Victron controller output seems to taper off. Overheating?

[SunSailor]

This is my first solar installation. Six 150 Watt panels are connected in parallel to a Victron MPPT 100v/50a controller. Ambient temperature is about 75F. I would appreciate any feedback on why the controller’s output cyclically fades away. Is it overheating? The box only shows a steady blue Bulk LED. After the third cycle I pointed a fan at the controller and it seemed to make a slight difference In the duration. Is a cooling fan a common addition to these systems?
Thanks

 

[snoobler]

Welcome to the forum.

If the heat is the problem, is the SCC mounted on a vertical surface that permits natural flow? If not, that's potentially a problem.

On your 2nd VictronConnect link, I'm seeing that you're likely at absorption, so it's natural for current/wattage to taper.

However, the trend graphs are well below absorption.

How are your voltages going so low in such short periods of time? 11V?

 

[chrisski]

Is this a vehicle or RV instaltion?

Almost as if something is triggering a self resetting CB when the voltage drops to 11.

 

[Rocketman]

I have 800 watts going into a Victron 100/50 and have never needed a cooling fan (even at 90 degree days - OK I needed the fan - not the controller).

Tell us a little more about the system. Batteries (quantity, type, etc.) ... loads... What other equipment do you have in the system?

Is it weird that every 10min or so the battery voltage drops to about 11 volts, then it seems like something shuts the load off, the battery voltage recovers and the charging recovers - then the load turns back on? Is there a load on the system?


I also noticed a small difference in the voltage at the battery (12.36v) and the voltage at the Solar panels (14.78v).
(I am guessing here...)
Your panels are all wired in parallel, One of the problems may be not enough voltage difference between panel voltage and the battery voltage. Try going to a serial arraignment with your panels - a 3s2p or 2s3p wiring arraignment. (I don't think you can go with a 6s - you will exceed the Solar controllers voltage and kill it). My guess is the solar panel may be heating up and then the voltage is dropping - it is dropping just enough to no longer charge. the charging stops - it cools down and can start charging again and the cycle starts over. If you wire in a 2 or 3 series and 3 or 2 in parallel, that will boost the voltage to 45 or 30 volts (at the panel) giving plenty of room for the MPPT to do its job without a slight voltage change bothering it. (I would probably go with 3 in series and 2 in parallel - if all things at the panels are equal).

SunSailer - when you get it figured out PLEASE report back what the problem was. That way we all learn.

 

[pollenface]

I can't answer the question of the OP but I do run temperature operated cooling fans behind the mppts, but it gets to 45c/113f at my place in summer.

 

[mikefitz]

As sugested by Rocketman the voltage is to low going into the Victron, may be volt drops in connections. Would be better rewiring panels in series pairs then in parallel to raise the input volts. Its not a overheating problem in the unit.
Mike

 

[SunSailor]

Thank you all for your input. It encourages me forward.
The panels are in parallel on the reasoning that at least one or two will not have the shadow of the mast or boom at any given time. The best I have seen so far is 500 watts from the 900 watt array.
I am also puzzled by the extreme voltage drop with charging current. The batteries are new.
Yes, there were one or two loads on at the time. The watermaker that draws about 8 amps and possibly the fridge or freezer. I will keep testing and let you know what I find.

 

[Rocketman]

Because of the boom shading - try wiring in two in series and three in parallel. See if that improves things - if not it you can go back

 

[SolarSailer]

Here is what this morning looks like. I believe the wave is caused by the boat dancing at anchor. With this morning’s wind, waves and current the dance cycle is about 2.5 minutes.

If you are not familiar, the boat at anchor is theoretically pointing directly into the wind in what should be a stable state. In reality, the wind will get slightly to one side or the other of the bow. It then pushes the boat to that side until the anchor line pulls the bow back through the wind and the cycle continues in the opposite direction.

 

[HRTKD]

There must be a heck of a load on the system to drop ~.5 volts.

The easy solution is to put all the panels on gimbals and set them up with an X/Y tracking mechanism. LOL!

 

[snoobler]

Okay. Shading makes a lot more sense. Had I understood this was a sailboat, that would have been #1 on my list, but something serious is going on.

8A on your water maker? Are you SURE about that? Something is hammering the crap out of your battery, and it's not long for this world.

I would also take the opportunity to check for loose connections at every point in the system.

 

[SunSailor]

Here is more data. Note that the drops to zero solar watts happen at midday, but not in the late afternoon at lower wattage. The drops seem to be at regular intervals With the same wave signature. Every other load, including fridge and freezer, were off. The watermaker does not have any kind of controller. It is simply 12v connected to an 8A motor that drives a piston which pulls raw sea water on the down stroke and pushes it at high pressure through the RO membrane on the upstroke. About 1 second per cycle, four gallons per hour. The last image shows the water maker turned off for a few minutes.

The panels and frame help up well in 30 knot winds yesterday.

This does not seem to seriously hamper overall performance. I am just concerned that it may be reducing the life of the controller.

 

[chrisski]

A piston driven to match the drop In voltage makes sense to me. It’s be interesting to see if you jhad a generator, which normally would provide plenty of power if you’d still see the voltage surge.

I would not like that continuous spike. A few times an hour to turn on and off a motor, perhaps, but not continuous. I’d be as worried about the battery as the SCC. When I see a voltage drop like that, I think it’s when I’ve turned the microwave oven on at night by accident. It’d make me think the piston in that RO filter may need 4 to 8 times the power of the wattage sticker for it. Not a problem when using a generator, those can handle quite the spike, but not a solar charge controllers.

I wonder if upsizing the battery bank would help in the same way that upsizing a battery bank would see less of a surge when a high wattage appliance is turned on, Also, lithium phosphate reacts better to large loads than Flooded lead acid, and I do not know what your bank is.

I think Capacitors are inserted into circuits to help with these circuits. I’ve not be a fan of them, but that is for dealing with running a motor with a smaller sized battery bank or inverter than needed That is started a few times an hour, but in this case, it may be a good thing,.

 

[MisterSandals]

Just as a test, is it possible to slow your water maker down to 2 gallons per hour to see how that affects things? Maybe it will show something that we can learn from.

 

[mikefitz]

If you change the plot for the solar controller to solar input volts, you will see what is happening. My suggestion is that the solar voltage is falling too low and the solar controller shuts down for a moment. The water maker and other loads are dragging the voltage down when the solar input is low. One of your previous plots showed a solar voltage of only 14 volts. Shading ( as the boat moves at anchor), I suggest, is momentary reducing the panels output, such that the controller shuts down. Having lived on a sail boat for 10 years I know that even a slight shadow on the panel from rigging can reduce the output considerably. The voltage of 14 volts also suggests that there may be voltage drops in the cables and/or connections.

With the Victron solar controller, the solar voltage has to be in excess of 5 volts for the controller to become active and it shuts down if the solar volts drops to within a volt of battery volts.

The batteries may be past their best or there is a poor connection somewhere in the system or an excessive load exists. The volt drops to 10v are not good news. Looking at the various plots, the battery bank must be almost flat in the morning, and is not reaching 50% by late in the day. I suggest you charge the batteries to full via the alternator or generator if you have one.

Your plots show that, with the 440 Ah of batteries, the controller is completing bulk and some absorption, but not going to float ( the third attachment). The next day, the lowest volts show 10.7 volts, assuming this is in morning, the batteries have lost ( in theory) over 400 Ah overnight. The conclusion is they have lost significant capacity with age and use.

You need to go through the system and evaluate what is wrong. Everything points to dead batteries.

Mike

 

[Rocketman]

Earlier I suggested that changing from 6 parallel panels to 2 in series and 3 in parallel ( or 3 and 2- if it makes more sense with your setup ). Have you tried that yet?

having a much higher voltage into your mppt will fix the problem of the panels going to zero - if that is caused by the solar voltage getting too close to the battery voltage. See if you get more total power into your batteries for a day or two- if not go back to all parallel.
Because of the mppt (multiple power points ) on your Solar charge controller - sometimes you can get more power out in series than in parallel - even with shading i way is to test it.

Can you get enough fresh water without running your water maker so much?

if the problem is caused by the water maker pulling the voltage too low for the solar controller to charge. Turning the water maker off for a while will allow the solar charge controller to catch-up on charging.
Good luck

 

[snoobler]

I haven't seen confirmation the MPPT is properly mounted on a vertical surface.

I haven't seen confirmation that all connections have been verified as secure and tight.

I haven't seen evidence that the battery is ever fully charged and suspect it's not getting there.

An 8A load should not hit a 440Ah battery that hard UNLESS it is in a perpetually low state of charge (which may be the case as I haven't seen a single plot where the battery voltage is above 14.0V). A 440Ah battery should be able to handle 22A continuous for 20 hours.

I doubt PV voltage is an issue for the drop offs UNLESS all panel are hit with shade at once or nearly simultaneous. Even in low light conditions, your panels can provide 21-22Voc. Even a single panel in full sun will provide the necessary voltage to stay +1 above the battery voltage. If you do go 2S3P vs. 6P, then it's very important that the panels in series with each other are in shade at the same time.

You're probably aware of it, but with your 50A controller, even with 900W of PV, you will NEVER see more than 700W, and that's at absorption voltage. Most of the time, it will be less.

What are your absorption and float voltages?



The above is pretty telling. Even with 20A coming out of the controller, you're only at 12.36V. This suggests that your battery is very heavily loaded - certainly well in excess of 20A.

 

[HRTKD]

If the root cause is shading, a 6p configuration sure isn't working out like most folks think it should.

 

[snoobler]

I can't help but think that with waves/motion and multiple shading sources, the constant power fluctuation might actually muck with MPPT controllers. I'm beginning to wonder if this is the one case where "dumb" PWM may be superior - no arbitrary +XV above the battery, etc.

The constant power drops are either 100% shade on all panels, or it's the MPPT "resetting" to track max power point.

(or a malfunction)

 

[SunSailor]

Again, thanks for all the input.

I am going to get in there and remove and clean each connection point once this rocking and rolling dies down. This Covid escape plan is a lot of work. Taking the dog to the beach on a new uninhabited island every day makes up for it.

As far as I know, the water maker is a binary device that is ON or OFF, so no way to vary output other than to cycle it on and off.

Here are some charts from this morning. The dip at 8:20 is the microwave oven for one minute. The 9+9 minute square wave is the freezer cycling on and off. Having no prior exposure to solar systems, I am assuming that the dips every 8 minutes and the noise in the solar V/A/W are due to the MPPT algorithm. No indication so far of the drop-out we are discussing.

Driven by wiring convenience, the Victron’s battery port is connected to the main breaker panel on the battery side of the main 12 volt breaker, so in fact it is three cables and five connection points away from the battery.

What is the acceptable voltage drop across a 12 VDC breaker before it needs to be replaced?

 

[SunSailor]

I think Capacitors are inserted into circuits to help with these circuits. I’ve not be a fan of them, but that is for dealing with running a motor with a smaller sized battery bank or inverter than needed That is started a few times an hour, but in this case, it may be a good thing,.

The concept of a capacitor for the watermaker is interesting. I remember reading about capacitors being used as power storage devices, which of course they are by definition, but for applications like car starting batteries. Comments?

People have been pointing to low battery voltages, although I have not seen anything to indicate that. In this month at sea, no device has failed to perform when I needed it, including the windlass (fairly heavy electric winch to hoist the anchor). I have run the generator maybe three times in total because I needed to run motors like the windlass or heat water at night. I think I need to clean all contacts, as suggested, and review all wire sizes. I have just assumed they were adequate, but sometimes manufacturers can be driven by the finance department.

The watermaker has been running all day and the spikes to zero solar watts have not occurred. Today has been cool and largely overcast.

 

[snoobler]

I haven't seen confirmation the MPPT is properly mounted on a vertical surface.

I haven't seen confirmation that all connections have been verified as secure and tight.

I haven't seen evidence that the battery is ever fully charged and suspect it's not getting there.

What are your absorption and float voltages?

@SunSailor

I'm convinced the downward spikes are due to variable shading as they are not present on a cloudy day.

You are posting data that shows low battery voltages. Just because nothing has failed to perform when needed doesn't meant there's not a problem. If your batteries need to be replaced in less than a year because they're being abused, is that a problem? I certainly see it as a problem. If you're good with that, then you can ignore the rest of my post.

I'm highly suspicious that you are operating your battery in a way that does not insure FREQUENT (daily) charges to 100% capacity. Operating a battery bank in this manner will shorten its life dramatically.

Please respond to the following:

1. Is the MPPT properly mounted on a vertical surface with sufficient ventilation (not a in a closed, sealed cabinet).
2. Are all connections between the MPPT and battery secure and properly tightened.
3. What is the gauge and approximate length of wire between the MPPT and battery.
4. Do you see daily evidence of the battery achieving absorption voltage AND the MPPT indicates it's in FLOAT mode
5. What is your absorption voltage?
6. What is your Float voltage?
7. What make/model battery? If flooded, are you monitoring specific gravity?
8. Did you conduct a complete energy audit to establish your energy needs and size the system accordingly?