Protecting batteries with load control

tomba
tomba Registered Users Posts: 21
Hi all. 2nd post. I'm successfully charging a pair of 12V 115 Ah Costco Marine batteries in series with a 180W 24 V panel and a 20A-24V Morningstar SunSaver (with LVD). The 1100W Exeltech 24V inverter coming in the mail can draw approximately 50 A, and as such will get connected directly to the batteries, not to the controller load. I plan to watch the batteries charge state assiduously and first run simply the inverter in on position for a few days, with some light AC components like a CD player. But in the long run, the idea is to have this inverter working continuously, with its scary low voltage disconnect of 19 Volts. This sounds like certain death for my batteries if I ever were ever to get there.

To avoid that ever happening, my plan was to run a 50A/24V automotive relay off the charge controller 20A load (something like this) and thus cut off current to the inverter when the Morningstar says. But the controller cuts the load at 23 V, which seems low as well. Survivable, but to be avoided. There's no modifying this setting, AFAIK. As a failsafe, is this tolerable?

If 23V is also a state to definitely avoid, the sure solution is to get a Xantrex C60 (or similar controller) and use it in its load controller mode, set to 24V or some such level. But isn't this overkill, for a very particular problem? There's no dedicated, quick and dirty load controller out there?

Better ideas are always appreciated. I already know the best idea is to not wreck your batteries. :D

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Protecting batteries with load control

    The problem is that it is not very practical to estimate the state of charge of a lead acid battery by measuring voltage while under load. You need to have the battery rest for a few hours before you can measure the voltage to determine the state of charge (or use a hydrometer).

    If you run the battery below 20% state of charge, you run the risk of cell reversal (one cell goes dead and then actually reverses voltage and begins to recharge via the energy of the other cells). That is pretty much instant death for that one cell (and the battery it is in).

    I really like Battery Monitors to estimate the state of charge of the battery bank. It measures the Amp*Hours going out and going back in. A few even have programmable outputs that you can use to turn on a genset, turn off loads, set alarm, etc. if you need an automated solution to protecting your battery bank.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Protecting batteries with load control

    in addition to what bb said you would find that the battery voltage can vary quite a bit depending on the load. if you were to draw that 50a at 24v the battery voltage would sink immediately and trip the voltage threshold of any voltage protection device even if the batteries are fully charged. remember the battery bank only has 115ah and 50a is a large % of that and i don't recommend you drawing that much current from that battery bank. at the same time if you were to draw say 1a then you would hardly see any drop in the voltage for quite some time.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Protecting batteries with load control

    The need to design a system around the expected loads rears its head again.

    If the battery bank is adequate to supply the loads between charging, the question of dangerously low Voltage doesn't come up.
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Protecting batteries with load control

    Let me restate the obvious.

    Using voltage alone is a very poor way of determining when to turn off the loads.
    Your 50 amp inverter, loaded fully will drop the voltage of your 330ah battery through the floor! (50 amps*2= ~1200 watts,, a huge draw on a small battery. Just for comparison, my 450ah (12vdc) battery, will drop to ~12.0 volts with a ~150 watt load.

    You really need to define the loads, and their duration, and then manage draw relative to recharge capability. (both from PV as well as charger). Most people spend a lot of time really learning the nuances of batteries and in the process they wreck one (or more) sets of batteries.

    I suggest you read the following links :http://www.windsun.com/Batteries/Battery_FAQ.htm#Lifespan%20of%20Batteries

    http://www.batteryfaq.org/

    Tony
  • tomba
    tomba Registered Users Posts: 21
    Re: Protecting batteries with load control

    Thanks for all the replies and links. The load I am baselining for is a bar frig. 50 A is just the absolute limit, 3-4 second surges. I'm just spec-ing a relay that would tolerate that max current of the inverter.

    Regular usage might be 5-10 A, with a frig pulling ~100W after starting up. If that is a gigantic load for the batteries I have, I can double those now. The panel provides ~20-25 AH, so using the 10% rule I've seen in other threads, a 230 AH battery bank is not too large. Getting the specific gravity of the solution in these batteries isn't really an option. An ammeter and logging capability would be better for me.

    So maybe 23V under load isn't a disaster. I think the Morningstar will allow that for 6 seconds, to allow for the surge of a refrigerator turning on for instance. So adding a relay to use the charge controller's LVD to control all battery loads is sounding good to me. For a $50 relay, I can't think of a cheaper way to stupid-proof the thing.

    I realize that upfront planning is the best way to avoid dumb outcomes. Right now, the inverter @ 20W will already use around half the energy supplied by the panel (on average). The frig will want the other half. So another panel and another pair of 12v batteries are already staring me in the face. Depending on how cold I want my beer.8)
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Protecting batteries with load control

    Up-front planning includes plugging that 'frige into a Kill-A-Watt and getting real-world usage numbers. Do not trust the spec plate! Small refrigerators do not draw a proportionately smaller amount of power compared to full-size units. They are notoriously inefficient, because the manufacturer's don't care about this in the small units.
  • tomba
    tomba Registered Users Posts: 21
    Re: Protecting batteries with load control

    Cariboocoot: Thanks, and yes, your point is well taken. A good sized chest freezer is more efficient than most any bar frig. There are some EnergyStar Haier 4.1 cft units that rate at 270 kWH annual vs 330 for most of the other ones (and the Sanyo I've been considering). So that's probably where I should start. Then again, I Kill-A-Watt-ed a small, no frills GE frig at work and found it only used 0.6 kWH per day, i.e. surprisingly good. So indeed, you only really know when you finally get the frig and check it out. When the inverter shows up and checks out, the frig is next.
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Protecting batteries with load control

    As Marc suggests,, small fridges are notoriously inefficient. Depending on how much you use it,, (part time, full time, weekends etc) another alternative is a Propane fridge. As we have discussed here often, a propane fridge will run a long time on a small amount of propane, allowing any given PV/battery system to be very much smaller.

    Your guess of ~100 watts running for your small fridge, at a 50% duty cycle would translate to ~1.2 kw/day. By my rule of thumb that would need a Pv system in the 600 watt range just for the fridge day in and day out. Your described 180 watt PV is going to be woefully short unless I have missed something.

    Tony

    PS:

    As I have often mentioned, a battery based Pv system easy calc. is to take name plate rating divide by 2 to account for all system loses, then multiply that number by the number of hours of good sun you can reasonably expect,,seldom more than 4 on average

    180/2*4=360 wh/day

    Some days you might get twice or even three times that, but many days none. Most people over estimate the amount of power they will actually get, and underestimate the loads they will draw. Even the simple radio/cd player might draw 15 watts. 4 hours a day is 60 wh, 1/6 of your daily average capacity, just for example.

    PPS. what I don't see mentioned in this thread is what it is you are trying to achieve. A little more information as to what your goals and objectives are might be helpful.

    T
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Protecting batteries with load control

    tomba,
    to clarify things are you using 2 180w pvs as you stated in another post or not? with 2 115ah batteries in series you have a 24v 115ah battery bank (not 330ah). 2 of those pvs will give in the neighborhood of about 10.1a and is a good matchup for the 115ah battery bank with a charge rate of 8.78%. if using only 1 of those pvs it will be half of that or about 4.39% which is below the recommended 5% minimum.
    if you get say 4hrs of full sun on average each day the pvs will deliver under 1kwh and this does not account for cloudy or rainy periods. more battery ah may also be needed to supply the power during the off times for 57.5ah is usable and at 24v represents 1380wh which may be cutting it close for daily uses and surges needed to start the frig.
  • tomba
    tomba Registered Users Posts: 21
    Re: Protecting batteries with load control

    Thanks for your evaluation of the system. First off, an overall 50% efficiency is not a number I was carrying, and if I should from now on, it's likely I'll need a second panel. These crummy frigs are 330 kWh annual, or 1 per day. So with a 50% efficiency you need to have PV production of 2 kWh per day, and at 5 sun hours, need 400W on the roof. My fancy single 180W panel gets called woeful and that makes perfect sense.

    The goal is to run an outdoor/backyard/hobby system running a small frig (at a to-be-determined duty cycle) and anything else that I have the capacity for (weed whip, radio, LED Xmas lights). Also, in the SF bay area (with 5 sun hours), this is my "solid state" backup in case of power outages. I know I could get a generator and be done with it. But what's the fun in that?

    If the frig has an adjustable thermostat, i could throttle it at the outset. It will be an outdoor refrigerator, so I am counting on the benefit of the outside average annual temperature being ~50 degrees,vs 68 inside my house. Wintertime draw from the frig will roughly descend with the insolation, etc. I am also expecting to be able to make seasonal adjustments to the array angle, given that I only will have one or two panels. If for now I have to turn off the frig Sunday through Thursday, fine. As long as the beer is cold by Friday night! With a second panel, things get much easier.

    I take it from the responses that using a relay to kill the power to unregulated battery loads based upon the charge controller LVD is not a completely dumb idea. Rather, expecting to run a frig full bore off a single panel and two batteries is. It's OK with me if it doesn't cook me breakfast as configured currently. I think I can get by without destroying my batteries, evaluate and decide upon the best upgrade once things start to run.

    I really appreciate the input and the reality check. :roll:
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Protecting batteries with load control

    since this is a "hobby" system, it seems to me, that I would look to provide the biggest bang for the buck. That said, I'm not sure I would chose it to be to power a fridge, especially since the small fridges use about as much power as a full size fridge. Given that, I might live with the beer in the house fridge, and try to power the stereo, or patio lights or the widescreen instead. 180 watts of panel isn't much, but it is enough to show you what you can do with PV. In point of fact, we live off grid with only about twice that much!

    180/2*5=450 wh/day,,, enough to run a stereo for many hours, or some landscape lights or a LCD widescreen for a few.

    Tony
  • tomba
    tomba Registered Users Posts: 21
    Re: Protecting batteries with load control
    niel wrote: »
    tomba,
    to clarify things are you using 2 180w pvs as you stated in another post or not? with 2 115ah batteries in series you have a 24v 115ah battery bank (not 330ah). 2 of those pvs will give in the neighborhood of about 10.1a and is a good matchup for the 115ah battery bank with a charge rate of 8.78%. if using only 1 of those pvs it will be half of that or about 4.39% which is below the recommended 5% minimum.
    if you get say 4hrs of full sun on average each day the pvs will deliver under 1kwh and this does not account for cloudy or rainy periods. more battery ah may also be needed to supply the power during the off times for 57.5ah is usable and at 24v represents 1380wh which may be cutting it close for daily uses and surges needed to start the frig.

    I have the single panel.

    OK, so I was thinking the baseline was 10% of battery bank charge per day (or 5 hours), not per hour. Thanks for clearing that up.

    I realized that 2 12V/115 Ah batteries in series is just one 24V/115 Ah battery. I was imagining life with another pair of batteries for 230 Ah and a single panel. I was in the clear with that when I was multiplying by 5. :blush:

    What I'm gathering is if I need more batteries, I have no business adding them without another panel or at least a charger.
  • solarvic
    solarvic Solar Expert Posts: 1,071 ✭✭✭✭
    Re: Protecting batteries with load control

    Your amp needed is higher than you stated. Minium of about 54 amp and about 64 amp need for surge for that inverter. S:Dlarvic
  • tomba
    tomba Registered Users Posts: 21
    Re: Protecting batteries with load control

    Solarvic: yep, 50 A is less than the rated limit of the Exeltech inverter. Maybe it will enjoy abuse? I'm finding efficient 70A relays cost much more (like 200 bucks, DIN mountable). I could try a cheap one to test the concept. That texasindustrialelectric site is just great and has several options. Eventually I'd like something that doesn't sit there and burn 5W in the coil, though all of them pale in comparison to the inverter's draw. :\
  • tvengineer
    tvengineer Solar Expert Posts: 31
    Re: Protecting batteries with load control
    tomba wrote: »
    Solarvic: yep, 50 A is less than the rated limit of the Exeltech inverter. Maybe it will enjoy abuse? I'm finding efficient 70A relays cost much more (like 200 bucks, DIN mountable). I could try a cheap one to test the concept. That texasindustrialelectric site is just great and has several options. Eventually I'd like something that doesn't sit there and burn 5W in the coil, though all of them pale in comparison to the inverter's draw. :\

    Getting back to his question about the relay... THe Exeltech inverters that I have seen (and the one I have) have provitions for remote ON/OFF... he could use a very small relay on the "load" output of his charge controller to turn the Exeltech off.

    Agreed on everything else in this thread.. I have a similar problem with voltage drop running my 300watt SureSine on only one battery... with the battery at over 50% charged.. the voltage drops soo much with the inverter load that it triggers the Low Voltage Shutdown within minutes.
  • tomba
    tomba Registered Users Posts: 21
    Re: Protecting batteries with load control
    tvengineer wrote: »
    Getting back to his question about the relay... THe Exeltech inverters that I have seen (and the one I have) have provitions for remote ON/OFF... he could use a very small relay on the "load" output of his charge controller to turn the Exeltech off.

    Hmm, that sounds like a *much* better option and a cheaper relay! When USPS finally brings the inverter, I'll check that out.

    But reading the SunSaver manual again, the LVD kicks in after 2 seconds @ 23V, not 6 seconds, and likely not long enough to get through a refrigerator startup. So the best plan is, as pointed out here, simply not connect loads that have the potential to overtake PV charging.

    I also ran into this product, linked from batteryfaq, that mechanically disconnects loads if the battery is below 23.4 V for a minute.
    http://www.prioritystart.com/ps_pro.html
    I just am unclear about how it would restart/reset on a PV type system as opposed to a vehicle.

    If spending 125 dollars on that, you might as well get a Xantrex C60, that disconnects (and reconnects) loads at an adjustable battery voltage level after a nice long 6 minutes. Long enough to saw something in half with a saws-all.

    Thanks for all the useful inputs.
  • tomba
    tomba Registered Users Posts: 21
    Re: Protecting batteries with load control

    I now have the Exeltech 1100 power controlled by a 24V relay with the coil voltage determined by the Morningstar load voltage connections. Actually, I have the whole first iteration of my "power center" pretty much finished.

    zi60148.jpg

    The relay is to the right of the inverter, with a wall switch inline with it's contacts. Otherwise, the only way to turn off the inverter is to disconnect (15A breaker) the battery from the charge controller (triggering the LVD), or disconnect (80A breaker) the battery from the inverter itself. With full batteries and full sun, I can run a 600W weed whip without tripping the LVD. How it might work whipping weeds at midnight has not been tested. And how it works with a refrigerator probably also will wait unit the 2nd panel goes in and batteries to match. Even if it eventually has to get pulled, it works for now and was worth figuring out.

    Anyway, this is the result of many hours reading posts here, product manuals, info across the net, and my own warped ideas of how things should work. The idea from tvengineer to use the relay on the inverter remote circuit and not the battery terminals really helped. This relay (an NTE R14) uses only 1 Watt.

    Al Lasher's Electronics in Berkeley is probably one of few places you can walk in off the street, babbling about your project, and get steered into the back room to behold a wall of boxed DC relays.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Protecting batteries with load control

    Now I know what's wrong with my system: no skull lights! :p
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Protecting batteries with load control

    tomba,
    just kicking a few observations around here i have to ask why you have the input voltage going through 2 breakers. the first is white with a black lever and the 2nd is black with a white lever and could possibly be a tandem trip? having a tandem breaker (which isn't a terrible idea) will still trip the input at the proper current level (if rated at that level) without that first breaker inline.

    i also see the lvd output going to something. is that just a monitor because if it is then you have negated the lvd output by recombining with the battery + output from the controller? these are not meant to be combined as only the lvd lead is protected and limited and is run separately from the battery +. the battery + line should really only go straight to the battery and loads should come from (with a breaker) the battery and not from the battery + of the controller. that means you need to run a wire from the battery's positive terminal to the breaker and then to positive loads.
  • tomba
    tomba Registered Users Posts: 21
    Re: Protecting batteries with load control

    Niel: Regarding the multiple breakers, the PV comes first into a (white) 15A breaker, then into a (black) 63A GFCI breaker pair, where the neutral and ground are wired to either end of a 0.5A breaker that switches in tandem with the larger load. I saw this recommended in a Xantrex manual, bought that part early and decided to keep it in line. The question now is whether this connection should be the *only* place where ground should be bonded to neutral, because I have those two bonded twice in two other places in the system, which would necessitate a ground fault of at least 1.5 amps to trip the breaker.

    I have a 120V AC line tester, and I could just remove the other neutral-ground connections and see if it still shows a properly grounded AC power.

    The end-to-end run of the LVD output line is a little obscured by the neutral line going into the inverter, it keeps going past the 80A marine breaker (where it looks like it's attached but isn't), below the inverter, and over to the relay to the right of the inverter. There's no other DC positive attached anywhere to the LVD. If you pull the LVD output line off the charge controller, the relay cuts the power as its supposed to.

    Now that it's working a little wire management is probably in order. :roll:
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Protecting batteries with load control

    very good as it is hard to depict those things from the low resolution pic.