Solar Panel System Troubleshooting - Charging Battery

[sophiecentaur]

latitude effect shows less than i expected

That latitude effect applies for a fixed installation over the whole year with the optimum alt az setting. Pointing the panel at the Sun on a Summer's day will give more than a fixed panel is likely to produce. Leaving the panel in that position will produce rubbish output in winter. It must always be worth while chasing the Sun if you can access the panel.

 

[Wolst73]

Ignore numbers on meters above. Borrowed pic off YouTube.

1. Hooked up as above without the volt measurement. Measured 0.02amps.

2. Hooked up from solar panel as follows. Positive wire from panel into charge controller. Negative wire of panel to red terminal of multimeter. Black of multimeter into negative of charge controller. Measured -0.02amps.

Voltage of panel showing about 19v. Voltage at charge controller terminals for panel and battery 11.8.

 

[sophiecentaur]

If I were in your shoes, I would be measuring pretty well everything that I could and under all conditions. You need short circuit current from the panel with full Sun, charge current with a charged battery and with a battery after it's been discharged to,say, half it's rated capacity. Measure, measure and measure again or the real problem will slip down a crack in your floorboards. We can't help if we don't know much more. Every combination of charge, sunlight and discharge may be needed here. (That's not really much trouble is it?)
I must say, it looks as if the regulator could be the problem but, without many more measurements, you (we) can't have a clue about the real situation. Trying to fault find by remote control is very hard for all the experienced brains that you are connected to here. (Myself not particularly included )

 

[jim hardy]

It's hard to troubleshoot from somebody else's observations
you're always not quite sure exactly what they meant..

2. Hooked up from solar panel as follows. Positive wire from panel into charge controller. Negative wire of panel to red terminal of multimeter. Black of multimeter into negative of charge controller. Measured -0.02amps.

something wasn't allowing current to flow from the panel into the charge controller.

Voltage of panel showing about 19v.
? 19 volts at one end of the wires ?
Voltage at charge controller terminals for panel...11.8.
11.8 volts at other end of wires ?

and battery (11.8) it's sure not connected to that 19 volt solar panel, is it ?

hmmmmm lessee here

Where's the disconnect ?
I'd check connections on back of solar panel

You're reporting 7.2 volts drop along a wire
that should be easy to find

then check for current by steps 1, 2, and 3

 

[sophiecentaur]

The dreaded 'remote control' problem, Jim.

 

[jim hardy]

The dreaded 'remote control' problem,

along the lines of

“ I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant”― Alan Greenspan

 

[Wolst73]

End of wires coming from solar panel show 19v. But when hooked up to charge controller with battery hooked up, they show 11.8 which is the same as what battery terminals show. Same as what this showed:


Raining now so may not get anymore sun for today.

 

[jim hardy]

End of wires coming from solar panel show 19v. But when hooked up to charge controller with battery hooked up, they show 11.8 which is the same as what battery terminals show. Same as what this showed:

Oh.
So the 19 and 11.8 you reported earlier were measured with two different conditions ?
19 with panel not connected, 11.8 with panel connected ?
That might explain why they are so different. okay
with everything connected , panel to controller and battery hooked up

At top between - terminal and + terminal (where i asked about 19 V,) what voltage do you see? About 12.1 ? __________________ V
At bottom between - terminal and exposed junction of diode&red wire, what voltage do you see? About 11.8 ? ___________________ VNext, with everything still hooked up,

What voltage do you see between - terminal and bare end of black wire ? Zero ? _______________ V...........................Now disconnect panel from controller by lifting the red and black panel wires from the controller
set your meter for 10 amp scale
move the meter plug to correct receptacle as she described at 1:08
post a picture of your meter if it's not clear which one to use , they're not all the same
connect meter set to 10 amp scale right across the two panel wires so panel delivers its short circuit current to the meter
what do you read ? ____________amp
how do you estimate sun ? Strong, weak, or middlin' ?
If sun is strong and you don't get most of an amp, the panel is not delivering current

If the panel delivers current

disconnect battery from controller
hook panel black wire to battery's nehative
put ammeter red wire to panel's red wire
put ammeter black wire to battery's positive
does battery accept current from panel ? _____________ amp

If battery is accepting current let's see if it charges up
With your other meter read battery voltage _____________ V
Wait an hour or two
read battery voltage again ____________________ V
Is voltage increasing ? That should narrow down the trouble

 

[Wolst73]

Couple clarifications needed.

connect meter set to 10 amp scale right across the two panel wires so panel delivers its short circuit current to the meter

I will have the red and black wires from panel and red and black from meter. Connect red to red and black to black?

If battery is accepting current let's see if it charges up
With your other meter read battery voltage _____________ V
Wait an hour or two
read battery voltage again ____________________ V
Is voltage increasing ?

Is this with panel hooked directly to battery or connected through charge controller?

Btw, ammeter says "at 10A range: for continuous measurement </= 10 seconds at intervals not less than 15 minutes." So I can only test for 10 seconds every 15 minutes even if I'm only measuring about 1amp (hopefully)??

 

[jim hardy]

Couple clarifications needed.

Thanks - it is so difficult to communicate in words, isn't it "

..................
connect meter set to 10 amp scale right across the two panel wires so panel delivers its short circuit current to the meter

I will have the red and black wires from panel and red and black from meter. Connect red to red and black to black?
Yes. But if you get them backward you'll just get a negative reading
.........
If battery is accepting current let's see if it charges up
With your other meter read battery voltage _____________ V
Wait an hour or two
read battery voltage again ____________________ V
Is voltage increasing ?
Is this with panel hooked directly to battery or connected through charge controller?
Panel to battery , no controller
next test will be to see if controller let's current through to the battery

..........

Btw, ammeter says "at 10A range: for continuous measurement </= 10 seconds at intervals not less than 15 minutes." So I can only test for 10 seconds every 15 minutes even if I'm only measuring about 1amp (hopefully)??
wow i never saw that before
okay
1 amp will make 1/100th as much power in the internal shunt as will 10 amps
SWAG here, not Uswag ,
at 1/100th the power it takes 100X longer to deposit same heat in the internal shunt
100 X 10 seconds = 1000 seconds which is longer than 15 minutes but not by much
i would say that 1 amp is the limit for continuous measurement
does it have a 2 amp scale ? If so use that.

Would you post the model of your meter so we can lookup the manual ?

 

[sophiecentaur]

Btw, ammeter says "at 10A range: for continuous measurement </= 10 seconds at intervals not less than 15 minutes." So I can only test for 10 seconds every 15 minutes even if I'm only measuring about 1amp (hopefully)??

If this worries you, you can short across the Ammeter contacts with a wire and just remove the short when you want to measure. (Perhaps you can now see the advantage of a non contacting tong type Ammeter.)

 

[jim hardy]

If this worries you, you can short across the Ammeter contacts with a wire and just remove the short when you want to measure. (Perhaps you can now see the advantage of a non contacting tong type Ammeter.)

why didnt i thinkof that ?
Oh- it was 3am...or leave meter on volts and make that shunt.

If i knew of an affordable Hall meter that goes down to 1 amp i'd own several.

 

[Wolst73]

https://www.amazon.com/dp/B00WR42XP2/?tag=pfamazon01-20

 

[sophiecentaur]

https://www.amazon.com/dp/B00WR42XP2/?tag=pfamazon01-20

That isn't a Clamp type meter.http://www.ebay.co.uk/itm/Digital-LCD-Clamp-Multimeter-AC-DC-Current-Voltmeter-Ammeter-Checker-Tester-UK-/152066355863?hash=item2367dc7297:g:mzgAAOSwjKpXHKn6 I think I paid a bit more, though.

 

[Wolst73]

That isn't a Clamp type meter.http://www.ebay.co.uk/itm/Digital-LCD-Clamp-Multimeter-AC-DC-Current-Voltmeter-Ammeter-Checker-Tester-UK-/152066355863?hash=item2367dc7297:g:mzgAAOSwjKpXHKn6 I think I paid a bit more, though.

Oops. Once again proving my ignorance, I didn't know that clamp meter was what you were talking about when you mentioned hall meter...I will know better for the next one!

 

[sophiecentaur]

Oops. Once again proving my ignorance, I didn't know that clamp meter was what you were talking about when you mentioned hall meter...I will know better for the next one!

They used to make AC meters, years ago but now, with the Hall devices, you can measure down to 10mA or less on a 400A DC meter. A brilliant device and not too pricy.
Edit: Treat yourself to one Jim!

 

[jim hardy]

They used to make AC meters, years ago but now, with the Hall devices, you can measure down to 10mA or less on a 400A DC meter. A brilliant device and not too pricy.
Edit: Treat yourself to one Jim!

10 ma ? I didnt know they had that kind of resolution

this thread did inspire me to find my old Fluke DC clamp-around , it was about three layers down in my sedimentary workshop
it makes 0 to ±2 volts at banana jacks on rear for 0 to ±20 amps through the window so you use it with any old meter, i like that because i can use an analog one
ten turns makes it ±2 amp range
it's good from dc to ~2khz
it's the top one in this photo

Fluke was a junkshop find i'd guess it dates from 1970's
bottom one is probably from 1930's, was a recent yard sale treasure
it's extremely handy around the car no batteries or wires.

i'll take a look at your link , a DC clamp-around with that sensitivity used to be just a dream.

BTW here's a related instrument
Gaussmeter is much newer but the similarity is obvious

obviously the Gaussmeter needs a zero adjustment

Oh, the curse of curiosity ! But it sure beats boredom, eh ?

old jim

 

[Wolst73]

with everything connected , panel to controller and battery hooked up

At top between - terminal and + terminal (where i asked about 19 V,) what voltage do you see? About 12.1 ? ________12.25________ V
At bottom between - terminal and exposed junction of diode&red wire, what voltage do you see? About 11.8 ? _______11.93__________ VNext, with everything still hooked up,

What voltage do you see between - terminal and bare end of black wire ? Zero ? ______0_________ V...........................Now disconnect panel from controller by lifting the red and black panel wires from the controller
set your meter for 10 amp scale
move the meter plug to correct receptacle as she described at 1:08
post a picture of your meter if it's not clear which one to use , they're not all the same
connect meter set to 10 amp scale right across the two panel wires so panel delivers its short circuit current to the meter
what do you read ? ______0.12______amp
how do you estimate sun ? STRONG (in and out of cloud but measuring when out)
If sun is strong and you don't get most of an amp, the panel is not delivering current

If the panel delivers current

disconnect battery from controller
hook panel black wire to battery's nehative
put ammeter red wire to panel's red wire
put ammeter black wire to battery's positive
does battery accept current from panel ? ______0.07_______ amp

Appears panel is issue?

Excellent instructions Jim. Thank you very much for all your time.

 

[jim hardy]

does battery accept current from panel ? ______0.07_______ amp

Appears panel is issue?

I think so.
If you want to do a final sanity check
connect ammeter on 10A scale right across those two screws on back of panel
be sure you're on the pane side of that diode not the red wire side
you should see real close to your same 0.12 amp

that's more to build your familiarity with how to test panels than anything else. I think you've circled the problem pretty tight.

Excellent instructions Jim. Thank you very much for all your time.

Nice job on the measurements, and the presentation of same.
and
Thank you for the kind words !

Whats important is to build skills.

Got another minute ?

One point should be mentioned here, for safety's sake
because we can mis-apply things we've learned
and there's a point about current measurements we didn't emphasize

In all this testing we did, there is only one device that can deliver very much current and it's your battery..
We used the ammeter with impunity on your solar panel because we knew the panel inherently limits current to around an amp, and you found more like 1/8 amp.
Notice we never attempted to measure how much current the battery could deliver by using the ammeter.
That's because the ammeter is a ten amp gizmo and Mr Battery can probably deliver over a hundred amps to it. And Mr Battery would feel no guilt about that.
That meter says it's unfused on ten amp scale , meaning it cannot protect itself against overcurrent.
While Mr Battery would probably only let the smoke out of Mr Meter's internal shunt and maybe set fire to his plastic case, we don't want that to happen.

But it would demonstrate a fundamental fact about ammeters :
FACT: AN AMMETER CANNOT LIMIT CURRENT
COROLLARY: NEVER CONNECT AN AMMETER WHERE THERE"S NOT SOMETHING ELSE IN THE CIRCUIT THAT LIMITS CURRENT, and a breaker or fuse does NOT qualify.

If that ammeter were to be connected across incoming power at your house panel , a fireball big as a garbage can would result likely engulfng and severely burning you..
It'd make an impressive shower of sparks at any regular household outlet.I belabor this because i have encouraged you to use an ammeter, without giving you that requisite safety briefing.
So don't use that thing on amps scale around housepower. We'd all feel awful if you had to get skin grafts.
Sounds melodramatic but it's a real danger .

thanks for tolerating me

and thanks also for how much i learned about solar panels and controllers from your adventure !
grumpy old jim

 

[Wolst73]

connect ammeter on 10A scale right across those two screws on back of panel
be sure you're on the pane side of that diode not the red wire side
you should see

Measured across the 2 screws. Got up to 0.35A in full sun but dropped rapidly to 0.12 as sun went behind cloud.

Can someone recommend a good panel that will fit my needs so I don't buy another P.O.S.? You get what you pay for. That one cost me about $50.

Hopefully battery isn't toast too!

Thanks for the caution. I am a long way from doing anything with electricity. I don't like getting zapped.

 

[Tom.G]

re: post #88
@Wolst73 Excellent data, but a few questions remain regarding the 0.12A and the 0.35A you measured from the panel.

In several ways, the numbers don't seem to compute.
For instance you have a 19V panel that has an open-circuit voltage of 12V. However your post #72 states 19V output.

Voltage of panel showing about 19v.

The short-circuit current is rated 1.23A but reads or 0.12A (or 0.35A).
Based on your present numbers, the internal resistance of the panel is 99 (or 35) Ohms, making it a 1W to 3W panel.

1) What was the panel orientation relative to the Sun when taking the readings? (directly facing the Sun? flat on the ground? facing due South? Other?)
2) Did your shadow, or another shadow, fall on the panel while taking readings? Even on a corner of it?
3) Time of day the readings were made?
4) Anything else that may or may not describe the situation?

Part (much?) of the descrepencies may be attributable to Latitude, time of day, and partially cloudy weather.

I hope @jim hardy can come up with some more queries for all this. He is much better at digging up the patience for detailed instructions than I am!

 

[Wolst73]

1) What was the panel orientation relative to the Sun when taking the readings? (directly facing the Sun? flat on the ground? facing due South? Other?)

1. Tried to aim the panel directly at sun to get the initial 0.12a reading. The 0.35a reading had the panel laid a bit flatter to the ground but high enough to get to the terminals.

2. No shadows other than clouds going by.
3. Readings were about 1:30pm.

Any other reasons why the output would be so low. Panel is slightly dusty??

 

[jim hardy]

@Wolst73 Excellent data, but a few questions remain regarding the 0.12A and the 0.35A you measured from the panel.

The difference is curious. I attributed it to variation in sunlight.

Fact that neither short circuit approached what the panel should do, 1.2 amp, shouts "panel"
and his open circuit of 19 not 22.4 shouts "panel"
and that panel terminal voltage drops from 19V at zero current to around 12V at just 0.07A shouts "panel" , that's not even a watt !

so i yielded to the old practical adage
"There comes a time in every project where you have to shoot the engineer and get on with it"
time for me to step aside.

I think we all 'synergized' as a group and did pretty good troubleshooting. We have a lot of evidence pointing to "panel" .

All i did was maybe help a little bit with communication... that was a big part of my job my last twenty years.

old jim

 

[Wolst73]

I will repeat tests next weekend. Hopefully be more of a normal sunny day then.

P.S. Jim, I ordered a 3 amp dc panel meter as per your suggestion. Went a bit bigger just in case I need the extra amp down the road. Toys! (wife is starting to question my sanity).

 

[Tom.G]

Maybe I missed it, but what is the make and model of the solar panel? Is there a link where the datasheet can be found?

I strongly suspect the panel is under-powered for your use, by a factor of two to six. Could be a defective panel or just the wrong model (too small.) Since there are too many variables that we can't control for the measurements, all we can do is attempt re-engineering the system from the data sheets. No guarantees though! Also realize that we are working at the best time of year for solar power. Winter will double or triple the needed panel size for the same lights-on time.

Other approaches I can think of is get a duplicate panel and compare those readings with the existing panel. If the new panel is substantially better, return the existing one as defective. If the readings are similiar, connect the two in parallel, Red to Red wire, and Black to Black wire. This will double the solar capacity; although that still may not be enough. Alternatively, find a local Solar Power company and see what recommendations they have. (I'm pretty sure they will want to sell you SOMETHING.)

Also, using the site that @jim hardy referenced (http://www.pveducation.org/pvcdrom/properties-of-sunlight/arbitrary-orientation-and-tilt ) the optimum tilt angle for best year-round performance is 72 degrees. i.e. the panel should tilt 72 degrees from horizontal, or facing 18 degrees upward from standing vertical. This will give you approximately equal power in the summer and in the winter. This is based on your Latitude of 48 degrees.

Sorry about all the 'could be' and 'maybe', it's been forty years since I've done any of this Solar stuff.

 

[Wolst73]

Maybe I missed it, but what is the make and model of the solar panel? Is there a link where the datasheet can be found?

Recap: 30A solar charge controller (also have a 10A), 12v 18ah battery (bought new but might be a problem too), 6 3W LED lights hopefully powered for 3 hours per night.

 

[OmCheeto]

I still don't see where we've done measurements for the following configuration:

And with the new ammeter, we could figure out the efficiency of the Luxdrive.

 

[Tom.G]

Well, I did some research and 'back of envelope' numbers with these results.
Panel Azimuth: 180 Degrees
Panel Tilt: 72 Degrees
Nominal panel Current: 1A

The following is for Winter configuration with roughly 6Hrs of sun and clear cloudless sky.
Corrections for:
Low angle during Winter: 0.67 (same value for Summer Latitude mismatch)
Azimuth correction during the day: 0.68
Battery charge Efficiency: 0.8

1A x 6Hrs x 0.67 x 0.68 x 0.8 = 2.18AH per day in winter
If the controller adjusts/tracks the maximum panel power output at around 18V:
2.18 x 18 = 39WH per day in winter.
So you will need two panels in the winter if there is full Sun every day.

For cloudy days corrections are:
High Clouds: 0.83
Med. Cloud coverage: 0.3
"Really, Really Overcast": 0.15 (rain?) 0.15 x 39 = 5.9WH per day

Bottom line is if you want to carry thru from two days of rain you will need at least ten times your current panel area.

For those of you that are willing to check these calcs, here are the references:
http://www.ftexploring.com/solar-energy/clouds-and-pollution.htm
http://www.ftexploring.com/solar-energy/clouds-and-pollution.htm
http://www.pveducation.org/pvcdrom/properties-of-sunlight/arbitrary-orientation-and-tilt

 

[sophiecentaur]

10 ma ? I didnt know they had that kind of resolution

LSD, iirc.
Quite good enough to see if you're doing worse or better.

This will give you approximately equal power in the summer and in the winter. This is based on your Latitude of 48 degrees.

If we are doing garden lighting the demand will (could) be greater (longer) in winter so it may be better to tilt the panel down a bit. (A real world consideration). Adjustment during the early cycle could give the best of both worlds.

 

[jim hardy]

Wow @TomG nice job with those calcs !

it's been forty years since I've done any of this Solar stuff.

you're one up on me - i never have...

Recap: 30A solar charge controller (also have a 10A),

I am curious about "efficiency"
a controller rated 30 amp 12 volt is rated to handle 360 watts
if it's spec'd 99% efficient it is entitled to consume 3.6 watts irrespective of actual load
which does not sound unreasonable , it takes considerable power to charge those gates on big power mosfets
so its idle current might well be ~0.3 amp
and who'd complain about 99% efficiency ?
The question to ask is "99% of what?"

that'd be worth measuring per Om's suggestion
when you get that ammeter and a better panel , try that ten amp controller too

can you return your panel on a warranty claim ?
Sounds to me like it has an internal short that's stealing power , except at zero terminal volts
that would explain the difference in Isc on opposite sides of the blocking diode
upstream of it you got 0.35 amp, that'd be zero across the panel terminals (well maybe a few millivolts across ammeter )
downstream of it you got 0.12 amp , that'd be about ¼ volt across panel
as voltage across panel increases the short steals more and more of your 1.2 amps

Try modelling it as a Norton equivalent: 1 amp source paralleled by 15 ohms ? Where's my slide rule...

 

[Tom.G]

If we are doing garden lighting the demand will (could) be greater (longer) in winter so it may be better to tilt the panel down a bit. (A real world consideration). Adjustment during the early cycle could give the best of both worlds.

That's sort of what I did. With the high Latitude of the installation (48 Degrees), I first optimized tilt for Noon on the Winter Solstice, Dec. 22, then checked the Summer Solstice results. They turned out within a couple percent of each other. At that point I said "the heck with it, optimize for worst case!"

Fudging the tilt in general depends on intended usage of the power. In this case there is no use for any excess power in the summer. If on the other hand this was for domestic usage as a supplement for mains power, then the tilt would be optimized to maximize total annual power. The total-power optimum tilt is accomplished by aiming the panel slightly down from the Latitude of the installation.

 

[sophiecentaur]

If on the other hand this was for domestic usage as a supplement for mains power, then the tilt would be optimized to maximize total annual power.

Absolutely. I'm surprised that people with grown up mains connected systems don't use some manual adjustment to improve their returns from the Feed in Tariff over the year. I'm sure you could get a result for the man hours put in.

 

[vadslram]

Why do you say no? The picture says CYCLE on the side. It is a starting battery.

It doesn't mean motor CYCLE. That is a battery that is used for UPSs and emergency lighting. The cycle count is the number of discharge/ recharge cycles that the battery can do and still keep the rated AH.

 

[anorlunda]

It doesn't mean motor CYCLE. That is a battery that is used for UPSs and emergency lighting. The cycle count is the number of discharge/ recharge cycles that the battery can do and still keep the rated AH.

I stand corrected.

 

[Tom.G]

The cycle count is the number of discharge/ recharge cycles...

Although a useful number, I don't recall anyone mentioning 'cycle count' in this thread. Can you point me to where it was brought up?
Thanks.