Tricky performance of 24V system

Saipro · August 2015

Caveat: These data aren't from within a lab so I can't say everything is absolutely constant nor are the variables fixed.

I can't seem to find sufficient data for comparison online and frankly, I'm simply too lazy to do the math.

I am running a couple of experiments on solar systems in a bid to find the perfect mix. That's not a big deal. What is however is the different results I get in a stationary system by changing nothing but the charge controller. It's maddening.

The experimental system has six 240W strung up in two strings of three (3x2) for 1440W. I'm getting adequate current into my batteries (Trojan T105-RE x 8 in two strings of 4 in a 24V system). Charging voltage is fixed at 28.8V for bulk/absorb and 26.4V for float.

My iTracer IT6415ND (MPPT) gives up to 1,200W. I'm not sure of the highest I've gotten yet but recall seeing as high as 51A (I don't remember the voltage at which it did that).

My Midnite Classic 150 MPPT (after adjusting/correcting the PV and battery bank offsets) marginally exceeds 1,000W and peaks at about 40A. Is it possible that I've messed up the settings?

And here's the confounder: my Xantrex C60 PWM will ALWAYS give the 8.17A x 6 = approximately 48A in full insolation, consistently topping the MPPT bros. Mind you, due to VoC limitations, I have to string all the PV panels in parralel when using the C60. Could someone please explain what's going on with my MPPT CCs? The orientation of the panels is the same in every experiment.

The second aspect is, currently shade temperature is typically less than 28 degrees Celsius, temperature of the panels is about 40 degrees Celsius in the sun. Assuming the coefficients are the typical, how come the classic can hardly squeeze more than 1,000W under the same conditions.

westbranch · August 2015

well first I would not trust that iTracer for its stated output... Have you compared the Watts you produce with the CL150and the X C60... have you verified with a good meter what the batteries are getting?

What are your charge settings on the 150, are you using the Local App? there are some pretty touchy settings to make...

hth

Saipro · August 2015

I think the iTracer might not be "telling the truth". Everything else seems okay with my meter attesting. On the accuracy of my meter, it's the typical store type but a slightly more expensive one (I've learnt to stay away from cheap stuff, the hard way; bargains too good to be true are often just that - too good to be true).

It would seem the Classic is the one telling the most accurate of the "truths". Just wondering why I'm getting a lousy 1kw from a 1.4kW setup in the tropics.

westbranch · August 2015

Saipro wrote: »

1kw from a 1.4kW setup in the tropics.

welll check out the derating factors listed on your panels for temp > 25C
C

Photowhit · August 2015

Saipro wrote: »

...Just wondering why I'm getting a lousy 1kw from a 1.4kW setup in the tropics.

Perhaps it is because you're in the tropics!

You didn't state which 240 watt solar panel you used, so I will take Solar World's sunmodule-solar-panel-240, They post their NOCT values (Normal Operating Cell Temperature)

Under Normal conditions they expect their 240 watt panel to produce 174.2 watts or about 72.5% of the panel rating. 1400 x .725 = 1015watts....

In the tropic's you might actually be hotter than the NOCT values...

Saipro · August 2015

Problem solved! My calculations gave me roughly 175W/panel (from what I was generating in peak sunlight). I figured it shouldn't be that low from a whopping 240W but your calculations seem to rhyme to the beat, down to the last digit. Faith restored. Seems the tropics are truly harsh on these panels.

PS: At peak sunlight, the day is really hot hence the likely decline rather than increase from bright morning values. My highest readings of ~50A are invariably before noonday - and on the iTracer alone. The Classic seems to be the most sane of the trio.

BB. · August 2015

Pretty much anything solar power related--If you are within ~10% of your expected/measured readings (and across meters, controller, etc.)--You are pretty much dead on.

Some charge controllers seem to "error" on the high side (5% to even near 10% over actual current*voltage)--So having a system that is doing 10% better than you expected is not always true either.

-Bill

Vic · August 2015

From what boB MN Solar Chief Engineer (or similar function) has stated elsewhere, that the MN Classic has a very accurate output current sensor (Hall-Effect?). Most other CCs do not have such accuracy -- believe that many others have very optimistic output readings.

FWIW, Vic

Saipro · August 2015

Vic wrote: »

From what boB MN Solar Chief Engineer (or similar function) has stated elsewhere, that the MN Classic has a very accurate output current sensor (Hall-Effect?). Most other CCs do not have such accuracy -- believe that many others have very optimistic output readings.

FWIW, Vic

I believe that's the phrase bobbing around in my head - very optimistic readouts

Saipro · August 2015

I like to give updates to my observations and inferences because I learnt from others who did the same.

On the Monday the 17th of August, three more PV panels arrived of the same make and ratings. I followed previous advice and used the 72.5% NOCT and all seemed well with my system when I put it up on Tuesday, the 18th.

240W PV panels x 9 = 2160W
NOCT adjusted PV power = 2160W x 72.5% = 1,566W

(6V @ 225Ah x 4) x 2 = 24V @ 450Ah
Max charging current for batteries = 450 Ah x 13% = 58.5

With a bulk/absorb voltage of 28.8V and assuming 100% efficiency,
1,566W PV panels/28.8 absorb voltage = 54.375A

54.375A is about 12.1% of maximum charge current during absorb so that's pretty okay for me.

Cold and cloudy mornings prevailed for the first two days but I was getting reasonably full current of up to 53A in the afternoons after midday thus I was content.

On the third morning, the sun broke through early andI was getting about 17A by 7:30am. By 9am, clouds appeared again and charging current dropped back to about 13A from 25A. Such is the pain we RE enthusiasts endure when everyone hopes/prays for cool, cloudy days whilst we seek sunny days. About 10am, a brillant sunshine broke through and I felt it must have been the "cloud edge" phenomenon I've often read about because all of a sudden, I was getting between 72A and 77A of charging current. Now, that represents a nearly 100% output from the panels (less minor inefficiencies) and a nearly 100 system efficiency. Using the trusty Midnite Classic.

The next day, this occurence spanned about an hour during the aftenoon and about 30 minutes the day after that. I'm sure glad I have a CC capable of handling such current. I guess that would have been the point at which I'd have been horrified to see my iTracer IT6415ND explode from being unable to handle current (it's rated at 60A). My daily avergage from the CC is typically like this:

Attachment not found.

Cloud edge gave:

Attachment not found.

Peak was at 76.8A transiently though I captured this:

Attachment not found.

DarkstormNH · October 2015

Saipro said:

I like to give updates to my observations and inferences because I learnt from others who did the same.

On the Monday the 17th of August, three more PV panels arrived of the same make and ratings. I followed previous advice and used the 72.5% NOCT and all seemed well with my system when I put it up on Tuesday, the 18th.
240W PV panels x 9 = 2160W
NOCT adjusted PV power = 2160W x 72.5% = 1,566W

(6V @ 225Ah x 4) x 2 = 24V @ 450Ah
Max charging current for batteries = 450 Ah x 13% = 58.5

With a bulk/absorb voltage of 28.8V and assuming 100% efficiency,
1,566W PV panels/28.8 absorb voltage = 54.375A

54.375A is about 12.1% of maximum charge current during absorb so that's pretty okay for me.
Cold and cloudy mornings prevailed for the first two days but I was getting reasonably full current of up to 53A in the afternoons after midday thus I was content.

On the third morning, the sun broke through early andI was getting about 17A by 7:30am. By 9am, clouds appeared again and charging current dropped back to about 13A from 25A. Such is the pain we RE enthusiasts endure when everyone hopes/prays for cool, cloudy days whilst we seek sunny days. About 10am, a brillant sunshine broke through and I felt it must have been the "cloud edge" phenomenon I've often read about because all of a sudden, I was getting between 72A and 77A of charging current. Now, that represents a nearly 100% output from the panels (less minor inefficiencies) and a nearly 100 system efficiency. Using the trusty Midnite Classic.

The next day, this occurence spanned about an hour during the aftenoon and about 30 minutes the day after that. I'm sure glad I have a CC capable of handling such current. I guess that would have been the point at which I'd have been horrified to see my iTracer IT6415ND explode from being unable to handle current (it's rated at 60A). My daily avergage from the CC is typically like this:

Attachment not found.

Cloud edge gave:

Attachment not found.

Peak was at 76.8A transiently though I captured this:

Attachment not found.

Saipro

Im trying to setup a system just like yours in Haiti, your setup seems to be perfectly what I need.
I have the 8 trojan batteries already. I was debating on the charge controller jumping between the classic like you or the one from magnum.
i'm saving this post to use as a setup. a few questions what system do you feel provide you a better bang for the buck the 24 or the 48.
the 6 panel you started with 2 string in series than parallel right? using a combiner box than to the cc?
can you send me your system diagram please I can use your setup i looking for all your post to check for your equipment list
please help

Thanks

vtmaps · October 2015

DarkstormNH said:

a few questions what system do you feel provide you a better bang for the buck the 24 or the 48.

I would be surprised if you find much difference. If you think the system might grow, then go 48 volts. If your panels are far from your controller, then go 48 volts.

One important consideration is cheap batteries... if there is a particular battery that is available cheaply in your location, you might want to use it. If that battery was (for example) a 200 ah 6 volt battery, then a single string at 24 volts would store 4800 watthours, and a single string at 48 volts would store 9600 watthours. If one of those numbers works for you, then use that voltage. If you know you need more energy storage than that, then you need to decide what larger battery to design the system around. It's always easier to design the rest of the system after the battery has been chosen. Of course, you can't choose a battery until you know your loads...

--vtMaps

Saipro · November 2015

As vtMaps has said, systems are highly individualized. Asides loads, local variables (such as sourcing for cost-efficient materials) determine a lot. I wasn't keen on headaches so I bought my batteries and CC from the States. The rest, I sourced for locally. I initially used a 3kW 24V inverter but traded it in for a 48V later on (added a couple of bucks). The system is growing and, with the worsening situation of grid supply in my area (initially frequent outages, tolerable but dirty supply with voltages swinging between 140V and 280V), I just might go off-grid in the heart of the city!

Funny thing is, there are 3 large supply transformers within a kilometer of my house each and I've connected to each at different times, each with its own headaches. But I stray ......, back on topic.

With my current setup (initially 24V and now 48V), I've gone up to a whole week off-grid at times due to wanting to avoid damage to sensitive equipment. Neither system appears more efficient or easier to handle than the other. The 24V setup had me lining my arrays in 3 strings of 2 PV panels (up to 74V VoC and 49V - 60V operating voltage). The 48V had 2 strings of 3 PV panels (up to 110V VoC and 68V - 85V operating voltage). This is because I noticed the Midnite Classic 150 doesn't like excessive voltage headroom and heats up sufficiently to get me worried.

The 24V system gave no hassles asides the fact that I fussed needlessly over my batteries for a while from the fear of imbalance between my two strings of batteries (the strings are a few months apart in age). With the switch to a 48V system, I worry less about my batteries (single string) and rely mostly on what the CC reports. I'm about to launch into another round of voltage and SG measuremnts to test a theory otherwise the system performs satisfactorily.

I've also added three more PV panels making 240W x 9 panels for a total of 2,160W in 3 strings of 3. I am able to use my air conditioner between the hours of 11am and 3pm (sunniest periods on sunny days) but not but not everyday. My fridge kicks up at 9am and goes off at 5 - 6pm (depending on the day's insolation). The freezer compartment is now permafrost. Most days, I finish my absorb cycle, often before 3pm. A few days, I don't at all.

As a finishing note, running DC equipment matters a lot too. Most of my DC equipment are now redundant as I painfully discovered there are ever so few equipment which run on 48V DC. 12V is lord over the others and 24V holds it own. All my floodlights and a few other stuff are 24V and are switched via the AUX of the the CC so they're unattended to. Wiring and hooking them up was a pain. Can't junk them now. I've ordered a 48V-to-24V DC-to-DC converter (and another 48V-to-12V for backwards compatibility

) to help me overcome the problem. In hindsight, the 24V system was perfect for my needs. Using foresight, it's a matter of time before I get restless and start growing the system again and the 48V setup starts looking quintessential. I hope I've done more to clear up cobwebs than create them.

Addition: My PV panels are wired straight to the CC

vtmaps · November 2015

Saipro said:

Most days, I finish my absorb cycle, often before 3pm. A few days, I don't at all.

Sounds like you have achieved perfect balance for an off grid system. That's what system design is all about. Congratulations!

Saipro said:

Addition: My PV panels are wired straight to the CC

What are you saying? No fuses or combiner box?

--vtMaps

DarkstormNH · November 2015

Saipro said:

I am able to use my air conditioner between the hours of 11am and 3pm (sunniest periods on sunny days) but not but not everyday. My fridge kicks up at 9am and goes off at 5 - 6pm (depending on the day's insolation). The freezer compartment is now permafrost. Most days, I finish my absorb cycle, often before 3pm. A few days, I don't at all.

Great info the air condition is that 120 or 240 volts? im assuming you cooled the house around that time to take advantage of the
excess power around that time?

also you don't leave your fridge on 24/7

last question on the panels you do not use a combiner?

Addition: My PV panels are wired straight to the CC

zoneblue · November 2015

>I hope I've done more to clear up cobwebs than create them

Your post is actually quite an apt description of how systems evolve for DIYers. Many of us have gone through similar evolutions as the system grows, as your experience grows. Aint life a wonderful thing.

Saipro · November 2015

DarkstormNH:
240VAC home use. I have no real need to leave the fridge on 24/7. I have teenagers in the house and God knows they're the natural enemies of almost any stable system and probably the worst enemy of RE (rated right next to ignorance). I'd rather spare the batteries of excessive wear. My panel energy production would have exceeded my energy requirements by 10am on most days so I'm wasting energy by midday. Suits me fine because I can go three days of really bad weather and my batteries would still not fall beyond 55% DoD. Indeed I've roughed it out for five days in which I had probably a moments of cloud break. Panels are cheap here and I got my 240W panels for roughly $205 a piece.

vtmaps
No combiner box but certainly I use both fuses and breakers. Fuses for protection and breakers for easy maintenance. After all, I keep fiddling with the darn setup, less and less each week though (haven't touched anything for a few weeks now on my primary system).

zoneblue
You're sure right on that one. Nothing beats learning on the job. I'm gradually turning pro-style and have even begun criticizing pro-jobs. Was blind, now I see.

PS: Today is one of those cloudy days. It's not noonday yet I've harvested 2.4kW. Good enough to raise my batteries out of the gutter.

vtmaps · November 2015

Saipro said:

No combiner box but certainly I use both fuses and breakers. Fuses for protection and breakers for easy maintenance.

I am stretching my imagination to figure out how you might wire to code three strings (with fuses), but not use a combiner box. Do you have a fuse on each of the three strings?

Do you have lightning protection? A combiner box is also a good place to put a lightning arrester.

--vtMaps

DarkstormNH · November 2015

Saipro said:

DarkstormNH:
240VAC home use. I have no real need to leave the fridge on 24/7. I have teenagers in the house and God knows they're the natural enemies of almost any stable system and probably the worst enemy of RE (rated right next to ignorance). I'd rather spare the batteries of excessive wear. My panel energy production would have exceeded my energy requirements by 10am on most days so I'm wasting energy by midday. Suits me fine because I can go three days of really bad weather and my batteries would still not fall beyond 55% DoD. Indeed I've roughed it out for five days in which I had probably a moments of cloud break. Panels are cheap here and I got my 240W panels for roughly $205 a piece.

vtmaps
No combiner box but certainly I use both fuses and breakers. Fuses for protection and breakers for easy maintenance. After all, I keep fiddling with the darn setup, less and less each week though (haven't touched anything for a few weeks now on my primary system).

zoneblue
You're sure right on that one. Nothing beats learning on the job. I'm gradually turning pro-style and have even begun criticizing pro-jobs. Was blind, now I see.

PS: Today is one of those cloudy days. It's not noonday yet I've harvested 2.4kW. Good enough to raise my batteries out of the gutter.

Saipro

Thank you for your great post I'm 100% certain I will follow your steps, I flag your post for the details
you have 9 panels in 3 strings, 48 volts 120/240 inverter ( magnum? pae) classic cc life is good
your setup would serve me perfectly, batteries are you using trojan L16 or trojan t105?
if you have any further details schematic please if possible send them to me , I would be eternally greatful

Saipro · November 2015

vtmaps
No DC surge protection. I'm in a region with very low odds of a lightening strikes. Almost non-existent for most of the year and the thunderstorms are there for 4 or fewer weeks every year. Begining and end of the monsoons/ rainy season. No surge arrestors on the AC side either. Decided to do away with a box altogether. Installation and wiring codes for those things are literally non-existent here thus I'm not in violation of anything I'm aware of. Saves me some bucks on those too. Limited equipment with minimal risk of a major fault. Might someday get around to buying a combiner box if I get a great deal online. Until then, ......

DarkstormNH
I'm not particular good with schematics so I'm afraid I won't be of much help there. The inverter is, interestingly, a Must Power IR series. Chinese product. 3kW 48V with 230VAC output. Power factor of 1. Works perfectly. Idle consumption is a bit worrisome at about 60W (from my tests). I wish the idle consumption could be lower. Serves me well otherwise. I'm rarely off-grid for more than three days and only recently had to go a whole week about thrice (due to really horrible grid supply which wasn't worth reconnecting to while it lasted), so the idle consumption isn't too much of a headache.

Batteries are Trojan T105-RE.

SPA · July 2016

Saipro Said: "I've also added three more PV panels making 240W x 9 panels for a total of 2,160W in 3 strings of 3. I am able to use my air conditioner between the hours of 11am and 3pm (sunniest periods on sunny days) but not but not everyday. My fridge kicks up at 9am and goes off at 5 - 6pm (depending on the day's insolation). The freezer compartment is now permafrost. Most days, I finish my absorb cycle, often before 3pm. A few days, I don't at all".

Hi Saipro, what type of AC unit do you use? Could you possibly tell us the model and faceplate specs? 120V, current consumption etc...
I might try to use a similar AC unit on the system I'm putting together.

Thanks for you post!
Seb

Saipro · June 2018

I've been away for too long. I'm back now
I hope I can still be of assistance to you.
LG makes DirectDrive and Inverter mode appliances including AC units, chest freezers and refrigerators

Tricky performance of 24V system

Comments

Categories