# Why are standard battery bank voltages so low?

I'm a mechanical engineer by training, and I've studied switching power supply electronics a lot in the last year. What I want to know is why don't PV systems get designed with higher DC voltage busses? Like 100V DC or 120-150V DC etc?

Of course, there are *very* few products designed and sold for DC voltages above 48V, so that leads designers and installers to design around 12, 24, or 48V most frequently.

Consider this.
Many PV systems for homes and offices are above 1kW, meaning there are at least 10 panels in the system, and if wired in series you could get 150 - 180V DC out of the string.
Also, consider that eventually the voltage will have to be brought up to 120V or 230V AC depending on your standard voltage where you live. In order for inverters to produce 120V AC, they must first transform the bus voltage as high as 170V.

So why isn't anybody building systems and products with higher DC (over 100V DC) bus voltages, where all batteries and all panels are in series, and you don't have all the losses associated with converting voltages from very low (24V DC) to very high (120-230V AC)? Every time you run through a transformer you throw away energy due to transformer losses. In addition, you can use smaller wires everywhere, and you can use components that don't need to be able to handle large currents.

I know of course that the higher voltages are more dangerous . . .
and that if you make 24V components, the pieces can be arranged more flexibly. . . .

Anyone have any feedback? Do US regulations demand that DC voltages stay below a certain level?

I've seen inverter products for grid-tie systems that have DC input voltages that start at ~120V DC, but never anything similar for systems that include batteries.

Thanks,
GNB

Re: Why are standard battery bank voltages so low?

I can give you some of the reasons--others here can fill in the details.

One is the FET's (field effect transistors) have stray capacitance, so the higher the input PV panel voltage, the more losses there are (this seems to apply to off-grid systems, but not as much to on-grid systems--Solar Guppy or others can probably add the details).

Which leads to Grid Tied Inverter systems (or Utility Interactive Inverters systems)--these typically do run at 200 to 600 VDC on the solar PV panel input bus--so they are using high voltage there.

Another issue is that the NEC and other international safety regulations typically use 60 volts (DC in this case) as the maximum "safe" voltage before isolation and other issues are required/tested.

Yet another issue is that DC current is much more difficult to interupt (fuses, circuit breakers, switches, shorts/faults). DC Arc welders are much easier to sustain an arc than a similar votlage/current AC arc welder. If you look at a breaker/switch/etc. you may find that they both have AC and DC ratings for the same switch. The DC rating is much lower in voltage and current.

Also, DC is actually much more dangerous vs AC as a shock hazard. Basically, any current over ~10 mAmps (0.01 amps) through the heart can stop it (i.e., right foot in water, left hand on water). The DC danger is that it can cause muscles to contract--such as cause your hand to grasp firmly the source of the current and you cannot let go (AC is easier to let loose). Probably another reason for the 60 VDC maximum voltage (reduction of current flow in damp skin).

You will find larger computer UPS systems out there with high voltage battery banks/buses (over 100 volts DC). It is an interesting question why we don't see solar RE off-grid battery banks with higher input voltages--but I would guess it gets back to the safety, training, rating, etc. issues on systems that, many times, are customer installed and maintained (no licensed electricians around).

Now, you mentioned the use of 24 volt systems and 150 VDC solar panel arrays... As I said, Grid Tied systems (which connect directly to the utility grid are high voltage).

But, 48 volts is also used quite a bit for Solar RE (12 volts is OK for ~1,000 watt maximum, 48 VDC is good for over 2,000 watt systems). Larger Inverters are typically available in 24 or 48 VDC units--with very little price difference.

And solar battery chargers (MPPT type--Maximum Power Point Tracking) are also, typically, configurable for 12 to 48 VDC use (really about 62 volt maximum battery bus voltage during some charging cycles). And--since the charge controllers are rated for maximum battery charging current (typically 60-80 amps maximum for the larger Solar RE units)--they can use 4x the number of solar panels on a 48 VDC battery bank vs a 12 volt battery bank (cost and complexity savings).

The input voltage for a off-grid MPPT solar charge controller is roughly 2volts + maximum battery bank charging voltage. So--for a 48 volt battery bank, the solar panels need a Vmp>~62 volts. And the typical large 48 VDC MPPT controller's maximum operating voltage is 140 VDC--so there are higher voltage off-grid systems available.

Now--there is an issue with running at 140 VDC on a solar charge controller... Solar panels are not "perfect" solar batteries... There output voltage increases as the temperature drops--and there is a voltage difference between a panel under load and one with no load...

That Vmp (hot/loaded) to Voc (cold/no load--batteries charged) is almost a 1:2 range--so if you want to run 140 VDC on a bright/cold day (in the snowy north), your panels will output only 70-80 VDC on a very hot sunny day (range depends on location/usage/etc.).

Yes, there are transformer and inverter losses--but the modern electronic devices are becoming very efficient at working with DC (you can get >90% efficiency for some now, and very low standby currents too). At this point--for many larger "home" installations, it is almost better to work with 120 VAC appliances and a True Sine Wave inverter (and accept the losses) vs trying to run a 12 VDC lifestyle (very short runs, large cables, "car type" adapters, more expensive DC devices, issue with Battery bank voltage from 10.5-15.5 VDC voltage range causes many "car type" DC devices to fail).

All the above being said--People continue to develop new equipment--I think I read here that we may have a solar charge controller that breaks the 140 VDC solar panel input voltage...

And, there are ways of using current equipment to push the 140 VDC barrier to 200-600 VDC using a solar GT inverter with a off-grid True Sine Wave Inverter--basically using the GT inverter to back-drive the off-grid inverter which will charge the battery bank (with added safety issuses and complexity--not something for the average user to tackle at this time).

-Bill
Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
• Solar Expert Posts: 200 ✭✭
Re: Why are standard battery bank voltages so low?

Bill hit on a lot of good points. I think low voltages are more or less a legacy from the "old days" of solar. When solar installations were first implemented, there was no grid-tie, and there were plenty of 12-volt appliances. There was really no need for higher voltages and the complications that came with them. From an electronics standpoint, practical switching power supplies (for use in controllers and inverters) were still in their infancy. Technology has improved dramatically since the 70's, especially in the case of FETs. While the parasitic capacitance of the FET has actually increased, it's been accompanied by a steep drop in on-state resistance, so they are now able to handle much more current and operate cooler and more efficiently. The biggest problem with a FET is that its conduction losses are nearly proportional to the current passing through them. With higher voltage FETs (which by nature have higher on-resistance) and larger currents, this can contribute to a significant loss in efficiency. So today we also have IGBT (insulated gate bipolar transistors) -- that combine the relatively constant voltage drop of the bipolar transistor with simple drive requirements of the FET.

I might also add that costs for components have come way down over the decades, so it is more economical now to design for higher voltages.

Marc
• Solar Expert Posts: 1,458 ✭✭✭✭
Re: Why are standard battery bank voltages so low?

Some legacy and some safety. Above 60 vdc gets dangerous.

Many UPS's for commercial use have 32 to 40 twelve volt batteries in series. That is in 600 vdc range. Just changing the batteries is a dicey operation.

fyi, hybrid vehicles are in 300 to 500 vdc battery voltage range. After a few more years when factory warranty gives out, expect to hear of some garage mechanics getting fried.
• Solar Expert Posts: 72 ✭✭
Re: Why are standard battery bank voltages so low?

120vdc systems were very common during the pre REA days of the 1930's, 40's, and 50's. During the 1970's and 80's when wind energy was making it's comeback, 120vdc was considered the Cadillacs and the old pre REA 120vdc generators were highly sought after and brought premium prices over the standard 32 volt generators. Just about every inverter manufacturer had 120vdc inverters available but they were very inefficient and used SCRs for switching . Midway through the 80's, Trace and Heart came out with much more efficient and dependable inverters using FETs for switching but didn't offer higher voltage units. My guess is high voltage FETs were not available at that time. The present inverter manufactures have dictated the use of lower voltage systems...why, I don't know. There is one maverick out there, Chad Lampkin of Michigan Energy Works in Sanarac, Michigan has been hand building 144vdc inverters since at least 1990. The last time I talked to him several years ago, he had several hundred successfully operating inverters operating in the field and he was making both pure sine wave and modified sine wave inverters. I personally have two of his modified sine wave inverters. They have been flawless and extremely efficient, better than 98% at all loads AND 0.01 amp no load draw. No other manufacturer can claim anything close to that. Needless to say, I'm totally sold on them.

Concerning circuit protection, I use Buss NON fuses. They are rated 125vdc. I've blown a number of them on 144vdc and they've always performed flawlessly. Fuse boxes for NON fuses are very inexpensive and even available at Home Depot. I'm sure there are higher voltage fuses available since 180vdc motors are common in industry.

I started using 120vdc in 1981, then later moved up to 144vdc and I'm still alive. I've been shocked many times by it and I don't think it's as bad as 120vac. In the almost 30 years I've been using high voltage DC, I've never heard of anyone being injured or killed by it. I'm not saying it can't happen but 48 volts can also theoretically kill.
• Solar Expert Posts: 200 ✭✭
Re: Why are standard battery bank voltages so low?

A look back at the electrical demands of yesteryear can be educational. Pre-REA folks were happy if they had a few electric lights and perhaps a radio. If you look at wiring from houses back to the 30's or so you'll find that they had about two circuits for the whole house -- which was adequate considering they didn't really have much demand for electricity. Contrast that with the typical household today and the difference is staggering. Makes one wonder how much of that extra demand is really necessary.

MOSFETs didn't really become practical until the early 80's. I think Motorola and Toshiba were two of the first companies to offer them. They were available with a breakdown rating of up to 1kV, but the on-resistance was on the order of 10 ohms or so...clearly not the best for efficient switching. And the Rds was probably only exceeded by the cost. SCRs have been around much longer and are available with much higher ratings that FETs (thousands of amps and tens of thousands of volts). Today's MOSFET is much more efficient than an SCR, though, so the only inverters that still use SCRs are very high power types.

I read somewhere a while back that the original 32v standard was based on the likelihood that a higher voltage could become lethal under some worst-case scenarios. I think that 32 volts is still considered to be the low end of what might be dangerous...but a lot of that depends on other factors, too (body hydration, path of current, frequency (for AC), etc.).

Marc
• Solar Expert Posts: 72 ✭✭
Re: Why are standard battery bank voltages so low?
lorelec wrote: »
A look back at the electrical demands of yesteryear can be educational. Pre-REA folks were happy if they had a few electric lights and perhaps a radio. If you look at wiring from houses back to the 30's or so you'll find that they had about two circuits for the whole house -- which was adequate considering they didn't really have much demand for electricity.

Marc

Not really, they had 32vdc electric mixers, toasters, irons, washing machines, refrigerators, deep freezes, vacuum cleaners, waffle irons, well pumps, milking machines, electric blankets, table and ceiling fans, electric drills, and yes, arc welders in addition to lights and radios that you mentioned. I remember seeing literally 100's of 32vdc and 110vdc 1/4hp to 3/4hp, 1800 rpm electric motors in an old farmstead in North Dakota years ago. The old farms were definitely very electrified.

There are several reasons 110/120vdc was popular.
1. Most 120 volt appliances available in the 1930's through 1950's were rated AC or DC so you could buy standard off the shelf appliances.
2. Much smaller wire was needed to keep voltage drop reasonable.
3. No series/parallel connections on batteries. Just one set of batteries in series. If one cell went bad, remove it or ignore it.....not a problem. Simplicity is beautiful.
• Solar Expert Posts: 200 ✭✭
Re: Why are standard battery bank voltages so low?
jacobs wrote: »
Not really, they had 32vdc electric mixers, toasters, irons, washing machines, refrigerators, deep freezes, vacuum cleaners, waffle irons, well pumps, milking machines, electric blankets, table and ceiling fans, electric drills, and yes, arc welders in addition to lights and radios that you mentioned. I remember seeing literally 100's of 32vdc and 110vdc 1/4hp to 3/4hp, 1800 rpm electric motors in an old farmstead in North Dakota years ago. The old farms were definitely very electrified.

Pre-1930's? Rural farmsteads had all of these appliances running off of windmills?

Marc
• Solar Expert Posts: 72 ✭✭
Re: Why are standard battery bank voltages so low?
lorelec wrote: »
Pre-1930's? Rural farmsteads had all of these appliances running off of windmills?

Marc

Wind generators were first introduced about 1928. Appliance manufacturers quickly responded to the market. Watch eBay.....you will occasionally see 32vdc appliances for sale there. Thrift stores are a great place to find 110/120 volt ac/dc appliances. I've found many there.
• Solar Expert Posts: 72 ✭✭
Re: Why are standard battery bank voltages so low?

Here's an 1930's photo of a dealers display.
• Solar Expert Posts: 200 ✭✭
Re: Why are standard battery bank voltages so low?

Very neat. It was always my understanding that the early wind-charged farms had little more than some lights and a radio. Maybe I'm thinking of an earlier period. Judging by the houses I've seen from that period that still have some of their original wiring intact (and granted, there's only been four that I can remember), I can't imagine that they would accomodate many appliances. Maybe one outlet and one light per room. Two to four circuits. What's even more interesting is some of the turn-of-the-century homes from the big cities, before electricity came into widespread use. Gas lighting fixtures!

Marc
• Solar Expert Posts: 72 ✭✭
Re: Why are standard battery bank voltages so low?

Not all farms were equipped with the better and larger wind generators along with all the appliances. It's just like photovoltaic and wind systems today, there are large and small systems. It all depended on how successful and how progressive the farmer/rancher was.
Re: Why are standard battery bank voltages so low?

Here is a good explanation about Delco-Light:
Written by Mike Brazeau
Charles Kettering and Edward A. Deeds formed the Dayton Engineering Laboratories Company (DELCO) in Dayton, Ohio. DELCO was originally a research and development company but began manufacturing to meet the demand of the automobile self-starter they had developed. In 1916 Billy Durant bought DELCO under his United Motors Corporation holding company and it was folded into General Motors in 1920. One of their other widely known inventions was the Delco-Light Plant. It was a gasoline powered electric generator that powered many farms in rural areas across the country before electricity was readily available.

By 1930, 90% of urban America had the benefit of electricity but only 10% of rural America had been electrified. My mother recalls, as a child growing up on her parents farm in rural Greytown, Ohio in the 1930’s, hearing the hum of the "Delco" (as they called it) and the rows of glass batteries lined up on the shelves in the garage. The generator would run a few hours a day and store electricity in the batteries for use as needed. There were some 100 different models built over the years but the 850-watt, 32 volt DC unit accounted for 75% of the production. Being a unique 32-volt system, DELCO also sold the appliances to go with the light plants.

There had been nearly 150 different companies manufacturing home light plants with General Gas & Electric, Westinghouse Machine Company, Kohler Company and Fairbanks-Morse being a few of the more famous, but Delco-Light was clearly one of the most successful. They had sold more than 350,000 units by 1935.

DELCO built light plants for the war effort during WWII and continued production after the war but Congress passing President Roosevelt’s Rural Electrification Act in 1936 had finally brought electricity to most rural areas and in 1947 production of the Delco-Light unit had ceased.

-Bill
Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
• Solar Expert Posts: 1,341 ✭✭✭
Re: Why are standard battery bank voltages so low?
lorelec wrote: »
Gas lighting fixtures!

Marc

Oh you ain't seen nothing yet!

I went and checked out this place years ago...and just about crapped my pants.

http://www.glendalehistorical.org/doctors.html

Gas AND Electric - IN THE SAME FIXTURE! YIKES!!

Imagine a brass lighting fixture, with two arms, hanging from the ceiling. It has two glass "tulip" style globes. One arm points downward, and the other turns upward.

I'm looking at this thing, and the downward turned arm has an electrical lamp socket in it, but the upward turned arm has a...a...a valve? What the hell?

So I ask the tourguide and he says that's the gaslight. For when the electricity is turned off at night.

So I'm studying this thing and I see that there is only a common brass tube between the two arms...

The wires run down that tube...and SO DOES THE GAS!

GREAT GOOGLY MOOGLY.

Made my skin crawl looking at that thing.

Attached is an image from the site. It's a small image, but it shows a 4 armed fixture that's the same design (different glass globes though). Note how the two upward arms are dark - those are the gas lights.

Edit: Found another pic that looks kinda like the fixture I remember.
• Solar Expert Posts: 72 ✭✭
Re: Why are standard battery bank voltages so low?

Sounds similar to today.....electric fuel pumps submerged in gasoline on modern cars.
• Solar Expert Posts: 200 ✭✭
Re: Why are standard battery bank voltages so low?

Interesting fixtures! Supposedly there wouldn't be enough oxygen inside of the tube for combustion if a spark occurred, but I'm not sure I'd want to experiment much with that theory.

If you're ever in the Seattle area, check out Gas Works Park. Originally it was a coal gasification plant that supplied the city and surrounding area with gas for lighting. It was turned into a park some years ago, but still retains much of the original machinery and equipment.

Marc
• Solar Expert Posts: 10,300 ✭✭✭✭
Re: Why are standard battery bank voltages so low?

i'm enjoying the stuff talked about here, but it is kind of off the original subject from the title. maybe bb would like to carry this further with you guys in its own thread?