Solar Controller for Low Voltage Panels? Not Victron!

[Msailor]

I'm looking for recommendations on a new solar controller for charging my 12v AGM batteries. I've got Renogy 100 watt panels, 3 years old, which unfortunately are no longer putting out their rated open circuit voltage. From a factory specified 21.2v its now 19.5v, consistent for each panel individually and when paralleled (no loss through the connections to the controller input). They've got to be paralleled because shading from the mast, sails, etc. is unavoidable.

My three year old Victron MPPT controller was doing its job and the system was working fine with the higher panel open panel voltage, but not any more. Now there is no charging at all. This is because Victron hard programs MPPT controllers to require a substantial difference between panel and battery voltages, and my system no longer satisfies that arithmetic. Victron literature says the differential is 5v to start charging, but that's not the whole story. There's a voltage drop caused by the controller itself, and in my system that's about 3v. The result is the panel-battery differential is not enough to start charging.

I'm hopeful that other sailors (those who can't wire in series) have dealt with low voltage panels and have found controllers to do the job. Any suggestions?

Note, new solar panels is not now an option. Thanks for any ideas.

[noelex 77]

Welcome to the forum Msailor.

The Victron controllers will look at the input voltage. Before the charging begins the current will be zero therefore any resistance in the wires and the connections will not produce any voltage drop. 19.5v should be OK providing there are not any other charge sources raising the battery voltage. Once charging commences the input voltage only has to be around 1v above battery voltage for charging to continue.

The factory specified 21.2v is for a solar cell temperature of 25° C. As the solar cells heat up the Voc drops and solar cell temperature is always well above the ambient temperature. A Voc of 19.5v is quite good in practice.

In short, I doubt the 19.5v you are measuring is the cause of the problem. I would at least measure the Isc (short circuit current) to see if the panels are capable of producing useful current before contemplating new controllers.

[Msailor]

Thanks for the analysis. A bit more info may be of help. The 3v drop occurs when the panels are connected to the controller, and is shown by the controller's software readings transmitted via bluetooth. The software measures what Victron calls "Solar Voltage," which moves around a bit, but averages about 16.5v. A few times it has hit 17 plus mid-day in perfect conditions (I'm in Florida), and that started a charging cycle, but most days I get no charging at all.

I'd like to measure short circuit amps but my current multi-meter doesn't have the capacity. However, I think it unlikely that all three panels would have the same VOC reading if one (or more) was deficient in producing current. It does seem that the panels are simply aging, perhaps more rapidly then expected, and so panel VOC has deteriorated.

Waiting for the sun angle to increase is one thing for a charging system, requiring perfect conditions is quit another. The bottom line for me is that Victron's voltage differential requirement is too restrictive for my panels.

[noelex 77]

Originally Posted by Msailor

The 3v drop occurs when the panels are connected to the controller, and is shown by the controller's software readings transmitted via bluetooth. The software measures what Victron calls "Solar Voltage," which moves around a bit, but averages about 16.5v. A few times it has hit 17 plus mid-day in perfect conditions (I'm in Florida), and that started a charging cycle, but most days I get no charging at all.

The solar voltage of 16.5v is the Vmp or the maximium powerpoint voltage. A 16.5v Vmp is typical of “12 v “ panels in the real world. This is not the voltage that the Victron controller uses to decide if it should start charging. The voltage that initiates the start of charging is the Voc which you have measured at 19.5v as long as the battery voltage is lower than 5v below 19.5v (14.5 v) in the morning when the solar charger wakes up it will try to start charging and as long as the battery voltage is lower than 1v below the Vmp which is reported as 16.5v (so 15.5v) the controller will continue to charge.

What status is the controller reporting around the middle of the day? The status can be found on the Bluetooth app. If you could take a screenshot around midday the other values are also helpful.

[Msailor]

Thanks again for your response. "Solar Voltage" may measure available controller output, but I can't find where Victron says that. 19.5 volts open circuit may be real world good, but it is less than when the panels were new and when the controller charged dawn to dusk.

Still interested in suggestions for a "low voltage" controller.

[noelex 77]

Victron are the only MPPT controllers that use a +5 v wake up voltage. Most companies do not publish their wake up voltage, but around +2 to 3v would be typical. There is not much difference between brands other than Victron. Some of the expensive controllers such as Outback and Midnite have adjustable start up parameters, but few people understand this feature and it is rarely adjusted from the default setting, which is a pity. Both waking up too early and too late wastes some energy.

When running, all MPPT controllers are similar (including Victron), requiring an input voltage around 1v higher than battery voltage to keep charging, so there is almost no difference with different brands of controller once the wake sequence has been triggered.

There are a small number of MPPT controllers that will boost voltage. They are usually designed for using 6v or 9v panels on a 12v bank. Genasun are the major manufacturer. However, having this boost feature reduces the efficiency when the solar panel voltage is high so this is not something that is a good idea for most systems.

The final option is to select a non MPPT or so called PWM solar controller. These will start and continue to work when the potential input voltage is only a fraction above battery voltage. There is some loss of efficiency when the solar panel voltage is high, but they are slightly more efficient when the solar panel voltage is low. This is a very rare situation for most systems, but as this is your concern they would be ideal.

If you are convinced that low solar panel voltage is the cause of your problem (I am doubtful this is the real issue) a PWM controller would be the best solution. There are many inexpensive brands of PWM controllers. As the electronics in this type of controller is very simple these cheap units are often OK, unlike inexpensive MPPT controllers that are usually poor. If you want a good quality PWM controller Plasmatronics make very good units.

As a test you can try connecting the solar panels in parallel directly to the battery bank (use a suitable fuse). This is exactly what a PWM controller will do when the batteries are not at a high state of charge. Monitor the battery voltage and check it does not climb too high while the panels are connected this way (it will only climb slowly so there is plenty of time to disconnect providing it is monitored). If this does not charge the batteries with reasonable current no conventional (non voltage boosting) MPPT or PWM controller will work in your system and its faults.

This above test (or measuring Isc) will show if the fault is really low solar panel voltage before spending money on new controllers.

[s/v Jedi]

If the bypass diodes in your panels are intact, you should try putting them in series and see what happens during partial shading.

If you lose output during partial shading, maybe you can repair the panels by replacing the diodes. I have had panels that gave simple access to the diodes but also ones with sealed connections.

Edit: worst case you can add external bypass diodes that bypass the whole panel that has partial shading… which happens with parallel setup anyway.

[Nicholson58]

I think a better solution is new panels!

New panel will have higher efficiency, greater output. Don’t skimp.

Our 3 Sanyo 220HT BTW are 8 years old and still capable of exceeding full rated output. Name plate says 660 watts max.

[marty9876]

Thanks for this discussion, I've always been perplexed by Victron's stance/description on how all this works and truthfully I still am. For my initial solar install a few years back, with flexible panels on a bimini, the the difference of the +5v was a razor thin margin. I could never get over what seems like a big deal and went with a different controller at the time.

I'm going to expand my system this year (installed davits so getting a proper fixed panel also) and need to re-visit this whole conversation. Victron is naturally the price/feature/performance winner these days for small controllers and I'll need to find a way to put my mind at ease with all this again

[s/v Jedi]

Originally Posted by marty9876

Thanks for this discussion, I've always been perplexed by Victron's stance/description on how all this works and truthfully I still am. For my initial solar install a few years back, with flexible panels on a bimini, the the difference of the +5v was a razor thin margin. I could never get over what seems like a big deal and went with a different controller at the time.

I'm going to expand my system this year (installed davits so getting a proper fixed panel also) and need to re-visit this whole conversation. Victron is naturally the price/feature/performance winner these days for small controllers and I'll need to find a way to put my mind at ease with all this again

There are many reasons to get the array voltage up much higher than the 5V above battery voltage. I had a 120V DC array before a hurricane ripped it off and am considering a 250V array as replacement. In that case, while underway I will slide two panels underneath three other panels, which is the ultimate shading and still expect the other panels to perform with all five of them in series

[noelex 77]

Originally Posted by marty9876

Thanks for this discussion, I've always been perplexed by Victron's stance/description on how all this works and truthfully I still am.

In practice it does not seem to be an issue. However, I would be cautious about using the Victron controllers in a parallel or single controller per panel system with “12 v“ panels where the Voc of the panels was on the low end of what is typical. This is especially true if there are alternate charge sources (such as wind) that may raise the battery voltage first thing in the morning. It is also potentially more of an issue in hot climates or situations where the panels do not receive much ventilation, such as flexible panels attached to the deck.

It should also be noted that the opposite problem occurs with solar controllers waking up too early before the solar panels are capable of delivering enough current to compensate for self consumption of the MPPT controller.

It is easy to see if your controller is waking up too early or too late although it means staring at electrical gauges instead of enjoying the early morning light. Ideally the solar controller would wake up just at the point where the solar panel production equals the self consumption of controller (or strictly speaking the extra consumption involved in supplying the tracking circuits). So at the point of waking up if the contribution of the solar controller is negative (ie it is consuming power) the controller has woken up too early. If at the point of waking up, the contribution of the solar controller is positive (ie it is producing power) the controller has woken up too late. You can also gauge how significant this error is by estimating the amount of energy wasted by a late or early start. This should be very small.

Unfortunately, if you do this test on 99% of controllers there is nothing you can do to adjust the wake up parameters. These are set by the manufacturer of the controller and are not user adjustable. If you did identify a severe issue of the controller starting late you can change the wiring to series, or as a temporary measure “fool” the controller into starting each morning by adding a brief high load.

[Msailor]

First, thank you Noelex for the market survey on controllers. Hopefully someone will get on this thread and tell us about their experience with low voltage systems that use some of these products.

A couple of points: Serial wiring pros and cons are well known, as are the practicalities and realities of putting panels on a sailboat--particularly smallish monohulls. As a way of illustrating, I'll add to this thread the results of the new test of serial wiring of my panels. With the same Victron controller model and the same panel installation, instead of the usual mid-day 15-20 amps I would get with the panels paralleled, I'm now seeing only 2-6 amps mid-day, and that's with the sails down and the boat oriented to minimize shading. My new trickle charge system.

Another point, for finding an installed system's starting voltage, I held test probes to the controller input terminals and monitored both the external voltmeter and Victron Connect "Solar Voltage." They were virtually the same (in the approximate 14 to 17 amp range) and so I think either works for determining voltage drop. For observing commencement of charging, Victron Connect shows charging status and battery voltage in the same view as Solar Voltage. I found charging on my system started when the Solar Voltage/battery voltage delta reached 5.

So, serial wiring and the fine points of Victron software are really not where this discussion should go. It should be about controllers which work well with panels that are on the low voltage side.

[noelex 77]

Originally Posted by Msailor

With the same Victron controller model and the same panel installation, instead of the usual mid-day 15-20 amps I would get with the panels paralleled, I'm now seeing only 2-6 amps mid-day, and that's with the sails down and the boat oriented to minimize shading. My new trickle charge system.

Parallel connection is generally better on a sailboat than series connection, but the difference should not be as drastic as you are finding. I suspect you have some blown bypass diodes.

Originally Posted by Msailor

Another point, for finding an installed system's starting voltage, I held test probes to the controller input terminals and monitored both the external voltmeter and Victron Connect "Solar Voltage." They were virtually the same (in the approximate 14 to 17 amp range) and so I think either works for determining voltage drop.

The Victron controllers are normally reasonably accurate so it is not surprising that the input voltage you are measuring matches the results the controller is reporting, but this is not “voltage drop”. The controller is adjusting the output voltage of the solar panel to maximise the power production. This is the Vmp. It is different to the voltage that determines if the solar controller is going to “wake up” this is Voc, or open circuit voltage ie when no current is flowing. Voc is typically around 3-4v higher than the Vmp in the real world for “12v" panels, but this difference will vary depending on conditions and other factors. Voc is only reported by the solar controller before wake up.

Originally Posted by Msailor

Victron Connect shows charging status and battery voltage in the same view as Solar Voltage. I found charging on my system started when the Solar Voltage/battery voltage delta reached 5.

This is exactly what it should be and this is typically spot on. More important is what current is the system producing at wake up. This will tell you if the wake up sequence is starting too late or too early in your system.

[marty9876]

Originally Posted by noelex 77

In practice it does not seem to be an issue. However, I would be cautious about using the Victron controllers in a parallel or single controller per panel system with “12 v“ panels where the Voc of the panels was on the low end of what is typical.

Thanks for confirming I’m not crazy and my analysis paralysis was pretty accurate - you hit my initial setup spot on - pair of flex panels in parallel. From memory, rigid panels I think tend run a little higher voltage than flex but naturally the parallel is the real killer. Based on your other comment perhaps I should revisit the serial vs parallel decision. I understand the benefits if higher voltage, the sailor in me went with the conservative approach

Ironically I was looking into using my current controller/pair of flex panels in parallel and adding a rigid but once I understood the potential power generation I’d be giving up (the flex were much lower voltage than possible rigid) that killed that idea quickly. I had all the right pieces just not put together in the right order in my head.

[smac999]

Originally Posted by Msailor

Still interested in suggestions for a "low voltage" controller.

if the panels can't get enough voltage your only option would probably be a pwm controller. I'm betting any mppt will want higher voltage.

I would stick in series and see what happens. even just to test and see if you get any power out of them. less ideal is still better then your current nothing.