Sunny Island/Rolls 530-S charging question

palausystem · November 2012

Hi all,
I'm following Rolls advice as far as charging volts go:
Float - 2.2v
Boost - 2.4v 2 hrs
Full - 2.4v 5 hrs
EQ - 2.58 - 2.7

Things are basically OK, but at the beginning of each winter when things cloud up, the Sunny Island does a little freak out, dropping from 70% to 20% state of charge, the "health meter" of the batteries drops a few points, and I have to do a corrective EQ...which is a two day affair what with doing one string at a time for all day.
I do a preventive EQ every three or four months on both strings together at 2.58v, and I can get about 6 hours at close to the target voltage before it drops a couple volts and the countdown clock stops.
I'm just wondering if I should be doing something different so that I don't have to do the "Corrective" EQ once a year.

Sunny Boy 3000
Sunny Island 5048U
18 Sharp 170w
16 Rolls 530-S

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

This is a strictly off-grid system, is it? I'm confused. Normally one would not use a Sunny Boy + Sunny Island for off-grid. That would be a grid-tie set-up with battery back-up and the SI would keep the batteries floating as long as the utility was up.

As an off-grid system it's a failure because 3kW of panel isn't enough for 800 Amp hours of 48 Volt battery.

Not sure what you mean by "corrective EQ" either. Equalization is done when cell SG is too disparate, not because the batteries are low. Batteries hitting 20% SOC are dead.

As I said, I'm confused about this whole scenario.

palausystem · November 2012

Re: Sunny Island/Rolls 530-S charging question

Yes it's strictly off grid. I'm a novice, and the installer seemed to know what he was doing but I don't know much.
From the manual:
"The Sunny Island is a bidirectional inverter (battery inverter and charger) for off-grid power systems.
The Sunny Island supplies loads on the off-grid side and charges battery banks with the energy from
grid-feeding units connected on the AC side."
I guess the Sunny Boy (and the generator) is a considered the "grid feeding unit".

What would be the "normal" setup for using the Sunny Island?

As far as 3KW of panels not being enough for the 800Ah of battery, what more can you tell me about that so I can complain to the installer/designer? What makes it a failure? The panels seem to charge the bank with little issue.

The 20% soc is more a monitoring anomaly of the SI.. the bank voltage was 46.5v at the time, so maybe near 50% soc. I just know that when there's a sudden adjustment in the numbers like that, the SI is unhappy with the batteries, I notice the sytem is sluggish, then I check some voltages and some SG's and find the need for an EQ.

"Corrective" EQ is Roll's term for all out maximum volts EQ to break up heavy sulfation. I do one string at a time at 2.68v and watch the SG's. Last year it was about 10 hrs per bank.

"Preventive" is their term for a lower voltage maintenance EQ, which I choose to do at 3 month intervals for 5 or 6 hrs, both banks together. I do these EQ's without paying much attention.

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

This is completely crackers.

Sunny Boy is a straightforward GT inverter meant to take whatever power the panels can produce and back-feed the grid.
Sunny Island is an off-grid inverter capable of synching to GT inverters to provide back-up power (and utilize the GT inverters) in the event of a power outage.

You can use the SI as a strictly off-grid inverter. Using the Sunny Boy as the array connection in such an application is nuts. Normally the batteries would be kept charged with a charge controller; a device specifically designed for the purpose. As it is the SB is AC coupled to the SI and back-feeds it to charge the batteries. This is ridiculously inefficient and expensive and as you can see, doesn't work very well.

Surrette batteries are over-rated in my opinion too, but that's a different issue. Have you got any specific gravity readings on these batteries?

My advice would be to get a proper charge controller and flog the Sunny Boy inverter if at all possible.

What I'd have for 800 Amp hours @ 48 Volts: target a 10% peak charge rate of 80 Amps @ 48 Volts (3840 Watts) plus factoring in panel & controller inefficiency: 4987 Watt array (+/-). What you can expect from (18 * 170) 3060 Watts: 49 Amps which would work well for 1/2 that much battery. The Sunny Boy configuration probably is less efficient and costing more power. Probably down around 5% peak rate. But I'll bet you haven't got good metering to see how it does work (like a battery monitor).

Check the SG on every cell and see what it's like. Then try charging half the battery bank at a time and see what the best results are. You've got to know how bad off the batteries are now before you can figure out what needs to be done. Checking your load requirements is in order too.

Only my opinion, but this system was stupidly designed and probably could have been done a lot better for less money.

palausystem · November 2012

Re: Sunny Island/Rolls 530-S charging question

What is the real world effect of sending only 40 amps to the 800Ah bank?

Ralph Day · November 2012

Re: Sunny Island/Rolls 530-S charging question

Sounds like someone had some Sunny products to sell.

Ralph

stephendv · November 2012

Re: Sunny Island/Rolls 530-S charging question

Hi,

Firstly let me say that there is nothing wrong with the components you have. There are hundreds of off-grid systems installed using this configuration of Sunny Island + Sunny Boy and it's SMA's preferred and recommended option for all medium to large off-grid systems. Having said that, it's not the cheapest nor the most efficient system for battery charging, but in general it works just fine.

With regards to the charging, your settings look OK in general.
How often have you set it to the "full" charge? Depending on how far you draw the batteries down every day, the 2 hours for boost could be too short to fully charge them. This is fine, as long as the full charge happens quite regularly, I'd guess about once every 7 - 20 days (told you it was a guess).

Eric L · November 2012

Re: Sunny Island/Rolls 530-S charging question

This is completely crackers.

I actually think these SMA system designs are pretty cool 8) although I agree with the others that it may not have been most cost-effective in this case (may not have been that bad, either).

The Sunny Island signals the grid tie inverter by varying the AC line frequency slightly so that the inverter takes only what the batteries and loads can use (for pure off-grid).

In some cases, it might be more cost effective to do it SMA's way. Example, a very large array with a long run between panels and inverters. The Sunny Boys can take 600 VDC in and outputs at 240 VAC (US spec), and the largest model can handle a 10KW array. So you could have some flexibility in where the grid tie was put relative to the inverter to optimize the wiring configuration. One big Sunny Boy plus one or more Sunny Islands could arguably be cheaper than buying several very high-voltage charge controllers plus inverters.

Even if this wasn't the most cost-effective installation, it is likely good quality and not crackers, IMHO.

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

palausystem wrote: »

What is the real world effect of sending only 40 amps to the 800Ah bank?

Deficit charging resulting in prematurely dead batteries. 40 Amps is a 5% peak charge rate. Add some loads and it goes below that. Result: you do not get the batteries fully charged most of the time.

Yes, SMA equipment is good. But no, this is not a good system. It sounds to me like the installer was used to putting in GT systems with a battery back-up and that's what he gave you, figuring "off gird" is the same as "long term power outage". Well, it isn't.

What's wrong?
1). The array. 18 panels to get a 3kW array? There are larger panels available for less $ per Watt that would have been a better choice. That's money, your money, wasted.
2). The SMA Sunny Island. About $5,000 by my info and it only puts out 120 VAC. Same money could have bought any number of other combinations which would provide the same or more power and left you with some spare change. Again, your money wasted.
3). The SMA Sunny Boy. Since this only produces during sun hours it doesn't really add to the inverter capacity (and do you need 8kW?) which just makes it an expensive connection to the array. Again, your money wasted.
4). The batteries. Are they sized for your needs? 800 Amp hours @ 48 Volts is up to 19 kW hours. That is a lot of stored power. Surrettes are expensive, and obviously this bank doesn't fit with the charging capacity of the system. More of your money wasted.

In short there is nothing right about the design here. You could have done better for less $. But now you're stuck and have to fix it.

Step 1): determine how much battery capacity you actually need. If you can get away with 400 Amp hours, rebuilding your battery bank using the best half of what you've got will solve most of your problems.
Step 2): determine how much charging you're actually getting. This is DC clamp-on Ammeter time, as you need to measure current going in to the batteries. Many people will say "get a battery monitor" but it's too late for that. These batteries have been abused and you cannot tell what shape they're in so you can't program the monitor accurately to begin with.
Step 3): ditch the Sunny Boy and replace it with an MPPT charge controller. This will be the biggest problem because the array will need to be reconfigured (need to know what it is now - I can only guess) and a combiner added with different wiring to the controller (likely you would be changing from 450 Volts @ 5 Amps to 105 Volts @ 30 Amps). You will also need the wiring and circuit protection to the battery bank.

I doubt you will get much help from the installer. Nor is it likely you would have any legal recourse against him. All he'll have to do is point out you didn't check the batteries regularly, et cetera, and he's off.

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

Eric L wrote: »

I actually think these SMA system designs are pretty cool 8) although I agree with the others that it may not have been most cost-effective in this case (may not have been that bad, either).

The Sunny Island signals the grid tie inverter by varying the AC line frequency slightly so that the inverter takes only what the batteries and loads can use (for pure off-grid).

In some cases, it might be more cost effective to do it SMA's way. Example, a very large array with a long run between panels and inverters. The Sunny Boys can take 600 VDC in and outputs at 240 VAC (US spec), and the largest model can handle a 10KW array. So you could have some flexibility in where the grid tie was put relative to the inverter to optimize the wiring configuration. One big Sunny Boy plus one or more Sunny Islands could arguably be cheaper than buying several very high-voltage charge controllers plus inverters.

Even if this wasn't the most cost-effective installation, it is likely good quality and not crackers, IMHO.

Good quality equipment, yes.
But it is crackers.

Let's see how it stands up to the tests:
1). Does it work properly? No, it doesn't. Instant fail.
2). Is it cost effective? Not by any stretch of the imagination. Instant fail.

Your contention that there is an advantage in using the SB 3000 because of its ability to run high Voltage is moot: the XW 600 MPPT 80 can do the same thing, costs $500 less, and is more efficient (no converting DC to AC then back to DC to charge the batteries).

This set-up doesn't fit the bill. It's crackers.

palausystem · November 2012

Re: Sunny Island/Rolls 530-S charging question

Thanks for the input, everyone.
I do appreciate Cariboocoot's observations regarding my system, but for some reason I would expect a "Moderator" to be a bit more, um, moderate. I mean, come on... "This is completely crackers...ridiculously inefficient...Surrette batteries are over-rated...replace the Sunny Boy...buy a new charge controller...There are larger panels available for less $ per Watt that would have been a better choice" ...etc etc.
So yes, I get that you would have done it differently. You can keep shouting "Crackers! Failure!" and everyone can join in an argument about SMA products and all that.
But I don't want to waste my time. Too often these exchanges end up sounding like a peeing (edited swear word. niel) match.

The one useful item I think I can take away from the Moderator's comments is about The battery bank size in relation to the 3K array.
Does anyone else have any input on this? Are we all in agreement that this is a major design flaw that needs to be remedied in order to avoid significantly premature battery bank failure?
Thanks...

BB. · November 2012

Re: Sunny Island/Rolls 530-S charging question

Moderators are people too.

-Bill

Photowhit · November 2012

Re: Sunny Island/Rolls 530-S charging question

I too think your a bit too light on your array to battery bank. particularly if there are greater charging ineffiecentcies as 'Coot says (he's usually right) Generally I'm OK with a 5% charge rate, if your not using the system a lot during charging. Heck that's pretty much what I have at my cabin, 1700 watt array feeding a 800AH 24Volt battery bank. ...but this system doesn't get much use. If your living and using the system, it's likely getting more use during the charging.

I don't think the age of this system has been determined? If it's relatively new and your finding your batteries need corrective equalizing once a year, this tends to be indicate cronic undercharging. If they are reaching full charge weekly (what I shoot for) and not drawn down below 70% state of charge, then they shouldn't require a corrective equalizing, likely in their first 5-7 years.

'Coot(and I) have seen many posts about problems with Rolls-Surette batteries in the past 4 years, so likely you can understand his thoughts on this premium brand of battery. Their big 2 volt cells have been known to go 15+ years in the past! Likely some of what we've seen is like watching the news... you only hear the bad stories.

The hard part about your situation, is what to do. I do think it's likely the installer balance your system like a grid tied system with battery backup. since the battery would get little use, the charge rate was fine at roughly 5%. but in an off grid system you likely want a higher charge rate, particularly since your having to do this annual corrective equalizing. Since you have a high voltage system this will be a bit challanging. I would try to add another string of panels with the same or slightly higher VMP. You may only have 1 string! In this case you might actually be cheaper to bring the array up to @5Kw and switch to a Midnite classic charge controller(or 2) which may give you greater charging from the same array.

Where I do have a problem with what 'Coot said is "Equalization is done when cell SG is too disparate" this is only what Trojan recomends, from what I have been able to assertain, all other flooded battery manufacturers recomend a maintainance equalizing, from once a month to once every 6 months. This does more than equalize the cells, it also mixes the electrolite preventing stratification (which can also lead to sulfation over sections of a battery. I try to put equalizing into mainanance and corrective catagories, I am surprised that you use 2 different settings for these, but that is another tale...

Hope this helps, 'coot really is a friend, what else would you call someone who is taking time out of his day to offer suggestions, not paid or required... Thanks to all the Mods!

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

That was moderate, for me.

I really get upset when I see people have been charged too much for a system that, worst of all, fails to meet expectations.

Repeatedly I've said there's nothing wrong with SMA equipment. I even recommended it, for the right application. For a strictly off-grid set-up it's a lot of money for what you get.

It is unfortunate that Surrette quality seems to be sliding downwards, but I don't think that's the problem here. That number of "20% SOC" is way below what you'd expect for any standard deep cycle battery, so you couldn't expect them to stand up to much of that.

Get the loads determined first, as I said before. Off-grid systems work better when you know how much power you need to supply in terms of maximum Watts and total Watt hours. Otherwise it's a guessing game with a very high percentage possibility of being wrong. If you don't actually need that massive amount of stored power and can reduce the battery bank by 50% everything else you've got will start working much better right away without spending a dime, providing the batteries haven't been ruined.

Photowhit · November 2012

Re: Sunny Island/Rolls 530-S charging question

I saw from your previous post that you have a generator that should kick in, is this a manual start to your generator?

Sounds like your installer wanted to keep you in the top 1/3rd of your battery capacity, this is a good thing! If your generator isa kicking in you shouldn't have had such a low levell on your battery.

How did you determine loads for your system? We kinda go dropped in the middle, if we were determining the needs, it's easier to figure out the proper array:battery bank...

Do you have any type of monitor to determine how much energy you are drawing from your system?

stephendv · November 2012

Re: Sunny Island/Rolls 530-S charging question

'Coot is right that if you're in the US then the SB + Sunny Island is an expensive system, and you could have installed other equipment for a better price and had split phase 240V/120V volt output. But since you already have the kit, I think there's little point in discussing it now. There are some advantages to using the SB, so I would definitely NOT recommend selling the SB to buy a charge controller. In fact there was a time where I was considering the opposite: selling my charger and replacing it with an SB so that I could use the Sunny Island's charging system.

The ratio of panel to battery bank depends on your loads. Without knowing your daily loads and how much solar energy you receive every day, there's no way to know whether you have enough panel or not. Panel to battery ratio is not as important IMO has panel to loads. If the panel to loads ratio is right then you have some freedom in choosing the number of days of autonomy you want from the bank. The 5% charging rate is a nice rule of thumb if someone is just starting out with the design of their system and they're looking for a ballpark of battery bank size or array size. If you already have the equipment and you want to troubleshoot a problem then energy in vs energy out is going to give you a much more accurate answer than a magic number which somehow applies equally to everyone on earth no matter what their loads are.

SolarMusher · November 2012

Re: Sunny Island/Rolls 530-S charging question

The ratio of panel to battery bank depends on your loads

I agree with Stephen and I don't understand why coot said this ratio was wrong. Hey guy, C/10 is just a rule of thumb, not the truth.
Is there someone on this forum who never had problems with batteries? Surrette, Trojan or others brands. Even with 4 or 5kw array!
Please, stop shooting the installers as you don't know what was the system he designed, why he made it that way and how his customer uses it.
There is not just one way to install solar, even if is the "Coot" way.
Moderator, you're 100% wrong!!!
Erik
Installer in Quebec

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

SolarMusher wrote: »

I agree with Stephen and I don't understand why coot said this ratio was wrong. Hey guy, C/10 is just a rule of thumb, not the truth.
Is there someone on this forum who never had problems with batteries? Surrette, Trojan or others brands. Even with 4 or 5kw array!
Please, stop shooting the installers as you don't know what was the system he designed, why he made it that way and how his customer uses it.
There is not just one way to install solar, even if is the "Coot" way.
Moderator, you're 100% wrong!!!
Erik
Installer in Quebec

Hey, no I'm not.

But people are free to disagree with me (providing you keep it civil - HINT) and waste their money as they like.

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

For those who haven't just earned a free week's vacation away from the forum ....

The OP came on here looking for help because his system wasn't working properly. This included SOC dropping to 20%, a need for corrective EQ to get the SOC back up, and typically occurring when the days were getting shorter. He is not in sunny Spain, alas.

So from this the only conclusion that can be draw is that it does not work properly. Saying there's nothing wrong with it is not correct.

I offered an analysis and solution based on my experience in correcting dysfunctional systems. No, I am not an installer - but I've cleaned up a lot of messes made people who claim to be.

If you have anything positive to contribute to the thread, by all means speak up. Telling me I'm wrong without offering anything factual (including alternate suggestions for correcting the problem) to back up your assertions is not helpful.

Photowhit · November 2012

Re: Sunny Island/Rolls 530-S charging question

SolarMusher wrote: »

Moderator, you're 100% wrong!!!

Well he's allowed his system to run down pretty low, and the installer might not have the correct figures to work with...

That said if he's serious about correcting the problem, there are a lot of people here, who have off grid systems, who would be glad to help correct that, if we can get some more information. If your an installer, I hope you can see that we are trying to help, I think 'Coot might be seeing the annual equalizing as a cronic under charging, yes it could be battery problems of other sorts, but that might well be a good guess...

What is your guess? Do you have something to aid him? I went back and checked his pervious post, he has a generator, not sure why that's not kicking in before he gets past 50% SOC? and we're not talking about being a little under 10% we're talking about feeding it 6% peek most of the time. Suggesting that it might not be getting that do to day time loads is a very valid point. Running a household on a 3KW off grid array is very minimal, how many off grid systems have you done that could run on such a small array, with a full time resident/family?

I do think most battery makers suggest a minimum of 5% (C/20) charge rate, If we can get some more information, we may find his system isn't getting that, Love to talk to him about "new loads" or how he figured his loads with the installer. If this is his first system, he may not have understood what information the installer needed, or under estimated his loads. Says his installer figured 5.9KWhs a day with a 2 day autonomy so it appears the installer figured 5.9 KWhs a day. with the right location, he might be able to do that on a 3KW array, my guess is he uses more than that, is new to off grid life. We need more info, a stay at home wife, doing laundry, cleaning, fridge running, lights, heat strips on his water lines, radio or TV, satalite, may all be running while he's wanting to charge his battery bank.

I'm a single guy and have lived off grid for more than 10 years, I'm figuring I'll use 4Kwh a day for the things of daily life, other than summer (I'll be running a couple A/C's) adding a good sized freezer would put me in the 5-6Kwh per day use... I can easily see someone who hasn't conserved and is a first time off grid person using more than that. If it's a family often they need to all get on the same page, and it would be easy to use more than that. Also not being aware of what uses lots of energy, coffee pot left on during the day, hair dryers, toasters... Perhaps he can give us some figures on what he's using, if your an installer I'm sure you have seen solar not meeting someone expectations.

As for battery problems, I've had none in the systems I've setup from a tiny pair of SLA for a bicycle trip to the current forklift battery, I've pretty much gotten what I expected. I have killed a set, left an inverter on, accidentally, with no or minimal charging.

stephendv · November 2012

Re: Sunny Island/Rolls 530-S charging question

A feature of the Sunny Island's charging is that it doesn't do a bulk and absorb charge every day, instead it may do successive float charges if the battery doesn't dip below a factory set value. I've not been able to nail the rules by which it works yet, down but it looks like if SoC is above 80% then it'll skip bulk and absorb for 2-3 days and just do a constant float charge all day. This is fine for summer, but in winter you might run into trouble with cloudy weather if it floats through the good days, and then tries to bulk-absorb on bad days.

If this is happening often, you could try changing the "full" charging interval to 1 or 2 days, so that it'll do a full absorb more frequently during winter.

Photowhit · November 2012

Re: Sunny Island/Rolls 530-S charging question

Well I think this thread is dead, I just don't like to have given bad info. I said you might only have one string, I made this assumtion from thinking this was a high voltage system similar to Xantrex's which I think runs close to 600 volts and down converts. I read a little and it appears this runs at lower voltage.

I still think it's likely your array is too small for you use/battery bank. Wish you would let us help! Think an additional string of panels might help, and more info would help us help you. If you provide more information. This really is a great collection of informative people, with installers, engineers who make some of the best charge controllers and have designed many inverters, and people who actually use this equipment willing to give information. Heck, we don't always agree but you would be in a far better set of hands here than most anywhere else on the web, including many/most professional installers and salesmen.

niel · November 2012

Re: Sunny Island/Rolls 530-S charging question

i'm no expert when it comes to backfeeding inverters like this, but it sounds more like the loads were exceeding the production from the pvs in order to drain off power from the batteries. if not that then there may be a glitch or wrong setting someplace. battery capacity most likely isn't the problem, but i can't say for sure as in very large draws the ripple voltage may have come into play. the ratio of how much is being fed to the acting grid inverter (from sb to si i believe) was touched on in the past by one of you guys.

palausystem · November 2012

Re: Sunny Island/Rolls 530-S charging question

Well, I do appreciate all of your input. I took some time out to learn more...
There are I think 2 strings of panels for 3.2Kw The maximum current I've see (from PV) charging the batteries is 40amps. (via the Battery Current gauge on the SI)The batteries are 2 strings, 48v for a total of 800Ah.
The problem is some erratic soc readings at the beginning of the cloudy season. And some faster than usual draining of battery capacity. I've had the batteries 3-4 years. We had to fire the installer (a long story) before the dust settled so I had to take over and learn the battery voltage settings, etc
The installer had never checked with Rolls to find the proper charge settings, he used the defaults, which were all too high (except EQ, which was too low to even heat up the batteries). So after 2 years of his settings, I changed them to the the recommended. The soc readings never seemed to make much sense, jumping up and down at night by 10% or more. I figured the batteries must've been pretty sulfated from improper charging. I learned from Rolls that the best way to do an EQ was to do each string separately at the highest voltage the inverter would allow. I did that last year about this time, about 10 hrs each string, and although I didn't do all the proper SG testing at that time, the system began to run smoother than it ever had, soc mostly in the 90's, slow drain at night...
I thought I was out of the woods...I would do a half-hearted EQ on both strings together, every few months - just run the gen for 6 or 7 hrs with the high voltage setting. After 4 or 5 hours though, the voltage would drop more than 1.5 volts below setpoint and the timer would stop, but I just figured it was good enough.
But then the same erratic behaviour is popping up - you know the Sunny Island is confused when the Soc jumps down from 70% to 20% at 8 in the morning and then jumps up again an hour or two later.

It makes sense that proper full charges have not been happening and that the first thing to do is a long EQ.

Rolls has a formula for absorption time Time = 0.42(800Ah/40A) so for my 800 Ah battery getting fed 40A max, would take 8.4 hrs.
The Rolls guy said about the formula "this is worst-case scenario with 50% dod cycling" Now I'm not sure why he would even mention dod in this equation because this formula is for a battery that is already at roughly 80% soc and is ready for the absorption phase...so what does it matter how low it was before? I emailed him and am waiting for an answer.

I've a book on PV system design and he says 40A would be the minimum for that battery
So that gave me some hope that the system is perhaps not unworkable, but will be a maintenance hog what with generator assisted full-charges and more frequent EQ's

So what I'm going to start doing in Winter months is at least every two weeks do a full charge with sun and generator together (65 or so amps) for 5 hrs.

This brings us to the Sunny Island charging behaviour -someone here mentioned how it spends a couple days in float then goes to a "Boost" charge which is really not a Full charge because it's only designed to bring the system up to 85% to 90%. Then it will do a "Full" charge at a set interval - you choose - which is just a longer absorption time. So all that is a little confusing...
The default Full time is 5 hrs, and the default Boost time is 90 min. That's a difference of 2/3rds. So, I figure if I need an 8.4 hr absorption (as per Rolls) for full charge, then a third of that would be about 3 hrs for Boost. This way I figure I'll be playing along with the SI intentions, however mysterious they may be.

Our soc typically stays above 70%.
Our needs are at about 6Kwh. We don't have a cumulative meter for that, most of our loads are in the evening what with pellet stove, lights, TV. What's the best way to read amps out-amps in? The Sunny Boy has some sort of cumulative total but I'm not sure what it's telling me or whether it's in the loop with the batteries...

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

A battery monitor would be the best way to keep track of the SOC of the batteries. Unfortunately they depend on being programmed with information including battery capacity, and it's entirely likely yours no longer meet their 800 Amp hour spec after three years (especially with the mis-charging). So your best friend is the hydrometer now.

What Stephen said about the SI skipping Bulk/Absorb for a couple of days is why it is perfect for grid-tie back up, and why it is imperfect for strictly off-grid.

Likewise the inefficiencies of using the SB (two current conversions and the minimum power needed to activate the unit) to couple the array are costing small amounts of power you could use. Vis: your 3kW array with 4 hours of "equivalent good sun" would be expected to provide your 6kW hour daily need just; 3000 * 4 * 0.52 = 6240 Watt hours. The SB/SI combination may not even have the 52% over-all efficiency used for standard systems.

The "worst case scenario with 50% DOD" the Rolls guy refers to means maximum EQ needed for a battery if it has been regularly discharged to 50%. This has nothing to do with the SOC before running the EQ, which should be as close to 100% as possible. It has to do with repeated deep discharges causing greater cell to cell imbalances that need maximum effort to correct. You may not need such an extreme.

You clearly have a problem maintaining any EQ charge long enough to be truly effective. Perhaps you should consider purchasing an Iota charger to get around the SI's quirky charging behaviour. Only 15 Amps and a lot of money, but could solve the EQ problem: http://www.solar-electric.com/iodl48vo15am.html

Some adjustment of the gen start may be in order to get it to contribute a bit sooner.

Cariboocoot · November 2012

Re: Sunny Island/Rolls 530-S charging question

Forgot to mention:

6kW hours AC daily is about 140 Amp hours @ 48 Volts. That's a 280 Amp hour 48 Volt battery bank minimum. On 400 Amp hours that would be only 35% DOD.
You don't really need 800 Amp hours of battery.

You could take one string out, get the biggest 12 Volt Iota charger you can afford, do some corrective charging on that string one 12 Volt section at a time, then swap for the other and repeat.

niel · November 2012

Re: Sunny Island/Rolls 530-S charging question

this one is hard to determine what is going on for sure. now keep in mind that sulfated batteries lose some of their capacity. it would be good to determine what you actually have there in capacity. i'm not sure if you took sg readings or not as i didn't look back. if your batteries have gone low in capacity then large loads can cause a large ripple to the batteries and things won't work right.

(remember coot that sg said a minimum of 100ah in capacity per kw of inverter to avoid ripple)

yes, do be sure to give the batteries a good wallop from both sources and be sure of your water levels too with distilled water. can you do this one string at a time and still power your normal loads from another source?

Photowhit · November 2012

Re: Sunny Island/Rolls 530-S charging question

palausystem wrote: »

Our needs are at about 6Kwh. We don't have a cumulative meter for that, most of our loads are in the evening what with pellet stove, lights, TV. What's the best way to read amps out-amps in? The Sunny Boy has some sort of cumulative total but I'm not sure what it's telling me or whether it's in the loop with the batteries...

I think some type of shunt meter at the battery, Trimetric, Pentametric by Bogart(?) or I think others here have used Victron battery monitors.

Off gird systems rely on an over abundance of current to be healthy, sounds odd but you want to reach full charge on a regular basis and waste energy. I don't know where your located, but here in Missouri I was planning on a 4Kw array for my daily average of 4 KW use in the winter, Our solar isolation only gets down to 3.25 hours in the winter, but I also will have my panels at a more summer time angle and I don't use a backup energy source. For some reason I couldn't open the Nrel.gov site tonight so I won't link to it.

If you have a few cloudy days, I'd go ahead and run the generator to be on the safe side.

palausystem wrote: »

Our soc typically stays above 70%.

I'm not sure where this number comes from. If it's from a confused charge controller/system It might not be a very good number. Most/all of the State of charge numbers are based on data you have recorded, type and size of battery bank, Unless it's a very simple voltage based system which can fluctuate wildly (I guess this might be what your seeing, starting a microwave while a coffee maker is on could be a huge morning draw that would allow a voltage based monitor to drop a bunch quickly and return quickly)

Photowhit · November 2012

Re: Sunny Island/Rolls 530-S charging question

I went back and looked at your new post and it looks like some of the voltage fluctuations could be explained by a voltage based SOC meter.

"The soc readings never seemed to make much sense, jumping up and down at night by 10% or more."

You may be able to create this by applying a large load and then remove it, something like a microwave, and a coffee machine or toaster. see if you pop some popcorn while toasting or brewing, if your SOC doesn't go down 5-10% and return (likely you'll need to apply a heavy load for 5 minutes or so, so perhaps a microwave isn't the best load, unless you want 2 bags)

"Soc jumps down from 70% to 20% at 8 in the morning and then jumps up again an hour or two later. "

This might well be morning loads and the sun hitting the array, with a voltage type SOC meter.

To be honest, a larger than 10% bounce in such a large battery bank might be indicitive of diminished battery capacity, (What others are calling a ripple, I think. I can get my large traction battery to do this with a relativly small load 1500 watts applied for 5 minutes, but my battery is half your size)

A shunt type meter just measures the amps in and out and can give you the current voltage but this isn't part of how it calculates the SOC.

Once you have such a meter, you would want to reach full charge. So doing an equalizing until the SG no longer rises, this should be uniform across the cells. If not go another hour or 2 to see if the Cell continues to rise, when it there is no improvment across the cells, I would record the SG for all the cells. If there is a single cell that remains low (I think more than .20 different from the rest) I would call Rolls for suggestions.

This is no fun when your also trying to live off the system. Hope this helps.

niel · November 2012

Re: Sunny Island/Rolls 530-S charging question

photowhit,
i think we agree on the batteries as suspect, but ripple is the artificial induced ac (actually pulsed dc) by a huge variable load to an inverter that is straining the batteries which would be undersized for the loads for whatever reason. sulfation can lower the battery bank capacity that would create large ripples under large loads and throw things off, but i think we agree that reduced capacity alone may be a problem even if not severe enough to cause the ripples to be too large for the inverter.

palausystem · November 2012

Re: Sunny Island/Rolls 530-S charging question

The bit about the Rolls absorption time formula - that was not about EQ, just about calculating the absorption time in general for a return to full charge. It starts by saying that Bulk brings the battery to roughly 80% soc and the other 20% will come from the absorption charge time, and then how to calculate that time.
And as you suggested earlier, with my 3K array it would take 8.4 hrs of absorption by that formula, which the sun ain't gonna do. So his dod comment still baffles.
When I did the EQ on each string separately, it went pretty well. But I will check out the charger you mentioned.Yes, I could use one string for loads and then Zap the other.

So if skipping bulk/absorb for a few days is not good on an off-grid, what is the best scenario? Full charge it every day? Why does the Sunny Island charging scenario make it better for grid-tied?

I do have the shunt,tri-metric, but it's also based on Ah capacity...but I think it might have a setting for checking cumulative Ah in and out, not sure...the hydrometer will be coming out more often now.

What is the 0.52 efficiency figure? Is that temperature compensation and inverter efficiency and...?

"yes, do be sure to give the batteries a good wallop from both sources and be sure of your water levels too with distilled water. can you do this one string at a time and still power your normal loads from another source? "

That's a good question. Last time I EQ'd one string at a time they received the 30amps that Rolls said was a good number for that battery, and powered our normal loads ...and they got almost too hot. So I think it'll be OK.

So as far as the 12v Iota charger goes, how would you do the EQ with that? Two batts at a time I guess? The maximum amp safe-limit for the S530 is 44. I think Rolls actually recommended 30 amps for EQ So would I get a 12v 75 amp charger and do two 6v batteries at once? Would that work?

Thanks

Sunny Island/Rolls 530-S charging question

Comments

Categories