trouble shooting system design

mtjag

I installed an off-grid system this past March and it has been doing well for the past several months until November. It is colder here now and the infloor heat is running much of the night. I am not getting enough power to recharge my batteries on any given day, despite having completely clear skies with plenty of sun. I did reposition my array from south east to due south now, and it does help, but still not getting enough amps to charge the batteries completely. My setup is a 48v system and is comprised of the following components:
12-ES190 Evergreen panels 32.8 voc-pole mount non-auto tracking wired in 3's
1-Outback MX60 PV Charge controller 60a MPPT, 12-60 vdc. 60a output
1-Outback VFX3648 3.6kw, 120vac output
16-8L16 Deka 6v, 370ah lead acid batteries
The panels are 39' from the inverter using #4 wire, the battery box is under the inverter.

My property is south facing in south central Colorado at an altitude of 8,600 ft. We have plenty of sunshine year round except for snow days.

The biggest draw on my system is my 1hp well pump which I manually turn on and off. It runs on a Xantrex Transformer.

I use LED lights through out the house, I have a 28 cuft refridgerator, flat screen tv, and lap top. We have propane fired in floor heat that uses three 1/4 hp pumps, and my water pressure pump is 1/2 hp.
My cooking range, heat, dryer are all propane gas fired. Our house is about 2000 sq ft. with two people most of the time.
I also have a 12kw propane gas generator as backup.

My charge controller on average shows me a reading of around 54 volts for the high of the day and 47 volts for the low of the day. However, recently, it shows a low of 43, so I have to run my generator to get enough amps to get to 58.8 volts to reach a float point I was told by Outback to set . It is set to do that for four hours.
I EQed my battery bank on the 12th of November.

My questions are:
1) do I need more solar panels?
2) are my batteries a match for my solar panels
3) is it resonable to expect to have to run my generator every other day to reach the float point?
4) Is there something obviously wrong with my setup that I can't see?
5) am I way off or just some minor adjustments
I would appreciate any help.

Thanks much.

hillbilly

Re: trouble shooting system design

What are your loads: any ideas on total KWH a day you're using? Winter time is pretty tricky to keep batteries fully charged without a generator, even with more PV. You probably would do well to add some more PV panels, but I think you'll still have to run the generator some to get the batteries up to full charge. We usually run our generator once or twice a week during December>February, and 2-3 times a month during November, March and April.
Without a more firm number on your loads, it's hard to give real specific numbers... so that would be a help to know.

niel

Re: trouble shooting system design

it's possible you are using more than is being generated for the time of the year as we here in the northern hemisphere suffer fewer hours of full sun, but i can't be sure without hard kwh data both in and out of the batteries. maybe a battery monitor and a kill-a-watt meter are in order to track power generated and used. i also am not sure if the pvs are the ones i think they are as you list the voc at 32.8v and i see the es b 190 listed as 21.5v voc.
http://www.evergreensolar.com/upload/MAY%202009%20NEW%20LITERATURE/ES-B/ES-B_US_010809_Lo.pdf
going by the above pv specs you are about at a 6% charge rate with 4x 10.92a = 43.68a. also, 43.68a / 720ah = 6.067% which is fine by itself, but with high loads on the system this could be drawn down to the point of not charging the batteries properly or at all if low enough. how long did it take you to recharge the batteries? if not very long then it may just be a matter of more pvs or supplementing every few days with the generator. be sure of your batteries' condition as in specific gravities and voltages. check all connections, wires, etc. and verify each string's output to rule out problems with any pvs.

mtjag

Re: trouble shooting system design

Sorry, you are right about the voc on the panels, they are 21.5v.
Since we have had snow for the past two days, I ran the generator 8 hours yesterday to get them charged, based on the controller (that was all the way up from 43). I ran the generator today for 2 hrs this morning (it was showing 49.7 this morning) and another hour this evening and again according to the charge controller, the batteries are fully charged.

mtjag

Re: trouble shooting system design

hey thanks for the input. I have ordered a kill-a-watt to understand my loads better and it should be here in a couple of days. After running my genset 8 hrs yesterday and another couple today, it appears the batteries are fully charged. By the way, we had snow the last two days, all day. Also, we have had about five days in a row with the below zero temps so my in floor heat pumps have been running alot. So, tomorrow, it should be sunny and then we will see how the batteries hold up. It does seem your accessment is right in that I will need to run my genset more frequently for the next couple of months.
Thank you again for your time and comments. If I do end up with another three panels, would you suggest positioning them fixed toward the south or slightly east as we have full exposure toward the south, east, and west?

BB.

Re: trouble shooting system design

You can use the PV Watts Website to estimate your system's output:

12*190w=2,280 watts = 2.28 kW of panels
Derating Factor=.77 (panel+charge controller eff) * 0.80 battery * 0.85 inverter eff = 0.52
Location=Alamosa, CO

"Station Identification"
"City:","Alamosa"
"State:","Colorado"
"Lat (deg N):", 37.45
"Long (deg W):", 105.87
"Elev (m): ", 2297
"PV System Specifications"
"DC Rating:"," 2.3 kW"
"DC to AC Derate Factor:"," 0.520"
"AC Rating:"," 1.2 kW"
"Array Type: Fixed Tilt"
"Array Tilt:"," 37.4"
"Array Azimuth:","180.0"

"Energy Specifications"
"Cost of Electricity:"," 8.4 cents/kWh"

"Results"
"Month", "Solar Radiation (kWh/m^2/day)", "AC Energy (kWh)", "Energy Value ($)"
1, 5.66, 211, 17.72
2, 5.97, 199, 16.72
3, 6.56, 236, 19.82
4, 7.08, 239, 20.08
5, 6.70, 226, 18.98
6, 6.74, 212, 17.81
7, 6.44, 207, 17.39
8, 6.67, 217, 18.23
9, 6.57, 211, 17.72
10, 6.56, 225, 18.90
11, 5.60, 197, 16.55
12, 5.28, 197, 16.55
"Year", 6.32, 2578, 216.55

So, right now in November with a fixed array pointing south, you should be getting 197 kWhrs per month or:

197 kWhrs per month / 30 days = 6.6 kWhrs per day = 6,600 Watt*Hours per day

One thing I did not see you type about was a Battery Monitor... I really like these to quickly identify the state of charge of your battery bank.

If you have flooded cell batteries, a good hydrometer with thermometer should be used too.

-Bill

blackswan555

Re: trouble shooting system design

MX, Do you have a bat temp sensor on it ? does it reach / end its absorb cycle, How long have you got set for absorb ?

Pumps, Have a look at these, http://net.grundfos.com/doc/webnet/poweredby/gpu/US/alpha2.html

Have a good one
Tim

icarus

Re: trouble shooting system design

I would consider adding a battery monitor, like the Bogart Tri-metric. I second the notion of getting a kill-a-watt. With no way of knowing how much you are putting in/taking out it is pretty hard to know where you are. Another question is are your hot water heating pumps the right size? 1/4 hp is fairly big for a circ. system,,especially three of them. If you use all three zones, might you be better of running one pump that is bigger?

Tony

sawmill

Re: trouble shooting system design

I agree with Tony about the size of your floor cir pumps. Three 1/4 hp pumps is quite a load overnight on a system. The Grunfos Alpha that Blackswan linked sounds great for floor systems and renewable applications.

I use floor heat that is powered by one Taco 005 1/35 hp, uses 60 watts when running. Even that 60 watt load is quite a strain on the system with these short cloudy days. You also might want to consider using one pump (propertly sized) and valves to control heat to different zones rather than turning a pump on or off for each zone.

Just a thought, might not be suitable for your situation.

Bill

blackswan555

Re: trouble shooting system design

Always nice as-well ( thanks Peterako) A bit of in the field usage and real user report`s
http://forum.solar-electric.com/showthread.php?t=6505

Have a good one
Tim

jeffkruse

Re: trouble shooting system design

What is/has been the SG of the batteries? I have the same batteries and it takes a lot of power to keep them "fully" charged. I have my absorb Voltage set to 61.5V and that seems to work. Our systems are similar.

hillbilly

Re: trouble shooting system design

jeffkruse wrote: »

What is/has been the SG of the batteries? I have the same batteries and it takes a lot of power to keep them "fully" charged. I have my absorb Voltage set to 61.5V and that seems to work. Our systems are similar.

Not to be too critical, but don't you think it might be a bit early to say that an absorb voltage of 61.5V "seems to work"? Didn't you state in another thread that your batteries are almost brand new? Perhaps you've found something that works well, I'm not trying to slam your method here... just not sure that I'd recommend quite such an aggressive charging strategy to someone else without a bit more history to go on. You are very correct that it does take a surprising amount of power to charge a 740AH, 48V battery bank...

Solar Guppy

Re: trouble shooting system design

Look at this another way ... you have a 37kWh battery bank that you having to run the generator 8 hours a day to get them charged ... looks pretty close to 70-80% discharge to me.

This indicates your using about 30kWh/day, which is not surprising with the constant pump loads

As BB showed, your generation is about 6kWh day from the solar so if your looking to run generator free, you need 5X more solar than you have now.

A battery monitor is a must and don't wait till the battery's are so run down before firing up the generator ... the way you are cycling them now I would be surprised if they last more than a few months. Battery's are rated in number of cycles and depth of discharge. The more discharged, the less amount of cycles .. your probably in the 2-300 range of cycle at your discharge levels daily.

icarus

Re: trouble shooting system design

With all due respect to the OP and his problem, this thread illustrates a couple of issues that seem to recur.

The first is people have spent tens of thousands of dollars on battery based systems, but they haven't spent a few hundred on proper battery monitoring hardware. Without knowing how much energy is going in/going out on some average daily basis, one has no way of knowing how their system is performing.

Second, and perhaps more importantly, retailers should have more responsibility to, if not educate owners, but at the very least, point them in the direction of how to educate themselves on the systems that they buy. I have this vision of some PV salesman saying "you'll be able to do this and this and this" and the sun will magically fill the batteries up every day" The reality is (as most here know) it is much more complex than that.

Most folks are woefully uneducated when it comes to batteries and battery technology(ies). How many times have we linked the two major battery info pages?

http://www.windsun.com/Batteries/Battery_FAQ.htm#Lifespan%20of%20Batteries
http://www.batteryfaq.org/

My point here, is that I often accuse folks of "Ready, fire, aim". and to some extent folks are often guilty of that but; I also think that retailers/installers can have a tendency to sell hardware because they have a customer who is willing to fire, without really asking them if they have aimed.

Tonmy

hillbilly

Re: trouble shooting system design

icarus wrote: »

With all due respect to the OP and his problem, this thread illustrates a couple of issues that seem to recur...

...My point here, is that I often accuse folks of "Ready, fire, aim". and to some extent folks are often guilty of that...

Tonmy

Hey, I resemble that remark :-)
To the OP, we're all still learning here. There are (at least) two big reasons I hang around here:
1) lots of great information to be had here, I learn a ton from the folks here who have a vast amount of collective experience
2) I hope that I can help people to not suffer some of the same errors or confusion I did in the beginning.

Tony is right on that the industry as a whole could do a LOT better in regards to customer education. One of the main reasons that I like to support NAWS when I can, this forum represents what I think may be one of the best examples out there of how a retail store can help to better educate the public (customers, potential customers, and even competitors). I absolutely HATE dealing with a sales person who clearly knows little about what they are selling (particularly with something like PV systems). I have dealt with sales staff who were almost as bad as what Tony jokingly described.

Solar Guppy

Re: trouble shooting system design

icarus wrote: »

Second, and perhaps more importantly, retailers should have more responsibility to, if not educate owners, but at the very least, point them in the direction of how to educate themselves on the systems that they buy. I have this vision of some PV salesman saying "you'll be able to do this and this and this" and the sun will magically fill the batteries up every day" The reality is (as most here know) it is much more complex than that.
Tonmy

The issue is a majority shop the absolute lowest cost. Explaining how to properly design a power plant is not a realistic expectation to put on any retailer. To have the knowledge is one thing, to expect thousands of retailers to have this is unrealistic, at best.

This is why more complex system require permits, licenses and professionals to make sure a job is done right. If someone wishes to do it themselves, they have the same right to destroy the equipment they bought. RE is not any different than say plumbing or HVAC.

Can anyone name another trade that offers free how toos? try asking a HVAC company to show you how to do schedule J's, installation best practices and then give you wholesale pricing on parts ... RE is not special in any way, its just another way to do things, not less, nothing more.

icarus

Re: trouble shooting system design

SG,

I agree that way too many folks shop just for price. I also know that DIYers are prone to making mistakes. That said, I also think that retailers/installers also tend to over sell. (Not unique to this industry!)

I also agree that end users bear the brunt of the responsibility to "RTFM" and understand what their system can (and cannot) do, as well as have some inkling as to how it does it, but it is one thing to buy a big grid tie system figuring that you will eliminate your utility bill only to discover that it only reduces it by 1/2 for a variety of reasons, it is another to buy a big battery based system, to discover that it doesn't cut it, for the same reasons, and then be faced with a pile of lead scrap that may cost many thousands to replace.

Buyer beware, and buy from up front, ethical sellers, and you are right, few industries offer free "how to's". The worst probably being car dealers who not only don't know their product make it a matter of pride to sell snake oil, and make real money at every turn of every transaction. Heads you win, tails I loose!

T

mtjag

Re: trouble shooting system design

All,
Thanks much for the input. It is refreshing to hear from those that have real "hands-on" knowledge. First, let me clarify a couple of points. My batteries are not getting down that low (43) every night. This has happened about two times in the past two months. I ran my genset 8 hours a couple of days ago because it was down so low. According to my charge controller, I am getting about 10 kwhs a day on sunny days. This past summer I got up to 16 kwh.
I do suspect part of the equation has to do with the pumps for my infloor heat and I certainly appreciate the link to the Grundfos Alpha.
I am trying to understand the whole issue around properly charging my bank of batteries, but can't seem to get it figured out.
The 16 Deka 8L16 370ah show that the charge is 2.35VPC and the float is 2.25VPC. So, how do I use this information to set my float point in my charge controller?
I am waiting for the delivery of the Kill-A-Meter so I can realistically determine my usage. Until then, I will be running my genset more to keep my batteries up.

hillbilly

Re: trouble shooting system design

As for your loads, the Kill-a-watt meter is a handy little tool for any plug in devices, but it won't measure all of your loads. I find a battery monitor like the trimetric to be a very helpful to be able to measure the total sum of all loads at once. The battery monitor has a lot of other helpful info too, so I think that they are well worth the investment. It also has let me know more than once that I had a load on somewhere when I was not aware that I had anything turned on. The Kill a watt is cheap enough though, and has some other handy features so I think that it's worth getting both.

As for those voltages they are volts per cell, with each cell being 2 volt nominal, thus you have 6 cells for a 12V system or 24 cells in your system. So those setpoints would be 56.4V and 54 for a 48V nominal system. I hope that answers your question.

niel

Re: trouble shooting system design

:roll:let's see, 3 cells per battery x 8 batteries in series is 24 cells.
2.35v x 24 = 56.4v
2.25v x 24 = 54.0v


tony,
as to this,
"Second, and perhaps more importantly, retailers should have more responsibility to, if not educate owners, but at the very least, point them in the direction of how to educate themselves on the systems that they buy. I have this vision of some PV salesman saying "you'll be able to do this and this and this" and the sun will magically fill the batteries up every day" The reality is (as most here know) it is much more complex than that."

why do you think this forum exists? windsun saw the need and you, i, and many others here are part of that solution. we need some smilies shown to doing hand holding;):p

icarus

Re: trouble shooting system design

Neil,

How many folks come here AFTER they have spent a ton of money on a system that doesn't perform the way they expected it to? If every one who is considering buying/building a PV system came here first,t here would be a whole lot fewer cases of disappointment brought on by "Ready, fire Aim!"

Tony

niel

Re: trouble shooting system design

icarus wrote: »

Neil,

How many folks come here AFTER they have spent a ton of money on a system that doesn't perform the way they expected it to? If every one who is considering buying/building a PV system came here first,t here would be a whole lot fewer cases of disappointment brought on by "Ready, fire Aim!"

Tony

you are right that this isn't good to see happen, but how many of those same people might buy a car without knowing how to drive (or drive well) and they don't tell you how to drive or all you may need to know to maintain it, let alone how to fix it? the problem in solar is that they often aren't aware of the help available to them, for whatever reason, so they can make a better choice. many drive blindly without knowing for sure what they are doing and many wing it quite well, but many make mistakes. such is life.

blackswan555

Re: trouble shooting system design

I have to run my generator to get enough amps to get to 58.8 volts to reach a float point I was told by Outback to set . It is set to do that for four hours.

Please explain the above a little more and what you have the MX set to for absorb time & V, But running float from a generator is a waste of time and fuel in my opinion,

I EQed my battery bank on the 12th of November

.
did you charge the bank FULLY before EQ`ing ?


Have a good one
Tim

mtjag

Re: trouble shooting system design

Outback told me to set my absorb to 58.8v and my float to 56.4v. According to the Deka my charge point should be 56.4 and my float should be 54.
3 cells per battery x 8 batteries in series is 24 cells.
2.35v x 24 = 56.4v
2.25v x 24 = 54.0v
So, I changed these settings this morning. I don't know how long to set my absorb or float points. Originally, Outback told me to set the absorb for 4 hours. A couple of weeks later when I called them they told me to change it to 2 hours. Then last week, they told me to change it to 4 hours, so I am completely confused about the time. Currently, it is set to 4 hours on the Mate for the genset and 2 hours for the charge controller....so really messed up.
Where should I set either or both of these in terms of time to absorb and float?

Regarding the EQing, I did not charge up the batteries completely before EQing last time. I intend to EQ tomorrow, so I guess I need to charge completely before the EQ?
I was gone yesterday, but we had sun, so when I got home after dark, the system showed 51.8v. I ran my genset for one hour after that. This morning, the system showed 49.0v. I ran my genset for one hour this morning and it was at 52.4V before I left again. The temp at my house over night was in the single digits, so I know my infloor head pumps ran alot last night.
I looked those pumps over and they are 115V, .71amps each so they are pulling 81.65 watts each....does that sound right? I have three of them on my system. I suspect this is a big part of my depletion. Your thoughts?

icarus

Re: trouble shooting system design

81 watts times 3 pumps times 24 hours =5.8 kwh!

If memory serves, you have ~2.3 kw of panels, which on an ideal day might yield ~4-6kwh of power. By my simple calcs, you are perpetually using more power than you are making, unless I am missing something.

T

hillbilly

Re: trouble shooting system design

mtjag wrote: »

Outback told me to set my absorb to 58.8v and my float to 56.4v. According to the Deka my charge point should be 56.4 and my float should be 54.
3 cells per battery x 8 batteries in series is 24 cells.
2.35v x 24 = 56.4v
2.25v x 24 = 54.0v
So, I changed these settings this morning. I don't know how long to set my absorb or float points. Originally, Outback told me to set the absorb for 4 hours. A couple of weeks later when I called them they told me to change it to 2 hours. Then last week, they told me to change it to 4 hours, so I am completely confused about the time. Currently, it is set to 4 hours on the Mate for the genset and 2 hours for the charge controller....so really messed up.
Where should I set either or both of these in terms of time to absorb and float?

I think this is a fair example of what Tony pointed out about poor help/education within the industry. Regarding what's right, or who's wrong here I don't really know as it seems everyone has some different strategies on how to charge batteries. Perhaps what might help you understand a bit better is to realize that all these strategies are good *depending on the specific situation*. For example I've had help offered on charging parameters from some experts who were used to dealing with a grid connected UPS back up battery; while some of this advice was very helpful, my stand alone off grid system has some very different demands on the batteries.

A really short simple way to try to mull all this together into some sort of baseline charging strategy (to be monitored and adjusted as needed) would be to consider your loads, and your charging capabilities: larger loads and/or less charging ability (think short winter solar hours) would probably need higher voltage settings... light loads, shallow discharge cycles, and long sunny days on a large PV array would dictate a lower voltage setting. This is rough, but gives you an idea where to start perhaps. Check your SG levels, water consumption, and watch your voltage levels on the charging and discharging to start to get a better handle on how well you're doing on keeping things charged... adjust voltage settings and times in small increments.
The voltage settings that Deka recommend sound a bit on the low side compared to what I am familiar with but they do know their batteries better than anyone else I would hope... Try calling Deka (East Penn) for some support, telling them as many specific details as you can and perhaps they might have more precise charging parameters to follow.

jeffkruse

Re: trouble shooting system design

This is all I got with my Deka's. 2.45*24= 58.8
What is the harm with setting your absorb time to max?

mtjag

Re: trouble shooting system design

according to my CC, I average about 10kwh per day. For instance, yesterday with full sunhine, my CC indicated 15.6kwh. Of course, I did run my genset for one hour

mtjag

Re: trouble shooting system design

that's a good question. I was thinking it would be better to lower the absorption time during the winter months with less sun intensity...

chevenstein

Re: trouble shooting system design

I second looking into the heating system. I have two Taco 007 circulators on a propane fired hydronic system in my off grid home and if I let the wood stove go out and the gas heat take the full heating load (ie: cycle on and off all night and day) I see my batteries drain quickly. Do you have trees nearby so you can burn wood? If you can go this route then you'll be able to shed a large portion of your winter daily electrical load.

Photowhit

Re: trouble shooting system design

I'm going to jump in the deep end though I only skimmed the replys by the very knowledgable people here.

I didn't hear where the batteries are stored, if in unheated area they will lose storage capacity.

Reading your voltage is a small window as to what is happening, your charge controler may well also have the current load?

If your checking your voltage and it's 44 volts at 8 in the morning, what is the current load? Someone sneaking bread into a toaster? someone nuking water for coffee? heavy loads will also greatly reduce the voltage level. It's cold in the morning, all those pumps running at once, TV on for the weather report, fridge catching up after people fixed breakfast, etc?

The 49.7 with a reasonable load is likely still 75-80% capacity. If you want to figure out the true state of the batteries from voltage, you would want to turn off all loads and charging for a period of time, maybe 2 hours! to let them settle out.

I think it is important to get a feel for this. It seams most people don't understand the readings from a meter like the tm2020. it measures the current in and out and can give an idea relative to the numbers you put in for capacity. I find most people expect the meter to be completely accurate, they are only accurate relative to the guide line you enter.

44 volts is scary, but you need to have a refence point. My batteries are at the end of their life, likely half their original capacity between being 4 year old 4-6v GC batteries, and that they live unheated. if I run the microwave (tiny 800 watt) I'd bet the system voltage would drop to @22volts short term.

I've gone to shutting down the system late at night, if we have more than a day of overcaste skys, I have a hard wired 2 watt outside light which is enough to keep the inverter running at 30watts (I'd like to wait until next summer to replace batteries) Last night I was some what surprised to see the voltage down to 24.4 with just a 23 watt light and my laptop on charger (30 inverter and a couple ghosts) my LEDS on the C60 showing just a bar above warning. when shut down I gained a bar and I came out to check (and water the lawn) and found 4 green bars indicating about 80% of capacity by voltage. This is a greater swing than I would expect with a newer set of batteries, but gives an idea of what I'm refering to.

Good Luck. Read the manual well to the trimetric!