Troubleshooting solar setup... or troubleshooting my troubleshooting!!

[thomm225]

Originally Posted by Paul Elliott

That panel will still be putting out 2A even if the battery somehow gets to 20V. Here's that link to the panel specs again (it works for me): https://www.renogy.com/template/file...ifications.pdf

To be brutally honest, I don't trust the OP's original measurements well enough to base any conclusions on them. It appeared that he was trying to test too many things at the same time. I've learned over my many years to simplify as much as possible when troubleshooting. Doing otherwise leads to chaos and confusion.

I'll check the link. Thanks.

Also thanks for the info on the Victron 75/15. I just bought one because I got too into this thread plus I wanted to experiment with an MPPT Controller. Now I've ordered the cable so I can connect to it with my PC. (I'm used to a display and buttons on the front)

I also agree with the brutally honest assessment, but Mr. OP don't feel bad. I have guys with degrees in electronics and computer systems that get confused also when troubleshooting and/or make simple mistakes. (and I've blown up lots of stuff as well or mistroubleshot things)

A couple months back one young tech submitted orders for 2 expensive obsolete motherboards for our old Aechelon Display System Computers. One of our older techs got bored and happened to check it after I told them the only boards available were in China, were used and the price was like $1,000.00 each and we'd have to use our CPU's. The problem was a plug on the motherboard was installed off center by one pin

Op just needs to do simple testing. I'd say forget the precision and just go for good / bad. One or zero. On or off. (but to each his own) Most times a problem found will fix any partial problem as far as performance. It's also not inconceivable there are two problems. One the controller (or a connection) and two panel # 2

I am used to my system so i know about how long it takes to charge from a certain point, but my system is quite simple. Two panels ( a 20 and a 65 watt) in parallel through a PWM controller. I've had the 65 watt panel on there for maybe 4-5 years

[Paul Elliott]

Originally Posted by thomm225

I also agree with the brutally honest assessment, but Mr. OP don't feel bad. I have guys with degrees in electronics and computer systems that get confused also when troubleshooting and/or make simple mistakes.
[...]
Op just needs to do simple testing. I'd say forget the precision and just go for good / bad. One or zero. On or off. (but to each his own) Most times a problem found will fix any partial problem as far as performance. It's also not inconceivable there are two problems. One the controller (or a connection) and two panel # 2

And I meant no disrespect to the OP with my comments. When troubleshooting my own stuff I occasionally get so confused that I end up throwing away my initial tests and starting over from ground zero. I don't take it personally!

An even older engineer I used to work with had this philosophy about precision:

Measure it with a micrometer, mark it with chalk, and cut it with an axe.

Most of the time that's pretty good advice!

[basssears]

No offense taken... having done two successful solar setups I thought I was further up the learning curve than this troubleshooting scene has made me realize.


I think the other thing to clarify is, while I'm interested in "good panel" vs "bad panel" I also want to get some decent output out the panels as space is limited and I don't want some huge panel (i.e. two 50 watts in series) that puts out 20 watts.


So in the "does it work or doesn't it" realm I believe I have 3 working panels (I still have another round of testing to do but that won't be for a while) and 1 not working, but am eventually hoping to verify just how well the 3 working ones are working.

[thomm225]

Originally Posted by basssears

No offense taken... having done two successful solar setups I thought I was further up the learning curve than this troubleshooting scene has made me realize.


I think the other thing to clarify is, while I'm interested in "good panel" vs "bad panel" I also want to get some decent output out the panels as space is limited and I don't want some huge panel (i.e. two 50 watts in series) that puts out 20 watts.


So in the "does it work or doesn't it" realm I believe I have 3 working panels (I still have another round of testing to do but that won't be for a while) and 1 not working, but am eventually hoping to verify just how well the 3 working ones are working.

Sounds like you have a plan

My point with the good vs bad (1 or 0) was that most times if you find a problem, any problem, that will usually fix your partial problem as in a 20 watt output vs 50 watts. This is why I keep saying keep it simple................this is knowledge gained from many years of troubleshooting

A connection with half or more of the wire accidentally cut off for example while removing the insulation then crimping can cause this. Continuity check will still be good but current will be affected which in turn will decrease wattage and make for a longer charging time. The same goes for a corroded connection and maybe even a loose connection (like a loose screw in the J Box)

As far as the 3 panels, I'd point at the controller before the panels. That's one check. R/R controller and test not extensive rechecking of panels, and current, and voltage, and sun position

[Paul Elliott]

Originally Posted by basssears

I think the other thing to clarify is, while I'm interested in "good panel" vs "bad panel" I also want to get some decent output out the panels as space is limited and I don't want some huge panel (i.e. two 50 watts in series) that puts out 20 watts.

I think that would qualify as a "bad panel".

Our point on the precision issue is that at this point you should be trying to fix the big, obvious problems (that's one way of reading the "ax" metaphor). Only once that's done can you carefully tunnel down into optimization. I suspect that once you find the big problems you will have few if any little ones to worry about. With good panels, reasonable lack of shading, a good controller, good wiring, good connectors, and a good battery, it's going to work as well as it should.

Then you get to work on your power budget. Of course, that might be the first order of business! Still, we usually solve these things by working from both ends of the problem and meeting in the middle.

[Paul Elliott]

basssears, we would love to hear about any progress / clues. In spite of our technical bickering, we would all really like to see you get this puzzle figured out.

[Stu Jackson]

Originally Posted by thomm225

The panels look fine. When I troubleshoot the panels if I have good VOC I go forward toward the batteries.

>>>>>>>>>>>>>>>>>>>>>>>

On my PWM Controllers, the voltage on the panel side equals the voltage on the battery side until the battery gets to the float voltage then the voltage on the panel side will start to increase as the controller decreases current flow to the battery as it approaches fully charged>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>>>>>>>>>>

Perhaps I am missing something important here, but doesn't current decrease because of battery acceptance rather than the controller "decreasing current flow?"

[john61ct]

Yes.

The SC may have the ability to set a max ceiling on amps output. Smart VRs on alternators may have dynamic current regulation to protect the alt from overheating.

To some degree same is true for quality mains chargers.


But in normal practice all charge sources is "striving" to make maximum current available from the panels' output.

(MPPT doing a better job than PWM in that regard, but let's not derail)

The "trailing amps" phenomenon is controlled by the chemistry / physics of the battery, as SoC climbs so does resistance, to increase current acceptance only option is to increase voltage.

But once the max V setpoint is reached, aka Bulk/CC - Absorb/CV transition - also controlled by the batt - then V is held there so A declines.

At this point the batts may be 85-90% Full, but still 5+ hours away from 100%.

LFP does not have this problem, will accept high current all the way, if enough is available can refill in well under one hour.

Bottom line is even the smartest charge regulator is really only controlling Voltage.

[curtis742]

It appears only one panel failed in the last test. I was wondering how you arrive at the watt output . Using your readings and the formula to figure watts three of your panels are at or above e 50watts. The formula is P=I x E where I is current, E is voltage, and P is the Watts. If the readings are jumping around this indicates a bad connection somewhere. On the bad panel open the box on the back and check for bad connections and the blocking diode. Sometimes these diodes short or go open circuit.

[thomm225]

Originally Posted by Stu Jackson

Perhaps I am missing something important here, but doesn't current decrease because of battery acceptance rather than the controller "decreasing current flow?"

It may depend but on mine with my PWM Controller, it stops the battery voltage charge at the Float Voltage I set it at.

Today that is 13.7, but on Thursday I had it at 14 Volts for the weekend sail. (I run fans and depth all night so I let it get up a bit higher)

After it gets to float the battery side of the controller stays at 13.7 and the panel side slowly goes up to near VOC or about 21 volts Sun depending....which sort of proves your/our point (see below). Current is decreasing from the panels allowing that voltage to rise on the panel side of the controller I'll have to check this on my MPPT after I get it installed to see if it is similar

Btw, I used your argument on a direct charge no controller argument (see above some place) but another poster checked the charge curve and said the solar current wouldn't get high enough for the battery's internal resistance to decrease the current flow as it neared full charge. I took him at his word and didn't research further.

I have a Victron MPPT 75/15 now so am looking forward to experimenting with it

[curtis742]

Short circuit current is supposed to be 2.84a:
Taking into consideration the temperature coefficient your readings would be very close to specs. Temperature of the cells will effect output. As the temperature increases above 70deg the output decreases thus the temperature coefficient.

[noelex 77]

Originally Posted by Stu Jackson

Perhaps I am missing something important here, but doesn't current decrease because of battery acceptance rather than the controller "decreasing current flow?"

This seems to be a common theoretical point of argument on internet forums. I am not sure why. The controller/regulator controls or regulates the power to the batteries. If the batteries could control themselves there would be no need for a controller/regulator.

However, the controller/regulator only decides the voltage. The battery decides the current it is going to accept at the voltage the controller/regulator is set at.

So by altering the settings on the controller/regulator and leaving the battery the same, both the voltage and current entering the battery can be changed. Alter the battery and leave the settings on the controller/regulator the same and the voltage will remain the same but the current will alter.

I would therefore suggest that while both the controller/regulator and the battery both have an influence, it is the controller/regulator has primary control of the charging process.

A simple example is when the controller/regulator drops from absorption to float. The current flowing into the battery will decrease dramatically. The controller/regulator has made this change. The battery is exactly the same.

[skenn_ie]

Be aware that a panel will output the current of the weakest cell. If one cell is damaged, or shaded so that only 50% of it is functional, the whole panel will only produce 50%. Do your panels have their own blocking diode ? I favour connecting each of the panels the their own blade-type fuse, so you can easily plug/unplug each panel. Fuses should be rated way above what you expect them to see. There is also a spray wax that is used on boats to protect electrics. I highly recommend it wherever you have outdoor electrics. Something else that I have seen is corroded cells if the seal wasn't perfect. The cells looked fine from the front, except for a little corrosion on the link strip. Have a close look for any evidence of that.

[noelex 77]

Originally Posted by curtis742

It appears only one panel failed in the last test. I was wondering how you arrive at the watt output . Using your readings and the formula to figure watts three of your panels are at or above e 50watts. The formula is P=I x E where I is current, E is voltage, and P is the Watts. If the readings are jumping around this indicates a bad connection somewhere. On the bad panel open the box on the back and check for bad connections and the blocking diode. Sometimes these diodes short or go open circuit.

The Isc test and the Voc test are very useful, but you cannot multiply these two numbers together to determine wattage.

At Isc the voltage is zero. Therefore the watts produced are zero.
At Voc the current is zero. Therefore the watts produced are zero.

[skenn_ie]

Originally Posted by noelex 77

The Isc test and the Voc test are very useful, but you cannot multiply these two numbers together to determine wattage.

At Isc the voltage is zero. Therefore the watts produced are zero.
At Voc the current is zero. Therefore the watts produced are zero.

Correction, at Isc, all the power is dissipated within the panel.
Also, instantaneous power = instantaneous V x instantaneous I. MPPT controllers aren't steady state devices, so the voltage produced by the panel fluctuates, as does the current. The better ones are Buck regulators, the cheaper ones, PWM. Buck regulators minimise ripple current and ripple voltage, on both sides of the converter. PWM are 100% on/off switches, where ripple current is totally dependent on a capacitance on the input side, and the load on the o/p side.
The best test, by far, is to connect a resistive load directly to the panel. The load resistance should be somewhere in the region of 50% the rated panel MPPT voltage divided b the max current (R=V/I) , and power rated at the max panel power. That is likely to be a quite big resistor. I often use a bar from a radiant heater, with one connection moved in to get the required resistance. Alternatively, a brass rod with screw clamps to connect the wires. Don't tr to hold the rod, or put it on anything flamable when connected ! Wood is actually OK, as it will char rather than burn.