Using a UPS and Deep Cycle battery(s) to make a reeaaal man's 120 VAC backup

[K5WLF]

12v vs. 24v vs. 48v inverters.  What's the advantage/ disadvantage.

Each time you double the input voltage, the input current is halved. Thereby giving more runtime from a battery bank of the same Amphour rating. Assuming equal loads and efficiencies.

[K6IC]

Generally,  the more power that the inverter system needs to supply, a higher  battery voltage is chosen.   Higher battery voltage yields higher system efficiency --  less current from the the battery.  This yields lower IR drops.  Also,  the battery charging system runs at lower current on higher battery V,  so the charging system is also more efficient.

For Solar battery chargers,  the maximum charging current is fixed,  so doubling battery V can allow twice the charging power for the same charger output current (within limits).

A 1000 watt 12 volt inverter could easily demand more than 100 DC Amps from the battery.  Many high power 12 V inverters are specmanship hype IMHO.

But for Tom Vu-san's backup system, he does not need tons of power,  and 12 V is perfectly reasonable.

The off-grid power system,  under construction, for AM sloop (not Tall Ship) power levels,  below,  FWIW.  Vic

[WD5JKO]

The lower voltage systems have higher currents, and as already pointed out, this creates power loss from conductor resistance. Another factor is the semiconductor transistors used where the "ON' voltage needs to be a very small percentage of the applied DC voltage. If the on voltage is more than a few % of the applied DC voltage, then the overall system efficiency suffers.

So these days you can get FET's with < 10 milli-ohm on resistance. But consider 100 amps at .01 ohm and we get 1 volt "on" voltage. This approaches 10% of a 12V battery that is near low voltage cutoff. The other choices sometime used is the IGBT where the zitter has FET input and transistor output. Here these can be sized to the application, and get approximately 1/2 volt VCE SAT ("ON" voltage).

With switchmode power supplies, the rectifiers used also are a big factor. A 5V DC supply will always be less efficient than a 12V supply. A silicon diode used in the rectifier might drop around 1 volt whereas a Schotky diode will drop about half that.


Case in point, some years ago I 'updated' an old Ham DC-DC converter that took 12V and made 800, and 300v to run a tube type transceiver in a mobile enviornment. I made a drive PCB using a PWM chip, and replaced the four PNP Ge transistors with power FET's. The output was regulated (LV B+), and everything. The part I did not anticipate was with PWM, the FET 'On' state current was over 50 amps when the average DC draw was around 25 amperes (50% duty cycle) off the 12V battery. So with the FET RDS "ON" resistance, and 50 amps, my "ON" voltage was over 2 volts, or 2 X 50 = 100 watts whenever the FET's were "ON". In contrast, the old Germanium PNP's had an "ON" voltage at just under 0.5 volt. I keep meaning to update this unit with more modern FET's in the hope of getting the "ON" voltage back to where the old PNP's were at.
Jim
WD5JKO

[K1JJ]

Question:

When charging the batt using  the 1.25 A  "smart" trickle charger, the battery tops out at 14.4 vdc and then the charger shuts off.  I then disconnected the battery so that there was no external draw on it. Within a day I measured the voltage and it had dropped down about 2 volts to 12.5vdc.

I then put the trickle charger back on it and the batt charged back up to 14.4v within 3 minutes.  What's that all about?   There's obviously not much power being trained internally if it takes only 3 minutes at 1.25 A to climb 2 volts.

Does the batt's cells naturally want to be at 12.5V?  So why is the trickle charger designed to top it off at 14.4v if it cannot hold a charge there?  I would think this is overcharging - and could shorten life if I kept the charger on all the time set at 14.4v, no?


On the last inverter test, the batt dropped down to 12 vdc after a healthy night of power draw. So maybe it's really dropping only about 0.5 volts.  (12.5 - 12.0)

T

[WD5JKO]

Tom,

   What you describe seems about right. A fully charged lead acid battery AFTER it has been off the charger for a while will revert to about 2.1 volts per cell, or 12.6V for a six cell battery. The charge too voltage, and float voltage are described in detail here:

http://batteryuniversity.com/learn/article/charging_the_lead_acid_battery

"Lead acid batteries should be charged in three stages, which are [1] constant-current charge, [2] topping charge and [3] float charge. The constant-current chargeapplies the bulk of the charge and takes up roughly half of the required charge time; the topping charge continues at a lower charge current and provides saturation, and the float charge compensates for the loss caused by self-discharge."

Way back when on my UPS escapades, we designed the charger as above, and the  had a long CMOS timer to indicate need for a cycle test at 1 year. So after 1 year unused at the float voltage, pushing a button ran a one time discharge, and then the charger topped up the 60V battery pack to 2.5V / cell. This seemed to help keep the battery back A-H rating from declining over time. Afterward the float voltage was reduced to around 2.25 volts / cell for another year.

If you have a two stage charger where the float voltage is on the high side, then it might be best to unplug the charger after a couple of days. You could hook it to a timer to turn it on for a while every week or so.

Earlier in this thread you asked about how low to go when the batteries were at 50% capacity. I said to stop at 1.75V / cell, or 10.5V for a 6 cell pack. What I said was true when the batteries are near 100% used up. The answer to your question depends on the battery you got, and the discharge rate. Paul, in post 35 gave you a link, and I post a graph of that here. It shows 80% gone at about 1.85V / cell. Also at around 1.7V / cell the curve goes vertical, and here the battery voltage drops like a rock. Don't go there.

Sorry for the confusion,

Jim
WD5JKO

[K6IC]

Hi Tom,

As batteries near fully charged and fully discharged the internal R increases by a considerable amount.   This higher R at the charge/discharge end-points causes the battery V to change fairly quickly with a given amount of current. This is what causes the small charger to drive the battery V up very quickly with its low current.

An exception to this is when doing an Equalization charge (EQ).  This charge for Flooded batts increases the charge V to about 15.5 V for 12 V flooded batteries,  and the current required to reach this overvoltage can be quite high -- limited by the charger's current capability.

EQing a battery is used to attempt to overcharge the good cells to try to fully charge the lower SG cells in a string of batts.

BTW,  here is some interesting info on batt V vs SOC,  and vs current for charge and discharges and so on.  It is from Home Power mag:

http://scubaengineer.com/documents/lead_acid_battery_charging_graphs.pdf

At some point,  you might want to consider a charger that has more current capability,  and one that has a Remote Temp Sensor to compensate charge voltage for the measured battery temp.  There is about 5 miliivolt change in batt V per batt cell in series,  per degree C,  IIRC,  so it can be significant,  even on 12 V banks.  GL,   Vic

[N0WEK]

Back when I was in the Navy I spent a couple of years on a couple of WWII diesel/electric submarines. That was a battery system! Two 126 cell batteries @ 1 ton per cell, charged by four Fairbanks Morse 10 cylinder opposed piston two cycle diesels @ 1,600 HP each. During equalization charges you could boil the water in the rinse side of the sink with the immersion heater that was there to keep the rinse water hot. Propulsion submerged or surfaced was by two 2,600 HP electric motors. Switching for all this was by BIG multi-bladed knife switches activated by big chromed levers back in manuevering. If you forgot to turn down the motor fields during a forward to reverse manuever you go a BIG fireball, melted copper ran down into the bilges through the slag chutes and you were the known as "Fireball" from then on!

Good times!

[K1JJ]

Thanks for the extra info, guys.

The links were excellent. I see there is a lot more to charging and maintaining batts than I thought.

I will put the system in the closet and put a timer on the trickle charger  to turn on every week.   Though the trickle charger claims it shuts down and automatically maintains the batt itself. I'll have to read about this unit some more.


OK on the "FireBall."    The captain probably got very POed... :-)


Thanks again for the advice.

Later -

T

[W1RKW]

I think the key is how the battery was maintained while on the shelf at the store. Don't know what the turn around time would be on a product like that but if it sits for a long period of time as it self discharges, can't be good.  

Good point.

Here's a piece of advice I read on the web and actually applied it to the Walmart purchase...

There were about eight of these Maxx-29 marine batts on the shelf.  I grabbed the first one and then remembered to look at the manufacturer's date code and pick the most recent one.  I noticed they all read, "04/2012."    But one in the back row read, "09/2012."   It's not much, but at least I gots one made only a month ago - pretty recently.  Why should I pay the same price for one that was almost 6 months older and probably sat dormant without a boost charge?


T

I went to a specialty marine store for two deep cycle 100Ah batts. I asked them if they were maintained while on the shelf.  They said yes.  Got them home and they were almost dead. Terminal voltage unloaded was less then 12v. Put both on 2A charger to slow charge them and it took 3 days each to bring them up. I haven't hard load tested them but they seem to be holding a nominal terminal voltage 2 days later.

Looking at the batts neither one of them had a date code but got a 1 year 100% warranty. We'll see what happens.

[W2PFY]

I went to a specialty marine store for two deep cycle 100Ah batts. I asked them if they were maintained while on the shelf.  They said yes.  

I think there is a federal law out there that states that a battery (lead acid) cannot be sold if it is over six months old. I doubt there is any retail facility that is equipped to trickle charge batteries on the shelf. Again that is another problem with the release of hydrogen gas without the proper ventilating equipment. While I am not an expert on batteries, I did make my living selling and servicing industrial batteries for over thirty years.

[KK4YY]

A possible deep cycle battery source for hams that are cheap like me...

I know someone who works for a business that services power wheelchairs, scooters, and the like. They cover a large service area and once they drive a few hours to a handle a customer complaint... well, they have to sell them something. How about new batteries? The old batts are taken back to the shop, put in a pile (no pun intended) in the back room, and eventually scrapped.

I have managed to get in between their back room and the scrap yard. I give my buddy $10 a piece when he's sent to get rid of the old batts. The majority of them are in very good condition, with recent mfg dates, and have many cycles left in them. The few bad ones, I can scrap myself for a "refund". Of course I'll eventually get to refund them all. Starting to sound like "free batteries".

I now have enough 12V batteries to make a "doomsday prepper" proud.  Mostly 35 AH, some 55AH, and a few 75AH (the physical size of a car battery). My source recently switched from gel cell to AGM style as they're get better pricing on them now. Great batteries, the AGM's. They're "no spill" (even if the case is broken) and that's a very good thing when the batt is in the house. Also most AGM's are the recombinant type and vent very little hydrogen gas compared to a wet cell.

So the thought here is to find your local power chair service center and pay them a visit at lunch time. Find to the youngest guy working there. He's probably the one that has to schlep the old batteries to the recycler. "Hey kid, you got any used batteries you don't need? I'll give you a few bucks to take them off your hands." It's old school ham radio part scrounging I'll admit but, as they say, it's so crazy it might just work.

For my part I'm hooking up the batts (2 at a time for 24V) to a 1000VA UPS with PV's to recharge them. The system is almost ready to go online (well, off-the-grid anyway). Just a few more wire nuts and... free power! I like the sound of that.

[W1RKW]

What are the pros and cons of recharging a deep cycle batt with a power supply vice batt charger?  Would there be any issue charging a battery with a power supply that has adjustable current limiting and adjustable voltage output?

[K6IC]

Bob,

Just be very careful.  Many of the Linear PSes  will fry regulator guts if powered off with the battery connected,  and some might even do that if the Vout setting is much below the batt terminal V.

A series diode should protect any PS against this.  Most switchers should be immune to this frying,  but why take the chance?

Please do not ask my just how I know this issue with Linear Lab supplies.

Good Luck with the WX.   Vic

[W2PFY]

What are the pros and cons of recharging a deep cycle batt with a power supply vice batt charger?  Would there be any issue charging a battery with a power supply that has adjustable current limiting and adjustable voltage output?

Maybe I can help here. The maximum start rate on lead acid batteries should never 25 amps per one hundred amp hours of capacity. I am talking about an 8 hour charge rate. the finish rate should not exceed 5 percent of the amp hour capacity. For example a six hundred amp hour battery would start at 152 amperes and the finish rate should be 30 amps. For batteries like trolling motor types, the start rate would be 25 amps if it is a 100 amp hour battery and the finish rate should be 5 amps.

Now for floating a battery and it doesn't matter what it's amp hour capacity is, the charge rate should be adjusted so the float voltage goes  no higher than about 2.33 volts per cell. All lead acid batteries will start gassing at 2.37 volts per cell so it's important to keep the float voltage below the gassing point otherwise you will come back at some point and see your acid level below the plate structure and exposing the plates to air. It is one real quick way to lower the amp hour capacity resulting in short life-ed battery.  It is also important to equalize your batteries at least once a month and that is done by adjusting your charger to 2.550 volts per cell for about 4-6 hours. Then you let the batteries settle down off the charger for about 24 hours and then re-adjust the charger for 2.33 volts per cell and you will be golden!

Another important note. Never add acid to a battery. Just distilled water. If you add acid to the battery you will over time be increasing the acid strength and this added strength will destroy the active materials on the plates and short life the battery. Sulfuric acid does not evaporate, only the water evaporates!. The only time acid is added is when you have had a major spill.

[K6IC]

The few things that I would add are;

If at all possible,  use the charging/discharge recommendations of the manufacturer of the exact battery you are using.

The nature of the use of the battery also dictates charge/discharge routines.

Use a Temperature Compensated Charger that measures battery temp.

Some Gel batts need charge currents be limited to about 5% of 20 hour Capacity.  Others can  accept 25-33% of 8 hr capacity.

For Renewable Energy applications,  most Flooded battery Finishing Currents are 1-2% of 20 hr C range.

If possible,  choose a Float V that just (barely) maintains SG.

The only consistant thing that I have discovered about batteries is,  that "it all depends".  Many,  many variables,  and many intreptations of what to do,  etc.   YMMV,  GL,  Vic

[KK4YY]

I've spent some time searching for the answer to this question. But to no avail.

Assuming I have lots of time to recharge a battery, which I do, could I apply the recommended float voltage to it and eventually get a full charge? If the float voltage maintains full charge shouldn't it also provide a full charge in due time? Manufacturers I've seen don't indicate this method but always supply the way that gives the fastest recharging time.

The setup would be a voltage regulated supply with current limiting to allow a very slow rate of charge. I would imagine this to be a gentle way to recharge causing little heating or outgassing possibly extending battery life.

The benefits would be not having to monitor the process, or change settings during it. Thus, it can be done unattended and without worry of overcharging while using about the most "dumb" charger at hand... a simple regulated supply most of us already have on the bench set at the float voltage.

Also, as a battery ages, a given cell may become weaker than others. I speculate that the manufacturers high current charging routine may cause even  further damage that a "gentle" charge wouldn't.

I am using gels and AGM's but I suppose this would work for a wet cell too.

Though slow, it sounds like a simple way to charge. But the more I learn about batteries the less simple they become!

Any thoughts?

Thanks,
Don
KK4YY

[WD5JKO]

Don,

   There seems to be nothing simple regarding battery chargers as there are always exceptions, or considerations. Here are a few that apply to Gel-Cels, and NIMH:

  I worked for a company that bought pallets of gel-cel batteries for use with exit signs. They 'stored' the batteries in a non-environmentally controlled warehouse in central Texas. When the batteries were needed over a year later, they were mostly dead, and many would not take a charge. We needed to make our monthly numbers, and there was no time or budget to replace those batteries. So, being in the R&D lab with 4 other hams, and an ex-ham, we scrambled for a solution.

  Many of those batteries went high impedance, and would not pull hardly any current from the chargers. We found however that a 10 ma constant current source with about 100 volts available would 'wake-up' the batteries in 1-3 days, and do so non-destructively. The ones that came back had a terminal voltage around 12 volts, and the bad ones stayed way up, sometimes over 50 volts (these were 12V batteries).

  The ones that 'woke-up' seemed to charge normally, and then hold a load. Doing an Amp-hour test showed the good ones had at least 80% of their rating left. Many, about a third had to be recycled.

  With NI-MH, and even Ni-Cads, old cells left discharged for long periods sometimes requires brief high current charging while monitoring temperature to get them going again. Ni-Cads often short out, and sometimes can be brought back by charging something like a 50,000 mfd cap to 12V, and dumping that charge across the shorted cell. Wear safety glasses and gloves when doing this as the battery might rupture!

Jim
WD5JKO

[K6IC]

Agree with Jim ...  there are many exceptions and variations in "best"  discharge and recharge approaches for any battery that I've used.

It is my opinion,  Don,  that  most sealed batteries will not be fully recharged if the charge voltage is limited to the speced float V.   Most sealed batts seem to have a higher Float V than do Flooded batts,  and this might help the recharge progress farther than would a flooded batt at its speced Float V.

It IS possible that the sealed batts designed for Float applications,  as in UPSes,  might also be more fully charged at their Float V spec.  Will try to find that info.

There appears to nothing consistant about even a given type of battery from manufacutrer to manufacturer.  Each has its own "special"  twist on things,  and to the extent that this twist is in the Chemistry,  and not just marketing crap,   it can affect the fine detail of charge and discharge parameters.

Deka is a large,  and good supplier of SLA batteries.  Here is their battery manual,  and is a good general read:
http://www.dekabatteries.com/assets/base/0139.pdf

Flooded batteries -- ones with removeable caps -- would do poorly being recharged at the speced Float V (at least the Lead Antinomy ones),  as they would probably only get to about 50% State Of Charge.  These batteries are designed for heavy cycling,  and will only reach about 85% SOC when the charge voltage just reaches the Absorption voltage.  This voltage is about 1.75 V above the speced Float voltage for a 12 V battery.  These Lead Antinomy batteries must gas to reach full recharge.  The last Sulphates in the plates will not be removed without this last part of the charge cycle.  And not removing the sulphates over an extended time period damages the battery.  And in time,  these sulphates harden,  become impossible to remove,   and essentially insulate the plates from the electrolyte,  reducing capatity.  This last part of the recharge is very inefficient, as it is breaking down the water in the electrolyte into Hydrogen and Oxygen  ...  and so on.

Personally,  I'd use a three-stage Charger for any Sealed Lead Acid battery.  It is almost guaranteed to fully recharge the battery,  and while it is rare,  the ideal 3 stage charger should have a Temp Sensor attached to the battery being charged,  as  sealed batteries are very picky about charge voltage,  especially Gells.

Good thing that things do NOT need to be perfect when dealing with batteries.  Flooded batteries are the most forgiving of errors in charging,  BUT they really dislike sitting around mostly or fully discharged.  Opinions,  YMMV,   GL,  Vic

[K1JJ]

Looks like this thread has turned into a very valuable summary about batteries, charging and UPS tips. Good stuff for the archives.


Well, as luck would have it, I had a chance to test out the new emergency battery/inverter (110 AH) system over the last four days due to the hurricane power outage.

I pulled about 4 amps each night over a 9 hour period. (36 AH)    After each session, the battery voltage stood at about 12.5 volts.  I recharged it with 5 to 10A for a few hours when the generator came on and all was fine for the next night.

So, it did the job and I'm happy with it.


Problems during the shake-down run:  

1) One of the crimp terminal connections pulled out making an intermittent.  Soldered and fixed.

2) The inverter kept beeping whenever there was a drop in current flow. Disconnected buzzer.

3) The inverter fan kept coming on and off with minimal current flow. Noisy distraction. Disconnected. Since the inverter is rated at about 25A and I need only about 4A, there is no need for a fan for now.

Now the system is quiet and works FB. Ready for the next "adventure" ...

T

[VE7 Kilohertz]

Hi fellows,

Very interesting discussion, as this is what I do for a living. Just found this thread. Thought I would throw in pictures of my "manly" UPS system. It's a whole house backup, as I live in the country and power outages are frequent in the winter. I saw Vic's pics and mine is very similar. I have two Trace/Xantrex 5.5KW stacked in // for 240VAC and 12, 850 Ah 8 volt loco batteries for the battery banks. The batteries were "freebies" from a pair of locos being scrapped locally, but I had to remove them myself, by hand. That was fun itself.    Once everything was installed, I made a test of the whole system, and was able to go for 3 days, whole house, on just batteries. I heat with wood and shut down the water heater and used the wood stove for hot water, but otherwise went about daily living as normal. I was pretty happy with the system. The inverters are 48VDC inputs and I used 6 batts in each bank, to allow a rest period for each while charging or using them. We had a power outage one day, and I didn't even notice. I was busy vacuuming and didn't even notice the transition to backup power as the inverters switch in less than one cycle. My neighbors called and asked why I had power and they didn't. he he I ended up running a long cord to their place for lights for the night.

Good thread!

Paul

[W1RKW]

Don't have any plans on charging batts with a PS but was just curious. I've heard those who say it's OK. I always had my doubts but never had a clear understanding why or why not.  These posts help.

[W1RKW]

I was looking for info on charging and determining when a battery reaches full charge and came across this nice little FAQ on deep cycle batteries.

http://www.windsun.com/Batteries/Battery_FAQ.htm

[K6IC]

Tom,  great that you had the wisdom to get your battery backup system on-line when you did,  and nice that it did the job.

Paul,  great to see you here again.  Do recall a number of years ago,   that you were in the process of moving to a fairly remote site to live.  Have never seen Loco batts ... looks like a very capable system.   Here,  we run the same or similar inverters -- SW+5548s stacked,  Surrette 4KS25s@ 48V,  about 1280 AH when the specmanship is reduced.  This has been a very good system,  altho being off-grid,  it needs to do most everything.  Need to run generators on occasion.

And,  Bob,  tend to forget about the Wind-Sun FAQs.  While their Forum has a commercial sponsor,  it is  by far,  the best general site for specific info IMHO.  It appears to me to be comprehensive,  and without any hidden  bias in the answers toward the sponsor.  They seem to be very honest to deal with and have some of the best prices around on name brand RE equipment-- no off-shore crap that Ive seen ...  and so on.

Good Luck to all those affected by Sandy and her aftermath.  Very wrenching.  Having a backup power system,  genset and fuel must make this storm much easier to deal with,  in  general ... Vic

[W1RKW]

Vic,
I read some where that a wet deep cycle batt is considered fully charged when it pulls about 2% of its Ah rating, for example 2amps for a 100Ah at around 14.5v.  What's your take on that?  It seems once a batt gets around 14.1v one can here it bubble. Once it begins to bubble is it considered charged?

[K6IC]

Hi Bob,

Again I am no expert on this.
Speaking of Flooded batteries only,  for 12 V systems,  Think you are correct about the Constant Volatge stage of charging  (the Absorption stage),  is normally is about 14.5 V at 77 degrees F.  The bubbling does begin a bit below this V,  but 14.4 or 14.5 V is good for Absorption.  But the Asbsorption  V should be held for quite some time,  and the amount of bubbling increases markedly toward the end of this stage.  The amount of time needed IS often determined by the Finishing current,  (also known as Return amps or End amps).  And normally,  this current is between 1 and 2% of 20 hour capacity.  On the batteries here,  it is almost exactly 1% at around an equiv.  V of 14.5 ish V.  This current does depend on the Abs V and the battery temps,  and the exact chemistry of the batts,  etc.

In general for flooded batts,  the first charge stage -- Bulk -- charges the batts to about 80-85% SOC.  Bulk is called Constant Current,  and ends when the target Absorption V is reached.  The Abs stage  can easily require several hours to complete,  depending on how deeply the battery was discharged prior to recharging.

The ending current value of 1-2% of 20 hr Capacity is a good rule of thumb,  but,  of course measuring the SG of the batteries is the gold standard.  SG measurements are a fairly easy way of determining just value of ending current is required.

This is just based on my experience.  ForkLift batteries behave somewhat differently,  and have different parameters for charging,  and sealed batts also have different charge voltages and ending currents,  etc.
Mileage May Vary.    GL   Vic