Batteries insufficient?

TwoJolly · July 2020

I’m having what I think are battery issues with my first solar system setup that I did last Fall. It’s a small 12V system to power a TV (AC) and a portable Dish satellite receiver at our off-grid ranch in Northern Arizona. I tried to keep my costs down since this is only for entertainment and we’re not there full time so the batteries sit unattended in a hot building while we’re away. The system consists of a Morningstar TriStar controller (TS-MPPT-30), one Canadian Solar 305 watt panel and two Costco Interstate GC2 6V golf cart batteries. The TV and satellite gear only draw about 70W (based on Kill-A-Watt meter reading) so I started with a Samlex 150W inverter (PST-150-12) so I could get more hours from my batteries – wanted to draw the minimum I needed. All went well through this winter – could watch TV for hours. But now that it’s hotter, the inverter needed to run its fan more and the Samlex started setting off its alarm. Continuous alarm – presumably overload since it would stop if I shut off the TV screen while the satellite still ran. So I bought a Cotek 200W inverter (SE200-112) to just up the power a bit. Just started using it this week and quickly hit issues. With the Cotek, it starts throwing a low voltage alarm within about an hour of use. So I pulled out my little Ideal voltmeter to see what’s going on. Here’s the scenario:

Battery voltage prior to starting inverter (we're running the TV at night after a full day of charging): 13.1V

Battery voltage as inverter starts running: 11.6V

Inverter alerts start when battery voltage shows 11.1V; frequency of alerts increases as voltage drops

Battery voltage after about an hour of TV but once inverter has been shut off: 12.8V

So now I’m wondering if my decision to go on the cheap with the Costco batteries is my problem. I really don’t want to invest in expensive batteries that will be brutalized by our hot building when we’re away. I could switch to an RV DC TV, like a Jensen, since the TV portion of the system takes about 55W – but they’re expensive for the bigger screens – so I’d want to be sure this is the best path before doing that. We're here less than half time so I'd really like to stick with lower costs. Any insights into what’s likely going on here would be much appreciated. Thanks!

softdown · July 2020

Sounds like the hot building destroyed the batteries though I'm surprised it happened that fast. Every 10 degrees F shortens battery life about 50%. Rather amazing statistic.

Perhaps the batteries suffered extended periods with low voltage (below 12.0 volts) somewhere in the past few months. Sulfation also kills batteries.

Take the batteries back while you can. Keep the next pair as cool as possible. Preferably under 75F - somehow.

Everybody kills their first set of batteries. Much of the reason for the expense of solar batteries.

The Costco batteries were never the issue.

Photowhit · July 2020

I would tend to agree with Softdown, Sounds like the batteries are toast. and I also doubt it's the Costco batteries. Most of this I get from the battery voltage returning to 12.8 after the inverter has been turned off, this would indicate minimal capacity but there are other things that could cause this. Some questions below may add clarity.

So have you been checking the electrolyte?
Has it always been topped off?
Never had plates exposed?
What size wire goes to your inverter? and how far is the inverter from the batteries? with a 70 watt load plus another 15-20 watts running the inverter You are drawing over 7 amps from the batteries, If you have a long distance to the inverter, you will have some voltage drop over small wires. If you are using 14 gauge house wire over 50 feet the voltage drop for a 7 amp load goes from 12 volts to 10.23 volts.

Once the load is removed the voltage would go back up, even measured at the inverter as the load would be reduced. This could also reflect the voltage drop and return. The results would be as you checked them, the alarm sounds when the TV was on, but when turned off (load removed) the alarm quit. With a 2-3 amp load (Sat and inverter) the drop is only 1/2 - 3/4 of a volt.

Also how is the inverter connected to the battery? jumper cable like pinchers are a pretty poor connections, all connections should be bolted down and checked. A loose connection can play havoc with tracking down problems. Be sure to check the connections as heating and cooling will loosen aluminum and copper connection.

mike95490 · July 2020

I think 300w is marginal for your loads and battery size, and that you have been undercharging the batteries for a long time.

You never say for how long you consume 70W from the system . 70w is only half load for a 150w inverter, heat should not be a issue.
70w for 6 hours is 420wh, plus overhead, so say 600wh consumed. Any lights you you will add to that , A larger inverter will have larger overhead losses.

300w panel is actually only 240w when in the field, x 5 hours is 1,200watt hours. Could be enough if you get solid 240w all day but the sun moves out of alignment, so call it half. now you are at 600wh. Just barely making it..
Any daytime loads, rob your batteries of recharge. Recharge takes 1.2x what you took out because of losses, so you really need 720wh to fully recharge.
( with 5,000w of panels, I can harvest about 12kwh in long days of summer - not as great as it would appear, batteries only take what can can absorb in a 6 hour charge cycle.)

TwoJolly · July 2020

Thanks everyone for your input. Here's my responses to your questions and an additional thought:

Re Photowhit’s questions:

*So have you been checking the electrolyte? Yes, once a month

*Has it always been topped off? Yes, most of my checks have also included topping off

*Never had plates exposed? No

*What size wire goes to your inverter? and how far is the inverter from the batteries? Also how is the inverter connected to the battery.

I used the alligator clip style wires from the Samlex with the Cotek since I disconnect the inverter when we leave (since nothing is using power while we’re gone) - that style has been easier to deal with. These have 16 gauge wire and with 200W at 12V, a chart I found on The Inverter Store indicated that that wire size should be sufficient. They are about 3 feet long. But from what you’ve indicated, sounds like I need to switch over to regular ring-type connectors and I’ll also increase the gauge just to help.

Re Mike95490’s points:

*how long you consume 70W from the system; Typically a few hours each night for the one or two weeks that we are there

An additional thought since the Costco batteries aren’t being flagged as insufficient – I’ve never run an equalization cycle on them because I haven’t setup the custom voltages for the charge stages that the Costco/Interstate batteries would require. The standard voltage setting that I selected on the Tristar controller via dipswitch was:

Absorp Float Equal

14.7V 13.5V 15.4V

Interstate told me the Costco batteries need:

Absorp Float Equal

14.8V 13.2V 15.6V

So maybe I need to do the custom settings that the Tristar supposedly supports and then do an equalization? The settings might require a laptop connection with the management software – can I just equalize using the standard setting that I’ve selected, even though the voltage is lower than Interstate says I should use. Would that be more beneficial to equalize with the existing settings versus waiting until I get the custom voltages all set up?

Thanks again for your help!!

mike95490 · July 2020

By all means, use the standard and EQ those batteries, they have been chronically undercharged and may be tough to recover. You may need to run a couple absorb cycles to get them to the point where they will support being EQ'd.

Photowhit · July 2020

I do worry about your batteries, but don't think you are drastically undersized. You live in the sunny southwest, so longer and more consistent sun. 300 watts, should be pretty good, I used a 220 watt array in the Midwest, with 2 GC batteries. They survived 5-6 years. I ran a laptop which drew 50-70 watts, 120 watt adapter. Used it 2-4 hours a night, and a couple 8-10watt fans in the summer.

General guidelines would be an array able to provide 10-13% of the battery capacity for daily use or 20-28 amps charging say 14 volts. So with a 75% derating 225 watts/14 = @16amps + a sunnier environment.

I take it you also have utility AC? or are you surviving without air conditioning?

TwoJolly · July 2020

Thank you both for your replies. I'll do an equalization tomorrow and see how things are after that. And I'll get a better wiring setup going.

Re the question on AC - I wish! This is really just a recreational camp for us - had it for 25 years but never took it to the full level of a "city" house - just a bunkhouse to stay and recreate in the wilderness basically. An electrician friend put in our first solar that runs RV lights only in 2015 - uses the same kind of Costco batteries and has lasted this long with no equalization because it's just on a low-end controller that won't do one. But it doesn't have an inverter so the usage is no doubt so small that the batteries continue to do fine. The system for the TV was my first attempt at doing my own solar system, after much reading. Having never run an inverter this is my first time seeing the difference that makes with a system.

Sounds like I should get the custom settings going to correct the chronic undercharging. I'll get to work on that after the equalization.

Thanks again for all the input!

Ralph Day · July 2020

Don't just do an eq charge without getting them up to full SOC first. All you'll do is over charge and possibly damage the batteries even further. EQ should be done to fully charged (or as fully charged as you can get them) batteries.

NANOcontrol · July 2020

I had a 70W deice I had to run from an inverter. I had a number of small inverters I picked up as I would buy damaged lots and fix them. I tried a number of them and 400W inverter had the best efficiency. Smaller inverters just are not designed for efficiency, just cost.

Aguarancher · July 2020

NANOcontrol said:

Smaller inverters just are not designed for efficiency, just cost.

That's not true. This inverter is very efficient and has extremely low self consumption. https://www.solar-electric.com/morningstar-si-300-115v-ul-inverter.html

Photowhit · July 2020

A key in learning about your battery would be to use a hydrometer. You can often buy a glass one at an auto parts store for $5-10, you will want to understand temperature adjustments. Or you could buy one like the Hydrovolt which does the adjustments automatically. Be sure to have plenty of distilled water on hand to clean out you hydrometer and top off cells.

MidNite Solar Hydrovolt Battery Hydrometer

MidNite Solar Hydrovolt Battery Hydrometer

https://www.solar-electric.com/midnite-solar-battery-hydrometer.html

https://www.youtube.com/watch?v=BkUBffSv2O0

NANOcontrol · July 2020

Nearly $300 for a 300W inverter, it ought to be special.

TwoJolly · July 2020

Good timing on your tip on getting a hydrometer - I read last night that I need one of those to do an equalization! Thanks!!

wellbuilt · August 2020

I second the morning star 300 watt inverter , I’ve had one running for 5 years now , I never even look at it it just makes power . And there is no fan .

I use the 215 ah Duracell batteries and I get 5/6 years out of them .

It seams like they like to be charged I absorb at 59 volts summer and 59.5 winter plus temp compensation .

I would add a second panel to the system , I don’t think you have enough power to run loads and charge

littleharbor2 · August 2020

Two questions.
1, Do you shut down the inverter when not needed?
2, With the light gauge wire from inverter to battery, is there any corrosion visible on the wire?

New_Mexico_Will · September 2020

Sounds like your main issue has not really been addressed, which is keeping those batteries cooler. If you can, I'd suggest burying them in an insulated box. Of course, you'll need to have a vent tube to the surface for FLA batteries, and you'll need an easy way to get to them once a month. Once you get below the surface, the temperature drops pretty dramatically. It's not the prettiest or most convenient thing to do, but your batteries will be happier. There are other chemistries, like Lithium Titanate, that are much more tolerant of high temperatures, but you wallet won't like them much. If you do bury them, you'll need to use larger gauge wire for the extra distance.

Wheelman55 · September 2020

One of our local installers came up with a simple and effective solution for the heat. Location is 29th parallel, just a few miles north of the Rio Grande in TX. Temps are over 100 F April through September.

Dig a pit in the shade, line the pit so it will hold water, put the batteries in the lined pit, add water to close to the top of the batteries, loosely cover. The cover needs to be above the batteries enough to allow evaporation. The batteries never get above 85 F using this water bath. You’ll need to add water occasionally.

Batteries are lasting much longer.

Best of luck...

garynappi · December 2020

softdown said:

Sounds like the hot building destroyed the batteries though I'm surprised it happened that fast. Every 10 degrees F shortens battery life about 50%. Rather amazing statistic.

Perhaps the batteries suffered extended periods with low voltage (below 12.0 volts) somewhere in the past few months. Sulfation also kills batteries.

Take the batteries back while you can. Keep the next pair as cool as possible. Preferably under 75F - somehow.

Everybody kills their first set of batteries. Much of the reason for the expense of solar batteries.

The Costco batteries were never the issue.

I've always wondered about this temperature statistic. My car batteries clearly experience more than 100 degrees for extended periods of time while my truck is parked in our hot Florida sun running cameras and a RANGE controller for my engine and clearly more than that while running hour after hour driving and they last 4 years plus.

Is there some explanation for this that I can wrap my head around?

mcgivor · December 2020

The likely explanation is a car battery for the most part has a very easy life in comparison to a deeply cycled battery, all it has to do is provide a large current demand on startup, then be immediately recharged. The construction of a car battery is significantly different having more but thinner plates than that of a deep cycle equivalent of equal capacity, therefore it's not a fair comparison.

Most automotive batteries in the tropics have a life expectancy of ~3 years where high.ambient temperatures only add to the under hood gains generated from the engine itself. Lead acid's enemy is temperature, the deep cycle ones I've used begin to decline at 2 years rapidly declining thereafter, it's one thing having high temperatures for a few weeks or months, but having temperatures average >30°C all year is very different.

BB. · December 2020

This explains a bit of the history and the Arrhenius equation/Activation Energy (which is fairly constant across many materials):

https://www.electronics-cooling.com/2017/08/10c-increase-temperature-really-reduce-life-electronics-half/

For me, in one of my Calculus classes, we derived the +10C -> 1/2 life as one of the examples of Calculus in "real life" usage.

There are other failure modes too... HALT (Highly Accelirated Life Testing) is used in engineering to qualify designs:

https://en.wikipedia.org/wiki/Highly_accelerated_life_test

From my own past life--I did disk drive testing/qualification for our company (back in the day when 5 1/4" disk drives (and various new tech tape drives) were a new thing and heald 10-100 MBytes or so)... I set up a chamber to cycle between 5C and 40 C, twice a day with an ~1 hour ramp of temperature change... Everything was within the printed operating specifications of the drive. At the same time, ran drive tests (seek/read/write) test to both stress the drives and show they still worked.

More or less, after 2 weeks of testing, a majority of the drives would fail (could be anything, drives would write cold, could not read hot, surface mount components with pop of the boards, flex cables to read/write heads would fail, etc.). If a drive passed 2 weeks of testing, I rarely had any fail after even another 2 weeks of tests.

In engineering labs, they would do HALT (and similar testing) to distruction to figure out what would fail and need redesign for improved product life.

Remember--For batteries--There are different values for what a failure is... Concorde AGM batteries (as I recall) used 20% loss of capacity as value for warranty failure.

In your car, the batteries are assumed to be only cycled for 15% discharge (shallow discharge)--So you could have 50% loss of capacity and the battery still is "fit" your your needs in Florida...Take that battery to a deep cold climate in Canada/Northern USA in winter), and you will probably find the battery could not reliably crank start a cold soaked motor (chemical reactions being dramatically reduced in -40C/F weather--Also temperature cycling causing cracks/failure of materials--And as batteries sulfate, the specific gravity falls (electrolyte becomes more "water like" leading to possible battery freezing in cold snap where a fully charged newer battery with high SG would not)

And we had folks in Canada that could get 12+ years from their battery banks that cold soak unused in winter (stored fully charged, perhaps with a small "floating solar array") where we would expect 5-7 year life in the lower 48 states.

-Bill

Batteries insufficient?

Comments