diy solar

diy solar

Multiple EPEver Charge Controller issue

JMH1783

New Member
Joined
Jan 11, 2021
Messages
56
I have two arrays of 1200 watts. One is 100 watt panels and the other 200 watt panels. Now that I'm monitoring and logging every detail of the system, I'm seeing a problem. The sun came out this afternoon. Both controllers started pushing about 8 amps to the batteries. I then watched the #2 controller slowly decrease until it was down to 0.1 amps. I went to the barn and disconnected array 1 from its panels. The amps crept back up on #2 until it was about 14 amps. I reconnected the panels for controller #1 and the same process happened again where #2 decrease to almost zero.

I have a third controller of a different make (panels aren't in place yet) so I switched the wiring for array 2 to controller #3. It then shot up to 14 amps and controller one was still doing about 8 amps on its own. The sun got even stronger and it went over 20 amps. Now the sun was varying the entire time so the numbers are fuzzy but I think you get the point. I've seen this behavior in the past but was busy with other things so didn't investigate. I've had the first two controllers in service for a bit over 2 years.

I'm thinking that something is wrong with controller #2. I know EPEver offers an external controller for putting charge controllers in parallel but I don't think it is necessary. Has anyone seen this behavior or have any suggestions? Btw, the battery bank is somewhere in the middle when it comes to charge 26.9 volts (LiFePO4).

John
 
Update. I thought I had found the problem. The Tracers have a separate set of wires for monitoring the battery voltage. I think one was broken internally. I replaced the wires on both controllers. It ran fine yesterday but it was heavy overcast. Today we are having a rare full sun cloudless day. Both controllers took off and pulled 20+ amps each. After about 2 1/2 hours they both started to throttle back. Controller #2 is down to 0.8A. Controller #1 is at about 6A. It isn't any less sunny than it was an hour ago. The voltage on the batteries has reached 27.9V but my threshold is set at 28.2V.

I'm going to go down there at lunch and plug a PC into one of the controllers and see if it gives me anymore information. Maybe it switched to float or something. If that isn't it, I'm going to swap in the cheapy third controller in and see what it does. I wish I had this information two years ago, maybe I wouldn't have lost almost $3K on batteries.
 
Evening update. Not good, but weird. I rebooted both controllers. They came back and did increase charge a bit but stayed with 27.9V. I increased the charge limit from 28.2 to 28.4V. Both reacted to that and briefly made an attempt at 28.4. The meter I have on the bank even came up to 28.4V. They then dialed down to very low amps which makes sense if the batteries are full. I watched it and my net ammeter bounced slightly positive to negative.

Then I added another hundred watts load. They didn't react much. I started to lose ground and by the time I had returned to the house and checked, the voltage had dropped to 26.9V. The afternoon was clear and sunny but the batteries continued to lose ground until sunset when the voltage was 26.5V. Am I missing something? Shouldn't the controller increase if their is increased load and a drop in battery voltage?

I've switched the wires on array #2 to use the cheap controller #3. I'll see what happens tomorrow.

Anyone with experience with more than one controller, I'd appreciate your input.
 
I am seeing this too on my system. The current climbs until it reaches a point about 70% of my panels’ full power current rating and then levels off. I have a Epever Tri Ron 3210. I will try to post graphs tomorrow
-Bob
 
Last edited:
Thanks for the feedback. I've been using a MakeSkyBlue 60a controller in place of my second EPEver 5415an (the offending one). The first EPEver and the MSB 60a seem to be working well together. I do notice something that seems odd when I reach battery capacity. Briefly I'll get 28.4 volts until both controller back off then the voltage falls into the 27s. I assume this is normal. At least with these two controllers they keep up with the load and don't let the battery discharge to any great degree as I was seeing before.

So BoBinGA, how long are your wires? Are you getting a voltage drop from your panels to your controller? What is the watts output vs amps?
 
To answer your questions first, I have a 200 ft. of 10 AWG cable from panels to Charge controller. I know this is not ideal, but even this long run does not answer why I am not getting the power I expect. I have 3 X 200W panels in parallel for what I thought would get me close to 600 watts at about 40 volts. They are the 24V Rich Solar panels. I have two 70Ah 24V batteries in parallel each with 70Amp BMS's to give me 140Ah.

So here are the graphs from the EPEVER software. This was on a clear blue sky day. The panels are in full sun from about 10AM until about 1:30PM. On the day I grabbed these graphs, the batteries reached full charge around 12:30PM, so that is why the power drops suddenly then, otherwise the power would slowly decline until about 1:30 before dropping too much:
Power:
03_04_2021_power.png
You can see that I am topping out at about 360 watts.

Current:
03_04_2021_current.png

Voltage:
03_04_2021_voltage.png
In the first half hour of this voltage curve, you can see the voltage dropping from just over 40V (the open circuit voltage) down to about 32V as the power peaks for the rest of the charging period.

So my questions are:
1. Why does the power stop rising suddenly when the solar current from the panels reaches 11A and the battery charging current reaches just 13A, which is only half of the charge controllers rated 30A current?
2. Is there a setting that I am missing that is limiting the power and current?
3. Is this caused by the charge controller limiting the current, or is this caused by the solar panels not producing anywhere near rated power?

I even changed the panel angle to 45* to match the solar altitude for that time of day on March 4, so I should be getting at least 500 watts. Even if the long cable run is stealing 20%, which I don't think it does, that should still give me over 400 Watts. I previously had these on a 100 ft. cable in a different spot and saw basically the same power limits.

Any ideas?

Thanks,
Bob
 
Hi Bob,
If the panels are in parallel they give you about 37.6V @ (5.36 * 3) or 16.08 amps. I'm using the "maximum power" numbers from the Rich Solar panels. With the 200 feet of 10awg you end up with a voltage drop of about 6.4 volts so that is limiting but not the problem. For your distance you might have been better off with the 12V version of the panels and put them in series. I think the controllers limit is about 100V. A different controller might give you a limit of 150V so you could put the 24V panels in series. Higher voltage, less loss over distance.

I guess the next dumb question from my end is how charged are the batteries? When the batteries get close to full the controller will back off. It is one of the weird things I've noticed with the EPEver. It declares that it hit 28.4 volts without my ever recording that voltage on my monitoring system. I originally was using a limit of 28.2 but that dropped off too early. My batteries never reach that plateau. It is always lower than the set value.

On my system I've seen this plateau-ing (word?) behavior which is partially what this thread was about. The root problem for me was I have a constant ~100W draw which I increased when I saw the batteries were full to ~200W. The EPEver once it had decided the batteries were full backed off and didn't adapt to the new load. My batteries started to drain despite a sunny cloudless afternoon. I may still play with settings for this. I'm wondering if the recovery boost voltage (or something like that) is what would kick it back into high gear.

I would try draining your batteries down below 40% and then with a nice sunny day see what happens. Which EPEver do you have?

John
 
I started with two panels and ran them in both series and parallel at different times. But either way, I was getting only 240 watts. I got the third panel just to get a little more because I was draining the one battery I had at the time and in December, the two panels couldn't keep up with the power drain on the one battery. Since, like you said, the controller can only handle 100 volts, I had to run three in parallel, but the extra watts and another battery got me to the point that the batteries stay charged and as you see, they are now usually at a high state of charge.

So from that standpoint, my system is working well. But I still want to understand why I am not getting anywhere near the stated power and current I expected. Since the panels are now in a location that gives me more sun hours than what I had in December, I will try draining the batteries and putting two panels in series again and see if I get closer to the stated power. If so, I guess the next step would be to get another panel and run them 2s2p so that I can use a higher voltage.

I agree that the CC does charge very quickly once I hit about 27.0 volts. But I keep an eye on it and I have the software set to stop charging at 28.5 volts. The CC does that fine and it only maybe goes over 28.5V for a minute of so before the current starts dropping fast and the voltage drops back to 28.5V and stays there until I put a load on it.

On a different day, to test the to see if the panels were OK, I unplugged one at a time and then added them back one at a time. Since the power dropped 120 Watts or so for each panel, I concluded that they are all the same and are probably not the problem.

-Bob
 
Last edited:
Well, the mystery is solved, I think. After running the battery down last night, today I put just two of the 200 watt panels in series to see what would happen. That helped a lot. Here is the power chart for today:
03_08_2021_power.png
Now just the two panels in series give me over 300 watts whereas all three in parallel gave me only 360 watts. So it looks like you really need to get the voltage up into the higher range of the EPEVER Triron 3210 charge controller for it to optimize its maximum power point tracking. I start getting tree shadows at about 1:30, so that's why the graph above drops then.

It looks like if you take the panels' 400W STC rating, de-rate them by 20% to NMOT levels and then deduct another 10 watts for the long cable run, it comes out to 310W, which is what I got today. I guess I should be satisfied with that.

So now the question for me is: Do I get a fourth panel so I can run 2s2p which would get the voltage up and 600 watts of power, or do I stick with the barely sufficient, but less than optimal performance with the three panel I have in parallel? Hmmm...

Thanks, John, for making me think a little harder and try some things.

-Bob
 
Hi Bob. Sounds like you are on to something. Good test. Reading your post made me think of something. In the manual for my controller (Tracer 5415an) there is a section on array requirements. It has these little charts with cell count and Voc. It cross references with the battery bank voltage and has Best and Max categories. For example for 24v bank and 60 cell panels with Voc < 38v it says max 3 panels and best 2 in series. Doing the math that would imply best voltage of less than 76v. If you work across all the cell counts for best voltage it gives you a working range. I get 62v to 92v for my setup. I would try to target the voltage range based on that. I hope that makes sense?

If you have shading then I would definitely recommend going parallel. If you have a half in and half out situation then maybe two controllers would make the most sense.
 
My voltage today peaked just over 80V (that's pretty much the open circuit voltage) and while the batteries were charging, the voltage dropped to just under 70 volts. That's all within the 62V to 92V range you came up with. Again, that indicates that two panels in series is the best idea.

I know that parallel is best for shading issues, but in my case, its mostly all or nothing as far a shade. I get almost 100% sun from 10AM to 1:30PM and then it drops as all the panels hit shade at the same time, so I have no real benefit from running parallel. It looks like series is the answer, but whether four is worth the difference over two for my needs is the question.

So what happened with your controller issues? Did you wire the strings going to both controllers to produce the same voltage and see if that got them to track each other more closely?

-Bob
 
Sounds good. Maybe just let it run a bit and see if you've got enough panels or not. Well, I've just kind of reconfirmed my problem. I lowered my boost voltage to 27v. The arrays hit it early on then kept that voltage for about an hour or so before turning off. They both went to zero and didn't kick in again until the batteries were drained down to the boost reconnect voltage (26.2v). But the weird part is they didn't go all in. They just produced a small amount of current above the load so the battery wasn't getting all that much in charging. I went down and rebooted controller #2 and it came up to about 500w. Controller #1 never came up like that. I'm wondering if these controllers will only do one charge cycle per day. I played with my settings again. We'll see what tomorrow brings.

John
 
Sounds good. Maybe just let it run a bit and see if you've got enough panels or not. Well, I've just kind of reconfirmed my problem. I lowered my boost voltage to 27v. The arrays hit it early on then kept that voltage for about an hour or so before turning off. They both went to zero and didn't kick in again until the batteries were drained down to the boost reconnect voltage (26.2v). But the weird part is they didn't go all in. They just produced a small amount of current above the load so the battery wasn't getting all that much in charging. I went down and rebooted controller #2 and it came up to about 500w. Controller #1 never came up like that. I'm wondering if these controllers will only do one charge cycle per day. I played with my settings again. We'll see what tomorrow brings.

John
So…just curious as to what you found out the next day after playing with the settings again?
 
Back
Top