Making multiple power sources play nicely

Hi,
Where we live, redundant power sources are essential - we have a lot of days of low sunlight (aka fog), and a hydro source which is vulnerable to storms and isn't 100% dedicated to our house.
So, the solar system is fairly new (praise be to replacement insurance policies!) and sends through between 10A and 15A when it's sunny. (Our batteries are 12V, 500AH capacity). It uses "standard" solar cells because everyone ignored our suggestions that amorphous cells would perform much better in our environment. The solar is on a Wellsee 50A solar controller (room for expansion!)
The hydro has a long cable run, since it is equidistant between two houses. So it has high cable losses. Typically we can get a background 2A 24/7 (leaving enough pressure for other users on the 1.4km pipe) or about 6A if we hog the water. The hydro is controlled manually - direct connection to batteries, and we walk up and turn the water on and off. FYI, hydro and wind power cannot run through a solar controller, because those controllers simply disconnect the load when the battery is full - and if you remove the load from a hydro or wind generator it spins free and wrecks its bearings fast.
Of course both systems are protected by diodes to prevent reverse voltages.
With these figures, the solar does the lion's share of the charging, with the hydro as backup, or to keep things ticking overnight. We have markedly less power to play with over winter!

Our problem is, that if the hydro is running, it puts out a floating voltage sufficient to charge the batteries (the voltmeter at the pelton wheel often reads over 15V). The Solar controller sees this voltage as the battery voltage, so it stops charging. If we go to town in the morning (as we do several times a week) it may well be cloudy/foggy enough to have the hydro on, so we risk missing out on all the solar power if the sun comes out (as it did today) or getting inadequate power from solar if we turn the hydro off and it stays sunny.

How can we isolate these systems from each other so that they can work together?

A battery bank is like a swimming pool. You can only add stuff to it until it gets full. If you generate more power than you use, then the rest can be burned off for other projects like water heating. You may want to consider using a "diverting controller" which runs a load when there is excess power. That could take inputs from both solar and hydro, and burn off any excess power.

As noted, you could use a charge controller with a dump load. The dump load is there to keep a load on the turbine (air or water) so it doesn't kill itself once the battery bank is fully charged.

Here is one example:

http://www.ebay.com/itm/Xantrex-C35-Charge-Controller-wind-generator-solar-Hybrid-Dump-load-/391183095361?hash=item5b14545641:g:svUAAOSwT6pVo76q


A hundred bucks and name brand.

You could run this completely independently of your inverter and/or your solar charge controller. It sits between the water turbine and the batteries.

Hook the turbine up to your charge controller as if it is a solar panel. If you hook it up that way the battery voltage is on one side of the charge controller and the solar/turbine is on the other so that the controller doesn't think your batteries are 15v anymore.

Ben House wrote:Hook the turbine up to your charge controller as if it is a solar panel. If you hook it up that way the battery voltage is on one side of the charge controller and the solar/turbine is on the other so that the controller doesn't think your batteries are 15v anymore.

That works if the controller is a "diversion controller" which redirects the water-turbine power to some purpose like heating water. It doesn't work with a normal solar charger because with charging switched off the turbine would spin way too fast, and burn up the bearings, or shake itself to pieces...

I see.

You could hook it up to the charge controller as a solar panel like I said, and on the battery side hook a normally open(?) 12v relay that is setup so when the voltage gets to full charge it makes a connection between the turbine and a load. An electric motor perhaps?

Dumping to an electric motor doesn't do any useful, maybe.

Diversion controllers usually dump to a heating device, either a water heater (every body needs hot water) or sometimes just a big resister that heats the air.

I have found that off the shelf relays are not super reliable. A diversion controller is (supposed to be) all solid state these days. Better reliability.

Ideally I think you don't want to run your dump load as described above, because it will make it hard to charge the batteries correctly and fully. The dump load needs to be big enough to handle the entire incoming charge with a safety margin. So, if the batteries are still charging, but nearing full, and you just turn on the whole dump load... too much load! It takes all the available current, and the battery gets nothing. So the battery voltage settles as the charging current is removed, and the diversion controller senses the lower voltage and shuts off... and then all available power is headed to the battery again and this is too much... battery voltage is above the set point again, and the dump load turns on...

How much of a problem this is would depend on the individual components... IE, I have no idea how problematic it would be in the real world with these bits!


Richard, I think there are a few possibilities.

1) A diversion controller, as everyone else suggests, sounds wise for the sake of the turbine... but as I read it, a diversion controller alone might not help get your solar array charging. If the solar is merrily charging away, and you turn the hydro on and leave, a diversion controller with a low enough setpoint to trigger while solar is charging the batteries(at whatever arbitrary state of charge exists atthe time) would immediately send the hydro power to the dump load. Then, if the sun goes away, it would detect the lower voltage and send the hydro power to the batteries. However, if the sun comes back, the solar charge controller would still be faced with the current dilemma, with the battery voltage showing high enough that it would never start charging. Plus, this would mean that the hydro power would be wasted while the solar was charging, rather than working in sync.

2) The best option, IMO, is to replace your charge controller with one that can specifically handle this application. It looks like blue sky energy and missouri wind and solar have options.
http://store.mwands.com/combination-wind-solar-hybrid-charge-controllers/
http://www.blueskyenergyinc.com/products/details/solar_boost_3024il_duo

This is not the same as controllers that can do *either*, such as some that Xantrex and Morningstar offer; with those you'd need a controller for solar and another for diversion controller. Unless...


3) Some folks connect all charging sources including solar(without a charge controller) directly to the battery bank, and then count on the diversion controller to handle the charging; the power from the solar panels would be diverted to the dump load right along with the hydro/wind power once the batt voltage gets high enough. Not sure where you start getting issues from having a panel voltage too high above your battery bank voltage....

In fact, it looks to me as if the missouri wind and solar controller is basically this setup.

Some of the 'diversion controllers' just turn the whole load on/off at a set point; I think I'd be looking for one that can vary the dump load to maintain a set bulk/absorption voltage, to provide a better charge and avoid the potential problem I mentioned up top.


4) Scrounge up another, smaller battery bank and separate the charging sources, while also separating your loads. You could wire things up so that either charging method could charge either bank, and manually control which is getting what from where when.

That last sentence was really fun to write!


5) Probably least practical, but with the best outcome if it could be done: add an electronically controlled shutoff for the hydro, so that at a given voltage on the solar panel output it will turn off the water, and the solar charger will kick in once it sees a suitably low voltage... and of course the reverse, opening the sluicegates if the sun goes away and voltage lowers to a triggering point.

This might be fiddly, given that the Voc for the solar array might be pretty high even on a dim day without much actual power available... Control would be from something like a morningstar relay driver, if the voltage range is right for your array. Not sure exactly what the electronic shutoff water would look like; solenoids would be easy to control, but waste power, and probably excessively limiting to flowrate... complicated all around, and especially problematic at the fairly low wattages described.


That's all that comes to mind immediately, hope it's of some use.



Unrelated: how are you liking that Wellsee 50A solar controller? I see that it claims to have MPPT, but it's REALLY small and cheap for that to be true; my understanding is that the smaller Wellsee 15A 'MPPT' is not in fact MPPT, and the size/price suggests to me the that 50A is just as dishonestly described.

There's a teardown of a 15A discussing this here:

On the other hand given the price of panels it doesn't necessarily make sense to go for a proper MPPT controller vs more panels in any case.

I have both solar and hydro in a not dissimilar situation to the OPs. I have a ridge to the south which the sun mostly disappears behind for a month either side of winter solstice and a spring-fed stream with a highly variable flow rate (depending on whether the winter rains appear or not) which powers a water wheel hooked up to an axial flux alternator. Every winter is different. In wet winters I'm laughing, but they're about one in three at the moment. Having added another couple of panels this winter, in dry winters I can just about make enough in the 1 hour 50 minutes I have sun at winter solstice to cover daily needs (though not complete a charge cycle). It's the in-between weather that's difficult - when there's not enough rain for the hydro to produce much but the skies are cloudy. Like now. Very occasionally I still have to run the generator.

The solar array runs through one MPPT charge controller and the hydro through another. As the batteries reach full charge, the solar controller will progressively reduce the charge it takes from the panels while the hydro keeps churning away until the float period times out, whereupon it dumps to a heating element. The two charge controllers are from different manufacturers and calculate battery voltage slightly differently so there was a little bit of tweaking of the bulk, absorb and float set points for each to get them in synch and then the hydro controller was set to marginally higher voltages so the solar will duck out first. It all works fine together. In good wet winters, I can use the hydro to keep 2 separate systems (one 12V, the other 24V) fully charged.

I'm envious that you have a hydro system to play with. Disclaimer: I don't know much about what can break them; however could you simply turn off the water automatically when the sun comes out (and leave the hydro system always connected to the battery through the reverse voltage diode)? If so, you could turn the water on/off with a drip irrigation controller (Lowes, THD, Amazon, etc.) managed through your smart phone (i.e. manually turn water on/off just not physically turning the valve), or rig up something through a dusk/dawn outdoor power outlet (for Christmas trees) that automatically cycles power on to the valve when the sun comes out and opens the water valve when it goes away. Maybe you can find a drip irrigation controller that has a solar cell in it to tell when the sun is out. If it were mine, I'd run the solar cell and the valve controller (and probably some other gizmos) through an Arduino controller to manage this automation, but it'd likely cost 5x as much as it needs to do your job. This doesn't take advantage of the water you have when the sun is out; however from your description of the problem, you're more concerned about missing out on the sunshine when you get it.

My idea to solve your problem is much simpler than the others, but it requires a change of strategy. I don't believe in the 'all of your eggs in one basket' method for power. I personally like the idea of having separate 'units' of power generation and storage for different tasks, similar to how houses use circuit breakers for different circuits. I would make a more robust solution for my refrigeration, and a less robust one for a non-essential circuit.

My idea for you revolves around the notion that it is considered unnecessary to use a charge controller if you are putting out a volume of charge that is less than what it takes to charge the battery bank in 20 hours time. If you split off 50AH or more of your battery bank and ran something such as a circuit for lighting and charging small devices or other low power electrical loads, then you should be able to simply run the hydro 24/7 @ 2A without a charge controller. If you had a bit more storage on this circuit (say 150AH or more), then you could turn up the water to the 6A output when you need it and still not need the controller. The internal resistance of the battery should emit the extra energy as heat and being spread across such a large surface area it wouldn't cause any damage. It would be wise to look into this idea from multiple sources to see specific recommendations on what charge to what size battery bank and so on.

The problem with using a valve/servo/whatever mechanical device to open and close the water source to the impeller as a means of controlling charge is that if it gets stuck open just one time you could end up destroying your generator. However it could be a possibility to design such a system for emergency shut off and check the functionality regularly. This would be a good idea no matter what method you go with to connect to the batteries, as it could save the generator from damage in the event of a wire disconnecting or catastrophic battery damage. A small Arduino Nano that cuts off water with dangerously low voltage (say 10 volts) and a cut off switch on the circuit for testing. You could go as mild or wild as you like with an Arduino.

I have one question about the hydro generator: Does it use a permanent magnet DC generator, or does it produce AC and step it down to DC at the water source? AC, particularly above low voltage (50 volts) works much better over longer distances than DC. If it does produce AC at the source, it would be best to send that AC across the long wire to the batteries, and then step down to proper DC voltage at the battery bank. You would see a significant increase in power output. You can also increase the wire gauge of the wires connecting to the generator to get more of your energy to the batteries instead of being lost as heat to the resistance of the wire. However if it is really that long of a run it may be cheaper to buy a 100 watt panel for $100 and get more bang for your buck there.

One last thing I want to note is that even though amorphous panels can do better in low light conditions, they also have a lower life span. I'm pretty sure I heard it on a video from Green Power Science and read up on it in other places. Approximately 10 years (variable depending on usage) before a serious drop off in power output. This is what also worries me about thin film panels, as they haven't been around long enough to test their lifespan. Mono and Polycrystalline panels degrade much slower, and I have heard of ones 50+ years old giving around 70% of their initial rated power. It is cheaper and greener to just oversize your array by a few panels with polycrystalline in many cases. If you have more than enough power in summer and too little in winter you can put a steeper angle on the panels to help moderate that issue. This will put more sun on the panel when the sun is low in the sky at the cost of less sun hitting the panel for a given moment in the summer. The extra daylight hours make up for it not being aimed optimally.

Hopefully this gives a few good ideas to people. Good luck!

Do both your solar cells and your hydro both run over the same long wire to the battery?

As for the type of solar panel you are likely not losing as much as you think from your power production. The power production at low light levels is so small that the greater efficiency of crystalline panels likely makes up for most of it during the times when they have good sun.

Daniel, thanks for raising the issue of valve failure. Reading back over that part of my post, it does kind of sound like I meant that valve to be the only cutoff for the hydro system. Oops! I wouldn't want to get rid of the dump load controller, but simply adjust so that it should not trip until *after* the flow should already have been cut off. Plus, if your dump load is useful(hot water?) you might want to disable the valve, if you happen to have plenty of water at certain times of year.


Your suggestion about multiple independent smaller systems reminds me of a system I read of recently, which relied on dedicated inverters on a variety of battery banks to power individual loads. Sounds way expensive to use decent inverters this way... but this was using re-purposed APC UPS units as inverters. Intriguing, since you aren't likely to find a quality true sine inverter anywhere near as cheap as a good APC with dead batteries...

The proponent claimed he'd had no issues with using them in this application. Elsewhere I've heard of overheating problems in extended use. Could be a benefit of the distributed nature of his system; less likelihood that any one UPS would see a lot of continuous runtime.


The main downsides that come to mind for multiple smaller systems aren't too terrible IMO, depending on the demands on the systems. For one, there's the difficulty of running larger loads. A 500AH/12V battery bank running a 3000W load for a short period is still only discharging at 0.5C. Split that up into 3 banks of similar capacity, and it would be more like 1.5C, not as good for runtime or battery longevity. The other downside is increased maintenance; more battery banks to water and monitor for correct operation, more parts to maintain/fail/replace.

The upside of a degree of redundancy is certainly appealing though; I guess everyone will have their own sweet spot on this continuum. I think mine would be more than one bank... but not more than 3. Perhaps 2... 2 is a nice number.


Though this might all be a case of a problem looking harder than it is, from what Wendy says. You might be just fine with a different solar controller and some tweaking... I've certainly heard of other setups done like this that work.

Different setups are a good way to add resilience - one reason why I have separate 12V and 24V systems at opposite ends of the farm - but (please correct me if I'm wrong - I'm no expert on the subject) doesn't charging direct to batteries rely on power generation within a much narrower range of output? And on a regular and predictable draw to prevent an overcharge situation developing?

In my situation, my generating capacity and my usage are too variable (anywhere between zero and excess, weather dependent) and I wanted to take advantage of the full range of capacity available to me while enabling complete flexibility in usage, so to me the cost of decent MPPT controllers was an expense well worth making. The biggest plus point of the MPPT controller on the hydro that when my stream volumes are very low, it can turn a lot of fairly useless volts into useful amps without stalling the wheel, considerably reducing its cut-in speed. And with so much else to be filling my head with in terms of developing this place, anything I don't have to think about much is a bonus. I only run the hydro in deepest winter or in periods of heavy rain (they tend to coincide) to extend bearing life. As soon as I've enough solar to complete a charging cycle it gets switched off. The whole system demands minimal attention. Apart from switching the hydro on and off once a year, topping up batteries, running the occasional equalisation charge and monitoring winter battery levels at times of low power production, I mostly don't have to pay it any heed.

Dillon Nichols provided the correct answer. Hydro is set up just like a wind turbine. All your power inputs hook to the battery and you use a diversion dump load controller from Missouri wind and solar, eliminating your current charge controller. I've purchased 50 percent of my system from mwands and they are top notch. My buddy has been using their diversion controller for years with his solar wind set up and it works great. You can also get the water heating elements and dump loads from them also.

I don't have enough wind so I'm all solar.

Hi guys great thread: you could always Run the load dump to an electric motor attached to the hydro turbine housing-when the batteries are full-the load dump motor lifts the turbine free from the water & saves bearing wear. When the battery voltage drop the motor is switched on to reverse the process dipping the turbine back in the water.

From the system description, it sounded to me like the hydro power is always electrically connected and they manually open the water valve to start generating power (up to 6A but usually run the flow to generate about 2A) and vice versa. If the water valve were to get stuck in the on position from the failure in an automated water valve, it wouldn't destroy the hydro unless the load (i.e. the battery) were also electrically disconnected, which they don't ever need to do except when they don't want the voltage produced by the hydro to flake out the solar panel charge controller (i.e. when the sun comes out and they want to gobble up as much of those solar rays as possible). Rather than consider disconnecting the electrical power of the hydro (to remove this charge voltage), consider to simply automate turning off the water to the hydro to both stop producing this charge voltage (on the battery from the hydro) and to stop the water flow on the hydro (to save the bearing life) at the same time. That way the battery voltage can drop (with no charging source), be identified by the solar charge controller, and allow the solar charger to work alone. This solution doesn't make the multiple power source work together (15A solar and 2A hydro), but it does keep them from fighting with each other (15A solar or 2A hydro) without having to spend your days hanging out around the water valve looking at the clouds.

As an aside, I saw that you could trickle charge a 500Ah lead-acid battery with up to 5A continuously (much higher than the typical 2A from your hydro) to maintain the battery at a fully charged state without overcharging (i.e. 5A is a low enough charge rate for the recombination reaction).

Wendy Howard wrote: Every winter is different. In wet winters I'm laughing, but they're about one in three at the moment. Having added another couple of panels this winter, in dry winters I can just about make enough in the 1 hour 50 minutes I have sun at winter solstice to cover daily needs (though not complete a charge cycle).

The solar array runs through one MPPT charge kcontroller and the hydro through another. As the batteries reach full charge, the solar controller will progressively reduce the charge it takes from the panels while the hydro keeps churning away until the float period times out, whereupon it dumps to a heating element. The two charge controllers are from different manufacturers and calculate battery voltage slightly differently so there was a little bit of tweaking of the bulk, absorb and float set points for each to get them in synch and then the hydro controller was set to marginally higher voltages so the solar will duck out first. It all works fine together. In good wet winters, I can use the hydro to keep 2 separate systems (one 12V, the other 24V) fully charged.

You will not get better advice from another hydro system expert than Wendy lays out above. Her street cred would be the renewable resource available onsite, and that her system and her operation of it is successful, in my humble opinion.

Controllers are good and the engineers have made them programable and flexible for us to integrate so that we are no longer forced to operate our systems like ww2 u-boat crew.

I know this is an old discussion but would it make sense to incorporate some type of recirculating stream/water feature once the hydro charges the batteries.

My property slopes towards the river and this could keep a load on the hydro as well as circulating water for ponds, hydroponics etc