12V vs 24V Battery System

ScoobyMike

I am looking for direction regarding a 12V vs 24V off grid electrical system, from the point of view of MPPT charge controller.  Pertinent system info:

 - 1KW PV with single charge MPPT controller

 - Four Interstate GC2 6V, 210 AH golf cart batteries

 - 150’ distance between PV and controller

 - Using 12V loads wherever possible with multiple 12V power points wired throughout the cabin, short runs (~30’) to each power point from central 12V power distribution.

 - the PV controller and inverter are <10’ from the batteries

 - the location is Maine, where cloudy days are common

My options are to wire the batteries as 12V, and distribute power directly, or 24V and step down to 12V either at the center of the 12V distribution or use smaller converters at each 12V power point.

I am leaning towards using three 290W to 330W 20V or 24V solar panels in series with a MidNite Solar Classic 200 Controller. The Midnite Solar controller sizing tool has a note stating that higher input voltage on MPPT controllers are less efficient.  I want a higher voltage so I can use a reasonable gauge wire over the 150’ from PV to controller.

Will I see significant controller power loss in a 12V system with a 100V – 120V input voltage?

Will this higher voltage help for low light days?

I would consider using 4 panels in a 2 string configuration to feed a 12V battery bank

Photowhit

We didn't hear about your loads. At a 1000 watt array and a new install, I would be looking seriously at a 24 volt system.

Don't know what your 12v volt loads are, if they are small, <20 watts, and momentary, vs continuous. I'd think 24 volt would be a 'no brainer'. Using individual wall warts.

The higher voltage will aid a bit in overcast, but not something I would concern my self with. don't know how significant it would be over losses converting from the higher voltage.

I'm not a huge fan of 12 volt loads, I used them with my early tiny systems, but gave it up after a couple years. I used some 10 gauge landscape lighting wire, might help you when looking for locally available twin stranded pairs of wire.

westbranch

I'm the same as Photwhit, been there done it in my summer cabin/ now guest house. Main house is all AC powered.  One line of 24V to power the WIFI and Sat modem , would have been AC > DC  instead of 24V DC to 5V DC. so if we are away  we dont hav to run the house Inverter.
Have you considered having the batteries closer to  the array and then sending 120VAC to the house?.. may be a lot cheaper in the long run , uses smaller wires from array to Charge Controller and  batteries and then send AC in Cu or Al wire...

ScoobyMike

Largest DC load is refrigerator, .3 to .5 KW per day. Then car stereo with Sat radio, cell phone booster, laptop & phone chargers. The Midnite note indicated the inefficiency was fairly minor, any idea of the magnitude? I'm guessing that is is less than the 85% inverter efficiency.

mike95490

Go with 24V and set your array to no more than 100V   1Kw @ 80V is only 12A..   But the other losses are in downconverting higher voltage to battery voltage.  I know, I have 2 systems side by side, one runs 180V other 120V and the losses in my Classic 200 are much higher, (controller runs hotter with a smaller array)  than the 120V system.  And I'm charging at 65V (NiFe bank - special programming)

Estragon

You may want to check the dc fridge to see if it can use 24vdc. If so, I'd consider 24v. Aside from wire and controller losses, going 24v allows for expansion without having to add controllers or replace inverter.

At 150', I'd consider sticking with a classic 150 and smaller string size. 3x300ish watt panels in series for Voc ~135v might work okay, depending on location/climate. IIRC a 150 to 24v bank is okay up to 174v.

Photowhit

Estragon said:

You may want to check the dc fridge to see if it can use 24vdc. If so, I'd consider 24v. Aside from wire and controller losses, going 24v allows for expansion without having to add controllers or replace inverter.

This is my thought as well, many will work on either.

ScoobyMike said:
...Then car stereo with Sat radio, cell phone booster, laptop & phone chargers. ...

The car stereo Might be a high enough wattage to be a problem, the others may be adapters them selves, Laptops are usually 18-21 volts and cell phone chargers usually 5 volts.

I've seen home 'boom box' type satellite radio cradles inexpensive as people choose to quit their subscriptions.

ScoobyMike

Thank you all very much for your insightful comments! I am coming around to the issue with a high voltage PV (although I would love to see some efficiency numbers), but I am not yet with you on the 12V vs 24V.  My hangup with 24V is that although I am new to PV, 40 years with RVs has given me a deep understanding of 12V systems and I very much see my camp as an RV without wheels rather than a traditional home.  Having said that, the combined experience of the folks recommending 24V is a powerful motivator to run some numbers using the Midnite calculator and a voltage drop calculator.  I will get back with my thoughts.  i believe that the basic requirement for charging comes from a Midnite note that the minimum PV voltage needs to be 130% of the battery equalization voltage, which I believe would be:

                   V-EQ            V-PV
12V            15.6V           20.2V
24V            31.0V           40.0V

I am already somewhat invested in my system:

This winter the 3-way fridge in my Outfitter Apex cab over died and I replaced it with a Dometic CRX110, which fit better than the Norcold did :-).  My annual trek across country will be in my truck with the Outfitter and I will transfer the Dometic to the camp for the summer. This reefer can operate on 12V or 24V.

I have pre-built and tested my AC/DC system.  I used a Cotek-SK700-112 that I found on clearance (I could use this to replace the cheap 2KW MSW in my Apex) and, perhaps my biggest hesitation is the IOTA DLS-55 I have used.

ScoobyMike

Here is my electronics and control system, there is also a 12V distribution fuse box:

Estragon

I believe there are efficiency curves on the midnite website which can give you a sense of losses in conversion at various imput voltages.

With short runs, 12v can work fine. I use it in my cabin for fridge, a small pump, various cigarette plug devices, and hopefully this year for lighting, as well as for a small inverter for night use. I might have used 24v, but I really wanted to use morningstar's small PSW inverter, which only comes as 12v, and the little gizmos are easier to find in 12v.

To me, the main reason to seriously consider 24v in your application as you've described it is expansion. At 1kw, you don't have a lot of output headroom on a single controller, and would be stuck with a 12v inverter, making expansion an expensive prospect vs going 24v now. That said, if you know the lifestyle from RVing, and don't anticipate a need to expand anytime soon, 12v might be fine.

softdown

Clear case for 24 volt by my estimation. 

If anybody wonder why I sometimes repeat what others say? Help with decisions to read several affirmations. 

ScoobyMike

I spent the afternoon perusing the Midnite site, mostly in the forum.  I could not find a good reference to show efficiency vs voltage in the controller.  I did find buried in one thread the efficiency of a Classic 150 with a 72V input voltage charging a 12V battery is 92%, which is very encouraging.

I have 2 personal traits that constantly battle one another (especially at Costco):

 1 - I am a cheap SOB
 2 - I am driven to maximize efficiency

Finding inner peace has been a lifelong challenge

I actually think .5 KW PV would be tolerable, but panels are cheap and 1KW would be great (and fully utilize the Classic 150). The Canadian Solar 260P are ~$160 apiece, 2 strings of 2 would give me ~60V into the Classic 150 which can handle either 12V or 24V battery bank. Since I already have my 12V inverter and shore power charger I will most likely start with a 12V system.  Changing to 24V would only require replacing the inverter and charger that I already have.  BTW, I drive right by AltE on my way to Maine, so $0 shipping cost

So, at 92% controller efficiency on a setup similar to the one I described has very little room to improve for a 24V battery.  So, is the consensus for 24V driven by something else in the system? Inverter efficiency perhaps?  Cable size for DC power distribution? Battery life? Something else?

westbranch

In your case, yes you would see a small increase in efficiency. The reason is that the closer the incoming (PV) V is to the V of the battery the less energy that is ( consumed by the electronics and ) lost to heat in the C C. 
So that is why the rule of thumb is 'keep the Array voltage as close to the bank voltage''  as possible
With GT panels this can be a bit difficult to do but using 'Nominal ' 12 V and/or 24 V PV's  makes it much easier.
And you have to make sure the Vmp of the GT array will be at or Greater Than the Vmp

706jim

I built my place in 1992 before inverters were common or cheap. Wired two complete systems, one for 12 volts DC and a second for 120 volts AC. The 12 volt circuits never get used as line loss with any sort of load was horrendous. My system is 24 volts and I presently just run the inverter 24/7 with little heed for parasitic losses from wall worts or the inverter itself. I would go with 24 volts for the batteries and would just transform 120 volts back down to 12 volts for those appliances that need it, odd though that may seem. And I have to put a plug in for my old Trace DR1524. It has run faultlessly since 1994.

Photowhit

Just to be clear, Charge controller work on amperage, so with future expansion, The Classic can handle about 1350 watts at 12 volts and 2700 watts at 24 volts.

Higher voltage works toward being more efficient. Mostly for the reasons you wel understand with the losses from you array to the charge controller/battery. 4 - 210 amphour flooded lead acid batteries would like a 10-13% charge rate, working backwards, a 420 ah 12 volt battery bank would like a 42 - 55ah charge rate for full time use. Solar panels produce about 75% of their panel rating, so 42 amps at 12 volt = 504 / .75= 672 watt minimal array size (660/.75=880 watt for 13 %), might add a bit for loss in efficiency. I think you are working in the right ball park. 

I always prefer to be somewhat over paneled. I don't use other means of charging so it helps.

SupraLance

ScoobyMike said:

So, at 92% controller efficiency on a setup similar to the one I described has very little room to improve for a 24V battery.  So, is the consensus for 24V driven by something else in the system? Inverter efficiency perhaps?  Cable size for DC power distribution? Battery life? Something else?

Yes, Yes, Yes, and yes.  Controller efficiency at 24v will probably rise 2% (just based on other controllers I've reviewed).  24v inverters are also typically about 2% more efficient than 12v inverters.  Line loss and/or line cost is a lot less.  The batteries are easier managed and tend to last longer all in series, which means 24v for the 4 6v batteries you have. 
When you add all the inefficiencies together, the 24v system is a lot better.  That said, you already have a 12v inverter, and your batteries are already in use at 12v, but anything new you buy should be with an eye toward going 24v either now or in the future.  And just doing it now will save you money on your wiring and is probably smarter from an efficiency standpoint.

Estragon

The chart I was thinking of:

http://www.midnitesolar.com/pdfs/classicPowerChart.pdf

ScoobyMike

Thanks guys, I very much appreciate all of the knowledge and guidance you provide! Presentation of all of the variables is sometimes difficult as our sometimes flawed logic sometimes masks the real issue in the form of a simple question.  I think the variable I omitted in this case is the site layout.

Last night I had an epiphany. I noticed people complaining about fan noise from the charge controller, andI have been focused on putting the power system inside the cabin (probably because that is how it is done in an RV).  My site actually has an 8' x 10' shed, 50' away from the cabin and the site for the solar panels is another 50' beyond the shed.  I am now thinking to use the same PV configuration, with converter, charger, batteries and inverter in the shed and push the AC and 24V DC over to the cabin. 

Estragon said:

The chart I was thinking of:

http://www.midnitesolar.com/pdfs/classicPowerChart.pdf

Thank you Estragon!  I saw this chart but was not quite sue how to interpret it. I thought it was showing the point Photowhit made about power capacity rising linearly with voltage.

Photowhit

I don't know how you use power, there will be diminished capacity with cold. My systems have all lived outside of my homes, but since having systems large enough to run air conditioners, My don't use a lot of capacity in the winter. Also I think Maine is prone to long spells of cloudiness in winter. Might mean running a generator a bit more to maintain a reasonable SOC for very cold weather. We rarely get below -10 here in Missouri. Charging and discharging will create some heat, so batteries rarely fall this low.

ScoobyMike

SupraLance said:

 you already have a 12v inverter, and your batteries are already in use at 12v

Actually, I don't have the batteries yet.  The cabin is on a remote island and I am pre-building as much as possible so I can bring the system up ASAP once I get there. I'll make a Costco run for the batteries next month.

I'll use the 12V system I built when I first get there, probably run everything on AC until I get the final system in place, which could take up to a month to do it right.

I have a lot of stuff that has been taking up space in my shed and garage to take to the Cabin, so I built an 8' x 7' box on my trailer. It is maybe 80% packed right now.  I need to haul all this stuff across 2 miles of lake in a 17' Lund when I get there, the batteries and power system will be in the first load.  I'll eventually use the plywood from the box for camp projects.

BB.

Don't let "minor" issues (like controller efficiency of 96 vs 98%) worry you about overall system design. If you are within +/-10% of your estimates (loads, harvest, etc.)--You are doing really well (10% is pretty much "perfect" in the solar energy world).

HOWEVER--For micro issues, that can matter... Say you have a charge controller passing 2,000 Watts with 2% losses--That is 40 Watts of heat that has to be dissipated. a 4% loss is 80 Watts of heat to get rid of--Need twice the air flow to keep the controller cooler at those higher losses (mounting in closet, under shelving, etc.) matters more with higher losses.

A good rule of thumb from Engineering and Physics... For every 10C (18F) increase in temperature, there is a 1/2 reduction in life (electronics, plastics, etc.) 70F "room temperature" vs 106F hot closet/under a sunlight filled window, 1/2*1/2 = 1/4 the estimated operating life.

Also, thermal cycling is very hard on devices too (40F in shed overnight vs 106F middle of the day)--Can kill devices in months (broken solder joints, fractured plastic transistor cases, etc.).

-Bill