Daily charging parameters

WaterWheel · July 2016

In my case with the new Conext 80-600 I didn't check SGs for about 2 weeks, my mistake, life has been very busy; instead I simply checked the battery monitor.     The controller came with a factory set 3 hr absorb but after the 1st 30-45 minutes in absorb the batteries only required about 5-9 amps to maintain the 58.1v (before temperature compensation) Crown suggested.      Since the batteries usually start the day (this time of year) at about 70% charge (according to battery monitor) (which SOC is obviously off with according to SG) with the charger only requiring about 8 amps to maintain the 58.1v (actually around 57.2v with temp compensation) the 3 hr factory set absorb wasn't enough.

I checked SGs yesterday after the controller came out of absorb and the SGs were about 1.220 across all cells; so after reading this thread today I increased the absorb to 7 hrs for today only.

This evening the cells were about 1.263 after the 7 hr absorb with resting voltage this evening (3 hr rest) at 58.4v.      I'm going to give it a few more days with a 5 hr absorb and see if the 5 hr absorb is where I need to be.

I'm learning why solar guys suggest an inexpensive "learning set" of batteries to start.

I realize I don't need to go to 1.277 SG or even 1.270 every sunny day but I've had tons of sunshine the last few weeks; bit of a drought around here this summer.     I suspect that the shorter absorb was making my batteries lose a bit each day.

Dave Angelini · July 2016

That sounds reasonable if that is the voltage crown wants. Almost all of the 90+ offgrid systems I did are loading their batteries at night.
I think that is one of the reasons the Surrettes are harder to break-in for some people. Even the folks that are not deeply discharging will turn on an electric heater to cycle. Use them or lose them!
If at first sun you are down to 49V with a couple hundred watts of load cycling and it looks like a good day, move on to something else. Life is too short! I don't use starter batteries because I design for the person/load and educate them to not make the bonehead mistakes.
Do your eq's and keep the tops clean and the water right and you should get the better part of 10 years. I bet you like that charger!
I sure do. Been using dozens of them and that is why I was hoping you would get some feedback from Schneider on their demise.
Nice to have all the data from a matched system also.

herodotus · July 2016

The evidence so far for me is that the length of the absorb charge is more significant than the voltage change. I say this because after running a couple more cycles with my new voltage settings, I can see that the SGs still don't make it above 1.260 unless I really push out the absorb charge a good long way. I'm currently back to terminating on an end amps of 3.6, and that winds up being about a 5.5hr absorb if the batteries are properly discharged. By that point my batteries are over 1.260, which I'm happy with for a regular charge cycle. I can push them up to 1.270+ maybe once a month on an EQ charge. Plus I guess I may see the capacity deepening over the next few months as well as the batteries break in.

When I get my voltage reference I will make sure that the actual voltage at the batteries is what it's supposed to be. The calibration problem aside, everything suggests that Dave hit the nail on the head here: just chill out and follow the manufacturer's recommendations rather than making a huge fuss about it all!

herodotus · July 2016

FWIW, Some more info from Steve @ Rolls:

For Renewable Systems, typically a fully charged SG of 1.265 to .275 at least once every 2 weeks will keep them in good health... You don't always have to have them fully charge, but the more often you do the less often that you will have to an EQ charge. I would start with lower voltages on EQ at first, but that may go up as they age.
EQ charges in general are intentionally boiling your batteries to remove sulfate, this is not so good for the batteries. If you are properly charging/discharging them in the first place you should never have to EQ.
After an EQ, the cells should be higher, but more balanced in SGs... If SG's are generally low, adjusting your bulk/absorb voltages and time is what you should be focusing on, rather than punishing the batteries with EQ cycles.

Vic · July 2016

Hi Lucian,

Thanks for the recent updates.

Believe that you are using the Hydrovolt Hydrometer, which should be accurate.

You may have seen this article on the Surrette site:
http://support.rollsbattery.com/support/solutions/articles/4347-measuring-specific-gravity

Following this procedure will give good, repeatable results in taking SG readings. And, RINSE -RINSE - RINSE, and rinse again, the Hydrovolt. This is very important in maintaining its accuracy.

FWIW Vic

WaterWheel · July 2016

Dave, Yep, Crown wants 58.1v for the 395 amp 6v battery. I'll update in a few days as I learn more with the 5 hr absorb time.
Conext has never given me any information about the 2nd controller I sent back. I had asked to be kept informed but no response.

WaterWheel · July 2016

In my case it seems that the 2% factory set absorb cutout settings on the 80-600 controller are what is cutting out the absorb. After emailing Conext last night about the factory 2% absorb cutout settings I've learned that this setting can not be adjusted. They suggested on the 80-600 controller screen that I cut my battery bank size from 395 hr to a lower number forcing the 2% absorb cutout amperage lower. By chance last night I had set that number to 200 hr in an attempt to do the same thing so the absorb should cut out at about 4 amps current; I'll adjust from there using SGs.

They also suggested listing the battery bank as 0 amp hrs and control the absorb using the timer like I had attempted to do last week not realizing the 2% current absorb cutout was my problem.

I'm surprised that they don't allow you to set the absorb cut out current.

westbranch · July 2016

with that kind of non adjustable setting,

that unit should be known as a Dinosaur!
no matter how good the other features are

Dave Angelini · July 2016

westbranch said:

with that kind of non adjustable setting, that unit should be known as a Dinosaur!
no matter how good the other features are

It is adjustable by setting the battery bank to a lower number than actual. This is nothing new and has been in almost all of the XW manuals for over 10 years. It is true that it requires someone be able to read or have the time to read the manual.
When in doubt read the manual.

herodotus · July 2016

Hey all,

So I feel like I'm starting to get to grips with my batteries, but I'm still attempting to puzzle out what's going on with my power output. My IR thermometer should be turning up in the next few days, but I'm pretty skeptical that it's really going to reveal anything that interesting.

During MPPT Bulk charging in the middle of the day, my CC tends to settle on a voltage of about 70V, which is a long way below the ideal 93 Vmpp, even allowing for the effects of temperature (Vmpp is supposed to drop by 0.35% / degree C). The array produces a maximum of about 35A at this voltage.

During absorb charging, when the battery bank starts to limit the current, the array voltage climbs to a bit over 90V (Voc would be 111V in ideal conditions). For example, right now it's 93.3V @ 7.9A.

I have tried:

measuring the current output of individual strings (all within 0.5A of each other)
running each string individually (no substantial variance in output voltage - within a volt or so out of 70ish)
turning off each string individually (no substantial variance in output voltage)
popping the backs off the diode boxes and measuring the voltage across each panel (in line with the output of the string as a whole)
Checking the voltage output at the panels, before the main cable run (within 1.5-2V of the voltage at the CC)
Manually playing with the MPPT set point (the CC is settling on the best point as far as I can tell - no other voltage produces more power)

There are no glaring voltage drops anywhere, so I'm a bit stumped. The panels are outputting 7A/string rather than 8A - which is pretty much what I would expect given that angles are not perfect, sky is a bit hazy sometimes etc. But I just feel like we should be getting 80V rather than 70V even given the fact that it's fairly hot here. As far as I understand it, voltage output is not really that closely related to irradiance levels above a certain threshold - the irradiance has much more impact on the current than the voltage.

If it's not voltage loss across connectors/cabling, and it isn't a rogue panel, what else could it be? Is there some way the CC could be faulty? Given my test of manually setting the MPP, I find it hard to see how this could happen unless it was actually burning off power internally and then misreporting the current, which seems pretty unlikely.

I guess I'm not going to know for sure the impact of temperature until it cools off here in a few months time - perhaps the panels are performing much worse in the heat than the specs suggest.

????

BB. · July 2016

I am a little confused... What solar charge controller do you have? A Conext 80-600 is a high voltage array charge controller that requires Vmp-array of 235 volts minimum to "turn on":

http://www.solar-electric.com/xaxwmp80amp6.html

Device Name

	XW-MPPT80-600
Battery Voltage	24 Volts DC
Battery Voltage	48 Volts DC
Input Voltage Range	195VDC to 600VDC (open circuit)
Cable Entry	7/8" and 1" knock-outs
Wire Gauge Range	14 AWG to 6 AWG
Weight	30 lbs.
Enclosure Material	Sheet metal chassis
Ambient air temp.	-20° to 45° C
Ambient air temp.	-4° to 113° F
Operating altitude	0 to 2,000 meters
Operating altitude	0 to 6,561 feet
Standards	CSA 107.1
Standards	UL 1741
Certifications	FCC Class B
Dimensions	30" x 8.6" x 8.6"

5 year warranty

-Bill

westbranch · July 2016

Bill, from his sig line i looks like am FM80

BB. · July 2016

Yea--I am not sure because: "In my case with the new Conext 80-600..." in the first post.

By the way--Does the FM80 allow somebody to program the Vmp-array voltage desired (rather than using the FM80 to figure it out directly).

-Bill

westbranch · July 2016

It looks like there are 2 conversations going on Waterwheel (80-600) and Herodotus (FM80)

herodotus · July 2016

BB. said:

Yea--I am not sure because: "In my case with the new Conext 80-600..." in the first post.

That's not me! That's another guy called WaterWheel who's chimed in with some related issues - it's confused because this is the 3rd page of the thread and his is the top post on the page.

By the way--Does the FM80 allow somebody to program the Vmp-array voltage desired (rather than using the FM80 to figure it out directly).

-Bill

Yes, it does. But as I mentioned above, doing so doesn't improve my power output. I admit that I haven't cross-checked the current with my clamp meter when manually setting the Vmpp, so I'm relying on the FM80's own power output readings to make that assertion, but I have no reason to believe (from other cross-checking I've done) that the current readings are significantly out - certainly not enough to account for being several hundred watts shy of the power output I would expect.

BB. · July 2016

Sorry--I try to check every thread (as a moderator)--Sometimes my eyes just glaze over.

-Bill

Dave Angelini · July 2016

Mine sometimes unglaze Bill.
You can program the sweep start point but that would have to be seasonal and the user would need to really understand what they are doing. Much better to set for auto sweep.
I would think if the OP really believes he has a problem, he should short each panel and measure short circuit current.
He would not be the first one to believe he should have more power a month or so after the Solstice. It is pretty hot out there!

herodotus · July 2016

Dave Angelini said:
I would think if the OP really believes he has a problem, he should short each panel and measure short circuit current.
He would not be the first one to believe he should have more power a month or so after the Solstice. It is pretty hot out there!

What would I be looking for in shorting each panel? I'm aware that it's hot, and that this will reduce output. It's just that I've run the sums based on the manufacturer's own temperature coefficients, and the cells would have to be at something like 90C for temperature to explain the voltages that I'm seeing on its own... so I figure there must be some other factor involved.

Vic · July 2016

herodotus said:

vtmaps said:

herodotus said:
In practice for reasons that aren't entirely clear to me yet (still some poking around to do on this) we're actually struggling to pull more than 2.5k from the panels at the moment
You have 3750 watts of array. With a 0.77 derate you should expect to see 2887 watts.

Yeah. And the 0.77 derate is obviously just an average for use in specifying a system. At the moment I'm hunting around for explanations of where *specifically* I'm losing all of that extra right now. I can account for some of it with temperature (although lacking an IR thermometer I'm not sure exactly how much I should really expect to be losing there), panel tilt (I'm at 40 degrees and solar normal is about 13 here at this time of year) and a bit of cable loss (~1.5%), but unless my cell temperatures are ~75C (nominal is 47C at 20C ambient with a light breeze) there's something else going on... but that's another question that I'm working on separately!

Hi hero..,

What is the elevation angle of your PVs.

What are the mid-day temperatures at your location, on clear, sunny days?

Did some searching on the PVs that you are running, looking for negative comments. Saw a very long Thread on some whirlpool solar-related site, but not any real data there before the eye glaze took over.

Believe that you had a Clamp DC Ammeter or order, or available ?? Forget. This is a useful tool to look at your Isc readings for each string. Would not hurt to also look at Voc s of each string with each Combiner breaker open (OFF).

Out of curiosity, wonder just where you are located, approximately.

FWIW, Vic

BB. · July 2016

Can you let us know the Imp of each string when the controller is MPPT mode (maximum power from array) somewhere around 11am to 1pm or so... Imp/Isc actually increases a little bit with higher cell temperatures--But we usually ignore it as it is not that much (less than Vmp/Voc temperature coefficients).

If Imp seems reasonable (around 80-100% of Imp rating when sun is near "right angle" to panels) on a clear day--The panels do not sound like the problem.

-Bill

herodotus · July 2016

What is the elevation angle of your PVs.

40 degrees

What are the mid-day temperatures at your location, on clear, sunny days?

At the moment I believe we're about 30C in the shade in the middle of the day. On a bad day it might be closer to 35, but that's not often.

Believe that you had a Clamp DC Ammeter or order, or available ?? Forget. This is a useful tool to look at your Isc readings for each string. Would not hurt to also look at Voc s of each string with each Combiner breaker open (OFF).

Haven't checked Isc, but Imp in MPPT Bulk at around solar noon (2.20pm here) is about 7A for each string, or 35A for the whole array. The rated value for Imp is 8.06, so I figured that was close enough, allowing for angle/haze/etc. Just the angle alone will be dropping the current by about 10% from nominal, since normal should be around 13 degrees here at this time of year. I can't remember what Voc was off the top of my head, I think it was about 95V, but I can check tomorrow.

Out of curiosity, wonder just where you are located, approximately.

On the southern slopes of the Sierra Nevada, in southern Spain, at an altitude of about 1250m

herodotus · July 2016

Ok, so new toys arrived today! I now have a calibrated voltage reference that tells me that my new clamp meter and my old amprobe DMM are both within 0.01 of a volt of each other and the calibrated reference. I also have an IR thermometer, but it turned up too late for me to be able to check the panels at a relevant time of day.

The FlexNet DC (aka the world's most overpriced voltmeter) turns out to be horrifically inaccurate (reads 0.4V too high), which I'm pretty unhappy about. It's not that important practically, I guess, since it doesn't control the charge voltages, but the device has one job (measuring battery statistics accurately) and you'd expect it to do it properly.

The FM80 is also over-reporting by a similar amount as measured at it's own output terminals. At least I can compensate for that, and I can now be confident that I have my charge voltages set correctly - and I guess that this was probably a factor (big or small, who knows?) in the undercharging of my batteries previously, along with the short absorb times.

Will report back on panel temperatures tomorrow.

herodotus · July 2016

So, finally managed to measure the panel temperatures today. It's a real stinker out there, and unsurprisingly the panels are HOT. The highest temperature I've recorded across the back of the array is 73C. Now I'm wondering what the actual cell temperature might be based on that, since that's the radiation coming off the back of the white board on which the cells are mounted. Anyone got any insight on this? If the cells could be as high at 80C that would go some way towards explaining what I'm seeing. I'd still expect my voltages to be more like 76V rather than 70V, but I guess it's a start!

Dave Angelini · July 2016

Most of the people on the forum knew temperature was one of your problems.
I have alot of clients running really nice solar in hot locations. They often run cooling for their homes and we track the sun to get them long hours of power. The other problem of lower panel voltage is ameliorated by using Panasonic, SunPower, and the defunct Sanyo modules. They all are superior at having less loss at high temperature, as well as the highest wattage per square foot.

vtmaps · July 2016

Dave Angelini said:

They all are superior at having less loss at high temperature, as well as the highest wattage per square foot.

And there is physics to explain why that is. All the sunlight that hits a panel is either reflected off the panel, transmitted through the panel, or absorbed by the panel.

The energy that is absorbed by the panel will create heat somewhere. If the panel is unloaded, all the absorbed energy becomes heat in the panel. If the panel is exporting electrical energy, the exported energy will become heat somewhere else, and the panel will run cooler. The more efficient the panel is, the more energy it exports and the cooler it runs.

--vtMaps

Dave Angelini · July 2016

vtmaps said:

Dave Angelini said:

They all are superior at having less loss at high temperature, as well as the highest wattage per square foot.

And there is physics to explain why that is. All the sunlight that hits a panel is either reflected off the panel, transmitted through the panel, or absorbed by the panel.

The energy that is absorbed by the panel will create heat somewhere. If the panel is unloaded, all the absorbed energy becomes heat in the panel. If the panel is exporting electrical energy, the exported energy will become heat somewhere else, and the panel will run cooler. The more efficient the panel is, the more energy it exports and the cooler it runs.

--vtMaps

I believe/know the increased output at elevated temperature has more to do with the electrical specification. These arrays work just as well unloaded or fully loaded. I am in the middle of testing this month for a very large company and it surprises my client how well they do. It has been near 100F for 5 days and my heatpump is working overtime. It has not been off the whole time...

Panasonic

Temperature Coefficient (Pmax)

-0.30%/ °C

Temperature Coefficient (Voc)

-0.126 V/ °C

Temperature Coefficient (Isc)

1.76 mA/ °C

Sunpower

Temperature Coefficients Power (P) – 0.38%/K

Voltage (Voc) – 235.5 mV/K

Current (Isc) 3.5 mA/K

NOCT 45° C +/– 2° C

herodotus · July 2016

Hey Dave,

Would be interesting to compare numbers in more detail if you have them. My panels are certainly a bit worse than the two brands you are using:

NOCT: 47 +-2
Temperature Coefficient (Pmax) -0.45%/ °C
Temperature coefficient (Voc) -0.33%/°C
Temperature coefficient (Vmpp) -0.35%/°C

Out of interest, what sort of output + Vmpp values have you been seeing with this client in the 100F temperatures? Do you know what the cell / panel temperatures are ?

Also, can anyone explain to me the science behind the fact that the Pmax coefficient is different to the Vmpp coefficient? Given that, if anything, the current should increase as the voltage drops, and the temperature increase should also, if anything, increase the current, not reduce it, I don't understand where the extra 0.1%/C loss comes from given that P = VI :-/

vtmaps · July 2016

herodotus said:
Also, can anyone explain to me the science behind the fact that the Pmax coefficient is different to the Vmpp coefficient? Given that, if anything, the current should increase as the voltage drops, and the temperature increase should also, if anything, increase the current, not reduce it, I don't understand where the extra 0.1%/C loss comes from given that P = VI

As temp goes up, current goes up and voltage goes down. But voltage goes down more than current goes up.

--vtMaps

herodotus · July 2016

vtmaps said:
As temp goes up, current goes up and voltage goes down. But voltage goes down more than current goes up.

Sure, but that doesn't answer my question. The manufacturer says that power output goes down by 0.45% for every degree C over 25. Fine. But they also say that voltage goes down by 0.35% for every degree C over 25. Which means that there's 0.1% loss due to temperature increase that is *not* related to the voltage drop. Since current goes up with temperature, it can't be that; but since power is the product of current and voltage, I don't see what else it can be. It's a DC system so it can't be phase... so what gives?

vtmaps · July 2016

herodotus said:

Sure, but that doesn't answer my question. The manufacturer says that power output goes down by 0.45% for every degree C over 25. Fine. But they also say that voltage goes down by 0.35% for every degree C over 25. Which means that there's 0.1% loss due to temperature increase that is *not* related to the voltage drop. Since current goes up with temperature, it can't be that; but since power is the product of current and voltage, I don't see what else it can be. It's a DC system so it can't be phase... so what gives?

I'm not sure. Probably the testing protocol. Voltage is probably measured instantaneously (with a flash of light) at a particular temperature. Power is probably measured over time where there is some further heating of the panel over that time period.

--vtMaps

Daily charging parameters

Comments

Categories