlet side would lift water into the float chamber, closing the valve again.

A pair of 1 hp liquid ring dental vacuum pumps would be perfectly suitable for this application.

A bit of thought would be needed to set up the water level and vacuum pump switching arrangement.

I have seen this arrangement used many times, mostly for lifting river or lake water into heat exchangers for cooling large air conditioning systems. Power plants use it to keep the main condensers primed. Also seen it used for lifting water over a dam which then flowed down into an hydroelectric plant's turbine.

pecmsg

RickDelta said:

What’s the operating costs for that?

sorry! ..... what is your question directed at?

3) 5-HP Pumps?

leonz

Rick, please check the PM I sent to you.

RickDelta

"@Pumpguy"
Hello! ...... very interesting read! Thank-you!

Can't I simply use my self priming pumps as the vacuum source in a "siphon assisted" pumping arraignment?

RickDelta

pecmsg said:

RickDelta said:

pecmsg said:

What’s the operating costs for that?

sorry! ..... what is your question directed at?

3) 5-HP Pumps?

.... operating costs (electric usage) I have no idea : (
5hp 240 3-phase units

RickDelta

Zman said:

Some random thoughts…

Pumps really prefer pushing over pulling. I would be reluctant to move them above.

A siphon like you are describing is very difficult to maintain unless you are able to keep both the intake and discharge pipes underwater at all times. The air always seems to work its way up from the discharge if it is in open air.

Have you considered putting VFDs on the pumps and sensing the water level with a strip that will tell the exact level of the water. You could program the VFDs to maintain the water level at a constant elevation with little or no short cycling.

This is an interesting post. Be sure to let us know how this works out!

"@Zman"
"Have you considered putting VFDs on the pumps ...... "
Yes! that was my first thoughts!
...... a very expensive solution through : (

But even doing it with VFDs was very problematic staying within the small given sump pit area (given the extreme dynamic ingress flows .... probably not even possible)

That's why I decided to simply flood the bottom tier and use simple "bang/bang" pump control with regular pumps.

RickDelta

..... an excerpt from message with leonz

My paramount quest in this pump station redesign is to never have to enter the bottom tier level to service the pumps.

Lets say in your proposal, everything in the small sump pit, the pump fails.
Think: at times of heavy ingress flow, the bottom tier , in about 5 mins, is already breaching onto the second tier! (this is were the 400 amp 240 3-phase electric service panel resides for the building)
At that point I would have to wade accross 2ft of rising water just to get over to the pumps in the pit to service them.
I can't shut off the power to any of the other back up pumps or Ill be standing in 5 ft of water! (but I not foolish enough to wade into 2ft of water with 240 3-phase electric power on)

So, my thoughts are to install the self priming pumps at street level.

Note: one year we lost power for three days during a snow storm .... we had 12ft of water in the basement! .... all tiers had flooded! : (

109A_5

Hello @RickDelta,
Is the discharge end of the two 4" pipes at the "Delaware Water Canal 330ft away" always under water when water is being removed from the basement?
And is it below the level of the sump pit ?

RickDelta

"@pecmsg"

1St the reason you're chewing up pumps is the short cycling. Get multiple pumps small med and large to better match the load.

By allowing the bottom tier to flood flood I can keep all the pumps the same 3" 300 gmp size.

Lets say, on a low ingress flow, this pump is just sleeping most of the day, til the water level reaches up to the second tier. The pump kicks in and drops the level all the way down to zero in lets say 5 mins run time. Then goes back to sleep til the next day.

Now, same pump, high ingress flow, turns on when bottom tier full, pump is pumping but water level is only maintained or dropping down only a 1/4" per day. The pump remains running for maybe three weeks like this!

This is all fine! .... the water level is contained to its allowed storage area.

RickDelta

109A_5 said:

Hello @RickDelta,
Is the discharge end of the two 4" pipes at the "Delaware Water Canal 330ft away" always under water when water is being removed from the basement?
And is it below the level of the sump pit ?

At times under water , other times not.
12 ft level differential from our basement to the canal

RickDelta

Thinking:
If you look at that aerial view of my twin 4" PVC 330ft discharge pipe over to the canal ...... that end discharge point is really dumping into a branch feed path to the nearby canal (maybe only a 2" or 3" level differential)

If I were to extend my twin pipes actually to the canal itself into and down into its lowest depths and turn its discharge end down stream ...... would this not create a "drafting assist" to the discharge flow??

ie: a venturi fitting that asperates (vacuum) the water in my pipe

I think the fire truck pumpers do this "Drafting"

109A_5

Hello @RickDelta,
Just thinking if the discharge end could be made to be under water, the pipes are air tight, just a vacuum pump could be used to prime the siphon system. Maybe similar to your post title and along the lines of what the @Pumpguy was getting at. Probably a lot less electricity used in the long run than actually pumping water. The Vacuum Priming Valve may not even be necessary if the vacuum line can be routed way above the water level and the vacuum is controlled.

RickDelta

109A_5 said:

Hello @RickDelta,
Just thinking if the discharge end could be made to be under water, the pipes are air tight, just a vacuum pump could be used to prime the siphon system. Maybe similar to your post title and along the lines of what the @Pumpguy was getting at. Probably a lot less electricity used in the long run than actually pumping water. The Vacuum Priming Valve may not even be necessary if the vacuum line can be routed way above the water level and the vacuum is controlled.

"@109A_5"

...... Brutal Juice!! : )

The highest point in my piping is 27 ft above basement floor level. then drops down 330 ft by gravity.

.... whats the spring check at the discharge end do ?? : (

109A_5

Hello @RickDelta,

RickDelta said:

.... whats the spring check at the discharge end do ?? : (

If there is a air gap between the discharge end of the pipe and the canal water level the swing check valve would close to allow the vacuum to build to prime the siphon (lift the water to the point it will flow out of the building).

If you used a water pump to prime the siphon you would still need some valve so you just don't pump the water back into the pit or a check valve to bypass the pump once the siphon has started.

Jamie Hall

Somehow I missed that the highest point is 27 feet above the basement floor level. That elevation is marginal for a siphon. You may have some difficulty priming it, and may have to use both a water pump and a vacuum pump to do so.

The underlying problem here is that you are doing battle with Mother Nature -- and she always wins. The open basement should never have been there, frankly, in the first place. But... that's just me.

leonz

Hello Rick,

The Essig venturi pump would not work for this either as the discharge height is too high and you would need a very large towed air compressor.

The Warren Rupp double diaphram pump would work very well for this application.

leonz

RickDelta said:

Thinking:
If you look at that aerial view of my twin 4" PVC 330ft discharge pipe over to the canal ...... that end discharge point is really dumping into a branch feed path to the nearby canal (maybe only a 2" or 3" level differential)

If I were to extend my twin pipes actually to the canal itself into and down into its lowest depths and turn its discharge end down stream ...... would this not create a "drafting assist" to the discharge flow??

ie: a venturi fitting that asperates (vacuum) the water in my pipe

I think the fire truck pumpers do this "Drafting"

------------------------------------------------------------------------------------------

Once the water is below the level of the syphon the vacuum breaks and any remaining water falls back.

The Hale and Waterous fire pumps require the use of an electric priming pump to fill the pump chamber before starting the mid mount pumping system on a pumper truck when suctioning water.

Pumpguy

I agree with @Jamie Hall on that 27 foot lift being too much for a vacuum lift to the highest pipe.

Keep in mind that the water column height to be lifted is measured from the surface of the water. As the water level drops, the lift height increases. The depth of the lift pipe below the water line has no bearing on the syphon lift principle.

Also keep in mind that its atmospheric pressure that pushes water up the lift. The vacuum pump just lowers the pressure in the lift pipe to provide the needed pressure differential.

Is there any chance the highest pipe can be lowered and still connect to the piping that runs downhill to the canal?

If a connection to this piping can be made that is 18 feet or less above the lowest point we need to lift water from, then we would be okay.

Also, as @109A_5 states, we would need a swing check valve or water seal in the discharge pipe that runs down to the canal. So long as this seal is located below the lowest level we need to drain from, we should be okay.

The volume of the piping between the water level in the lift pipe and the check valve or seal below the lowest level we need to drain from would dictate the needed CFM capacity of the vacuum pump. Any vacuum pump would do the job eventually. Its just a question of priming time from when the vacuum pump starts to when the syphon flow begins.

RickDelta

"@Pumpguy"

I'm at 15ft suction lift hight with the submersible now.


Installing self priming  pumps at street level forced me to discharge up another 12ft before I could go horizontal gravity down.

RickDelta

Pumpguy

As drawn, the only thing preventing the vacuum pump primed syphon lift is a check valve in the down flow pipe to the canal.

Again, if conditions are as drawn, and the discharge pipe opening is below the water level of the canal, this will provide a water seal so no check valve is needed.

Not shown in your drawing, but if other connections that are open to atmosphere are present, we won't be able to pull a vacuum on this pipe.

The nice thing about this arrangement is its economy of operation. You would only have to operate the vacuum pumps on an as needed basis to keep the system primed. Once its primed, gravity takes over and no pumps are needed.

Pumpguy

Looking at your drawing again, I just thought of something.

If you had a bypass pipe with shutoff valve between the inlet and outlet of the self priming centrifugal pump, and after flow was established, you could open the valve in the bypass pipe and the established syphon flow would continue with the pump turned off.

Wouldn't that be wonderful?

RickDelta

"@Pumpguy"

My thoughts at this time:
Flood the bottom basement tier to use as a massive storage sump reservoir (1,500 gallon capacity - 2ft in depth).
I realize this would create a real "death trap" for service personal having to work on the existing submersible pumps.
Eliminate all electrical and pump equipment from this flooded tier.
Install four 3" 300 gpm self-priming pumps 2ft above street level directly over the existing small (3' x3') sump pit well.
PLC (Programable Logic Controller) to monitor discharge flow rates and water levels.

The primary "water pumping" will always be the self-priming pumps at the ready even in a "siphon assisted" paradigm to the pumps.

Hopefully, I can program into my PLC to start up the pump, run for 60secs (prime), shut down pump, check for siphon flow.
If no siphon flow, resume pump.
If siphon detected, sleep pump.

If siphon flow less than water ingress, resume pump.
If water level drops to near "siphon break" , turn on discharge pipe solenoid stop valve.

If water level again rises, release stop solenoid valve to resume siphon flow (no pumps).

If siphon only flow allows water level to reach top of tier, resume pump.

Pumpguy

Only comment I can make now is if the syphon flow is through the pump, it will be significantly restricted. If that restricted flow is still adequate, fine.

On the other hand, if you can use a bypass arrangement I described above, the flow will be unrestricted and will be to the full capacity of the pipe size at whatever pressure differential is developed.

How ever you get this problem resolved, please let us know. All to often we never learn how, or if, the problem was actually resolved.

pecmsg

Pumpguy said:

Only comment I can make now is if the syphon flow is through the pump, it will be significantly restricted. If that restricted flow is still adequate, fine. On the other hand, if you can use a bypass arrangement I described above, the flow will be unrestricted and will be to the full capacity of the pipe size at whatever pressure differential is developed. How ever you get this problem resolved, please let us know. All to often we never learn how, or if, the problem was actually resolved.

3) 5-HP PUMPS is a significant flow!

109A_5

Hello @RickDelta,

Since you already have a bit of infrastructure in place, just two check valves or just one if you can get the discharge under water. With the size of your pipe without a discharge check valve or without the discharge under water the air may sneak in the discharge and break the siphon.

The pump control would another interesting issue.

109A_5

Hello @RickDelta,
Maybe an Eco friendly receiver in the canal to raise the water level for the discharge pipe.

RickDelta

"@109A_5 A_5"
Can you explain a bit more as to why and when I would need the receiver please? (low canal level at times?)

If this "siphon assist" develops a significant flow, over the course of the year the pumps will be in sleep mode 75% of the year (gut guessing).

The pumps turning on for just 60 seconds to recover from a lost prime or ingress flow over the siphon capacity.

Question:
The siphon "pull" is there even during the pumps run time, doesn't that aid in the reduction of the total dynamic head, thus higher discharge flow?

RickDelta

"@109A_5"
...... By your graphic additions to my control screen, I take it your a PLC programmer as well??

RickDelta

Thinking:
If water level again rises, release stop solenoid valve to resume siphon flow (no pumps).

Why even use a solenoid valve in the discharge pipe?

Just run pump for 60 seconds every time when storage basin level is full,
Check for a siphon (magnetic flow sensor),
Sleep pump if siphon is there,
if not,
Pump down to empty basin.

RickDelta

pecmsg said:

Pumpguy said:

Only comment I can make now is if the syphon flow is through the pump, it will be significantly restricted. If that restricted flow is still adequate, fine.

On the other hand, if you can use a bypass arrangement I described above, the flow will be unrestricted and will be to the full capacity of the pipe size at whatever pressure differential is developed.

How ever you get this problem resolved, please let us know. All to often we never learn how, or if, the problem was actually resolved.

3) 5-HP PUMPS is a significant flow!

The design capacity of the 4 pumps (300gpm each .... @optimal TDH)
My actual service gpm is around 230 gpm with head loss.

One of the pumps is just a back-up in sleep reserve (just to service the other pumps or in a catastrophic event, at time of max. ingress ever and we burst a 120 gallon water tank all the same time)

Generally speaking, just "2.5" of these pumps are required during the extreme water ingress periods.

109A_5

Hello @RickDelta,
Once a siphon is established two things that can stop it is, stop the flow (a valve) and air ingress from either end. Since your piping is prominently horizontal on the discharge side I can see air getting in relatively easily and working its way to the top of the piping and then your siphon is done, requiring a restart. With the Receiver it would prevent air from getting into the discharge pipe regardless of the canal water level, and you would not need a discharge check valve, a possible maintenance issue.

As @Pumpguy suggested with a valve to totally stop the flow the siphon could remain primed if the air can't get in from either end. So theoretically with proper control you would only need one prime ever. If the single siphon can't keep up with with the water ingress to the basement the second siphon could be activated then if needed the pump(s) could be run in stages.

A PLC programmer's dream.

109A_5

Hello @RickDelta,

RickDelta said:

"@109A_5"
...... By your graphic additions to my control screen, I take it your a PLC programmer as well??

I just use MS Paint for the edits. I have done some PC and Embedded programing projects over the years and have repaired PLC type equipment.

RickDelta

I still have no clue as to what the "siphon only flow" gpm would be.

When I shut off the pumps and noticed the water was still discharging long after, then putting my ear up to the pipe and touching it, I could tell a very large, solid volume of water was flowing thru this pipe!

...... when I pulled out the drywall screw again, the air ingested into the hole lasted for a very long time, gave me a sense of the length of the siphon's effect in the 330ft of pipe ..... was amazing!!