Low Vmp, problem with panel or controller?

[Mike90250]

Alright, so I went out and tested the current and voltage of each string of 18 cells on the panel. They both produced 2.95 amps. One string was 10.9V and one was 10.7V. However I also got my FLIR out and took some thermographic images .......

hell with panel envy, I've got a case of FLIR Are their prices any closer to less than selling a kidney ?

[TechnicalLee]

hell with panel envy, I've got a case of FLIR Are their prices any closer to less than selling a kidney ?

https://www.amazon.com/FLIR-ONE-Ther...dp/B00W5PRY52/

Price is slightly less than one kidney now.

[Sunking]

Well let's clear up one fact first. Your Controller is a BUCK Converter. Both can easily be done but requires more parts and complexity which translates to $$$ a lot more $$$. It would also defeat the purpose. The whole point of using MPPT controllers is to use much less expensive higher voltage Grid Tied panels instead of antiquated expensive low voltage battery panels and all the losses associated with them. Now having said that Genasun is one of the very few who make a Boost Converter but it will not Buck either for the same reasons above. They make that controller specifically for 36 and 48 volt golf carts using a 100 to 350 watt panel. I know this for fact because that is what they specify and advertise.

Is it possible you have a panel with a shorted cell lowering Vmp voltage? Absolutely Yes. Two panels not likely, but if you have a shorted panel in parallel with a good panel, you have two shorted panels. Think about that for a moment while you read.

But extremely easy to determine especially if you have two panels. The fact is you are misusing your panels, and have them wired in parallel silly, stop that. . You just threw away most of the gains of a MPPT controller. Anyway if you had the panels wired in series, and see the input voltage to the controller 1 or 2 volts higher than the battery voltage, you know for certain you are in PWM mode or have a shorted controller.

Sounds like you have some electrical knowledge? If so then I hope you understand a PWM controller is nothing more than an electronically controlled On/Off switch. When ON or Closed Switch connects the panels directly to the batteries which means Panel Voltage = Battery Voltage, and Input Current = Output Current.

The characteristics of a PWM Controller voltage input is you will see a voltage as low as the battery when at 100% duty cycle and Voc when at 0% duty cyle or turned off. All voltages are valid from battery to Voc. No if and or butts. That is the Gospel.

MPPT input voltage is Vmp at MPPT and up to Voc. Voltage will never go below Vmp if your tracker is operating correctly. Only time you will see voltage go below Vmp is when the Controller is in PWM Mode. No if, ands or butts.

So based on those two known conditions the only thing you can conclude with voltage is when the Controller is in PWM mode and operating near 100% Duty Cycle. That condition exist only when the Input Voltage falls below Vmp because Vmp cannot go lower in MPPT Mode.

Only way to tell if you are in MPPTT mode takes both a current and voltage measurement of input and output. Voltage on the input will be at or near Vmp, and Input Current will be lower than Output Current. If those two conditions are not TRUE, you are not in MPPT mode.

OK back to you with two panels in parallel. If you have a panel with a shorted cell, means the Vmp and Voc will be lower than normal. Thus pulling down the good panel in parallel Vmp and Voc to match the lower panels voltages.

Moro to the story, wire the panels in series like they should be. If you see anything less than 35 volts at the input you know you are in PWM mode or a defective controller. FWIW a Voc test on each panel separately will tell the tale of the panels.

Have a nice day.

[TechnicalLee]

I have a single panel (perhaps better called a module) connected directly to the controller using the MC4 connectors. So I don't think I have it wired wrong considering that I only have one panel... The panel/module, like most, has multiple strings of cells wired in series (two strings of 18 cells that each produce about 10ish volts open circuit in this case) with bypass diodes in parallel to each string. The open circuit voltage of the entire module is 21ish volts. The tests I did today involved measuring the voltage and current of the two individual strings of cells by making my connections in the little weatherproof box on the back so maybe that's the source of confusion? The open circuit voltage and short circuit current of each string check out (as does the entire panel of course).

Yes, I do understand PWM.

I ordered a 100W panel that has an output guarantee, so hopefully that will work better. I'll post an update on how things are working when I get it.

[Sunking]

I have a single panel (perhaps better called a module) connected directly to the controller using the MC4 connectors. So I don't think I have it wired wrong considering that I only have one panel... The panel/module, like most, has multiple strings of cells wired in series (two strings of 18 cells that each produce about 10ish volts open circuit in this case) with bypass diodes in parallel to each string. The open circuit voltage of the entire module is 21ish volts. The tests I did today involved measuring the voltage and current of the two individual strings of cells by making my connections in the little weatherproof box on the back so maybe that's the source of confusion? The open circuit voltage and short circuit current of each string check out (as does the entire panel of course).

OK my bad, I misunderstood and thought you had two panels. My appology. Your test results indicate your panel is OK

Yes, I do understand PWM.

OK then apply that knowledge and match it up to how a battery charges. There are 3 off the top of my head possibilities.

1. Your battery has seen it better days or is with in 80% SOC voltage. For your controller to go into MPPT mode two conditions must be met. One the internal timer has timed out meaning it is the next day at sun up. Two battery SOC is less than 80% Float SOC.

2. Defective controller

3. Defective panel.

Once your controller has cycled through Bulk, will not go back into Bulk until the next day, and voltage is 80% Float. You can power reset the controler and discharge the battery to test. Start eliminating the possibilities. That will lead you to the problem.

For the future build you an Arduino Dual Input Current Meter. One shunt on the input, and one on the output. Write a little Sketch that says if:

Input Current = Output, print PWM
else Input Current < Output Current print MPPT

Or use the Protein Computer.

[SunEagle]

hell with panel envy, I've got a case of FLIR Are their prices any closer to less than selling a kidney ?

The models that connect to a smart phone are only a couple hundred bucks and are pretty accurate. The one shown by Lee on Amazon is pretty good and a whole lot less then their stand alone IR camera.

[PNjunction]

Before we go any further, we need more facts, and I'm thinking the op is looking for problems in the CC where there isn't any.

The GV4 controller has a fixed CV-Absorb at 14.2v. Time limited to 2 hours. 13.8v float. For daily cycling with a large agm and your existing panel, this is insufficient and will result in a sulfated battery pretty quickly. If you are aren't cycling daily, then we might be able to overlook this relatively low cv, and short timed absorb. If running daily with a large batt, the 13.8v float won't do anything substantial when the sun goes down shortly thereafter.

The op is running an agm. What are the particulars? Is it new or used? What is the C/20 ah rating? Exactly what manufacturer and model is it? Is it a single unit, or paralleled?

How is this agm being used? Daily cycling? Weekend warrior?

Are you upgrading to a new controller, now that you have the 100w panel, or are you running the GV4 at the max?

Thing is, we can be sidelined by a mere technical exercise, when the actual application is destined to fail despite the quality of the CC, if the variables above aren't known - especially for an agm!

So TechnicalLee, let us know these things to see if we are concentrating on the REAL issue.