Single-battery system design, sanity check?

Ineffable

I've searched extensively online and not found detailed answers on this, just hints here and there that I've pieced together.  So, I'm seeking a sanity check on the design.

My goal is to have a small solar system in my vehicle without a separate house battery.  The reasons for not wanting a house battery are mostly not wanting to keep a battery inside the vehicle or have to find another place to put it, as well as simplicity.

Estimated load is < 80W for 6-8 hours, centered around peak daylight hours.

So, my thought is:

• Replace my current starting battery with a dual-purpose battery like the Exide Edge Marine Dual-Purpose AGM .  This battery is the correct group size for my vehicle.
• 100-200W 18V panels that can be tilted and rotated (IOW, not mounted to the vehicle).
• 20A PWM charge controller: https://www.amazon.com/gp/product/B06XWTKYDC/
• Run most of my load through the "load" circuit of the controller, to take advantage of the LVD feature (staying well within the 20A limit of the controller, of course).

I'm aware that this battery won't give me much reserve for cloudy or short days. I plan to work around this limitation by finding a plug when needed.

Am I missing anything important in this design?  If you can point me to any posts or discussions on a single-battery system, I thank you in advance.

Photowhit

If you have a small economy model car, just replace with a deep cycle battery if you battery box is large enough. Dual use are not really designed for deep cycling nd will die rather quickly. If you have a larger car/truck you should just get a battery isolator and run a pair of golf cart batteries for best long term cost, in my opinion.

Loads aren't well defined, and it would seem odd that you will mainly have daytime loads? No lights at night? No music/phone charging etc at night... With a cheap square wave inverter a 8x80= 640 watt load will quickly become 800+ watt load. at 12 volts that's 800/12= 67 amps, a higher regular drain that we would normally recommend for a pair of 6 volt golf cart batteries in series with a storage capacity of 215 amps...

Just to replace the 67 amps, assuming optimal 1 time positioning of your solar panels each day, and an average solar isolation of 4 hours (more in summer less and perhaps much less in winter) You would want about 400 watts of array on a PWM charge controller. 400 watt array will normally produce about 15 - 18 amps on a PWM charge controller per hour for the 4 average hours during the day.

Lvd are generally not recommended to use in conjunction with an inverter are your loads all 12 volt?

BB.

Understand the issue of not wanting a second battery and all of its complications... However, if you do run the battery "dead" (or it fails in 1-2 years do to deep cycling) and the vehicle won't start--Does that leave in in a bad/dangerous situation? Or just call a  friend/autoclub for a jump start and buy a new battery at the autoparts store?

80 Watts * 8 hours * 1/12 volt battery bank = 53 AH

That most smaller car batteries are not much more than 50-70 AH of capacity.

-Bill

Estragon

The LVD value is generally set to protect the controller, not the battery. Running the battery down to LVD (~10.5v?) will be hard on it, and prevent starting the engine to charge or move. You may want to consider getting a lithium booster pack to enable starting in case this happens.

mike95490

Yeah, what they say.  Battery goes dead in your location, no solar to charge with, how to get a jump start ??

Ineffable

Thank you for the replies.

Here are some clarifications.

If you have a small economy model car, just replace with a deep cycle battery if you battery box is large enough. Dual use are not really designed for deep cycling nd will die rather quickly. If you have a larger car/truck you should just get a battery isolator and run a pair of golf cart batteries for best long term cost, in my opinion.

My thought is that I wouldn't do a lot of true "deep cycling".  I expect to shuffle my load around according to available sunlight, and that the battery will be mostly backup for fringe periods.

It's a Toyota Sienna, 3.3L engine.  I think Toyota recommends 585 CCA min.  The BCI is 24F / 34R.  Based on your rec, I'll look harder for proper deep cycle that will fit.

Where would you put the golf cart batteries?  I definitely don't want a battery inside the van, and I can't imagine where else I'd put them.  There's some space in the engine compartment, but I'd need a mount (which I imagine would have to be fabricated).

Loads aren't well defined, and it would seem odd that you will mainly have daytime loads? No lights at night? No music/phone charging etc at night... With a cheap square wave inverter a 8x80= 640 watt load will quickly become 800+ watt load. at 12 volts that's 800/12= 67 amps, a higher regular drain that we would normally recommend for a pair of 6 volt golf cart batteries in series with a storage capacity of 215 amps...

At night I'll use a headlamp running off AAA batteries (this is my norm for camping, and even at home at times).  I don't often listen to music at night, but I'll have the phone if I have an urge for music. If the phone (or system) doesn't have enough juice, then I'll abstain from music...  The phone won't normally need to charge at night, b/c I plan to charge it during the day and turn it off whenever I need to conserve it.

Lvd are generally not recommended to use in conjunction with an inverter are your loads all 12 volt?

Yeah, I expect my typical load to be all DC.  So, no plans to use an inverter at night, and rarely during the day.  

I have a basic 120W inverter that I think would cover my occasional need.  The LVD circuit on the controller I mentioned claims to have over-current protection, but I'd also have a 15A fuse on it.

However, if you do run the battery "dead" (or it fails in 1-2 years do to deep cycling) and the vehicle won't start--Does that leave in in a bad/dangerous situation? Or just call a  friend/autoclub for a jump start and buy a new battery at the autoparts store?

Yeah, I'd have a backup plan and not bet my life on the battery.  Even now, I don't expect my auto to be 100% reliable and do my best to include a safety net in my plans.

But I also didn't realize that these dual-purpose batteries are so "delicate".  Why do they advertise them as good for use with trolling motors?

The LVD value is generally set to protect the controller, not the battery. Running the battery down to LVD (~10.5v?) will be hard on it, and prevent starting the engine to charge or move. You may want to consider getting a lithium booster pack to enable starting in case this happens.

This controller has a settable LVD, so I can set it at 11.7V or higher.  But either way, I plan to monitor the DOD (if necessary, with a battery monitor).

I've definitely considered a lithium jump starter.  If only lithium batteries didn't have that nasty spontaneous explosion (thermal runaway) potential...

mcgivor

To me, your plan is pretty well thought out, the battery would merely be a buffer for a passing cloud for example, the loads would be powered by the array as long as it's large enough, remembering a 100W module will not give you 100w, more like 80W, for this reason I would suggest your 200W option. By description the battery would never/rarely  be used so keeping the existing starting  battery would be fine as long as diligence is practiced, during cloudy/overcast would be when the battery may be called upon, but the vehicle could always be started to provide charge.

The only thing missing is the 6 to 8 hours, not many locations will provide useful sunlight for that many hours, you may get away with it nonetheless if, repositioning or tracking is used, try and see how it goes no experiment is without risk 

With regards to the lithium jump start, there are different chemistries with lithium batteries, LiPo4 are safe, some others not so, but they all get painted with the same brush.

Ineffable

Thanks all for the feedback.

After more research, it appears that this would be like a single-battery marine system, but with the addition of solar, and using the LVD on the controller.

Similar to this (photo credit: bluesea.com):

Wheelman55

You still haven't told us what your loads are. Everyone is guessing until you tell us what you want to run. 

I've been doing what it sounds like what you want to do for five years and with three vehicles. I run two iPhones, two tablets, two rechargeable lithium headlamps, a laptop using a 12v dc plug and an ARB fridge. Here's my take from experience:

Get the best battery that you can. I suggest this one.  It's what I use. 

https://www.batteriesplus.com/battery/rv/dual-purpose-(starting_cycling)/12/sli24agmdp

It's 9 pounds heavier than the Exide and has a 60 month free replacement warranty. They've got a 10% discount going plus a $20 mail in rebate, which brings the cost down to not much more than the Exide. 

Get a quality AGM battery charger, a Lithium jump pack and a plug in DC voltmeter to monitor state of charge. 

For solar consider one of the suitcase style units by either Renogy or Zamp. They come with charge controllers. 

Unless you are driving for many hours a day, your vehicle's alternator is not likely to keep up with your demands.  Once again, tell us what you plan on running. 

The better solution is a dual battery system. I've now got these in two vehicles. Note that you'll still have the run time and charging issues with the second battery, however your starting battery will not go flat. If you put the second AGM battery inside a battery box and strap the box down, it'll be safe inside your vehicle. 

If you want more info on how I set up these dual battery systems let me know. 

Best of luck!

BB.

There is a reason I am very worried about connecting your "alternator" (any engine driven alternator) to the "load" side of a switch...

Many alternators have very poor voltage control when not connected to a good size battery--When the alternator/engine is running and not connected to the battery bank, you can get 50-100+ VDC out of the alternator--And wipe out your connected electronics and lighting.

With most charging systems, the battery bank really does the voltage regulation. The "regulator" (for alternator, solar charge controller, etc.) for charging devices is designed to control the charging current to the battery bank, not "hold" a stable 13.8 VDC.

I personally would suggest two (or more) disconnects). One side of each disconnect should be directly connected to the battery bank, and the other side should go to the Charging source (for one disconnect) and a second disconnect to the loads.

That way, you can never have a direct path from the charging source to your loads.

-Bill

Ineffable

BB. said:

There is a reason I am very worried about connecting your "alternator" (any engine driven alternator) to the "load" side of a switch...

Many alternators have very poor voltage control when not connected to a good size battery--When the alternator/engine is running and not connected to the battery bank, you can get 50-100+ VDC out of the alternator--And wipe out your connected electronics and lighting.

Thx.

I don't plan to ever run the vehicle with the battery disconnected because of the voltage spike potential.  My understanding is that this can fry all of the vehicle's electronics.  

I'll also have a simple disconnect between the charge controller and the battery.  So, if necessary I will follow your suggestion and make a point to fully disconnect my solar and all electronics that are attached to it before starting the vehicle.

But for the sake of my understanding:

If the battery is always connected (except during normal maintenance), would putting the electronics on the controller output make them more at risk from the alternator than they currently are?  My vehicle electronics (radio, etc) are already connected to the vehicle charging system, and at times I plug other electronics into the 12V accessory socket (like a phone connected to USB charger, laptop power adapter, etc).

In other words, how does connecting the solar charge controller directly to the battery positive terminal and plugging the aux electronics into the load output from the controller put them at greater risk from the alternator?

BB.

You should be fine.

The load output of the solar charge controllers tend to be of lower current capacity, so make sure your load current through the LVD (low voltage disconnect) does not exceed the rating (generally any AC inverter should not go through an LVD).

Bill

Ineffable

Wheelman55 said:

You still haven't told us what your loads are. Everyone is guessing until you tell us what you want to run. 

Ah, ok. From reading how systems are designed (load watts, storage needs) I thought I gave enough information.

If the device specifics make a difference: it's just various electronic devices. As I mentioned earlier, occasionally I may use a small inverter, but only if / when the system can handle it. I still expect the total current draw to be well within the controller's LVD circuit 20A rating, and will also inline a 15A fuse. If the system won't handle it, then I won't use it.

@mcgivor summed it up well that the battery is more of a buffer. Ideally the battery would also give me some wiggle room in the morning and evening when sunlight is low.

Per @mcgivor's suggestion, it sounds like I may need to increase my array. But if the battery doesn't give me wiggle room in the shoulder periods, I'll have to accept it as-is. I'm not expecting to be self-contained solar.

Wheelman55 said:

Get the best battery that you can. I suggest this one.  It's what I use. 

https://www.batteriesplus.com/battery/rv/dual-purpose-(starting_cycling)/12/sli24agmdp

Thank you for the battery recommendation.

Wheelman55 said:

Get a quality AGM battery charger, a Lithium jump pack and a plug in DC voltmeter to monitor state of charge. 

Are you talking about an AC-powered battery charger, so that you can charge the battery from AC when necessary?

So, in your experience, the DC Voltmeter has worked OK to evaluate SoC? I've read that using a simple voltmeter to measure SoC is inaccurate.

Wheelman55 said:

Unless you are driving for many hours a day, your vehicle's alternator is not likely to keep up with your demands.  Once again, tell us what you plan on running. 

Hmm, that's something to consider. 80W at 12V is 7A, right? Even if I double it to 14A, wouldn't that be on-par with typical auto use (headlights + blower motor, for example)? My "aux load" won't exist when I'm actually driving.

Would a standard alternator be able to keep up with that? What am I missing?

Wheelman55 said:

The better solution is a dual battery system. I've now got these in two vehicles. Note that you'll still have the run time and charging issues with the second battery, however your starting battery will not go flat. If you put the second AGM battery inside a battery box and strap the box down, it'll be safe inside your vehicle. 

I may eventually upgrade to a multi-battery system. But I want to start with a single battery system.

Estragon

A DC voltmeter can work ok as a SOC meter (with AGMs, it's about all you have), IF voltage readings are taken when the battery isn't significantly charging or discharging , and ideally hasn't been for some time (hours). If left with no charge/discharge current for several hours, resting voltage can be a reasonably accurate estimate of SOC for a healthy lead acid battery IMHO. Using voltage when actively charging or discharging can be misleading though.

Alternator output will likely be a function of RPM, and temperature. Typical auto alternators are designed to recharge a shallow cycled starting battery, and may cut back output if it hot running near rated output for long. 7a + start recharging isn't likely a problem for a typical auto alternator.

It sounds like you aren't planning to regularly draw the battery down much though, so shouldn't be an issue IMHO.

Wheelman55

Yes on the AC battery charger. Yes on the plug in voltmeter. Never go below 50% to 60% SOC on your battery. 

Can't help any more until you tell us your loads. 

Ineffable

Estragon said:

A DC voltmeter can work ok as a SOC meter (with AGMs, it's about all you have), IF voltage readings are taken when the battery isn't significantly charging or discharging , and ideally hasn't been for some time (hours). If left with no charge/discharge current for several hours, resting voltage can be a reasonably accurate estimate of SOC for a healthy lead acid battery IMHO. Using voltage when actively charging or discharging can be misleading though.

Ok, that helps. I was wondering if a current-measuring battery monitor would be much more accurate.

Since you said "significantly" discharging, does that mean that you think the vehicle's standard key-off parasitic drain would not affect the reading much?

The main problem I foresee using a voltmeter is the required wait time. But, I'll research it more to get details.

Estragon said:

Alternator output will likely be a function of RPM, and temperature. Typical auto alternators are designed to recharge a shallow cycled starting battery, and may cut back output if it hot running near rated output for long. 7a + start recharging isn't likely a problem for a typical auto alternator.

Thx. I imagine the vehicle's nominal electronics would also be a factor. Some initial research indicates the OEM alternator is 110A or 130A. Sounds like something I need to look into further.

Ineffable

Wheelman55 said:

Yes on the AC battery charger. Yes on the plug in voltmeter. Never go below 50% to 60% SOC on your battery. 

Thx.

Wheelman55 said:

Can't help any more until you tell us your loads. 

I don't know what else to tell you about my loads.  Misc DC devices expected to be typically less than 80W total.  Possibly a ~15W AC device through a 120W inverter on days that I have enough power.  The specific devices will vary, but I will adjust the load according to available power and can shut off the auxiliary loads if I really need to.

These loads look like they're less than running the air conditioner in the car.

Estragon

A current meter would certainly be helpful, as it helps put volt readings in context. Some shunt based meters, such as the Trimetric, also keep track of variables enabling it to provide an approximate SOC indication. I use them as my main day-to-day SOC measures in my cabin systems.

I use voltage as my main SOC indicator on my boat, mostly successfully. Over time, I got a sort of feel for how voltage related to various loads etc starting and stopping, so could make an educated guess on roughly what resting voltage would be on my house bank. When I get time to use the boat more, I may add a Trimetric type meter though.

"Significant" is an admittedly squishy term, with lots of variables. Battery capacity, type, construction, and ambient temperature are some. A cool flooded deep-cycle battery may sag more with a given load than a warm AGM starting battery of equivalent capacity, for example.

The ratings on OEM alternators are likely just for starting. Recharging a shallow cycled starting battery, likely in a cool engine compartment, for a few minutes is different than sustained output over hours next to a hot engine. That said, your loads sound unlikely to be a problem.

Ineffable

@Wheelman55 - On your single battery system, can you still start the engine at 50% DOD?  What size is your engine?

Wheelman55

I’ve got two of the batteries that I linked to in a ‘93 F250. That has a 460 cu. inch V8. I’ve started it at 50% SOC when down in TX.

BB.

One of the issues is how to ensure that you stop the "house loads" from discharging below 50% state of charge... Using battery voltage is a bit of a guess (lead acid state of charge is not easily determined by just voltage measurements).

There are lots of devices out there... "Shut based" battery monitors (they actually measure the current in/out of the battery bank). And there are some voltage only units that estimate state of charge based voltage/temperature profiles).

https://www.solar-electric.com/catalogsearch/result/?q=battery+monitor
https://www.solar-electric.com/bogart-engineering-tm-2030-rv-battery-monitor.html
https://www.solar-electric.com/midnite-solar-mnbcms.html
http://www.smartgauge.co.uk/products.html

I have not used any of these meters (and there are many other good brands too)--Just a starting point for your education. Some meters have an output contact to signal low battery, others may have a digital comm output.

Of course, adding one of these monitors, is another point of complexity/failure if you try to automate the energy control...

-Bill

Ineffable

BB. said:

One of the issues is how to ensure that you stop the "house loads" from discharging below 50% state of charge... Using battery voltage is a bit of a guess (lead acid state of charge is not easily determined by just voltage measurements).

There are lots of devices out there... "Shut based" battery monitors (they actually measure the current in/out of the battery bank). And there are some voltage only units that estimate state of charge based voltage/temperature profiles).

@BB.

Thx.  My questions to @Wheelman55 and @Estragon about the voltage meter were because I've read in so many places that voltage isn't very accurate for SoC unless you use OCV and let the battery rest.  Sounds like it's working OK for them, though.

I've been thinking of going with a shunt-based monitor eventually.  

For me it brings up the question of how to install it for the best reading, including all of the load and charging from the vehicle.  For that it seems that I'd need to git a high-amperage unit and install it to the entire electrical system (like on the battery negative terminal).

It looks like I'll need to do some research and learn a lot more -- like about voltage sag and the parasitic draw on the vehicle.

BB.

Yep, you install the shunt between the negative battery connections and the negative load/system ground connections (you want "ALL" battery current, both charging and loads, to go through the shunt).

You can get 100/500/1,000 Amp shunts through our host (make sure your meter is "calibrated"/has a setting for the proper current/millivolt shunt rating).

https://www.solar-electric.com/catalogsearch/result/?q=shunt

I suggest that systems that have more than ~200 to 250 Amps on their DC bus--That you look for the next size up for the battery bank (i.e., 150 Amps @ 12 volts would be 75 amps @ 24 volts)--You don't really want a 250 Amp system on a 12 volt battery bus. You really only have a 0.5 to 1.0 volt of "available voltage drop with 12 volts--That means you have to have very short and heavy cables (and large breakers/fuses) between the battery bus and heavy loads (i.e., 12 volt battery drawn down to 11.5 volts under load, and 10.5 volt inverter cutoff voltage = 1.0 volt drop--And most inverters will run for a few seconds to a minute or so at 2x rated power, so that means 2x higher current and higher voltage drop--I.e., design current (not surge current) 0.5 volt drop suggested maximum).

Higher voltage battery banks have higher "supported" voltage drop (~1.0 volts for 24 volt bank, and ~2.0 volts for 48 volt bank)--Makes wiring a bunch easier and less costly.

-Bill

weejub

So get any deep cycle with the CCA you need to start your car.  I like the X2Power products, too.  Sealed etc.

Then carry and keep charged one of these cheap emergency start LiOn things if you go below your ability to start your car.  Jump start it, run it to recharge.  Done.