Battery Equalizer Circuit

http://www.google.com/search?sourceid=navclient&aq=hts&oq=&ie=UTF-8&rlz=1T4GZAZ_enUS396US396&q=Battery+Equalizer+Circuit

Found this...

https://www.google.com/search?q=smartspark+energy+pdf+batteq+1&channel=linkdoctor

Thanks for the links,guys.They all look somewhat adaptable for my needs, but I am looking for a schematic to build my own.

The circuit I remember seeing only had 3 or 4 components for each battery, all solid state, no mechanical contacts.Seems like I recall a zener diode,transistor,resistor and fuse for each battery.The circuit did not compare voltages from adjacent batteries, it simply switched at a predetermined voltage to by-pass the battery and allow weaker batteries to have more charge.

Anyone got any home-brew schematics for this application?

Of course by now, they probably have a single chip to do this.....?

This simple circuit may work for you. As the battery voltage increases above the desired maximum value, the transistor will turn on, shunting the charge current around the battery.



WARNING! The power dissipated in this circuit could easily cause it to self destruct. The transistor must be properly rated for the full charger current AND it must be able to dissipate the power calculated for your specific application. Proper heat sinking and a cooling fan may be necessary.

Example:
30A charger
6 volt battery string in series
one battery fully charged (or damaged)
the transistor will attempt to shunt 30A @ 6V and will dissipate 180W

I've shown a darlington transistor in the sketch. The two transistors are typically combined in a single package and, because of the high dc gain, can be driven directly off a 5k ohm pot (for R1 and R2). Good luck!

Thanks!It has been over 30 years since saw the circuit, and I can't seem to find it..I remember filing it under "Miscellaneous"(which is where I put everything).

I will take your advice into consideration and I will probably use a ntc resistor to limit current based on temperature, as well as fusing of output and adding an indicator LED for bypass mode, and one for full charge, that way a bad battery would be very obvious.

It would be nice to find some miniature relays with low contact resistance that would handle the load (30 A. DC) instead of a high pass transistor and large heat sink.

Any suggestions on relays?

As always,appreciative of your help.

We use Potter Brumfield SSR-240D50 Relays.

They can switch up to 50 Amps at 240 Volts DC and are solid state.

Coil voltage can be 3-32 vdc.

Good idea, but cost too much$($60) each.

Thanks for the feedback though.

Would it be too costly to use 6 automatic six volt chargers simultaneously?

Maybe a changeover contact would be better so that the battery would be disconnected, any bypass would short circuit the battery which was fully charged.

The transistor circuit above might get current from the battery as well as the charging circuit which would discharge the battery again. A diode in series with each battery would stop the reverse current but it might also stop the voltage across the battery being detected??

Needs more thought for an electronic solution......

I am presuming a fully charged set of good batteries to start with.The problem with charging batteries in series is imbalance of all batteries, which interferes with proper load sharing and capacity.When the standard charge has completed, I would switch to balance mode, and allow the batteries to find equilibrium.True, it would shunt higher voltage from a battery as well as from the charger, but that is ok, it will turn off as soon as the batteries are equal in voltage.There is an adjustable threshold voltage to control the switching voltage.Perhaps I can find a zener diode to provide reference voltage.May be hard to find a 6.6volt zener, so I may have to combine some in series to get what I want.

It would be ideal to put a smart charger on each battery, and charge them in parallel, but it is cost prohibitive.

I am still thinking about the best way to solve this problem, and
I have had some good suggestions, but I have not settled on a final solution yet.Thanks for all the great feedback.

When 12 V flooded flat plate lead acid batteries( say 150Ah) repeatedly discharged to 10.5 Volts,cutoff voltage,the acid gravity goes down at top the acid gets stratified.An equalizing charge is needed at higher voltage ( say 16V) and at low current( say 2A) for say 8 to 10hrs, the overcharge leads electrolysis and bubbles move up. The acid gravity almost gets even allover.But it reduces life of battery and so also loss of electrolyte.My question is how to know the optimum level of Charging Voltage,Amps,Duration with least damage to battery ? Can the charging period be reduced to say 4hrs?

The damage to the flat plates is caused by electrolysis inside the plate material causing bubbles that break up the material, so the lower the current the less damage you get, but the time to reach full charge increase in reverse proportion to the current. The charge voltage is determined by the battery and the charge current, so if you want to charge at 2A the voltage will probably start at about 12V when fully discharged and gradually increase to about 14.3V when fully charged. If you need to put in 150Ah this will then take at least 75 hours which you probably don't have the time for.

I asked for reviving a battery which after repeated charge/discharge has stratified acid,

and even after charging at 14.4 Volt charger,it has not got charged it to its full capacity owing to stratification and heavy sulfation on plate surfaces. Can this battery be revived by a equalizing charge of say 16V and 2 A current for a longer duration say 8 hrs. The overcharge will cause electrolysis of electrolyte and bubbles formed will help to equalize the acid gravity and also lead to removal of Sulfation. Overcharge causes weakness to battery and as such ,such charging be optimum. Any thoughts on optimizing this equalizing method ?

I am considering using an opto isolator to separate the batteries from the charger,or an opto-coupled transistor, powered by the reference circuit.I haven't worked out all the particulars yet, still gathering wool, as the saying goes.....

OK.

Take the transistor circuit above, stabilize the R1 & R2 with a zener, make R2 a pot so as to adjust the volts where this thing starts to bleed current, and put a hi wattage load resistor in series with the darlington transistor. Perhaps a couple of LEDs to indicate what's happening.

Calibrate with a voltmeter across each cell to start conducting around 13 - 14v

This way there is no loss of volts in a bypass, the volts across each cell will eventually be the same if it's calibrated right.

If I have several cells in bypass mode, the junction loss of each transistor will accumulate, will it not? That is why after due consideration ,a dry contact relay may be the best way of by passing.I just need to find a small relay capable of handling the approx 30 amp DC.This will eliminate the need for a large heat sink.

The circuit above can be used as a reference for switching the relay.

Still open for suggestions....

Adding a zener reference should provide a sharper switching threshold... and yes, the simplified circuit as shown will draw some battery current all the time. If R1+R2 is 5k ohms and the transistor is effectively off at 6.4V, the current drain will be about 1.3 milliamps (0.0013 A).

Unless you design the bank so that each battery can be switched out of the series string, parallel shunt devices (any type) MUST be able to dissipate the peak battery voltage (Vmax) times the shunted charging current. If you attempt the bypass using a relay, you will still need at least 1 resistor that can handle this power dissipation. Unfortunately, 1 resistor does not work as the charge current tapers to zero.

The transistor circuit will act as a "variable" power resistor that limits the maximum battery voltage over a large range of charge currents.

There is no simple/cheap way to perform this function without excessive power dissipation, which is why you don't often see this technique used in a high current battery string.

The idea of using a charger for each battery does cost more, but it can be very energy efficient AND it is the best way to maximize battery condition and life. How much do your batteries cost?

There are other schemes that use complex switching techniques (with inductors, capacitors, and diodes) to shunt the excess energy from one battery to another battery in the chain. By the time you get one of these designed, built, and working properly (and safely), I'd bet a charger for each battery will be the easier and cheaper solution.

For more information, try searching "active loads" as used in "electronic test equipment".

http://en.wikipedia.org/wiki/Active_load#Test_equipment

If I close a set of contacts across the fully charged battery, and open a set that connects the battery to the charger output,(Break before make), this will appear to the charger as less feedback voltage and increase it's output to the other batteries.

This should result in a faster initial charge to the other batteries in the string.I would probably have to add a regulator to keep the charger from overloading itself if more than a couple of batteries are bypassed.(O L protection may be built in, I will have to check charger specs).

The batteries cost around 160 dollars eachX6ea.,so I want to make sure to maximize service life.

These will be in standby mode 95% of the time, so a maintenance charge is all that will be needed most of the time.

Thanks for the link.

"If I close a set of contacts across the fully charged battery, and open a set that connects the battery to the charger output,(Break before make), this will appear to the charger as less feedback voltage and increase it's output to the other batteries."

This is where I think we have a breakdown in communication. If you perform the action described above, the relay will be vaporized by the discharge current from the battery.

You are correct.Poor description of my intention.I intended to disconnect battery from charge circuit, and pass the circuit on to the next battery in series.

However, I am now considering a circuit similar to a tv high-voltage tripler, that allows caps to charge in parallel, and discharge in series, except modified to use batteries instead.

I need to think on this approach a little more before I decide.I have a bunch of high current stud-mount diodes lying around, so this may be my cheapest method.

As always, your valuable feedback is appreciated.

The charge in parallel and discharge in series circuit is familiar to me when applied to capacitors. It is usually referred to a Mark-Bank or

Marx_generator

This works very well for generating high voltage transients, but I haven't seen a practical application used on high current batteries.

The Voltage multiplier circuit is called a Cockcroft-Walton generator or

Cockcroft_Walton_multiplier

I don't see how this can be used in your application, but please feel free to investigate further.

Thanks for the feedback!Looking at these type circuits, what if I substitute a batttery for each capacitor.One set of diodes would be used for charging(in parallel) and the other set used for discharge( load.)Instead of an AC input, the charge would be similar to the positive half (parallel) of the cycle, and the discharge(load) would be the (series) half.Would this work?I realize the diodes would carry the full charge/discharge load, but I have some(6) large stud mount diodes and heat sinks from an industrial 300 amp battery charger lying around collecting dust.

A slight modification to the multiplier circuit will be required to work with batteries.The High voltage output will go to load as the negative side of load.

The ground end will be removed from ground, and used as a negative for charger input.The positive terminal will be the common positive for both charging and load.

Load and charger will never be connected at same time,load connected thru solenoid switch.

This will require some rewiring of battery jumpers, but I think it will work.

Your valuable opinion is appreciated and welcome.

Sadly, you've lost me. I am unable to see a useful battery charging configuration resembling the voltage multiplier circuit. I'm recommending a charger for each battery as the best solution and the active transistor current bypass as a second best alternative.

At this point I can only suggest you carefully work out your circuit on paper first. Large gauge copper wire is expensive. Emergency burn treatment is also expensive and quite painful. Been there, done that, don't recommend it for anyone. Seriously, please be careful. A wiring error on a large battery bank can spray you with molten copper. Just the arc flash alone can have enough radiant energy to burn you and damage your vision. If I find another option, I'll post back to this thread. Otherwise, best wishes on your project.

OK, I am sure you are right,but please advise my of my error on this schematic.

The charger I am treating as the positive going pulse of an AC input.A very LONG pulse, during which time the batteries are charged in parallel.

The negative half of the pulse (another very long pulse) is when load is applied to discharge the batteries in series.

The charge is thru D1,D3,D5.(parallel)The discharge is thru D2,D4,D6(series).

The load and charger are never applied at the same time, isolated by a plug/socket and solenoid contactor.

Perhaps I should take a few more shots of Dewar's and look again.

Your comments and opinion are appreciated, as always.

If I'm interpreting your drawing correctly, you have 6 batteries and 6 diodes. Battery #5 is dead short through D2 and D3 while battery #4 is dead short through D4 and D5. BOOM! There are other problems, but these won't matter after the meltdown.

Methods to equalize a battery bank have been suggested by myself and other responders. You are trying to find a new way to perform this function that millions of people, well versed in the art, have researched over the last 150+ years. I strongly urge you to stick with one of the standard methods suggested and proceed with caution to avoid property damage and injury. Best wishes.

You are absolutely correct, as usual.My bad.. I will follow your sage advice.

Thanks to all contributors for their input.

Why not just have a separate transformer output for each battery with a full wave bridge, a capacitor and a series resistor for each? It could be 8-12V raw for 6V batteries. The series resistor value would be based on the current you want.