Running a 115vac well pump, battery voltage is being pulled down tell inverter shuts off

suprasolar · July 2017

I will start with what my system consist of
9 canadian solar 270 watt panels wired 3 strings 3 panels in series
midnite classic 150 charge controller w/wiz bang jr. making a average of 65 amp to charge 24 volt battery bank
8 ub121100 110AH batteries series in pairs for 24 volts and then paralleled to a large copper buss bar with 00 cable making a 440ah bank
reliable 3000 watt 6000 peak inverter also wired with 00 cable

Ok so everything is good as far as charging, the problem I have is I am trying to run a hallmark 1hp deep well pump. The inverter has no problem starting the pump and running it as long a the sun is out and putting 30 or more amp to the battery bank. but if the charge controller is turned off then the battery voltage slowly drops (about 10 mins) untell the inverted alarms at about 21 volts and then shuts off at 19 to 20 volts. Once it shuts off it only takes about 5 seconds for the voltage to rise again to 25 volt and the inverter will try to restart but with in seconds drags the volt back down and shuts off again. The pump when running is drawing 8 amps a 120vac not possitive shut what the start draw is but seems to be around 25 amps but once again it has no problem starting and even running as long as the charge controller is putting a decent amount of power to charge the batteries, on the the input side of the inverter I am see a 55 amp draw when the pump is running.

So my question is how is the the voltage dropping so quickly without the controller putting amps to the batteries? The cables are plenty big and the batteries are rated at 60 amps each max draw so that should be 60 x 4. The batteries a only losing a few % soc before the inverter starts shutting off.

thanks for any ideas

Estragon · July 2017

I would start by checking a few more things.

1. Measure the voltages of each individual string under load. If any are low relative to others, check individual batteries.

2. Check and retorque all connections. If you have a laser thermometer, check to see if any connections are getting warm.

3. When the inverter kicks off at LBCO, see if any of the batteries are warm.

littleharbor2 · July 2017

How old are the batteries?

BB. · July 2017

You have AGM (sealed) batteries, so you cannot measure the specific gravity. That leaves measuring the voltage of each battery (under load, under charge, at rest) to see if you have "weak" cells or batteries--Or if the whole bank is behaving badly (you are looking for differences between batteries).

Your system seems to have "enough" solar panels (for the AH@voltage battery bank). And at ~65 amps peak charging current, it appears your solar panels+charge controller+battery bank is working correctly. Using our "rules of thumbs":

9 * 270 Watts (2,430 watt array) of panels * 0.77 panel+controller deratings * 1/28.8 volts charging = 65 amps typical maximum charging current
440 AH * 28.8 volts charging * 1/0.77 panel+controller derating * 0.13 rate of charge = 2,139 Watt "cost effective" maximum array

Your load (1 HP well pump+inverter) seems to be working correctly... And your battery bank is being drawn down at:

8 amps * 120 volts * 1/0.85 AC inverter eff = ~1,129 Watts (or a bit less)
1,129 Watts / 24 volts = 47 Amps
47 amps / 440 AH battery bank = 0.11 = 11% draw on battery bank

Then there is the question of how much energy per day you are drawing from your battery bank. Typically (for full time off grid) one would draw about 25% of capacity per day from the bank:

440 AH * 0.25 = 110 AH per day
440 AH * 24 volts * 0.85 AC inverter eff * 1/2 days storage * 0.50 max discharge = 2,244 WH per day average load (recommended)

Then there is the question about the amount of sun you receive... Say near Logan UT, ~48 degree array (from vertical), fixed:
http://www.solarelectricityhandbook.com/solar-irradiance.html

Logan Utah
Average Solar Insolation figures

Measured in kWh/m2/day onto a solar panel set at a 48° angle from vertical:
(For best year-round performance)

Jan	Feb	Mar	Apr	May	Jun
4.05	4.56	5.54	5.83	6.05	6.49
Jul	Aug	Sep	Oct	Nov	Dec
6.51	6.24	6.09	5.42	3.99	3.83

Right now, middle of June with ~6.49 hours of sun per day, a 2,430 Watt array would (long term average) supply about:

2,430 Watt array * 0.52 average off grid system eff * 6.49 hours of sun per day = 8,201 Watt*Hours per day (June)

That is about the maximum continuous power (8 amps @ 120 volts) I would like to see (basically C/8 discharge rate or ~12.5%). AGM batteries do handle high loads just fine, and your system should be able to run that load for 3-4 hours for 50% discharge.

It sounds like your batteries are starting to fail (sulfation?). Typically this happens because of:

over discharging (lots of deep cycles)
under charging / deficit charging (you simply are drawing a bit more energy from the batteries than your solar system can restore in a day)
Over charging (AGM/Sealed batteries when over charged can overheat/vent--And being sealed, you cannot add water back in).
Simply batteries are at cycle life. There are batteries designed to float with occasional usage (UPS type batteries) and others that are designed to "deep cycle". A UPS battery may deep cycle a few 10's to a 100's of times. A deep cycle battery may last 300-1,500 deep cycles.
There is also "aging" life... If a Lead Acid battery is simply left on float charge--A UPS Battery may last 2-3 years. A true deep cycle battery may last 3-5 years ("golf cart" type flooded cell) to 5-8 years (larger/better quality) to 15+ years ("fork lift/traction" type batteries).
Operating/storage temperature of batteries has a big effect on battery life too. At 25C (room temperature)--For every 10C over 25C room temperature, your battery bank will be 1/2 the life (i.e., at 45C, the battery "ages" 4x faster or 1/4 the life). And conversely, for every -10C under 25C, your battery will age 1/2 has fast).

So--What is the charging voltage for your bank (28.8 is "typical" for AGM) and you want to hold that charging voltage for ~2-6 hours per day (deeper discharge, longer "absorb" time).

I fear that your batteries are at end of life.... Whether that happened because of how you used time (see above), or they simply "wore out" from usage (UPS batteries may not support a large number of cycles).

You might want to try flooded cell "golf cart" batteries. 6 volts @ ~220 AH x 4 in series x 2 parallel string (8x batteries total). Typically flooded cell batteries are about 1/2 the cost of AGM/Sealed batteries, and you can use a hydrometer to measure the state of charge of the battery (specific gravity of the electrolyte).

AGMs are very nice lead acid batteries. They are much "cleaner" (no acid fumes, less corrosion of battery connections/cables), and more efficient. Plus you do not have to check water levels (and add distilled water every 1-3 months or so).

However, AGM tend to be unforgiving and have a bit shorter life than a "similar" quality flooded cell battery, you cannot check specific gravity, you cannot add water, they do not like high charging voltages.

How you wire the battery bank can affect the life of the batteries--Here is a good explanation:

http://www.smartgauge.co.uk/batt_con.html

But it sounds like you have heavy/short cables--So that is probably not an issue here.

-Bill

suprasolar · July 2017

The batteries were bought used and had been on a hospital lighting backup, they are 3 years old but basically had no cycles on them They show all signs of being healthy, 12.7 volts still showed on them for over a week without receiving a charge, They will run other things like a compressor I have that says it is a 15 amp motor for well over 10 mins only dragging the voltage down to 24 volts which then will go back up to 25 or more once it stops. I'm guessing the compressor is a brush motor not a inductive, but still from everything I've read this should work. I could see if after a hour it starts to have problems but this just seem so quick

suprasolar · July 2017

this pump is currently the only thing drawing from the battery bank other then a amp or less to run a few 24 volt fans set to run when charge controller is from a output single on the control to turn on a 25 amp ssr and yes it is a dc ssr

Dave Angelini · July 2017

Large loads on a 24 volt bank require 4/0 battery cables. You have 00 ?

littleharbor2 · July 2017

A 1 horsepower deep well pump has to have a heavy starting surge. As it pulls down the voltage on your battery bank the amperage required to start the pump will peak. Sounds like Your batteries, alone, aren't up to the task.

Estragon · July 2017

It could be something as simple as a connection that's worked a bit loose, and just causes problems at higher currents. I'd start there. IMHO if the batteries are okay, they should be able to handle the load fine.

Solray · July 2017

I'd try a smaller pump and see if it will run longer for you.

jonr · July 2017

If the actual, measured at the battery terminals voltage is dropping, then it's not a connection problem,

mike95490 · July 2017

1,129 Watts / 24 volts = 47 Amps
47 amps / 440 AH battery bank = 0.11 = 11% draw on battery bank

This does not allow for Power Factor, most pumps are 0.6, so 80-100A @ 24V might be expected

The batteries were bought used and had been on a hospital lighting backup, they are 3 years old but basically had no cycles on them

They are likely standby GEL batteries. They were replaced because their calendar life timed out. Standby batteries have about 100-200 deep cycles in them at best. Start shopping for batteries and/or a smaller pump, if your plumbing can work with a smaller pump, often the well fellow, just installs the pump with the best margins for him.

suprasolar · July 2017

http://files.upgi.com:8086/UPGFileService.svc/GetSpecSheet/D5751

These are the batteries there not gel and are capable of way more then 200 cycles based on the specs

The pump is drawing right around 55 amps dc through the inverter when running I've check with a cheap meter and a $600 fluke meter and it runs just fine as long and the controller is put 2/3 the need amps out to charge the batteries. I guess I really need to recheck the batteries but when I got them they all read a good 12.8 volts even after sitting for a week before I got them on a trickle charger.

If possible one or 2 batteries where going south would it do this and drag the whole system down? should I unwire them and let them sit for a day and check each one by its self? making sure they each read 12.7 to 12.8? simple I guess I'm not getting how 1 or 2 batteries with a let say 90ah capacity could do this. please keep the info coming

thanks all

mcgivor · July 2017

Batteries that have spent most over their life in float/standby can appear, by measuring voltage, to be in good condition, but voltage in itself is not an indication of capacity. Capacity can only be calculated by discharging with a known load over a given time whilst observing voltage, a time consuming exercise, this is the reason why standby batteries are often replaced according to time in service as routine maintenance, the cost of labor to test capacity would exceed the replacement cost.

If each monoblock were individually load tested, according to the specifications, it may reveal if there is in fact one or more affecting the bank as a whole, being that they are not new there is that possibility. Actually a capacity test should be conducted before stringing in series, especially, because a lower capacity battery will drag the higher one down to its level in a short period of time. Perhaps someone who deals strictly with AGM batteries can provide more insight @Marc Kurth is an excellent example, pay attention to his advice on the subject.

jonr · July 2017

AFAIK, fully charged battery voltage isn't a reliable indicator of battery AH capacity. I see that the above comment has been edited to reflect the same idea.

Solray · July 2017

3 year old batteries that you do not know the actual history of, just what you were told. I suspect that they are not at their peak of performance any more. Perhaps they were drained a couple of times when the power was out for a while. Perhaps they just wore out.
I'd see how they do with a smaller pump first and then see about getting a new set of batteries, not a different used set.

Estragon · July 2017

> @jonr said:
> If the actual, measured at the battery terminals voltage is dropping, then it's not a connection problem,

The voltage could be dropping because a bad connection and/or undersized wire means the battery is not only supplying current to run the load, but is also supplying current to heat wire and/or connections. As the voltage sags, current increases, which makes voltage sag even more, and so on.

This is why I suggested checking individual battery voltages and temps under load.

If you have the meter, checking current from each string of batteries would be a good idea also.

Dave Angelini · July 2017

What was your answer on post #7?

Solray · July 2017

The answer to post 7 was in post 1:
"8 ub121100 110AH batteries series in pairs for 24 volts and then paralleled to a large copper buss bar with 00 cable making a 440ah bank
reliable 3000 watt 6000 peak inverter also wired with 00 cable"

Horsefly · July 2017

I only have a 1/2 hp pump and a 310AH battery bank. If the pump starts before there is much solar (and if there were some loads the previous night), the battery voltage drops while the pump is running, and as far as I can tell my Schneider SW4024 issues a low voltage warning at 24.6V. The battery seems to never get below 24.5V, and the inverter stays on. Then the solar kicks in and the batteries are in float pretty quickly (noon to 2:00pm, depending on the clouds).

Dave Angelini · July 2017

Solray said:

The answer to post 7 was in post 1:
"8 ub121100 110AH batteries series in pairs for 24 volts and then paralleled to a large copper buss bar with 00 cable making a 440ah bank
reliable 3000 watt 6000 peak inverter also wired with 00 cable"

I would rather the OP answer, he may not know that 4/0 is typical and not know that 00 is not acceptable for large loads.

suprasolar · July 2017

00 or 2/0 is rated for 195 amps and that is continuous. I'm running 55 amps continuous with less the 2 feet in between each series of batteries and the inverter.

mcgivor post #14 I am thinking have the best answer for me, I work for a company that has computers with a bunch of small ups backups using

https://www.amazon.com/Power-Patrol-SLA1075-compatible-replacement/dp/B004KO2W1C
There are basically a smaller version of what I have, manufacture in the same plant.

Well I got curious today and went to where the batteries to be recycled are and started checking there voltage.......I was like wtf a bunch were reading around 12.8 volts. So I took a few that read good put a couple of 12 volt fans on them and watch them fail within mins to run a 1/2 amp load.

so now I'm confused and pissed off how do you measure these batteries state of charge by letting letting them sit?

Well in a couple of days I will be going up to my plot of land and checking these batteries after I put a load on them, so I can see which ones drop to 12 volts or under and then slowly rise back to 12.8 making you think there fully charged and good to go.

thanks everyone

bsolar · July 2017

you measure them really with a carbon pile load tester .. batteries can be funny .. but the bottom line in your case is you shouldnt try to run a 1000+w load on batteries .. hate to scold you, but what we're doing is 'solar' power .. not battery power .. if you want a 'battery' system visit the local battery supply shop and invest in about 10 grand worth of batteries, make a huge bank, and get the shop charger from hell to keep them charged, what do you want with 'solar'? ... if you do want to have a solar system re-think your approach ...for example if you could fill a large tank during the day with your pump you could feed it to your house ect with a 12v rv pump that pulls about 30 watts ...

mike95490 · July 2017

Horsefly said:

..... Then the solar kicks in and the batteries are in float pretty quickly (noon to 2:00pm, depending on the clouds).

Batteries recharging quickly could indicate sulfated cells

mcgivor · July 2017

bsolar said:

you measure them really with a carbon pile load tester .. batteries can be funny .. but the bottom line in your case is you shouldnt try to run a 1000+w load on batteries .. hate to scold you, but what we're doing is 'solar' power .. not battery power .. if you want a 'battery' system visit the local battery supply shop and invest in about 10 grand worth of batteries, make a huge bank, and get the shop charger from hell to keep them charged, what do you want with 'solar'? ... if you do want to have a solar system re-think your approach ...for example if you could fill a large tank during the day with your pump you could feed it to your house ect with a 12v rv pump that pulls about 30 watts ...

A carbon pile battery tester is commonly used on automotive batteries to check cranking amp capacity, not something used to test deep cycle, personally my approach would be to use the constant current charts in the specs, choose the 20 hour rate ~5amps or 60 watts for 110Ah and leave that load on for 10 hours, if the terminal voltage after 10 hours is around 12V, I would consider it OK, then fully recharge. If the voltage after 10 hours is significantly lower than 12V, it would go to the recycling pile. As mentioned before it's a time consuming exercise, but a more accurate method than subjecting a battery to a massive load for a short duration, although that method may quickly eliminate the obvious weak ones. Additionally they should be fully charged before testing, put the closest ones in series with on another, some see recycling as a good opportunity, obviously not for everyone. My opinions, others may differ, FWIW.

jonr · July 2017

I agree (10 hours draw down for the best test). Might be a good idea for anyone getting "new" batteries - draw down to what should be 50% SOC and test voltage. But note that resting voltage to SOC conversion is only approximate. I see references from 12.06V to 12.3V for 50%.

Dave Angelini · July 2017

I would use a marine appliance for a DC load. They use to make a 12V heater and it was perfect for a slow discharge or what mcgivor said above.
The reason I was pesting the OP is from experience, people can mix up 2/0 and 4/0 with gage 0. The quality inverter/chargers from Outback and Schneider usually want 4/0.

Estragon · July 2017

> @bsolar said:
> you measure them really with a carbon pile load tester .. batteries can be funny .. but the bottom line in your case is you shouldnt try to run a 1000+w load on batteries .. hate to scold you, but what we're doing is 'solar' power .. not battery power .. if you want a 'battery' system visit the local battery supply shop and invest in about 10 grand worth of batteries, make a huge bank, and get the shop charger from hell to keep them charged, what do you want with 'solar'? ...

Ummm... I run 1000+ watt loads regularly. The submersible pump to pressure tanks alone draws around that. I try to avoid running multiple loads at once, but if someone takes a shower there's a good chance both the pump and the septic lift pump run at the same time. My 350ah 48v bank was about $3000 CAD IIRC.

mcgivor · July 2017

Healthy batteries will support a heavy load, no doubt, mine are not, so I use the generator once every few days to fill up the 3000 L tanks, then use a 100W Hitachi pump to supply demand,

Horsefly · July 2017

mike95490 said:

Horsefly said:

..... Then the solar kicks in and the batteries are in float pretty quickly (noon to 2:00pm, depending on the clouds).

Batteries recharging quickly could indicate sulfated cells

Mike - In my case I'm pretty sure that isn't the case. These batteries are only a couple of months old, and have never been below about 80% SoC. I was just pointing out that a big load (like my well pump) can temporarily / artificially bring the battery voltage down, especially with no solar to offset the draw on the batteries. Once the load turns off, the battery voltage comes back up relatively quickly.

If the voltage gets so low that it causes a low battery cutoff as in the case of the OP, the bank may be too small for the load. In my case I had my LBCO set pretty high, but my battery bank is just big enough for my loads.

Horsefly · July 2017

I just was re-reading the original post. If this is really a 1hp pump, I think the OP's numbers are wrong. The rule of thumb is that for 120V single-phase, it's about 14-16A per HP. He says the current is 8A, but that can't be for a 1HP pump. If we assume an optimistic 14A and ignore the power factor, that's 120V * 14A = 1680W. If we assume an 85% efficiency for his inverter, the draw while the pump is running should be 1680 / 0.85 / 24 = 82A which is quite a bit. He's doing this for 10 minutes, which is a fairly long time for that big of a load. I guess I'm not surprised it pulls down the battery voltage like he says, independent of how good his batteries are.

My 1/2 HP well pump only runs for 2-3 minutes each time it fills the pressure tank, and when it is running (with virtually no other loads) my inverter indicates about 1,000W.

Running a 115vac well pump, battery voltage is being pulled down tell inverter shuts off

Comments

Logan Utah
Average Solar Insolation figures

Categories

Running a 115vac well pump, battery voltage is being pulled down tell inverter shuts off

Comments

Logan UtahAverage Solar Insolation figures

Categories

Logan Utah
Average Solar Insolation figures