Upgrade Dilemma -- Please Weigh In!

openplanet

My girlfriend is moving in, so electrical loads will increase. Getting an apartment sized heat pump dryer, small mini-split heat pump for backup space heating, small microwave, etc. So I need to double my array and battery bank, and am facing a number of decisions that I suspect many others, in one form or another, may be also be facing. So I’d very much like to hear people’s thoughts. I’m totally open to suggestions, advice, and accusations of ignorance or stupidity. After all, that’s why I’m posting!

Here’s my existing system:

*Two parallel strings of 3, 295 watt panels in series, for a 1.79 kW array with a Vmp of 135 volts

*Midnite Classic 200

*Four Rolls S530 6 volt batteries in a single, 24 volt string for 10kWh of storage (3kWh usable based on a 30% dod)

*24vdc loads: overhead light fixtures (led bulbs), ceiling fan, pressure pump, SunDanzer ‘fridge.

*110 vac loads powered by a 2kw Samlex inverter: Staber clothes washer, notebook computers, medium sized flat screen TV, 40 watt receiver, C-pap machine, and a few lamps (led bulbs).

*Meanwell 24 volt battery charger and Honda 2000eu generator for the relatively few days per year when the battery bank has dropped below 70% soc.

*I consume an average of 2kWh per day. When it’s sunny the battery bank reaches float by noon. But it’s often...not super sunny.

As you can see, I’m running a number of critical load on the 24vdc side of the system. I do like the fact that those loads aren’t dependent on the inverter (a.k.a. “a fancy gizmo that can fail”). The few times the battery bank has dropped below what the inverter needs to operate, those critical loads have kept right on cranking. And leaving aside the “All a.c. vs hybrid” debate, I have an investment in 24vdc loads, so even if I decided that all a.c. is the best approach, it’s too late to change.

I know it makes total sense to double my battery capacity by moving to a single, series, 48 volt string. This would allow me to double the array (just repeating the current panel configuration and putting the two arrays in parallel) and keep my single Classic 200. One downside right off the bat is that I’d be retiring my current bank of Rolls betteries, which are only 2.5 years old and in pretty good shape.

One approach is pretty straightforward, but expensive:

*Buy eight new 6 volt batteries to make a 48 volt bank

*Upgrade to a 48vdc inverter / charger (tempted by the AimsPower’s features and low price, but thinking it may be too good to be true; that is, that at such a low price it can’t possibly approach the reliability of, e.g., a Magnum or a Schneider).

*Buy a dc-dc buck converter to power the 24vdc loads.

But here’s the thing…there are other options. I could...

*Simply replicate the system I have—array, charge controller, and battery bank—and use a great big A-B switch to run the house off one bank or the other. Result—no need to buy a new inverter and dc-dc buck converter, and I only have to buy 4 new 6 volt batteries instead of 8. Yes, I’d have to buy another Classic 200 charge controller...but you have to admit, there’s something about two completely redundant systems that adds a pretty significant degree of fault-tolerance.

*Buy 4 new 12 volt, 200ah batteries for a 48 volt string of the same kWh capacity of my 400ah 24 volt string, plus a 48 volt inverter/charger. I’d then make use of the 24vdc string by charging it from the 48vdc bank with an MPPT charge controller. In this case I’d run a.c. loads from the 48 volt string, and dc loads from the 24 volt string.

*Just go ahead an mix the old and new batteries into a 48 volt series string, and accept that the new ones won’t last as they would if the whole string was new.

So many options...that’s what makes it challenging. But on the positive side, it’s and interesting problem. Ultimately I’ll probably end up going with a new 48 volt bank, 48 volt inverter/charger, and a dc-dc converter for the 24 volt loads. But I suspect people may well have ideas I’ve not yet hit on. So...bring ‘em on. And thank you.

Estragon

A couple of points to noodle on.

- I've also considered running low voltage loads with a DC buck converter, but have recently read of another poster here having one fail in a way that exposes the low voltage loads to the high voltage source. Not ideal, especially for "critical" loads.

- Your new loads don't seem all that well determined in terms of peak current required, run times, etc. As you probably know, the better you can quantify loads, the better your system design will work. You can work from specs on the new appliances, but unfortunately girlfriends tend to be harder to get reliable load estimates on.

Since you already have a working system, why not add the loads (and girlfriend) to your existing system for a while and get some real-world experience to help with getting the upgrade right? You might end up running the generator more for a bit, but the info you get might be worth the price of a few gallons of gas?

If you really want to upgrade pre-girlfriend, and absent better load info, I would lean towards getting 8 more of the same batteries you use now for a new 48v bank, and enough panel to charge them. It may be more than you need, but you may find you grow into it.

Using the same batteries may allow you to use them as spares to keep one string or the other going for a time when the other string approaches replacement time. Using 4x12v batteries in parallel may cause balance issues, and would be problematic if you need more capacity.

BB.

For near term--I am tempted by paralleling the two battery strings ("old" and new) and live with 24 volts. I am a bit more happy with paralleling old+new in 24 volts vs old+new in series for 48 volts (as batteries age, they have higher self discharge--To keep the "old batteries" properly charged, you have to "over charge" the newer batteries). Just a personal opinion--Not sure it will even buy you a cup of coffee.

Check the details on the heat pump drier... That could, by itself, almost use 2x as much energy as you are using today (somewhere around 3.5 kWH per load?).

I remember that you live somewhere in Maine?
http://www.solarelectricityhandbook.com/solar-irradiance.html

Bangor
Average Solar Insolation figures

Measured in kWh/m2/day onto a solar panel set at a 45° angle:
(For best year-round performance)

Jan	Feb	Mar	Apr	May	Jun
3.35	4.16	4.74	4.71	4.66	4.78
Jul	Aug	Sep	Oct	Nov	Dec
4.82	5.01	4.63	3.77	2.96	2.65

You get cold weather--So drying in the home (wood stove?) can be pretty quick in the winter.

If you have a garage--Perhaps during sunny weather it gets warmer? My father, decades ago, made a clothes drying rack that he would crank up into the rafters of the garage. Even living on the cloudy coast just south of San Francisco California (moss, mold, etc.), it did a pretty good job of drying clothes in a day...

You might need a larger battery charger (and genset?)--But you might try running on your battery bank as is and see how things go. If you only run the drier (other heavy/optional loads) a couple times a week--You might do OK with a larger solar array and a bit more genset run-time.

Otherwise, a propane (if already on property) might work well too (especially if you can pre-dry your clothes, and just touch-up in the drier before putting clothes away).

I don't think you are looking towards a ~6 kWH per day life style (your partner may have different ideas).... Assuming 3 hours per day of sun:

• 6,000 WH per day * 1/0.52 off grid system eff * 1/3.0 hours of sun per day (Nov break-even month) = 3,846 Watt array

And your battery bank would look like:

• 6,000 WH per day * 1/0.85 AC inverter eff * 2 days storage * 1/0.50 max discharge * 1/24 volt bank = 1,176 AH battery bank @ 24 volts

That is a pretty large battery bank--10% rate of charge would require ~120 Amp charging--Two of the "standard" MPPT charge controllers.

• 1,200 AH battery bank * 29 volts charging * 1/0.77 panel+controller derates * 0.10 rate of charge = 4,519 Watt array nominal
  

I would really push the conservation (low power appliances, alternate fuel sources). You have a decent size battery bank. If your heavy loads are during the week and only a couple loads per week (dark clothes one day, lights a few days later)--I would just look at maximizing your solar array (and charge controllers) and stay with the 24 volt @ 600 AH bank you currently have. And see what happens. Running a gallon of gasoline per week during winter/poor weather to run the heat pump drier (assuming your eu2000i can run it) is not the end of the world. You may end up having to do the wash (if your partner is not into starting the genset).

The largest "cost effective array" and "over arrayed" for your bank would be:

• 600 AH battery bank * 29 volts charging * 1/0.77 panel+controller derates * 0.13 rate of charge = 2,938 Watt array "cost effective" maximum
• 600 AH battery bank * 29 volts charging * 1/0.77 panel+controller derates * 0.20 rate of charge = 4,559 Watt array "over arrayed" maximum

With the over-sized array, you need the remote battery monitor and you may want to use the Wizbang Jr. to reduce charging current to 13% if the battery bank gets too hot during sunny weather charging/deep cycling.

The up sizing of the panels and charge controllers are something that you can keep even if you finally decide to upgrade to a 48 volt bank later (or parallel more batteries). Your present bank is good for ~3 kWatt maximum inverter (1kW inverter per 200 AH @ 24 volts rule of thumb).

I would suggest spending the money on a larger solar array + charge controllers and keep your present bank. Simply doubling the bank capacity with only 1.79 kW of solar panels is not a good solution here--Especially if you 2x the battery bank capacity. You will be under charging the battery bank unless you use more genset (and assuming full time off grid, increased loads). Batteries will wear out (panels should last 20+ years, charge controllers+inverters should last ~10+ years or so).

• 1,200 AH * 29 volts * 1/0.77 panel+controller derating * 0.05 rate of charge = 2,260 Watt array minimum for your 2x battery bank (5% rate of charge for seasonal/weekend off grid cabin)
  

-Bill

openplanet

Thanks so much Bill for the super detailed and informative reply.  Do note that if I double the array and go to a 48 volt battery string, my one Classic 200 will suffice, whereas doubling the array and staying at 24 will indeed require a second Classic 200.  Admittedly, that alone probably isn't enough of a savings to warrant upgrading to 48 volts given the cost of batteries.

Something I definitely don't have a good sense for is what kind of efficiency gain the Classic will get by moving from a 24 to a 48 volt battery bank.  I know the Classic won't be working as hard, but by what percentage, ballpark?  I guess I should post that question on the Midnite forum.

Your suggestion to double the array, add another Classic, and stay at 24 volts is compelling.  I'll give that approach some serious thought.

BTW, after scouring the 'Net for the most efficient electric dryer, I came up with this:
https://www.ajmadison.com/ajmadison/itemdocs/DHP24412W-specs.pdf
I hope the specs are honest, and 900 watts is truly its power consumption.

Thanks again.

jonr

> The few times the battery bank has dropped below what the inverter needs to operate, those critical loads have kept right on cranking.

An alternative wording to this might be "when the battery DOD dropped to damaging levels, the loads kept draining it".

I agree with putting as much as possible on propane.  Certainly for things that can't be operated only when the sun shines (ie, a HP dryer used only during a sunny afternoons might be OK).

Estragon

The 24/48v decision has a few aspects. Yes, the classic will be a bit more efficient going from a given array voltage to 48v than 24v, but it's the current that gets to be more of an issue.

Basically, you run twice the current for a given wattage at 24v. That means 2x controller capacity and 2x inverter capacity and 2x wire capacity. 3500 watts is ~ 29aac. It's 73adc @48v or 146acd @24v.

Estragon

Also, had a quick look at the dryer. Looks pretty efficient. I have a Blomberg dishwasher and it seems pretty well made and works well.

Note that the dryer appears to be 230v though. At 24v there will be fewer inverter choices (schnieder sw2524 2000w 120/240?) than 48v, unless you use an autotransformer.

BB.

The efficiency difference for the mppt controller is probably only a couple percent difference. I would not base my design choices on that. Add one more solar panel and you have swamped any extra losses.

Plus your wiring losses will be on the order of 1-3%.

48 volts lets you use 1/2 to 1/4 the amount of copper. That can be a big deal when trying to bus 250+ amps around your battery bank+ac inverter+breakers vs 1/2 that if you double the bank voltage.

- Bill