Siphon Feeding a turbine

[Lowhead]

Hello all.  I am the owner of a former millpond dam in the metrowest area of MA and am beginning the approval process for a low head turbine.  There is 9.4 feet of head and generally 10-12K GPM.  It varies as low as 1k in August to sometimes over 60K GPM in the spring. Since there is no pipe through the dam, Paul at ES$D turned me onto the idea of siphon feeding over the top of the dam into an open box that would hold the LH1000 turbine.  My plans have an 8 inch PVC pipe feeding into the welded steel box (about 20 foot run) and then the LH1000 uses a short length of 6 inch pipe to extend from the draft tube up to the turbine.  At this point I don't have FERC permission so I can't grid tie yet.  I was thinking to keep costs down and to keep it a sort of proof of concept that I might just use the energy to heat water as a first pass - I already have a radiant floor system that should tie in nicely.  Once all the kinks are worked out, then I am thinking I will spend the $$ on an inverter etc...  Has anyone had any experiences with siphon feeding?  The cheapest valves I have found are still big bucks.  I was wondering if I could extend the siphon penstock into the box enough that I might be able to just cap the pipe off with a conventional PVC cap (not bonded) or use one of the molded rubber plugs/caps.  This would be to get the siphon started using a vacuum pump.  I would think that if the siphon system works out, then I would spend the $$ on a valve.  A couple of other thoughts are to make the box big enough to hold 2 LH1000's for future expansion.  I could run larger 10 inch piping into the box or split the box in half and run two different 8 in feeds in.  I have designed the pipe to be self draining if the siphon stops.  How I am going to support the box and pipe nearly 10 feet in the air has not yet been worked out.  I have access to welding and machine shop equipment.  Concerns of mine: Freezing, minimizing wetlands disturbance, reliability of the siphon and vandalism (should I use steel piping?).  Any thoughts or suggestions would be appreciated.  It has cost me a lot of money so far just to get the plans and application process started for the conservation committee.  At this point we are going with a high level drawing but I will need to come back with more detailed plans (if they give a preliminary nod to the project)and would like to get it right the first time.  The conservation commission is going to be focused on the wetlands disturbance - not energy conservation so I would like avoid pouring any concrete footings in the wetlands if possible.  I was thinking I might be able to drill some anchors into nearby boulders (no shortage of rocks in New England) and then weld a frame to them.  At this point the plans just show a short tailrace trench that needs to be dug to get the water back to the river (I don't own the river or the millpond - just the perimeter/dam and hence my FERC issues).  Sorry for the long rambling post.

Thanks

Low Head

[miro]

I've seen a siphon fed turbine on the Muskoka River between Bala and the Ragged Rapids power dam. There used to be a small power station there but it was decommissioned in the 1950s. Once the energy crisis hit in the 1970's the power utility installed a siphon arrangement to keep costs low. It is an all-weather arrangement with no buildings.. As I recall, the pipe is about 30 in diameter.

It seems to be working OK.

[Warren]

Hello,

I have no experience with siphon feeding a turbine but with the big variation in your flow you may find you are having to restart your siphon fairly often as the flow decreases.

If you use plastic pipe make sure it can handle the vacuum required to start the siphon.

A gravity system would seem simpler. Is there no way you can put a pipe in around an end of the dam?

Warren

[wdyasq]

I did a low head siphon years ago just to get some water over a low dam for a project. It was interesting. It can be done.

We used valves 'downstream' and a valve at the 'peak'.. The lower valve would be shut, the upper valve would be opened. A small, hand held vacuum pump (actually a bilge pump for a dingy) would be used to fill the pipe by vacuum. The valve at the peak would be closed and the lower valve opened. Water would flow.

Gas in the water, a dead fish or other things would sometimes stop the flow and they would require restarting. We were using 4" PVC. PVC will stand the low vacuum pressures you have mentioned.

One could easily build a flapper style stopper valve at the lower end and lever operate it. If one figured an 8" round flapper valve and 10' of head, the opening pressure would  be ~250 pounds. A four foot long bar would be plenty long to open the valve, less if mechanical leverage was increased.

A small amount of water and a venturi suction device could be used to keep the siphon active as long as water flowed.

Ron

[thirteen]

A pipe around the end of the dam sounds better to me. Your pipe above ground will need to be covered for protection and for stability. You could always put poles in for support. Sounds like you have enough volume. You will need to make a shutoff plate on the intake of the pipe. Plus a screen.  Cables tied to the larger rocks works but the base for everything to sit on will need a solid base plus a small shelter to protect from vandle's and the weather. I do not know thw water inlet for the milpond towards the dam but will the pond fill ok taking that much water out of it? After the turbine you might go down farther and put a waterwheel in the stream. Where will the wires be mounted on poles or in the ground going to the power house. Copper wire is at a salvage priece that is high.  If you put cables in place get some crawling vines to cover the cables or plant some trees. Be sure and put a turnbuckle on the cables so they can be tightened as the years go by. Put antisieze(sp)on the bolt threads so they can be turned over in several years.  Just some ideas.

[Lowhead]

Thank you all for your above comments.  I apologize for the long block of text and will try to keep this one short and use better punctuation.  Piping around the dam is not an option.  It is all wetlands and disturbing it as little as possible is key to approval.  There is no way I can bring in any sort of power equipment either - it is pretty much all swamp and the Conservation Committee would go nuts.

There is little change in the pond level due to flow rate changes.  I plan to get the suction strainer at least a foot below the lowest historic level.  The pond is only about 3 feet at it's deepest but only about a 1 foot deep at the dam so I will be using a post hole digger to drill a hole that I will line with a pipe (shhh - don't tell the ConCom or the millpond owner).

I like the idea of a lever operated flapper valve.  Somebody suggested I need a valve at the intake end but I don't think it is required - just a screen.  I was thinking the intake end would be tapered out like a draft tube to reduce intake velocity.  I believe the intake vortex might suck in air and stop the siphon.

I can't put a waterwheel in the stream because I don't own it.  My original plan was to use a crossflow turbine in the stream until the deed research showed that the river is privately owned.  Because of that, I can't put the draft tube in the river and have to dig a short tailrace (by hand).  That is going to give the ConCom heartburn but I think it will fly if I have done everything I can to minimize disturbance.

As far as the wiring goes, I can't dig any trenches so it will have to run on poles.  Anybody have thoughts on stringing through trees with lots of slack?  I have about 300 feet to the house.  I'm thinking 48 volts or maybe even the 120v option - especially if I am going to heat water.

Ok, I think that is long enough for one post.

Low Head

[wdyasq]

As for voltage I would 240V 3PH if possible. It may be practical to 'overdrive' a 3PH and grid-tie (check the laws in your area). You will need to consult tables and see what sort of power you will be able to harvest.

Good luck,

Ron

[scottsAI]

Hello Lowhead,

I like to keep things simple.

Hang the pipe on the dam, 10 feet the pipe should be able to support itself.

Main Pipe can be several smaller dia pipe to keep it cheap. More head losses... balance cost to benefit.

I assume your using a Banki turbine? (Looking at your other post I don't think it is.)

The nozzle should be a sliding nozzle to handle the varying flow rate.

Make this nozzle able to close all the way. You now have your valve.

To fill the pipe use a bilge pump and fill it up. Not a vacuum pump.

No need to get a vacuum pump. Add a second much smaller Siphon to suck from the top of the main pipes. This second Siphon will keep a lower pressure up there. It must discharge below the turbine to make sure it is a lower pressure, I do not know how much air is in the water, 1/2 to 2" pipe may be needed. Size it correctly, don't need to waste any more water than you have to. This secondary Siphon needs to have enough flow it operates with the air in it, if too small the accumulated air could stop it from working. You might even find you can close this secondary Siphon, let flow once a week or as needed. The secondary Siphon is started the same as the main.

Startup

Using several smaller main pipes can benefit by starting one, use the suction to startup the others. A Siphon will work as long as the down stream water weights more than the up stream. With 10:1 you should be able to operate with LOTS of air (for start up) If the flow is fast enough air will not be able to accumulate at the top bend. Normal operations this will not work, expect turbine to restrict flow such that air could accumulate at the top bend...

With some thinking, one small siphon can be started, use it to fill the main pipes. How the turbine is connected will dictate if this could work.

Voltage

For a given wire gage and length results in power loss, use whatever you can afford. Thinner wire is possible with higher voltages. Balance the various cost to determine what works for you!

Freezing

Flowing water does not freeze, with that said very cold weather can chill the pipe and freeze it from the boundary layer on the outside of the pipe inward, restricting flow so it no longer flows (should not burst the pipes). The water in the stream must be above 32'F, with a good flow and 10 feet of pipe, I would be surprised it could freeze. Use 1 inch layer of insulation if your worried about it. In the winter if the water stops flowing, make sure you empty the pipes of water. (Or the pipes will burst!-)

Have fun,

Scott.

[Lowhead]

Thanks for the advice.  I'm sorry I wasn't clear on the freezing.  The piping is designed to be self draining so that isn't the concern.  My concern is with the suction end that is in the pond.  I am assuming that the water around it won't freeze when it is running but if the siphon stops, the water around it would freeze and cause structural damage.  

I am planning to use the ES&D LH1000 turbine (www(dot)microhydropower(dot)com.  Originally I was going to build a Banki but I had to make a strategic decision to go with a smaller unit that would have less wetlands impact.  I plan to build the siphon box big enough so that I can add a 2nd LH1000 at a later time.  If would like make the siphon pipe large enough to feed both but I don't think an 8 inch pipe will make it.  

As far as voltage goes, I believe the LH1000 can be special ordered with 120 volts but is normally configurable as 12, 24 and 48 volts.  Since I was hoping to start with just heating water (sort of getting my feet wet).  I assume that a resistive heating element doesn't really care if it is DC or AC?  If I did choose the 120V version, at some point (when I get FERC permission) I would like to buy an inverter to grid tie and have a couple of options.  I believe the 120V option requires the rectifiers to be moved out of the turbine and located after a transformer that will step the voltage down.  I think I could also use an MX60 from Outback that is rated at 150V and leave the rectifiers alone.

Any suggestions on supporting the wiring to the house using the trees?  I'm afraid the Conservation Commission will frown on digging holes for poles.  I would be afraid of just running direct burial cabling on the surface and digging a trench is a not-starter.

Thanks

Low head.

[scottsAI]

Hello Lowhead,

At those prices I would build my own Banki and use an induction generator. Then again I am so cheap that sometimes I don't get things done. Work on selling the savings in CO2 emissions as a benefit of the slightly larger turbine... You know your FERC better than I, at least yours is talking to you.

I talked to a guy that dug a couple inch trench and laid the wire in, was perceived to be less impact than hanging the wires. Not sure if freezing will be a problem have not talked to him since. I did lay a 700 foot cable on the ground for years, winter was not a problem.

Induction generator has one less step in power conversion, generation is slightly less efficient, with no follow up conversions it is better over all. Speed control is listed as a limitation of this, as long as the turbine is not allowed to over power the motor then the motor will limit speed and speed control is not required. Reverse power test is required to keep from motoring the turbine if the water flow is too slow. Induction generator (motor) directly connectors to power grid with a breaker.

Keep it simple Scott.

Have fun,

Scott.

[Lowhead]

Thanks Scott.  I was absolutely going to build the Banki and still might.  In MA by Riparian law, if you own riverfront property and it is "non-navigable" then you essentially own to the central point of the river.  In my case, it turns out that the Governor of MA in 1798 granted the mill rights to a sawmill that was on my property and that the mill rights were then sold to the cotton mill on the opposite side.  That mill went out of business in the 1930's.  

Because of all this, I don't own the spillway where the Banki would have been located but do own most of the dam.  The LH1000 will be the easiest to get approval on since I will only disturb a few square feet of wetlands.  I do hope for future expansion to be able to maybe go with a Banki in the old spillway.  First I would need to work out an agreement with the other owner - Mass Bay Transportation Authority.  That is a huge government bureaucracy so I am not going to wait.  

I have spoken with FERC and they were surprisingly cooperative.  It turns out that there are two paths to approval.  One for a new dam that is essentially a dead end and the other for if you have a small preexisting dam.  It is pretty much a rubber stamp process as long as you have control of the entire perimeter.  You don't actually have to own it, you really just need a letter of permission.  I also talked to my utility and backdriving a synchronous motor would require about 10K worth of safety / monitoring equipment.

BTW, I found a copy of the auction brochure from the cotton mill liquidation in 1933 and it had a 15 inch Leffel turbine driving a Westinghouse 80 Kw rated generator.  Looking at the  Leffel specs, I think it was only good for about 15 Kw.

Low Head

[scottsAI]

Lowhead,

Induction motor does not require approval to connect, its a motor.

You will just run it faster and make it a generator. If the Grid power fails the induction motor/generator will stop outputting. One of the safest ways you can connect a generator to the grid. You will want a reverse power detection so as not to run it as a motor if the flow stops.

When the grid fails you will need a transfer switch with caps and a load on it to keep the output going or the generator will spin twice as fast or more... could be a problem. Unless you can add a stop at the intake.

To get net metering approved add a couple solar panels with grid feed... don't mention the motor/generator, add later.

Amassing what you can find when looking around. My sister recently bought a house requiring fixing up. Looking through the cabinets found a early picture of the home. Found the fire hydrant is where the driveway is now. Fun.

Have fun,

Scott.

[Lowhead]

I understand your rationale for the "motor" but NStar (our local utility) didn't buy it.  They pretty much want an inverter with a UL1741 rating before they will set up net metering.  I do like your idea of a few token solar panels to get the net metering and then tying in the turbine.  I was thinking I could use an NC solenoid valve at the top vacuum port on the siphon.  If the grid goes down, it shouldn't be difficult to actuate the solenoid and let some air in to stop the turbine.

[scottsAI]

Lowhead,

Vacuum brake sounds good to me. Simple!

Just checking, your not going to use a vacuum pump...?

UL1741 - makes it impossible to build something yourself and connect.

Now if you said it was a pump... you would be OK.

Have fun,

Scott.

[Lowhead]

I haven't gotten far enough really to know exactly what I will use to get the siphon started.  I like the idea of a bilge pump to get started and keep costs down.  Paul from Energy Systems and Design had suggested an electric vacuum pump.  For now I'm just planning on a port at the highest point.  What I end up connecting to it depends on what I can scrounge.  

Low Head

[Lowhead]

So I finally got approval from the local Conservation Committee to install the siphon fed ES&D LH1000 turbine on my preexisting dam (9.4 feet of head).  I want to size the siphon box and siphon piping so that I could add a 2nd unit at a later date.  Does anyone know of a cheap place to buy a 10 inch valve that will work with schedule 80 PVC?  Where is the best place to buy the pipe and fittings?  I was also thinking to keep wire costs down (and be able to add the 2nd LH1000 without adding or redoing cable) I would run 48V now and change to 120V and use an MX60 charge controller when I install the 2nd.  Outback doesn't recommend it for Hydro.  My understanding is the MPPT disconnects the hydro once in a while and the voltage can shoot above the 140V rating.  Any thoughts?

Lowhead

[ghurd]

This is not going to help with the math,

but pipe is always cheaper to buy local.  Probably a valve that size will be too.

At least that makes it a bit more simple.

G-