12v panel to keep 48v bank floating

Estragon · January 2017

I'm thinking of adding panels to keep batteries charged over the winter. There's about a month when no sun hits existing arrays, and depending on the weather they can get covered by sticky snow early in winter that won't melt until spring. I don't get there much in the winter.

One option for 48v bank is to get a 12v panel and put it somewhere that it would definitely see sun in the depths of winter, hang it vertically. and use a 12v to 48v converter to charge. The converter puts out 56v, which is a bit high for float, but at 3a on a 350ah bank of flooded L16s, I don't know if it's a problem. I'm a bit concerned that it could gas with the higher voltage and I wouldn't be there to check/add water.

The converter I was looking at:
https://www.solar-electric.com/12vopato48vo1.html

Obviously the other option is to get a couple of big panels for ~80vmp and a small controller.

Any thoughts?

inetdog · January 2017

How big a panel would you plan on using?
You would need to use some sort of charge controller as even a 3A trickle charge could eventually kill an unattended FLA battery.
The alternative, IMHO, would be to use two or more higher voltage grid tie market panels in series (more watts for the buck) and a good CC, leaving out the rather expensive 12 to 24V converter.
Nominal 12V panels are a lot more expensive per watt because of the lower demand among other things.

mike95490 · January 2017

If the place is un attended, and no loads, if the batteries start off full, and your charger is set for short absorb times, you should get through several cold months no problem.

I would not bother with the extra panel and charger - your array even under snow and ice will make more power than 1 panel. and if you have no loads connected, batteries will be fine,

Estragon · January 2017

@inetdog - I was thinking a 12v panel around 100w. They are more $/watt and the converter is expensive compared to a smallish 48 controller but 12v panel might be cheaper to ship for just a couple of panels. Will have to check local pricing for larger panels.

@mike95490 - the only load is the controllers but they take about 3-4 watts, which adds up. Last year was unusually warm in Dec so when I tried to get out in early Jan the ice wasn't safe. I got out in mid-Feb, by which point batteries were badly discharged and nearly frozen. Luckily the cold seems to have prevented permanent damage and I was able to recover.

softdown · January 2017

What was "badly discharged"? Meaning..what was the mid- Feb voltage level?

56 volts is 14 volts/battery. I'm not certain that would gas your batteries. Short days during the winter...

mvas · January 2017

How about using an adjustable DC-to-DC Boost Inverter with the new Solar Panel ?
Then you can set the output voltage to ~53 volt output = 13.2 volts ( Float Voltage ) x 4

Something like this ...
http://www.ebay.com/itm/DC-DC-600W-10-60V-to-12-80V-Boost-Converter-Step-up-Module-car-Power-Supply-OY-/281797179064

softdown · January 2017

mvas said:

How about using an adjustable DC-to-DC Boost Inverter with the new Solar Panel ?
Then you can set the output voltage to ~53 volt output = 13.2 volts ( Float Voltage ) x 4

Something like this ...
http://www.ebay.com/itm/DC-DC-600W-10-60V-to-12-80V-Boost-Converter-Step-up-Module-car-Power-Supply-OY-/281797179064

I am not seeing an adjustment knob...

DConlyGuy · January 2017

the 2 blue box's with the gold screw on top is the adjustment for this controller

http://www.ebay.com/itm/DC-DC-600W-10-60V-to-12-80V-Boost-Converter-Step-up-Module-car-Power-Supply-OY-/281797179064

mvas · January 2017

The Blue Multi-Turn potentiometers on the front right edge of the device are labelled as follows ...

I-ADJ = Amps Adjust ( not used, set to max amps for this application )
V-ADJ = Volts Adjust ( set to your Float Voltage ~53 V, as needed )

This particular DC-to-DC boost inverter is just one example.

If the existing Solar Panels are snow covered and the existing Charge Controller draws 3-4 Watts
then would it makes sense to disconnect the existing Charge Controller over the winter
and just FLOAT the batteries at ~53 Volts ?

Is it better to Float the battery at 100% SOC for two months vs
draw down every night and then recharge the next time the sun appears?

Estragon · January 2017

softdown said:

What was "badly discharged"? Meaning..what was the mid- Feb voltage level?

56 volts is 14 volts/battery. I'm not certain that would gas your batteries. Short days during the winter...

IIRC voltage was down around 10v or maybe less. It was low enough that the Outbacks wouldn't come on to charge (LBCO ~10.5) with generator power. I put a 12v car charger on a pair of batteries to get the inverters to see enough voltage to come on and start charging.

The battery gassing I don't know either. In the dead of winter a panel might see an hour or two. Even later in the spring wouldn't see much more than that as the area I'd put it is wooded with a narrow clear area to the south.

Estragon · January 2017

mvas said:

The Blue Multi-Turn potentiometers on the front right edge of the device are labelled as follows ...

I-ADJ = Amps Adjust ( not used, set to max amps for this application )
V-ADJ = Volts Adjust ( set to your Float Voltage ~53 V, as needed )

This particular DC-to-DC boost inverter is just one example.

Certainly a lot cheaper than the solar converters one, and adjustable output too!

If the existing Solar Panels are snow covered and the existing Charge Controller draws 3-4 Watts
then would it makes sense to disconnect the existing Charge Controller over the winter
and just FLOAT the batteries at ~53 Volts ?

That's sort of what I had in mind. Use 12v panel and boost to float.

Is it better to Float the battery at 100% SOC for two months vs
draw down every night and then recharge the next time the sun appears?

I'm assuming it would be better to float than to have small daily cycles, but not sure. If I couldn't make it out at all in winter, it could be more like 6-7 months. Late Oct/November gets too cold for In/Outboard boat, freeze-up usually runs from late Nov to early Jan. Break-up can be anywhere from 2nd week April to late May, so worst case could be mid-Oct to mid-May so ~7months. I could shut the controllers off too and hopefully self discharge wouldn't be so bad, but I was also thinking of setting up some security/monitoring gear at some point.

softdown · January 2017

If your batteries were relatively unharmed by sitting ~10 volts for several weeks, that would indicate that cold weather low voltage is almost harmless compared to warm weather low voltage. Perhaps there is still a point where low voltage destroys the cold battery?

I have not seen this possibility discussed here.

Estragon · January 2017

Although the cold apparently significantly slowed sulfation there is a much bigger risk of freezing damage when badly discharged. The electrolyte hadn't frozen solid but was sort of slushy.

softdown · January 2017

Freezing destroys the battery.

mike95490 · January 2017

Today, under 1" snow and heavy cloud cover, my 5Kw array peaked at about 300w harvest was just over 1Kw for the day,
Harvest under snow would be low, but should still be enough to recharge and handle losses.

But you CANNOT connect a DC-DC convertor to a PV panel and expect it to work, It needs to have a MPPT in it or it will collapse the PV voltage, because 95% of all gear is designed to work with a voltage source. PV panels are CURRENT sources,

Image: https://us.v-cdn.net/6024911/uploads/editor/uk/s7u2jwua2ens.png

mvas · January 2017

The OP stated that his snow covered Solar Panels cannot maintain 100% SOC.

No, we do not need an MPPT controller to Float Charge a battery using a Solar Panel.
We have been using Solar Panels to charge batteries long before MPPT technology even existed for PV.
You may be old enough to know that fact.

Under light loads, when the Output Voltage is greater than Vmp then the Solar Panel acts more like a Voltage Source
Under heavy loads, when the Output Voltage is less than Vmp then the Solar Panel acts more like a Current Source.

The I-V graph is not a pure current source since the I-V CURVE transitions
from a current source, to a voltage source, near the Max Power Point.

Look at the I-V CURVE ...
The CURVE is horizontal at high amps / lower volts = constant current mode but
then the CURVE is nearly vertical at higher volts / lower amps = constant voltage mode.

The OP will be operating his Solar Panel in the "high volts / low amps" Constant Voltage Mode portion of the I-V curve.
At light loads the Output Voltage of a Solar Panel will be greater than 80% of Voc.
The voltage from the Solar Panel will not collapse when a Float Charger is connected.

softdown · January 2017

Couple used 24 volt panels? Hard to know folks budget here. We have Angelini selling to the "rich & famous" j/k?. We have softdown who budgets eating at $2/meal. Heard he is a south Florida landlord....bless his soul.

Shipping 24 volt panels is big bucks.

mike95490 · January 2017

mvas said:

The OP stated that his snow covered Solar Panels cannot maintain 100% SOC.

......
The voltage from the Solar Panel will not collapse when a Float Charger is connected.

OK, My bad. I saw DC-DC converter, not a step up charger AND I did not pick up on the panels being in shade too.

First some thoughts to not spend any $.
rewire in winter to have both arrays feed 1 controller and mothball the 2nd Cuts the controller losses. (I know they are not
insignificant, the classics run warm even at night.) this may harvest enough to keep batteries full, or at least not draining them,
Look into updating the controller firmware, it may improve things
I had to select Legacy P/O solar mode, or the controller did not track MPPT well.

bill von novak · January 2017

mvas said:

The OP stated that his snow covered Solar Panels cannot maintain 100% SOC.

No, we do not need an MPPT controller to Float Charge a battery using a Solar Panel.

The OP will be operating his Solar Panel in the "high volts / low amps" Constant Voltage Mode portion of the I-V curve.
At light loads the Output Voltage of a Solar Panel will be greater than 80% of Voc.
The voltage from the Solar Panel will not collapse when a Float Charger is connected.

You are correct; you do not need an MPPT controller to float charge a battery using a solar panel. An array with a Vmp at or above float voltage will do a good job of floating the bank.

However, the voltage absolutely will collapse IF you use a boost converter to try to boost the voltage. Here's why:

The boost converter attempts to maintain its output voltage by pulling more current from the source. Early in the morning the panel will "wake up" and allow the converter to start working. At this point open circuit voltage is high (panel cold) but available current is very low (not much light yet.)

The converter will attempt to reach its output voltage setpoint - let's say it's 56 volts. The batteries have dropped to 50 volts overnight, so the converter ramps up its input current to attempt to make the output 56 volts. The small amount of current available from the panel is unable to do this. So it draws more and more current, bring the panel voltage lower and lower. Finally it reaches the point (5V? 4V? depends on the DC/DC converter) where the converter shuts down. The panel voltage rises and it starts up again and the same thing happens. It hiccups there all day.

Since the panel never gets above about 4V, maximum power available from a 100W (18V Vmp) panel is 100*18/4=22 watts. This won't be enough to float the panels.

How to fix this? Easiest way is to get rid of the buck converter and use a 60V panel with a cheap PWM controller. Next easiest is to do a "fake MPPT" on the boost converter. You can do this by choosing a voltage (say, 15 volts) and putting a circuit on the boost to not attempt to operate below that voltage. Now the system still hiccups, but it does so at 15 volts - and that gives you 83 watts out of your panel. What you have done is "set" the panel's operating voltage to 15 volts. This isn't as good as a true MPPT (which changes that voltage constantly to adjust for changing panel voltages) but for a maintenance function should be OK. Also have to ensure that the boost converter will survive that sort of hiccup operation.

Estragon · January 2017

So if I'm understanding this correctly, the (much cheaper) boost converter on Ebay linked to above won't work, but the Solar Converters one will?

The firmware is more or less up to date (last summer).

The mode thing is interesting. Looking through logs, there's zero measured aggregate Kwh but Vin is regularly over 90v and occasionally over 100v. High temp (which is FET temp IIRC) is also >0C at times, suggesting the controller may be trying to produce something?

On the rewiring thought, I wonder if rigging up a simple PWM controller like Schnieder C40 which has much lower self-consumption would be an option. I assume it wouldn't overcurrent, but voltage limit of 125v is not that much higher than 120 or so I see in the logs. The classics are okay with reasonable overvoltage, but I doubt these would be.

This shows shading on Dec 29 2014. That winter there wasn't much snow stuck on, so the little bit of sun melted it. Last winter there was 4-5" still there in mid-Feb. even though there's decent sun before then.

Image: https://us.v-cdn.net/6024911/uploads/editor/nz/d0ye093bdl2m.jpg

bill von novak · January 2017

Estragon said:

So if I'm understanding this correctly, the (much cheaper) boost converter on Ebay linked to above won't work, but the Solar Converters one will?

I don't have any experience with the Solar Converters product. Genasun makes a product to do this, though; it is both a boost converter and an MPPT circuit that avoids the collapse problem. (I've used it to charge 42 volt li-ions from 12V panels and they work well.) Google Genasun GV-BOOST. It's also a little cheaper.

BB. · January 2017

Here is one of the Genasun products:

https://genasun.com/all-products/solar-charge-controllers/for-lead/gvb-8a-pb-solar-boost-controller/

Or, get 4x 12 volt panels and run them in series with a 48 volt capable PWM charge controller.

-Bill

mvas · January 2017

bill von novak said:

mvas said:

The OP stated that his snow covered Solar Panels cannot maintain 100% SOC.

No, we do not need an MPPT controller to Float Charge a battery using a Solar Panel.

The OP will be operating his Solar Panel in the "high volts / low amps" Constant Voltage Mode portion of the I-V curve.
At light loads the Output Voltage of a Solar Panel will be greater than 80% of Voc.
The voltage from the Solar Panel will not collapse when a Float Charger is connected.

You are correct; you do not need an MPPT controller to float charge a battery using a solar panel. An array with a Vmp at or above float voltage will do a good job of floating the bank.

However, the voltage absolutely will collapse IF you use a boost converter to try to boost the voltage. Here's why:

The boost converter attempts to maintain its output voltage by pulling more current from the source. Early in the morning the panel will "wake up" and allow the converter to start working. At this point open circuit voltage is high (panel cold) but available current is very low (not much light yet.)

The converter will attempt to reach its output voltage setpoint - let's say it's 56 volts. The batteries have dropped to 50 volts overnight, so the converter ramps up its input current to attempt to make the output 56 volts. The small amount of current available from the panel is unable to do this. So it draws more and more current, bring the panel voltage lower and lower. Finally it reaches the point (5V? 4V? depends on the DC/DC converter) where the converter shuts down. The panel voltage rises and it starts up again and the same thing happens. It hiccups there all day.

Since the panel never gets above about 4V, maximum power available from a 100W (18V Vmp) panel is 100*18/4=22 watts. This won't be enough to float the panels.

How to fix this? Easiest way is to get rid of the buck converter and use a 60V panel with a cheap PWM controller. Next easiest is to do a "fake MPPT" on the boost converter. You can do this by choosing a voltage (say, 15 volts) and putting a circuit on the boost to not attempt to operate below that voltage. Now the system still hiccups, but it does so at 15 volts - and that gives you 83 watts out of your panel. What you have done is "set" the panel's operating voltage to 15 volts. This isn't as good as a true MPPT (which changes that voltage constantly to adjust for changing panel voltages) but for a maintenance function should be OK. Also have to ensure that the boost converter will survive that sort of hiccup operation.

I agree with you but only if there is way less than say 5 watts of sunlight (essentially no sunlight) falling on the 100 Watt Solar Panel.

So, let's say there is just enough sunlight to generate a minsicule 10 watts ...
The DC-to-DC inverter has just hiccuped to OFF.
The Output Voltage will instantly jump to 12 volts Voc.
When there is zero (0) amps from the load then Max Voc will be produced instantly.
This is how a Solar Panel works per the I-V Curve.
The Solar Panel will raise its output voltage to 12 Volts Voc faster than the DC-to-DC controller can restart.
The DC-to-DC controller needs way less than 1 amp to Float the battery at 53 Volts.
10 watts / 1 amp = 10 Volts.
So, now the DC-to-DC Inverter pulls the 12 Volt output down to 10 Volts.
The DC-to-DC converter turns ON, and stays ON, as the sun gets higher and higher in the sky

The Output Voltage of the Solar Panel is specified by the I-V Curve and sunlight.
When there is enough sunlight to produce at least a feeble 10 watts of power
then DC-to-DC converter cannot draw enough amps to pull the Output Voltage below 10 Volts and
the DC-to-DC Inverter turns on, and stays ON.

In your scenario you state that the the DC-DC Inverter is pulling the Output Voltage down to 5 Volts
So, how many amps would the DC-to-DC Inverter need to draw to do this?
10 watts / 5 volts = 2 amps
The DC-to-DC Inverter will not need 2 amps to float a 100% SOC battery.

We are not bulk charging the battery, we are doing a Float Charge.
My 150 AH flooded lead acid battery floats at a miniscule 0.025 amps !
So, where in heck are you calculating that DC-to-DC Inverter needs 2 amps ???

I have disconnected my 150 AH Battery from the float charger,
I will re-connect and measure the milliamps in-rush tomorrow afternoon.

Like I said before,
we running the Solar Panel in Voltage Mode (less than 1 amp), not Current Mode, not 8 amps.

Did you not do solar cell experiments in grade school?
Connect a motor, a bulb or a radio to some solar cells.
With just enough sunlight, eventually the motor will spin, the bulb will glow and the radio will play music.

You claim that the voltage will get stuck at 5 volts, or some "turn on" voltage,
and none of the above devices would ever turn on and stay on, but yet they do, everyday.

The DC-to-DC Inverter does not need MPPT, real or fake, to operate.

baphenatem · January 2017

Estragon said:

I'm thinking of adding panels to keep batteries charged over the winter. There's about a month when no sun hits existing arrays, and depending on the weather they can get covered by sticky snow early in winter that won't melt until spring. I don't get there much in the winter.

One option for 48v bank is to get a 12v panel and put it somewhere that it would definitely see sun in the depths of winter, hang it vertically. and use a 12v to 48v converter to charge. The converter puts out 56v, which is a bit high for float, but at 3a on a 350ah bank of flooded L16s, I don't know if it's a problem. I'm a bit concerned that it could gas with the higher voltage and I wouldn't be there to check/add water.

The converter I was looking at:
https://www.solar-electric.com/12vopato48vo1.html

Obviously the other option is to get a couple of big panels for ~80vmp and a small controller.

Any thoughts?

Option #3: Four of the 3W panels connected in series, mounted to a common frame and placed where the sun will hit them. Add a 1 amp fuse in series at the 48V battery bank. Total cost; about $90 with shipping, (plus wire) which is less than half what the 12 to 48VDC converter would be. https://www.solar-electric.com/solarland-slp003-12u-3-watt-12-volt-solar-panel.html

mike95490 · January 2017

baphenatem said:
Option #3: Four of the 3W panels connected in series, mounted to a common frame and placed where the sun will hit them. Add a 1 amp fuse in series at the 48V battery bank. Total cost; about $90 with shipping, (plus wire) which is less than half what the 12 to 48VDC converter would be. https://www.solar-electric.com/solarland-slp003-12u-3-watt-12-volt-solar-panel.html

12 W @ 50V (0.24A ?) is not going to float charge a 370ah 48v bank

AND - you are going to need a BLOCKING DIODE if you don't use a controller

For solar (4 hours) float charging of a roughly 400ah 48V bank, extrapolate what a 12V 100ah car battery needs and see where that goes.

I like the idea of switching the arrays to a dead simple PWM controller for the winter, better than trying to kluge around a Classic

baphenatem · January 2017

Depending who's numbers you want to use, FLA batteries lose about 1% of their charge in AH per day at room temperature, but less at lower temperatures. Run the numbers and select the panel sizes accordingly. The point is that the job can be accomplished with a minimum of additional equipment, even with a blocking diode added to the mix, but even that component has its advocates and detractors.

baphenatem · January 2017

BTW, here's a link to an article to just how little a cold battery will lose. So maybe I will stick with the 3 watt panel sizing estimate, keeping in mind that the battery needs to be fully charged going into this scenario. http://www.pbase.com/mainecruising/self_discharge

mvas · January 2017

I had reported on Tuesday, that my 150 AH 12 Volt Battery was floating at 13.2 V at 0.027 Amps or 27 milliamps.
After taking the measurement, I disconnected the battery from the float charger to let it "rest".
After 18 Hours of "resting", the battery voltage dropped from 13.2 volts to only 12.85 volts.
Next, I hauled this equipment outside ...
* A 210 Watt Solar Panel ( Voc = 23.1 Volts and Isc = 12.2 Amps )
* A Buck-Boost inverter (BBI)
* The 100% SOC Battery
The output of my BBI was previously set to 13.2 Volts.
I attached the BBI DIRECTLY between the PV Panel and the battery.
As I turned the PV Panel towards the sun ... the BBI turned ON.
As I turned the PV Panel away from the sun ... the BBI turned OFF
I did this ten times in row - the BBI worked every time, without fail.
If you don't understand how a PV Panel can energize a Buck-Boost Inverter then just do it for yourself.
I measured the milliamps every time the BBI turned ON - it was only 60 milliamps
After one hour of charging from the PV panel, the current was already down to 42 milliamps.
Now the sun was getting too low.
So, back in the garage, after another hour on the 13.2 V Float Charger, the current was down to 35 milliamps.
As you clearly see, it takes very, very, very little current to Float a 150 AH battery.

So, who else has actually measured the milliamps required to float charge a 100% SOC 150 AH battery?

The root cause of Estragon's battery drain is more likely the drain from the 3W - 4W consumed by the Controller with snow covered PV panels, than Self-Discharge.

Also, make sure you bubble and mix the electrolyte thoroughly before winter, to eliminate any stratification.

Estragon · January 2017

@mvas - yes, it's definitely the controller tare that is the problem. Self-discharge is mitigated by cool temps as the banks are in an insulated but unheated crawlspace area, but flooded L16s do supposedly have fairly high self-discharge rates.

I wasn't sure if disconnecting everything or leaving controllers on would be best when I first shut down for winter with the system. Consensus here was that leaving controllers on to charge would be best, and that worked fine the first year. There was a period of little production because of solstice shading as expected, but recovered okay after. Last year weather was different which both left panels covered in 4" of snow, and late freeze-up prevented me from getting out there. Weather has been similar this year, so I'm expecting the same. Fortunately (!) we've had lots of sub-zero F weather recently, so I should be able to get there next week. Will clear snow and charge. Sun should be high enough soon to melt any new snow for this year.

I did fully charge and EQ banks before leaving in Oct.

For future I plan to leave Classics off. The question will be how to offset self-discharge and maybe some small security/monitoring loads.

What would happen in your test above had the battery been down to say 70% SOC?

littleharbor2 · January 2017

What I do for summer off-season charging in Baja is use a single panel and basic controller to keep a maintenance charge on my batteries. I do this to keep my Classic 200 from working needlessly in the heat all summer long. In your case a strategically hung vertical panel would stay snow free and pick up all that reflected light.

mvas · January 2017

Estragon said:

@mvas - yes, it's definitely the controller tare that is the problem. Self-discharge is mitigated by cool temps as the banks are in an insulated but unheated crawlspace area, but flooded L16s do supposedly have fairly high self-discharge rates.

I wasn't sure if disconnecting everything or leaving controllers on would be best when I first shut down for winter with the system. Consensus here was that leaving controllers on to charge would be best, and that worked fine the first year. There was a period of little production because of solstice shading as expected, but recovered okay after. Last year weather was different which both left panels covered in 4" of snow, and late freeze-up prevented me from getting out there. Weather has been similar this year, so I'm expecting the same. Fortunately (!) we've had lots of sub-zero F weather recently, so I should be able to get there next week. Will clear snow and charge. Sun should be high enough soon to melt any new snow for this year.

I did fully charge and EQ banks before leaving in Oct.

For future I plan to leave Classics off. The question will be how to offset self-discharge and maybe some small security/monitoring loads.

What would happen in your test above had the battery been down to say 70% SOC?

RE: 70% SOC
Sorry, but my simple and cheap design is a Float Charge for a battery with no loads.
It would take too long a time to recharge a 70% SOC battery with a float charger.

I see now, that you are stating you actually have loads on the battery.
What is the minimum Voltage that your the Midnight Classic 150 needs to properly charge the 48 Volt battery bank? 130% ?
Maybe wire three (3) vertically mounted PV Panels in series and reconnect one existing Midnight Classic 150 to them.

Would (something like) three of these PV Panels (or bigger), connected in series, turn on one of your Midnight 150?
https://www.solar-electric.com/solarland-slp020-24u-20-watt-24-volt-solar-panel.html
http://https//www.solar-electric.com/solarland-slp030-24u-30-watt-24-volt-solar-panel.html
Is that enough to offset your LOADS?

I treat sulfate like cancer. It must be removed immediately. It is best to prevent any sulphation. When sulfate crystals harden, it is almost impossible to convert them back to lead & sulphor. It is possible to "boil" off hardened lead sulfate crystals to expose fresh lead using equalization. But then the lead sulfate crystals simply fall to the bottom of the battery. You have lost that lead forever, and you have lost AH capacity forever.

12v panel to keep 48v bank floating

Comments

Categories