4 Solar Controllers in Parallel?

[neelie]

Is it feasible to connect the outputs of 4 identical solar controllers in parallel to the batteries?

I was thinking that a "weaker" one or two will throttle back because they sense the higher bus voltage from the stronger one and their logic assumes the "battery" is fully charged. Worst case scenario would be 3 of them throttling back if one was putting out a much higher voltage.

Or have I, as usual, got this concept completely wrong?

(In case anyone's wondering, I am thinking of 4 x 140W panels, each with its own MPPT controller into 2 x 360AH LFP).

Thanks.

[Sumner]

With the Blue Sky 2512i(X) like we have on our 200 watt array you can use up to 8 controllers if you wanted. The first is the master and the others are setup as slaves. It is explained here...

http://www.blueskyenergyinc.com/uplo...E_SB2512iX.pdf

We have a 3024iL on the other array (480 watts) and it can be setup the same.

You need to check with the controller you want to use to see if it is similar or not. About the only time I see this being beneficial is if you know some panels are for sure going to be shaded at times and others not. Also if the array is too large for a single controller.

We have 480 watts on our 3024 which is about max for it, but there are controllers out there that could handle your whole array of 560 watts.

Our 200 watt array...

Macgregor 26S Outside Mods page 33

and 480 watt array...

Endeavour 37 Electrical Mods Index

Sum

[noelex 77]

It should work well without any problems. It has the advantage that each panel is kept at its Vmp rather than taking an "average".
The drawbacks are that the power needed to drive the electronics in the controller is higher and these losses can be significant with some controllers. The other drawback is that good MPPT are usually expensive. Cheap MPPT controlers often have very poor tracking of the Vmp. One good MPPT controller is likely to be much better than 4 cheaper controllers.

There are not usually any great conflicts with voltage regulation when you have mltiple charge sources in parrallel . Most controllers use a simple timed absorption phase which will still work normally.

[neelie]

Originally Posted by Sumner

About the only time I see this being beneficial is if you know some panels are for sure going to be shaded at times and others not.

Thanks for the BluseSky link.. I'm trying to digest it now.

And yes, no more space for more panels, so trying my best to minimize shade losses and the added benefit of redundancy...

The extra cost at the end of the day is marginal in the scheme of things.

[Pete7]

We have two panels through two PWM regulators. Seems to work okay and in full sun achieves the full rated output. The reason we have 2 is that is was so much easier to fit two rather than have wires running all around the boat.

Pete

[neelie]

Originally Posted by noelex 77

It should work well without any problems. It has the advantage that each panel is kept at its Vmp rather than taking an "average".
The drawbacks are that the power needed to drive the electronics in the controller is higher and these losses can be significant with some controllers. The other drawback is that good MPPT are usually expensive. Cheap MPPT controlers often have very poor tracking of the Vmp. One good MPPT controller is likely to be much better than 4 cheaper controllers.

There are not usually any great conflicts with voltage regulation when you have mltiple charge sources in parrallel . Most controllers use a simple timed absorption phase which will still work normally.

Thanks, Noelex

For sure, I will endeavor to get the best quality MPPT controllers - the added cost, as a percentage of the project is not that much. I have no more space for extra panels, so I have to get the most out these four, allowing for shading of one.

I don't understand the following.....:

"...There are not usually any great conflicts with voltage regulation when you have mltiple charge sources in parrallel . Most controllers use a simple timed absorption phase which will still work normally.."

Why will there not be any conflicts with multiple charge sources?

Will not the highest voltage on the bus be sensed and presumed to be battery voltage?

Manufacturing tolerances don't guarantee exactly the same voltage outputs and besides these may vary for other reasons.

My electrical knowledge is basic at best - so forgive the naivety of the questions. Its just that whenever I think I understand something, someone pops up with a curve ball.

[noelex 77]

Originally Posted by neelie

Will not the highest voltage on the bus be sensed and presumed to be battery voltage?
.

Yes.

The good news it that it will be the battery voltage (less the small voltage drop in the wires)

[delatbabel]

My electronics knowledge is pretty good but I wasn't aware of the internal workings of charge controllers (MPPT and not), so I posed this question to a local guy who is an electronics wizard and he assures me that it's no problem at all to have multiple controllers in parallel.

One point I can set you straight on is, which is the crux of the issue is this:

Originally Posted by neelie

Thanks, Noelex
Will not the highest voltage on the bus be sensed and presumed to be battery voltage?

It comes down to the power output of the panels and therefore the drive from the controllers, and the input requirements of the battery banks.

Let's say you have a 80W panel, and 800Ah of severely drained battery (all at 12V). The full output of that panel through the controller might be something like 5A at around 14.4V give or take a bit depending on efficiency of the controller, but that sort of current going into a discharged bank isn't going to drive the terminal voltage up to 14.4A, it's only going to get to about 12.9V, 13V, 13.2V or something like that. Your controller is going to be working overtime for some time to push the terminal voltage up to the full charge voltage of the bank (14.4V or thereabouts depending on battery type).

If you had a 160W panel going into the same controller, that's going to be more like 10A into the battery which is going to push the terminal voltage up higher, perhaps to 13.8V. It's only going to have to work for half as long to get the terminal voltage up to 14.4V.

Now go back to the 80W panel through the charger, and put a second charger in parallel. The second charger isn't going to see the terminal voltage at 14.4V when the two start working, it's only going to see the 13V or thereabouts which is where the first panel is going to drive it to, and in turn is going to push the voltage up a smudge more, perhaps to the same 13.8V mark that the 160W panel would have done.

All smart controllers have an initial stage where they try to push the voltage of a drained battery up as high and as fast as they can to get it up to the proper charge voltage. They don't know the total Ah capacity of the battery, they don't need to know, they just work on the fact that the higher the capacity and the more severe the drain the longer it will take to get there. In effect having the two controllers working in parallel approximately halves that time, in about the same way as having twice the wattage of panel (and hence amperage output of the controller) would do.

As someone else has pointed out, the subsequent stage charges often work on a time basis -- hold the voltage up to a level for 48 hours or so, and for that time it doesn't really matter how many controllers you have in parallel either because they will all be regulating their output to what the batteries require. If you put two smart controllers in parallel and they are running in trickle charge mode then effectively each will approximately halve their output based on the fluctuations they see in voltage on the output sensors.

So the answer to your question is: Yes they do, and no it doesn't matter.

[noelex 77]

If we use an example it makes it clearer how multiple controllers will work.

For the example lets assume just controllers.

Both are set to maintain an absorption voltage of 14.4v for 2 hours then drop to a float voltage.
As you say despite the same specs there will some miss adjustment so the two units will not be identical. Lets assume one has a real absorption voltage for 14.3v for 1:50.

During the bulk phase both will do no regulation.

When 14.3v the second controller will cut back the power from the solar panel to try and maintain 14.3v. At first this will work as the second panel will not put out enough power to hit 14.4v, but as the battery becomes more charged the second panel will raise the voltage to hit 14.4v.

So the absorption voltage will stay between 14.3 and 14.4v, which is fine. The absorption time will be slightly longer as the second controller will not start is 2hr countdown until 14.4v is reached.
On the better solar controlers the absorption time is adjustable and should be set so on average the battery drops out of float at the correct time.
The factory default setting is an arbitrary figure anyway which may be anywhere from 1 to 3 hours depending on the whim of the manufacturer.

This example shows that basically two solar controllers work fine together. The only slight difference is the battery will tend to receive the most aggressive of all the settings if the controllers have a slight difference in adjustment. The example can be scaled up to 4 or more units with identical conclusions.

Do consider the self consumption of the controllers however. Manufactures are coy about revelling this data, but many of the sophisticated solar controllers with the best tracking can consume the output of a 5-10w solar panel or more. This is not much a factor with a 400w array, but is more significant if divided into 100w for each regulator. Four controllers needs a very big solar array to justify the self consumption of the better MPPT regulators. There are some very efficient designs of contorlers that have been developed for solar cars, but these are not readily available to the general public. The closest commercial product is some of the Gensun range of controlers that still track reasonably, but with lowish self consumption.
No one has tested 4 of these controllers against say a single outback controller. I suspect they come out reasonably even, but this is an educated guess only.

[Sumner]

Originally Posted by noelex 77

...The drawbacks are that the power needed to drive the electronics in the controller is higher and these losses can be significant with some controllers...

Ok, I never checked this and you made me go look .

According to the specs the 2512 and the 3024 Blue Sky MPPT controllers we are using are .45 watt on standby and 1.0 watt typical charge on.

That's not too bad, but like you said this might be a good spec. to check before buying a controller,

Sum

[Dsanduril]

Parallel controllers generally can play well together in the manner Noelex describes, even without the master/slave arrangement. The one complication is when one controller is offline for some reason and comes online while the others are near the end of bulk charging. The controller coming online in this situation may see that voltage as indicating full battery and move to float, cutting off its contribution. If all the controllers come up together with the morning sunrise and see the voltage rise together then they should be happy. If you take one offline and then bring it online mid-day when the voltage is above the float point (~13.8) that controller may (depends on the controller) go straight to float until the voltage drops below that point (usually late afternoon or the next morning). Some controllers are built with a "test" mode when they come online in these conditions, they will see if full output for a small period of time brings the voltage up quickly (indicating real float) or not. Others do not have this "test" mode and may get stuck in float for the rest of the day.

This situation should be pretty rare, but can (and does) occur if you disconnect and work on one panel/controller set during the middle of the day. It's a small thing, but something to remember when you do do that kind of thing and can't figure out why things don't seem to be coming back online.

[Sumner]

Originally Posted by Dsanduril

... even without the master/slave arrangement.....

With the Blue Sky this isn't hard. 2 wires from one controller to the next on provided terminals and change a dip switch on them one time and you are done. Since they have this feature I'd just do that and be done with it if I was using the Blue Sky,

Sum

[noelex 77]

Originally Posted by Sumner

According to the specs the 2512 and the 3024 Blue Sky MPPT controllers we are using are .45 watt on standby and 1.0 watt typical charge on.

The Blue sky self consumption is reasonably low (if their figures are to be believed its a bit less than 1.5AHrs a day), but their MPPT implementation is not really state of the art.

[Captain Bill]

Charging sources do interfere with each other unless they communicate. That is why the Bluesky controllers are such a good idea. The master is the only one that neasures the battery voltage. It shuts down the output of the others every once in a while so it can measure the voltage without interference. The only problem is that Bluesky does not provide solutions for all your charging sources. To give you an example of the interference between sources, I cannot get a full charge on my batteries using my genset and battery charger. The genset has a 10 amp starting circuit alternator with a fixed regulator. It works fine when the battery is low and I get a full 50 amps, 10 amps out of the alternator circuit and 40 amps out of the charger. The problem is that the battery charger shuts down it's output every 10 minutes or so to measure the actual battery voltage, but the alternator circuit does not. The charger reads the voltage coming out of the alternator circuit as battery voltage and interprets this as fully charged and shuts down. I am left with the alternator circuit fixed voltage of 14.2 instead of the 14.4-14.6 I need at winter temperatures. The only way to get a full charge is to plug into a dock where the alternator circuit is not interfering. It is even difficult when I have a sunny day in the fall and winter and the solar is putting out a lot because the charger quits so early. I once wrote to Bluesky and suggested that they make regulators and battery chargers that plug into their communications circuit so that all the charging sources on the boat could work together, but got no response.

[Sumner]

Originally Posted by noelex 77

... but their MPPT implementation is not really state of the art.

Are you talking about their current line of MPPT controllers such as the 3024....

Blue Sky Energy Inc. | Solar Boost 3024iL & 3024DiL

I agree that there are a number of good controllers out there from various companies. Have you seen any independent head to head comparisons between them?

Sum