Greenhouse system not charging

Hawkeye3677

I have installed a solar system for my greenhouse since January 2020.  Last Thursday the outside temp was over 95 and the exhaust fan (runs about 95 watts) ran a lot.  The 2000 watt inverter had an error and shut down.  I reset the inverter but the panels will not charge. I disconnected the panels and the MPPT charger and did a system reset but it still will not charge. I have four 100 watt panels, a Epever 30 watt 3210an controller with MT50 display and two gel 12 volt batteries in parallel. The voltage shows 11.5 volts. What can I do to get it back online?

BB.

Welcome to the forum Hawkeye,

It sounds like your AC inverter shutdown because of low battery voltage--A typical AC inverter will shut down at ~10.5 volts and won't restart until 12.7 volts or so (or you may need to turn of DC power and turn it back on again to reset inverter).

But, the problem is that either your solar power side is not generating and/or your AC inverter simply took so much current (Amps*Hours) of "energy", it killed the batteries....

So you need to take a volt meter and check the Vpanel input of the solar charge controller... Depending on how the panels are wired (and what panels you have), you should see >20 volts on the input to the charge controller. If you see less than ~16 volts, check the wiring/connectors for the solar array.

Next. check the Vbatt output terminals of the charge controller--They should be within a few 0.10 volts of being the same as the battery bus terminal voltage (i.e., if you see 11.0 volts at the battery, you should see near 11.0 volts at the charge controller). If voltages are "different", check wiring connections.

Last simple thing to do is reset the charge controller. Disconnect or turn off the solar array first (Vpanel input to charge controller). Then turn off or disconnect the battery bank (Vbatt output of controller). Wait 5 minutes, then 1) reconnect Vbatt, and 2) reconnec Vpanel (in that order--This makes sure that the charge controller has a "good reset/restart".

Let us know what happens next... You do need to get those batteries recharged pretty quickly (turn off AC inverter, or run fewer fans until the batteries are recharged).

-Bill

Hawkeye3677

Thanks for the answers Bill.  The inverter took so much power that the voltage dropped below the lower limit of the charger. I disconnected the batteries and put a charger on them for about 4 hours each and connected everything up the next day and it is charging again. I didn't realize that the charge controller would't charge the batteries if they got below a certain voltage.  Do you think I need to add another battery to give more capacity?  I haven't run this system through the summer yet, but expect it will be running the exhaust fan a lot more as the temperature gets over 100 for weeks at a time.

Photowhit

You might check to see that the solar panels are wired correctly. The Epever 3210an is an MPPT type charge controller, and would like to see the panels wired as 2 strings of 2 panels, assuming they are 12 volt nominal panels.

If you are where the temps regularly get above 100 degrees, you might consider removing sections of your green house during the summer months so the fan system isn't over worked. It sounds like it already is!

A battery is NOT a unit of measurement. You might let us know the capacity of the batteries. AGM batteries as well as flooded lead acid (and I suspect Gell batteries) don't like extreme heat, and need to be protected from very high temperatures.

BB.

Typically, charge controllers seem to need at least ~9 volts to properly boot and run. But there is no standard, so anything below ~10.5 volts is probably suspect.

The problem with taking your batteries "dead" is that they tend not have very long life after being recharged.

Adding batteries, once you are at ~2 days of storage and 50% discharge (i.e., ~4x your daily energy usage) with battery capacity, adding batteries is not going help you much. Adding solar panels is the best solution for harvesting more power and keeping the batteries "happy".

Creating more square footage of vents (both high to let hot air out, and low to let "cool" air in), is the cheapest method.

Another method would be to run solar panels straight into DC fans directly. Batteries are always the weak/expensive part of solar energy. And when you have a system that runs without monitoring, the chances that you will over discharge the battery bank sometime/someday is much higher.

Adding some sort of alarm (when battery voltage falls below ~11.5 volts under load, or for some inverters, you can set the low battery cutoff to a higher value (like 11.5 to 12.0 volts would better protect the battery bank).

-Bill

Hawkeye3677

I checked the Vpanel and it is .12 v and the battery is 12.16 v. The connectors seem tight and no obvious wire damage to the panels. It was charging this morning but now it shows night mode. Don't understand what could have happened to the panels or wiring since this morning.

I have 2 vents on top and a low vent on west wall that are open. Haven't been through a summer with this greenhouse.

BB.

Just to be clear.... Are the panels at 0.12 or 12.0 volts?

The panel Vpanel input should always be equal to or greater than Vbatt output voltage.

If the panel is 0.xx volts--Sounds like a failed wiring connection. You may have to take apart each connection and test/inspect.

There is always a chance at a short circuit on the array side. Connections reversed on the Vpanel input (reversed polarity)?

-Bill

Photowhit

Photowhit said:

You might check to see that the solar panels are wired correctly. The Epever 3210an is an MPPT type charge controller, and would like to see the panels wired as 2 strings of 2 panels, assuming they are 12 volt nominal panels.

WOW! I get to quote my self...

So LOOK AT YOUR MANUAL!!!

Here's a link;  https://www.epsolarpv.com/upload/cert/file/1811/Tracer-AN-SMS-EL-V1.0.pdf


And shows a chart below that looks like this;


This would indicate that your charge controller is looking for 30-40 volts input from 72 cells in a string. Or 2 of your 36 cell (if there are that many) solar panels.

Hawkeye3677

This would make it a 24 volt system, right? I have 4 100 watt panels. Each has 36 cells. Putting 2 panels in series and the connecting the others (2 in series) would give 24 volts to the charge controller.  I could put my 2 12v batteries in series, but would have to get a new 24v to 110v inverter. 

This 12v system has been working for 4 months. Really don't want to buy a 24v inverter. 

BB.

Just to be clear, it looks like your controller is an MPPT type... They can take a wide range high voltage/low current arrays and efficiently down convert to low voltage/high current needed to charge the battery bank. If you are "happy" with your present array size and battery bank capacity/voltage--You would not need to change anything.

If you wanted to have a larger (higher wattage) array, the simplest way would be to change from 12 to 24 volt battery bank. Because P=V*I, making the battery bank 2x the working voltage, means that you can have a 2x larger solar array with the same charge controller).

There are other things to take into account too... With a higher voltage battery bank, you do need a higher minimum Vmp-array voltage too (array voltage always has to be higher than actual battery bank working voltage). And for MPPT type controllers, the MPPT function needs a bit higher Vmp-array voltage for their MPPT function to work properly.

-Bill

Hawkeye3677

I disconnected the panels today and each panel shows 20 volts. I disconnected the panel supply wires to the controller, jumped them together on the controller side and checked for continuity between all the connectors. I checked each + plug of the 4 to 1 connector to each of the - plugs (4 to 1). They all checked out.  Any suggestions as to what to check next?

BB.

It is not uncommon for folks to mix up + and - on the solar panel pigtails and wiring back to the charge controller input...

What was the Voc array voltage at the controller input? It should be straight forward and Voc-array (voltage open circuit) should be the simple addition of each panel... Two panels in series is 20+20=40 volts array, etc.

If you have a DC current clamp dmm, you can measure the current in each panel of the array--With the array wired up but not connected to the solar charge controller input, there should be no current flow in the array... If you do measure significant current (amps), then you probably have a +/- polarity/wiring issue in the array.

-Bill

Photowhit

Hawkeye3677 said:

This would make it a 24 volt system, right? I have 4 100 watt panels. Each has 36 cells. Putting 2 panels in series and the connecting the others (2 in series) would give 24 volts to the charge controller.  I could put my 2 12v batteries in series, but would have to get a new 24v to 110v inverter. 

NO, ONCE AGAIN, READ YOUR MANUAL!!!

Or read the chart that I have posted!!! It suggest the best ways to connect panels to your charge controller! 
Here I'll post it again;



Notice where it says "System voltage" and suggests the string size for 36 volt panels. 
DOES IT SAY ANYWHERE, OR SUGGEST ANYWHERE THAT "1" would be a good setup?

Hawkeye3677 said:

This 12v system has been working for 4 months. Really don't want to buy a 24v inverter. 

Obviously it's NOT working, or you wouldn't be here. If you had any load on it previously, the colder temps may have allowed some charging (as it does now) until the panels warm up and voltage drops. The nature of solar panels, read up about Normal Operating Cell Temperature values for solar panels, You will find the voltage drops as they warm up.

FWIW - you are measuring the VOC when you say the panels are reading 20 volts when they aren't connected to the charge controller. Under the "36 cell" in the chart, the recommendations are for VOC<23V.

Photowhit

Here's an explanation about the needs of a MPPT system from Midnite;

... Another important factor to consider is the input voltage.... 
...  Most  all MPPT controllers will want to see a minimum of 130% of the actual high battery voltage. For  example, we have a 48v battery and it has an Equalize voltage if 62.3 volts. We would multiply that  Equalize Voltage by 130% (62.3 x 1.3) and we would see that we need a minimum of 81 volts on the  input on the hottest day of the year in order to have enough headroom for the MPPT to work.  

I always try to differentiate between system voltage and battery voltage, but it's hard for people to understand. But when charging, you will want VMP of 130% of the charging voltage for the MPPT charge controller to work properly. When charging a 12 volt system, you will be charging at around 14.5 volts. so 14.5 x 1.3 = 18.85 volts VMP (voltage maximum POWER) When you read 20 volts that is VOC (Voltage Open Circuit) or voltage with no or minimal current (NO POWER) Most 36 cell solar panels have a VMP of around 17.5 volts which isn't high enough for the charge controller to work properly.

Hawkeye3677

I connected 2 sets of panels in series then connected the sets in parallel. I show 42.1 voc. I reset the controller then connected the panels. When I check the input voltage of the panels at the screw heads on the controller I get 2.4 volts. The battery voltage is 12.1 volts at the screw heads. The display shows night time with no charging taking place.

Photowhit

Provided you have connected things in the proper order, usually battery bank first, then solar array, it sounds like either there is a polarity problem, or your charge controller is shot. Might check to see that 1 set of panels or both aren't being hooked up with switched polarity.

Okay, I guess it could be an issue with the panels under load as well, but this is very rare. 

BB.

1) Get the batteries on charge... Lead Acid batteries sitting at 12.1 volts will sulphate and die pretty quickly (weeks, if not days).

2) Sounds like an array wiring problem--Or possibly a failed panel (panels do fail open... And if a panel was ever connected backwards polarity to a battery or large solar array--It can immediately burn out the panel).

Do you have a DC current clamp meter?

https://www.amazon.com/UT210E-Handheld-Digital-Resistance-Capacitance/dp/B06XSMDMYV (lower cost meter)
https://www.amazon.com/gp/product/B019CY4FB4 (mid-cost meter)

You may be forced to check every panel for Voc and Isc (open circuit voltage and short circuit current)... That will find 80% of the failures... If still issues, then you may need to connect the panel under test to a load (light, 12 volt battery, etc.) and measure current under load.

But seeing 2.4 volts at the controller would indicate a major array/wiring issues. If you disconnect the Vpanel input to the controller, what is Voc-array voltage? (it is always possible the input to the charge controller is short circuited).

-Bill

Hawkeye3677

I don't have a clamp meter, checking with some friends to see if they have one. Voltage on panels disconnected from controller is 40.7 volts.
Panels came with MC4 connectors so don't think polarity is the issue. Charging batteries now, they are gel 200 ahr. Sounds like this may be the charge controller. 

Hawkeye3677

Batteries are 12 v 125 Ah gel.

BB.

You do want to measure the array current into the controller... If it is many amps, then the controller input is probably shorted.

If you are seeing much less current (less than 1 amp?), then you still may have an open connection somewhere in the array wiring (digital volt meters take almost zero current--So a poor electrical connection won't be detected).

One quick check is to get a (for example) 120 VAC filament lamp (anything from 4 watt night light to 40 Watt lamp would work) and connect your array +/-... Then measure the array voltage. The lamp will provide enough loading and give you a "true" voltage reading from the array or highlight a corroded/loose connection.

Is this a 12 volt or 24 volt battery bank?

-Bill

Hawkeye3677

It is a 12 volt battery bank, 2 in parallel.

Hawkeye3677

I put a 40 watt bulb on the + & - from the panels, measured 2.6 volts. Without bulb, voltage jumped from 2 to 24 volts, up down up down.

BB.

Sounds like a wiring or solar panel failure (assuming full sun on the panels)...

P=V^2/R
R= V^2/P= 120v^2  / Watts  = ~360 Ohms filament resistance---Probably much less (hot filaments have higher resistance--use 180 Ohms cold)
Say I = V/R = 2.4 volts / 180 Ohms = 0.013 Amps (very rough guess)

You can test each panel at a time (isolate failed panel). At this point, your panel(s) and wiring providing very low current (bad panels, bad wiring connections, etc.). Look for browned/corroded/melted of heat/water access...

Unfortunately, no simple way of finding bad panels/bad wiring/bad connections/reversed connections. You have to take apart and test each component.

-Bill

Hawkeye3677

Thank you, I'll look at that next.

mcgivor

From the description, low voltage under load but normal in open circuit would indicate to me that there is a poor connection in the PV wiring, check all connections for tightness and look for any signs of heat damage . Also check each panel for output using a load, having had the same problem in the past I discovered  the MC4 connectors were the cause, this is more likely if they have been connected/disconnected multiple times and especially if done without the correct tool.

Hawkeye3677

A friend had a spare PMM controller (40 amps) and we hooked it up.  It showed no current coming in, so that eliminated my charge controller as being bad. I finally found a DMM that would read current.  I tried two DMMs I have and they would not read current, but my cheap one worked.  I disconnected all the panels, checked the voltage on each (approx 21 volts each) and short circuit amps (each approx 6.1 amps).  I hooked up two sets in series, then the sets in parallel.  The voltage after the merge plug is 54 volts and 11 amps.  It has to be the wiring going under gound to the greenhouse. Showed zero amps inside the greenhouse at the controller. I'll order new wiring and replace it when it gets in.  Thanks to everyone that helped.  Still can't figure out why the DMM won't read amps.

BB.

It is not uncommon to blow a fuse in the DMM's current path (assuming standard measuring current through the leads type) when a meter is connected to a > 10 amp power source (i.e., set to measure current and a person tries to measure the battery "current" or voltage.

Also some DMMs have three connections, one common connection, and two positive connections. One is for volts/and "low current" readings. And another for 10 Amp range (only) current readings...

Another reason why Current Clamp DMMs are nice... No cutting of wires to measure current, no damage to meter if set wrong or clipped on the wrong wires (CC DMMs still use leads to measure voltage, resistance, etc.).

It is not common for wiring in the middle of a conduit run to fail... Many times it is at J-Box--Water intrusion to wire nuts, or a "sharp" edge in a box or transition/fitting the cut the insulation (shorted, let water in, etc.).

It an also happen that when pulling the cable, a sharp edge somewhere stripped some insulation, and let water in too... 

-Bill