diy solar

diy solar

Are my panels broken? My charge controller? Low wattage..

musashi1707

New Member
Joined
Jul 22, 2020
Messages
31
Hi,

I have two 260 watt panels, paralleled to a 40 amps epever tracerBN(mppt), connected to a diy 280ah lifepo4 battery.

The battery is at a low SOC, it's a sunny California day at around noon.

I'm getting 17 amps max :-\

I check the voltage at the panels, it reads approx 31 volts. (Vmpp = 30.7)

I check the voltage at the solar panel leads to the charge controller (when unplugged), it reads approx 31 volts.

I plug in the panel leads into the charge controller, voltage goes to ~26 volts, both with multimeter and the mt50 display.
(is this normal for voltage to drop when connected to the charge controller?, my wires are kind of chincey but would voltage drop only be present when connected or wouldn't it be lower when testing it unplugged, if that makes sense?)

I tested the short circuit amps of a panel, it's showing 7.5A (I may have to retest this due to undersized multimeter fuse issue)... (Panel rated current = 8.50 A, Short circuit current = 9.01 A)

I tried to configure the charge controller, but I may have to redo this, and I have yet to connect the PC to gain full access to the settings.

My thinking is that since the panels are reading close to rated voltage, and the short circuit amps is..fairly close.. to the short circuit current, that means the panels are OK, right?

Now, could my chincey wire be causing all this issue? I think they're CORRECTION 14awg CCA ( 16ft) but I'm running all four wires to the charge controller atm and paralleling them at the terminals... The issue persists with only 1 panel plugged in, i'll be lucky to get 9 amps..

The battery bms is a JBD 150A, the settings are right to accept a great deal more than 40 amps.

I think I've got 8 gauge wire(1-2ft) going from solar charge controller to battery, and 4 gauge (1-2 ft)going battery to inverter.. No change when under load..

Only other thing I can think of is that I had the charge controller connected to a lead acid battery before my lifepo.. but the same current was happening (I had thought it was because of limitations of the lead acid, but that is obviously not the case.

The converted voltage for the battery I think is reading 13.9v and 9a for one panel. Just about exactly twice the amps for two panels..

So I'm basically getting half the wattage out of the panels.. but whyyyy!!??

Thanks,

-mus

Ps: a couple of edits later and I'm honing in on the 16ft, 14awg, CCA mc4 wires I got off amazon as being the likely culprit. I'm going to cut them and temporarily rewire so that they are much closer to the charge controller and see if that has an effect.
 

Attachments

  • solarpanelspec.JPG
    solarpanelspec.JPG
    227.5 KB · Views: 4
Last edited:
"The converted voltage for the battery I think is reading 13.9v and 9a for one panel."
Your battery shows 13.9V, that is basically fully charged LiFePO4 battery, as battery gets full the charge controller will draw less power from the panels, it will pull as much as it needs up to the maximum charging current.
Try running heavy load and see if the charging current will go up.
 
"The converted voltage for the battery I think is reading 13.9v and 9a for one panel."
Your battery shows 13.9V, that is basically fully charged LiFePO4 battery, as battery gets full the charge controller will draw less power from the panels, it will pull as much as it needs up to the maximum charging current.
Try running heavy load and see if the charging current will
That's the charging voltage, sorry, and it's a guesstimate of the voltage because that's the only number I didn't make note of. My battery is definitely at low SOC, and adding a large load to the battery makes no difference whatsoever.

Believe me when I say that the voltage/amperages that we're talking are going into a low SOC battery.

Thank you for your reply.
 
You say that you had it connected earlier to a lead-acid battery, and if the epever is defaulting to lead-acid, then temperature-compensation is likely in effect (lowers the CV voltage the hotter the ambient temp).

Until you can get the Epever controller set properly for LFP, (like disabling temp-compensation), then this is just going to be a frustrating endeavor throwing darts at the dartboard.
 
Can you describe how you come to conclude that its a low Soc battery? Voltage? Meter? Warm fuzzy feeling?

Happen to have a pic of your battery and wiring? A few sets of eyes are usually a good thing for trouble shooting.
Because I discharged the battery at about 40 amps for about 4+ hours, it has no constant charging source at this time (solar is not connected yet, doing testing when I am present at the location).

I don't have a pic handy, no. I may be able to get one uploaded in the next couple of days.

Again, if you were to take on faith what I have told you to be true, perhaps you might offer me some assistance in the form of say, charging profiles or configurations for the epever charge controller, which contrary to some of your previous posts, there doesn't appear to be readily available information on. I have searched and scrolled through previous posts, enough to take note of your comment linking "searching for epever", but not enough to find someone's detailed instructions for configuring epever on PC or on the MT50.

In fact, so far there is more information in a quick youtube search for epever configuration than on this site, which, I can't help but chock up at least in part to the fact that new users commonly receive snide answers and are brushed aside into the search function which, unless I am mistaken yields no results because people like you who know how are too busy giving snide responses :p

Food for thought buddy, and thank you for your continued willingness to help ;-)
 
You say that you had it connected earlier to a lead-acid battery, and if the epever is defaulting to lead-acid, then temperature-compensation is likely in effect (lowers the CV voltage the hotter the ambient temp).

Until you can get the Epever controller set properly for LFP, (like disabling temp-compensation), then this is just going to be a frustrating endeavor throwing darts at the dartboard.
This is very helpful, and thank you. I think you are absolutely correct.

I have not yet connected the temperature sensor. I am wondering if this and the setting you mention are part of it.

I have an update, which is that I significantly shortened the 14awg CCA wire from the solar panel, and it would appear that the voltage has risen to approximately 28 volts. I believe there was some voltage drop occurring.

That said, there appears to be no impact on the current. I can't be 100% sure of this as my recent test was later in the day, but the amperage was the same before and after the wire was shortened. I think this jives with science but my understanding of electricity is limited. I know that when I was using sub-par wires on my bench power supply to charge the batteries, I would achieve significantly better amps (double) when using much better wire. Is this relevant to a solar charging application as well?

Perhaps you could tell me... Is it normal that upon connection to a charge controller, the charge controller would initially read the panel voltage at their rated 30.5 volts, and then would proceed to drop down in steps to 15 volts, and then come back up to about 28 volts holding steady? All this occurring over a matter of 30 seconds? Is this a part of an mppt learning function of some sort?

Perhaps someone can tell me, if I have a single 260W panel and I want about 15ft between it and the charge controller, what awg wire should I be using? It's carrying 30 volts at a rated amps of 8.5.

Thank you for your time.
 
For a point blank question, perhaps someone who knows can answer..

If a panel is rated for 30.8 volts, and a multimeter shows 31 volts on the panel

and..

A panel is rated for a current of 8.5 amps, and a multimeter shows a short circuit amps of 7.5 (sun at slight angle, a little dirt on the panels, approximately 5 years old)

then..

Is that a good panel?

Thanks.
 
"I check the voltage at the panels, it reads approx 31 volts. (Vmpp = 30.7)

I check the voltage at the solar panel leads to the charge controller (when unplugged), it reads approx 31 volts"

Did you just say the panels are rated 30.7 Vmp, and you're reading 31 Voc?
If so, 31 Voc measurement is low. But what is panel rating for Voc?

measured 7.5A Isc?
Not bad, easily dirt, off-angle sun, hot day could give you 7.5A measured vs. 8.5A spec.

It is possible to have low voltage/current for Vmp/Imp even if Voc and Isc look good. Multiple failure modes in panels.
Have a suitable resistive load to try? I used a couple space heaters in parallel:


But if you have an SCC programmed for lead-acid and you're using lithium, it could just be tapering off. In that case, use a lithium setting and customize it to your battery according to vendor specs.

"14awg CCA"
Do you mean Copper Clad Aluminum?

If you suspect a bad wire or connection, use a DMM (with extension wire if needed) and measure voltage drop.
Have a clamp ammeter? Measure current from each panel. Or disconnect, test one at a time.
I found a wire pulled out of a wire nut after measuring zero amps for one string.
 
"I check the voltage at the panels, it reads approx 31 volts. (Vmpp = 30.7)

I check the voltage at the solar panel leads to the charge controller (when unplugged), it reads approx 31 volts"

Did you just say the panels are rated 30.7 Vmp, and you're reading 31 Voc?
If so, 31 Voc measurement is low. But what is panel rating for Voc?

measured 7.5A Isc?
Not bad, easily dirt, off-angle sun, hot day could give you 7.5A measured vs. 8.5A spec.

It is possible to have low voltage/current for Vmp/Imp even if Voc and Isc look good. Multiple failure modes in panels.
Have a suitable resistive load to try? I used a couple space heaters in parallel:


But if you have an SCC programmed for lead-acid and you're using lithium, it could just be tapering off. In that case, use a lithium setting and customize it to your battery according to vendor specs.

"14awg CCA"
Do you mean Copper Clad Aluminum?

If you suspect a bad wire or connection, use a DMM (with extension wire if needed) and measure voltage drop.
Have a clamp ammeter? Measure current from each panel. Or disconnect, test one at a time.
I found a wire pulled out of a wire nut after measuring zero amps for one string.
In order:
Yes I did say I'm reading about 31 volts open current (if i understand open current correctly), perhaps as much as 33. The Voc rating is 37.8 V.
Can you tell me, I take it voc means when there is no load on the panels, just the wires voltage being tested straight out of the panels? should the Voc remain constant provided there is some light on the panels, and the current is what changes, or do both voltage and current fluctuate throughout the day?

Yes I measured 7.5 isc.

When you say a resistive load, do you mean that I would plug in a space heater (or a heat gun for example?) into my inverter that is connected to my battery? Or are you saying I would wire the panels up directly to some resistive load? I don't want to take up too much of your time explaining this, but why is a resistive load important or what can it show me?

Yes I mean copper clad aluminum.

I do not have a clamp meter, but I will do some further tests on potential voltage drop, and a clamp meter that can test DC is definitely in my immediate future now that I'm realizing how many myriad uses they have and how relevant they are to this field.

Perhaps you can tell me, if the battery is accidentally unplugged from a charge controller when the panels are connected, can that damage the solar panels? This may have occurred when the system was plugged in to a cheap $13 pwm charge controller. What about a secondary charging source like a generator providing current to the same battery that the panels were connected to? Can a generator (through a battery) somehow "spike" the panels?

Your comment regarding "multiple failure modes for panels" is most insightful.. I will read the thread you provided, thank you for that. I have attached the sticker from the back of my solar panels.


Might I ask, in your mind what is the fastest procedure to test if my panels are the culprit in this situation? (I will be reading the thread you provided, but I am inclined to ask nonetheless) Keep in mind, they are both performing identically. I purchased them used, but have not connected both of them together until now.. I'm trying to figure the odds on both of them displaying exactly the same symptoms of low voltage and current.. My thinking says it's got to be my charge controller (unit or settings) or my wiring because of both panels having such similar under wattage characteristics. I suppose it is possible that I did something identical to mess up the panels on separate occasions, which would have been something like: starting up a generator to provide converted power to that battery while said battery was connected to the panels through a pwm charge controller OR the possibility that the battery was disconnected while the solar panels were live to the pwm charge controller.

Thank you for your time,
-mus

Ps: Does disconnecting the panels from a system when the sun is shining damage the panels?
 

Attachments

  • solarpanelspec.JPG
    solarpanelspec.JPG
    227.5 KB · Views: 4
"I check the voltage at the panels, it reads approx 31 volts. (Vmpp = 30.7)

I check the voltage at the solar panel leads to the charge controller (when unplugged), it reads approx 31 volts"

Did you just say the panels are rated 30.7 Vmp, and you're reading 31 Voc?
If so, 31 Voc measurement is low. But what is panel rating for Voc?

measured 7.5A Isc?
Not bad, easily dirt, off-angle sun, hot day could give you 7.5A measured vs. 8.5A spec.

It is possible to have low voltage/current for Vmp/Imp even if Voc and Isc look good. Multiple failure modes in panels.
Have a suitable resistive load to try? I used a couple space heaters in parallel:


But if you have an SCC programmed for lead-acid and you're using lithium, it could just be tapering off. In that case, use a lithium setting and customize it to your battery according to vendor specs.

"14awg CCA"
Do you mean Copper Clad Aluminum?

If you suspect a bad wire or connection, use a DMM (with extension wire if needed) and measure voltage drop.
Have a clamp ammeter? Measure current from each panel. Or disconnect, test one at a time.
I found a wire pulled out of a wire nut after measuring zero amps for one string.
Update:
After having read your provided thread, I will be changing gears in my diagnostic method due to the complexity of panel diagnostics.

I was of the mind to try to rule out the panels as a failure point first so that I could then diagnose the charge controller and wiring.

Instead, now I will attempt to rule out the charge controller, then the wiring, and will leave the solar panels for last. I will be ordering a clamp meter asap.

Thank you for your help.
 
Disconnecting panels from SCC while system is running won't damage anything except, if current is flowing, it will draw an arc and burn the contacts you unplugged. Could shock you if the arc reaches your finger. Recommendation is to turn off current flow with a suitable switch (or cover panel to reduce current to near zero.)

Generator spike isn't likely to go high enough to damage a PV panel; that would have to apply more than Voc and more than Isc to do anything.

"I check the voltage at the solar panel leads to the charge controller (when unplugged), it reads approx 31 volts."
So you measured open-circuit voltage (Voc) of 31V? And panel label says Voc = 37.8V? In that case your reading seems rather low.

Clamp meter needs to be DC, not just AC, of course. Lots of models are AC only.
There are cheaper ones, and more expensive, but I paid $100 for Harbor Freight 1000A/600A/60A meter. Resolution is 0.01A

If panels are bad, would expect different values from them, different amounts of degradation. If they behave the same, likely due to illumination or test conditions.

Copper Clad Aluminum sounds like junk, likely to corrode. But maybe OK if insulation protects it well? As I said, check voltage drops. If you have a "Y" cable paralleling two panels, you can disconnect one to measure voltage of remaining one at panels, and at SCC, to determine voltage drop (which should be minimal.)

Testing PV panels, ideally we expose it to one full sun and sweep current draw, measuring voltage. I connected two AC space heaters in parallel to the DC output of a 24V PV panel. The heaters were about 6.5 ohms, which was a reasonable load for the 165W panel. As the link I gave shows, out of 12 panels I found 5 degraded. But only some had reduced Isc. Thermostat and switch in heaters aren't meant to interrupt DC, so I turn them on/high first, then used a DC rated switch to connect panels.

People have had issues with some models of Epever, had to reset them. Search the forum for that issue. I don't have any experience with them.

Can you connect two 12V lead-acid batteries in series and connect PV panel to them? That would be about Vmp, so should deliver about Imp (except, those are STC ratings, PTC is typically around 85% as much power at typical temperature)
 
Disconnecting panels from SCC while system is running won't damage anything except, if current is flowing, it will draw an arc and burn the contacts you unplugged. Could shock you if the arc reaches your finger. Recommendation is to turn off current flow with a suitable switch (or cover panel to reduce current to near zero.)

Generator spike isn't likely to go high enough to damage a PV panel; that would have to apply more than Voc and more than Isc to do anything.

"I check the voltage at the solar panel leads to the charge controller (when unplugged), it reads approx 31 volts."
So you measured open-circuit voltage (Voc) of 31V? And panel label says Voc = 37.8V? In that case your reading seems rather low.

Clamp meter needs to be DC, not just AC, of course. Lots of models are AC only.
There are cheaper ones, and more expensive, but I paid $100 for Harbor Freight 1000A/600A/60A meter. Resolution is 0.01A

If panels are bad, would expect different values from them, different amounts of degradation. If they behave the same, likely due to illumination or test conditions.

Copper Clad Aluminum sounds like junk, likely to corrode. But maybe OK if insulation protects it well? As I said, check voltage drops. If you have a "Y" cable paralleling two panels, you can disconnect one to measure voltage of remaining one at panels, and at SCC, to determine voltage drop (which should be minimal.)

Testing PV panels, ideally we expose it to one full sun and sweep current draw, measuring voltage. I connected two AC space heaters in parallel to the DC output of a 24V PV panel. The heaters were about 6.5 ohms, which was a reasonable load for the 165W panel. As the link I gave shows, out of 12 panels I found 5 degraded. But only some had reduced Isc. Thermostat and switch in heaters aren't meant to interrupt DC, so I turn them on/high first, then used a DC rated switch to connect panels.

People have had issues with some models of Epever, had to reset them. Search the forum for that issue. I don't have any experience with them.

Can you connect two 12V lead-acid batteries in series and connect PV panel to them? That would be about Vmp, so should deliver about Imp (except, those are STC ratings, PTC is typically around 85% as much power at typical temperature)
Good information, I am still digesting much of it.

Your idea about connecting two lead acids in series is brilliant and I think something I can do.

So the method would be to take two 12v lead acid batteries to a lower SOC, then, series connect them to achieve approximately 24-26 volts. Then I would use my newly aquired clamp meter to measure the current going into them, and voila, I've got a simple but effective way in which to test how much current is coming out of the panels. This would bypass any "complicated computer stuff" to do with charge controllers and BMS.

Is my thinking correct on this? It's my kind of solution, tests with what I have on hand and should yield "close enough" results. Essentially if I'm getting anything over 4 amps out of a panel into those series'd batteries, then I know my specific issue is with my charge controller or elsewhere in my setup..

I take it that relatively healthy so-called marine 12v batteries at a low SOC should be able to accept that kind of amperage (10-20a)? I really don't know nearly enough about electricity, my computer background is allowing me to limp through this.

Much obliged,
-mus
 
May or may not even need to lower SoC first. If they are FLA, over-charging is "equalizing". You would only do it briefly, anyway.
Car batteries typically accept 10 to 20A or more for a little while from alternator.

If you don't have a DC switch, just cover the panels while making/breaking connection.
(voltage can still be high enough to shock, but current low enough not to damage connectors.)
Make/break away from battery outgassing, of course.

Charging to 15V per battery x2 and 6 to 8.5A would put it real close to maximum power point, so close enough.
 
If ... a panel is rated for a current of 8.5 amps, and a multimeter shows a short circuit amps of 7.5 (sun at slight angle, a little dirt on the panels, approximately 5 years old)
then..
Is that a good panel?

IMHO, yes. Quite surprisingly good, in my (limited, but of 30+ years) experience.
I would guess that even with the sun at a perfect angle, spotless panels, and normal temperature, you couldn't get much more than that.

I think the way they rate panels... they take them to the top of a mountain, where it's like below 0ºF, the atmosphere is as thin as it gets, wait for solar flares maximums, micro-tune the angle, put heavy metal music on at max volume to stimulate... or they calculate what would happen if they did all that, add a half-amp or two for good measure, and get the rating. IRL, I don't think anyone ever got an actual amperage like the rated one :)
 
IMHO, yes. Quite surprisingly good, in my (limited, but of 30+ years) experience.
I would guess that even with the sun at a perfect angle, spotless panels, and normal temperature, you couldn't get much more than that.

I think the way they rate panels... they take them to the top of a mountain, where it's like below 0ºF, the atmosphere is as thin as it gets, wait for solar flares maximums, micro-tune the angle, put heavy metal music on at max volume to stimulate... or they calculate what would happen if they did all that, add a half-amp or two for good measure, and get the rating. IRL, I don't think anyone ever got an actual amperage like the rated one :)
I am realizing that solar panels are a more flexible affair than I had previously thought. If I can get 90% efficiency out of the panels then I would be satisfied, but at 45% I figure I have work to do. After I make the modification to the temperature settings that Substrate mentioned above, and if that yields no effect, after I run a test on some good old durable lead acids in series like Hedges suggests, I'll start taking the panels down off the roof and getting them ready for our trip to Metal-Mountain ;-)
 
If I can get 90% efficiency out of the panels

I guess you mean 90% of rated power. If you can get that under close-to-ideal conditions, I would say you're doing OK.
but at 45% I figure I have work to do

But then I don't think the fault is with the panels, at all. If you get an actual 7.5A out of an 8.5A max rated panel, for my experience, the panel is just fine.
 
Back
Top