nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Child Item

Off grid polytunnel, needs power

ch750536

So I have a situation where I need power in a polytunnel which is about 300m from my house.

Over the winter I wish to continue to grow lettuce \ wheatgrass and I have LED lights which consume around 110w sqm.
I have a switch that automatically comes on based on light levels.
Although the tunnel is large (30m * 6m) I would only use a max of 10sq in the winter.

So if we eliminate the darkest 20 days as extreme, the rest of the time we would need led's for max 14 hrs a day.

14(hrs) * 110(w) * 10(sqm)= 15,400 watt hrs
1 day's storage needed so lets say battery depth of 32000 watt hrs.
LED's are 230v
Battery requirements are 160 ah (ish)

Q1 - Are my numbers right?
Q2 - What is the best (cost effective) way to do this?

Find more posts tagged with

Quick Links

Comments

[Deleted User]

Best measure the led consumption, don't take the rated consumption as an absolute that's only a maximum threshold.

How are you going to charge the battery?
Are these LED grow lights or work lights? If they're work lights you could use PIRs or switches.

Assuming the load is actually 15.4kWh
15.4kWh @80% efficient (inverter losses) = 19.25KWh *2 = 38.5kWh + 10% overspec. for battery ageing = 42.35kWh

/48 = 882Ah @ 48V

/24 = 1764Ah @ 24V

/12 = 3529Ah @12V

freddyuk

110w per sq metre is a massive amount of LED light? Please check this first as it is critical. Are the lights low down to increase effectiveness (if growing lights). Working lights would be tiny in comparison. You could light the whole tunnel for under 200 watts! LED's need to be DC 12/24v or you are wasting power using an inverter.

ch750536

They are 110w using a power meter. This is reasonably low power for grow lights.

How are you going to charge the battery?

That's the fun part. We have plenty of wind. Once I pin down demand \ storage then we can work out the generation requirement. It doesn't have to be 100%, am happy with much better than not growing over the winter.

I have access to cheap deep cycle batteries (forklifts etc) so I'm hoping to utilise those.

[Deleted User]

That's a good recipe.
Hrm...perhaps a grid tied turbine and a 240V - 120V battery would be best.
Those sort of DC voltages are not for the faint of heart though, pretty lethal and expensive switchgear.

ch750536

It has to be safe. I'm handy enough with a pair of pliers but there is lots of water running around in a polytunnel.

[Deleted User]

Well it's as safe as you make it. The danger is that it's lethal while you are working on it much like mains except with the added excitement of sustained arcing.
Just a suggestion. It'd mean you could use a super cheapo grid tie turbine and inverter and then save yourself the expense in cost and efficiency of a battery inverter.

Probably the best solution is a 24v system as Freddy suggests or a 48V with 24V lighting in series pair if that's possible.

[Deleted User]

Can you send a link to the grow lights? Chances are they have a LED driver in the back which can just be circumvented for an efficiency boost. I'd expect they're a native ~30VDC.

ch750536

The lights are self made using some cree lamps, the sort you would use in the home. They are mains bulbs but I can replace them if it helps the overall system.

[Deleted User]

Make them the same voltage as your battery nominal and you won't have conversion losses.
Generic domestic LEDs don't have the spectrum emission to make suitable grow lights.

air

Why not just run a cable and supply it from the grid?
I'd probably work out a lot cheaper and more reliable.

[Deleted User]

More reliable is certainly arguable.

ch750536

Sir Liamalot wrote: »

Make them the same voltage as your battery nominal and you won't have conversion losses.
Generic domestic LEDs don't have the spectrum emission to make suitable grow lights.

Already tested & work fine.

ch750536

Sir Liamalot wrote: »

Make them the same voltage as your battery nominal and you won't have conversion losses.

So realistically using old forklift batteries (24v I think) I should use 24v LED's, cutting out hardware & losses? Not a massive issue, the bulb cost is minor in the long term. At the moment the test panel holds 46 220v bulbs but I think I can source equally good 24v ones.

ch750536

air wrote: »

Why not just run a cable and supply it from the grid?
I'd probably work out a lot cheaper and more reliable.

Long way over a road & farmland.

[Deleted User]

ch750536 wrote: »

So realistically using old forklift batteries (24v I think) I should use 24v LED's, cutting out hardware & losses? Not a massive issue, the bulb cost is minor in the long term. At the moment the test panel holds 46 220v bulbs but I think I can source equally good 24v ones.

Forklift cells are 2V so 12 in series for 24V or 2 parallel x12 in series etc.
Yeah better off; battery - fuse - switch - light. Anything else is introducing expense, inefficiency and electronic self-consumption.
Those lights sound impressive. Grow LEDs have IR and UV clusters and actually power photovoltaics.
Ideally you'd only discharge a lead acid system 10% per day to allow roughly a week between full charges.

ch750536

Sir Liamalot wrote: »

Those lights sound impressive. Grow LEDs have IR and UV clusters and actually power photovoltaics.

Plants grow in sunlight. Simulate sunlight and plants will grow. How well you simulate sunlight will determine how well the plants grow.
Large parts of the green spectrum are not used by the plant for growth, that doesn't mean they don't need it - they just need it less. It is also true that as long as you supply the wavelengths it needs to some degree it will typically grow well.

I have run tests with lettuce & chilli, both of which performed well under the artificial only light. In the polytunnel there will be some winter sun along with the lights, hopefully enough.

I'm also thinking about reflectors on the outside of the polytunnel but I have concerns with the weather destroying them.

ch750536

Sir Liamalot wrote: »

Ideally you'd only discharge a lead acid system 10% per day to allow roughly a week between full charges.

If I had lead acid connected to a turbine would this suit?
Would this still be 24v DC compatible?

[Deleted User]

If you had a large enough turbine with a head wound for 24V, a three phase full wave rectifier, a charge diversion controller and a dump load it would.
Also probably want a tower, gin pole, guy wires, bottle jacks, stakes, winch, anchor, electronic brake, isolator, swa and conduit. You could use a monopole and solid foundation instead of guys. It looks far neater but will cost extra.
Suppose it all depends on the scale what's best.

You could set the lights to turn off/on too at set battery charge thresholds with some more gubbins.
Two TriStars would work well one set to divert another set to load control.
Relay drivers would work too, cheaper but not as refined.

Come to think of it you could use a tiny battery to begin with and stage the lighting to come on only when the turbine is working and according to it's output.

If you're handy you can make a turbine far cheaper and better than most you'd buy.

Personal preferences;

Small to mid sized
DIY: Hugh Piggott build.
Off the Shelf: Leading edge

1kW+
2 x Hugh Piggott builds
Chinese: Yang-Shen with a serviceable head and some overspin reduction modifications.
Best you can get: Proven/Kingspan

Before you even think of buying a turbine it'd be advisable to fly a data collecting anemometer at hub height preferably monitoring the season you will want it most.

If you want to maximise your harvest you might consider two machines: a high wind and a low wind.

Get a HAWT the rest are snake oil.

air

ch750536 wrote: »

Long way over a road & farmland.

If it's your own farmland then it'll still likely be cheaper and easier than an RE solution.
Pretty much the only technology which can compete with grid power on cost is PV and that's going to be useless for you given that you want power when PV will be producing nothing.
What's the nominal wattage of all your LEDs?
Are you using constant current drivers?

ch750536

Sir Liamalot wrote: »

Before you even think of buying a turbine it'd be advisable to fly a data collecting anemometer at hub height preferably monitoring the season you will want it most.

I can see 138 turbines from my house - it's a windy spot alright.

[Deleted User]

Yes but how windy is the question. If you know your mean wind speed you can get a turbine with a maximum efficiency in that envelope. Cheapo weather station jobs with data logging and warranty in Maplin, just plug the display into a computer every 3 months to graph the mean speeds...pays for itself.

ch750536

Sir Liamalot wrote: »

Yes but how windy is the question.

Sustainable ireland has good maps, 6m\s at 20m height so around 4m\s at 3m

[Deleted User]

ch750536 wrote: »

Sustainable ireland has good maps, 6m\s at 20m height so around 4m\s at 3m

Not accurate enough to your site might as well be licking yer finger and offering it to the breeze. You can't see the turbulence (gusting) or clean flow. Plus if you can't get 5 m/s it's hardly worth it. If you're hoping to be making kilowatts get her up at 7m - 9m hub height . €0.02.
7m/s average and yer laughin'.

ch750536

Sir Liamalot wrote: »

Not accurate enough to your site might as well be licking yer finger and offering it to the breeze. You can't see the turbulence (gusting) or clean flow. Plus if you can't get 5 m/s it's hardly worth it. If you're hoping to be making kilowatts get her up at 7m - 9m hub height . €0.02.
7m/s average and yer laughin'.

This will only be winter use though and 'just enough' is the best fit here. If costs spiral then I might be better off running a cable 300m.

If the lights need a construction project to power them then it's probably a non starter.

What will be just enough to keep 1764Ah @ 24V?

[Deleted User]

I think you are looking at this the wrong way round.

Think of a car, the battery doesn't run the appliances, the alternator does. The alternator does to an extent charge the battery but that's almost a tertiary application.

You don't need a battery, you have a load that's non essential and ultimately driven by a turbine. Having a large battery introduces charge inefficiency losses and deterioration of chemistry if not maintained correctly.

You'd probably be better off with a long mains cable that's ampacity rated ignoring voltage drop (not entirely kosher) because you can afford to lose a hellova lot before it has any detrimental effect on ELV lighting. ...but where's the fun in that?

What I would do is fly a 600W > 1.2kW turbine or two, or two + one low wind with a 60Ah battery and run (full spectrum) lights in "direct drive".
Turbine -> rectifier -> battery -> relay driver -> lights.
You don't even need dump loads because the lights are the load control. {EDIT: as long as max. load exceeds max. combined turbine output}

All the battery is for is to voltage clamp the turbine, it does not run loads. Soon as the relay driver sees a charging voltage it turns on the loads until the voltage stabilises. The more power your turbine produces the more lights it turns on. When the turbine stalls the relays open, lights turn off and don't discharge the battery. Simples! No charge losses, no sulphating ¾ ton of lead acid, no battery maintainence, more efficient too but your lights run at the whims of the wind which they were going to do anyway...realistically you'll never be able to charge a battery that big every day with wind alone.

You can build relay drivers with Arduino or similar, or buy one; MorningStar Corp. do an off the shelf solution, if you find an electrical engineer they could probably make you a PLC panel for similar.

The lights would be staged so one comes on before 2 etc. this means to get the most benefit from them you will need to rotate the plants or the lights or the light operating priority.

air

That wouldn't work Liam without PWM control. The steps between banks of lights coming in would likely lead to voltage spikes which would blow the lights. A battery bank would dampen these spikes considerably.

[Deleted User]

PWM only works for resistive.
a 110W load will drop a healthy 60Ah @ 24V battery from 25.6V to ~ 25.1V. Hardly what I'd can a spike...or even a dip...

air

Sorry I misread, thought you were proposing not to use any battery at all.

[Deleted User]

Yeah I kinda said that but not what I meant. The turbine or lights or both will die a flaming death without the battery to voltage clamp it. I meant you don't need a battery for the lights. Need a little battery for the turbine though.

ch750536

Sir Liamalot wrote: »

I think you are looking at this the wrong way round.

That's interesting.
I think with that approach I would turn on some lights over each sq. m to even out the effect.
Each area would then get some benefit meaning no swapping needed.
More power, turn on 10 lights in each area.