It is currently Mon Dec 11, 2017 11:57 pm

Battery charger design.

 
Posts: 1141
Joined: Mon Jul 25, 2011 11:37 am
Location: Willington, Co Durham.

Battery charger design.

Post by malcscott » Fri Apr 04, 2014 10:49 am

Hi, i plan to construct a battery charger to charge 12v, 6v and 2v lead acid batteries. I have obtained a NOS Douglas MT5AT battery charging transformer which has 2/6/12 volt outputs. Do i use a bridge rectifier or half wave rectification? Cheers, Malc.

 
Posts: 6972
Joined: Sun Aug 14, 2011 11:02 am
Location: Co. Limerick

Re: Battery charger design.

Post by Michael Watterson » Fri Apr 04, 2014 11:32 am

Plus a suitable series resistor. In very old chargers the Metal Rectifier was the series resistor.

Oddly to recharge Alkaline cells (only works if less than 1/2 discharged) apart from monitoring what is happening, the more pulse like the better.

 
Posts: 763
Joined: Sun Aug 07, 2011 5:17 pm
Location: Edinburgh

Re: Battery charger design.

Post by Alistair D » Fri Apr 04, 2014 11:34 am

From the fact that the transformer is NOS it may have been designed for a Selenium rectifier. Using it with a silicon bridge will raise the charging voltage an therefore the current. You may need to put a little bit of resistance in series with the output and more closely monitor the amount of charge given to the battery or the end point charge voltage.

Sorry Michael, just cross posted.

Al

 
Posts: 6972
Joined: Sun Aug 14, 2011 11:02 am
Location: Co. Limerick

Re: Battery charger design.

Post by Michael Watterson » Fri Apr 04, 2014 12:06 pm

Just to clarify,

ANY charger with modern rectifiers, with ANY transformer needs either a full charge management system or a resistor to limit current to maximum for type of cell assuming about 1.5V per cell load. If the cell is less than that it has been massively over discharged or dead. The current can be more than maximum for a short period. The cell voltage rises quickly.

Alternatively the strategy is to pick series resistor so current is trickle rate when battery is about 2.3V per cell.

Historically with the "correct" transformer voltage the same value "resistor" (the Selenium Rectifier) does give maximum current at 1.6V per cell and trickle at about 2.2V.

What does the simplest management system do?
It has current limit protection for short or totally dead battery
resistor limited between 1.6V and 2.1V per cell
Constant trickle charge above 2.1V per cell.
Then there are more complex regimes.
But in most cases Correct voltage AC, Fullwave (may be CT winding so then two diodes not bridge) and Resistor performs as well for Wet Lead Acid cells.
Gel cells may benefit from a much more complex controller especially if over discharged.

A third type charger is designed for equipment in use where battery is standby (Mobile base station etc, transmitter, UPS). It Fast charges then trickle charges when equipment is off. It will periodically stop trickle charging if programmed for Gel Cell as it's bad for Gel Cell life. When equipment on it doesn't trickle charge at all unless battery voltage drops. It will respond to load spikes to stop battery discharging. A battery can last for 10 years in such an online smart charger and it could be set to avoid over voltage on vintage direct filament valves, which are normally not designed to be connected to a battery on float charge (per cell voltage rises for 2.35 on "calcium" impregnated maintenance free wet types).

 
Posts: 1141
Joined: Mon Jul 25, 2011 11:37 am
Location: Willington, Co Durham.

Re: Battery charger design.

Post by malcscott » Fri Apr 04, 2014 1:11 pm

Would i need a different resistor for each seperate output? Does it have to be in the AC feed to the rectifier of the DC o/p? What value/wattage?

 
Posts: 6972
Joined: Sun Aug 14, 2011 11:02 am
Location: Co. Limerick

Re: Battery charger design.

Post by Michael Watterson » Fri Apr 04, 2014 2:31 pm

The resistor value depends on

(AC (RMS) rectifier voltage - (0.7 or 1.4, ct vs full bridge) - Flat battery voltage*) / max Current

* So if it's a:
6V battery, Flat battery voltage = 1.5 x 3 = 4.5V (recommended end point is 5.25V though)
12V battery, Flat battery voltage = 1.5 x 6 = 9V (recommended end point is 10.5V though)

Say max charging current is 5A

If VAC = 18 for 12V and no Centre tap
Then R = (18 - 1.4 - 10.5) / 5 = 6.1 / 5 = 1.22 Ohms

At 14.2 V (fully charged maintenance free wet cell like car battery)
As R is 1.22 Ohm then current drops to (18 - 1.4 -14.2) / 1.22 = 2.4 / 1.22 = 1.92A

If VAC = 15 + 15 for 12V and Centre tap
Then R = (15 - 0.7 - 10.5) / 5 = 3.8 / 5 = 0.76 Ohms
At 14.2 V (fully charged maintenance free wet cell like car battery)
As R is 0.76 Ohm then current drops to (15 - 0.7 -14.2) / 1.22 = 0.1 / 0.76 =0.13A

So the "best" solution is an (AC voltage - diode drop) just a bit above peak fully charged voltage and then choose resistor for max current on a flat cell.

Depending on desired cell currents and AC voltages it's possible to have only one value resistor. My vintage charger gives SAME current on 6V battery or 12V battery flat (about 2.5A to 3A) because they chose the "correct" AC voltages for the Selenium Rectifier. The "trickle charge" is about 0.2A, but if one cell is bad in a Wet battery it might not drop below 1A. If a Cell is bad in Gel battery the opposite happens, the current drops to a very low level quite quickly.

If for your transformer the trickle current is too high due to too high AC voltage then add another rectifier in series with each rectifier and reduce series resistor till you get desired max and trickle currents.

Wattage is the current at flattish battery (1.6V per cell) squared x Resistance.

You can use a large dummy resistor of (1.5V per cell x number of cells x max current) as a test load for max current. For testing trickle current you need a charged battery.

In real labs they have "magic" PSUs that can sink or source current with programmable voltage and series resistance. A home brew alternative is a variable shunt regular (no power needed though) to simulate flat or charged battery.

 
Posts: 209
Joined: Mon Jun 04, 2012 11:40 am
Location: Newcastle on Tyne

Re: Battery charger design.

Post by Ed Dinning » Fri Apr 04, 2014 5:01 pm

Hi Malc, simplest way is a variable resistor of suitably high wattage and an ammeter in series so you can see the charge rate you are setting. This current will decline as the cells come up to charge. Another way would be a string of 5 or so 1 ohm 5W resistors and a switch or series of sockets; the same can be adopted for the voltage selection.

Ed

 
Posts: 6972
Joined: Sun Aug 14, 2011 11:02 am
Location: Co. Limerick

Re: Battery charger design.

Post by Michael Watterson » Fri Apr 04, 2014 5:59 pm

Yes, if you want to charge lots of different kinds of batteries from 2AH to greater than 36AH. Most traditional chargers from the Metal Rectifier Era are for 6V or 12V (maybe sometimes 2V and 4V if pre 1946) at 15AH or larger size.

The single fix resistor approach is only good for a minimum battery AH rating.

 

Re: Battery charger design.

Post by XTC » Wed Apr 16, 2014 10:30 pm

It rather depends which sort of lead acid batteries you are charging.

There are two basic types; flooded and SLA.

Flooded will put up with crude charging schemes, but you need a current limit and you can't let the charging process continue until electrolysis reduces the electrolyte level to below the level of the plates so they dry out.

SLA are different. They put up with a certain amount of overcharging because they can recombine the oxygen and hydrogen produced, to a certain extent. They can't tolerate much overcharging or they dehydrate or start to leak. Then there are a couple of sorts, those intended to sit there with their float voltage applied and rarely be called to discharge, as in burglar alarms, and those intended for cyclic use, as in halogen lanterns.

There are two basic (sensible) schemes for charging them. Bear in mind that the maker's data sheet should be consulted for the float voltage and the cyclic charging voltage. There's also a gassing voltage which you ought to avoid with SLAs. All of these voltages are slightly temperature dependent, and vary slightly from maker to maker, although there is a latitude. Burglar alarms I've seen don't

Fixed voltage, current limited. The voltage is set to the float voltage and the current is limited to below what the maker recommends. The voltage is something like 2.25 to 2.30 per cell @ 25C and increasing by 4mV/cell/degree C as you get colder. Unless you have to deal with fluctuations in temperature of that sort, you can get away with an accurate fixed voltage current limited charger set to , 2.3V per cell based on an LM317 or an L200 voltage regulator chip. Ideally, the float voltage would be temperature compensated to track the ideal float voltage, which really rules out a wall wart type charger.

Two level chargers. There are chips such as the UC3906 which you can set up with resistors to implement a charging regime. They monitor the battery voltage and do an exploratory charge if it's below a certain level, that's to avoid catastrophes from charging batteries with dead cells. They then charge at a bulk rate, which you determine with a resistor, they switch to a controlled overcharge mode, apparently necessary to reclaim all the capacity of the battery with the voltage set at 7.3 to 7.4V for a 6V battery, then when the current drops to 1/10th of the bulk rate (50mA if the bulk rate is 500mA), they drop the voltage to the float level. They are temperature compensated. I'm not sure if they are designed to work with 2V SLAs. Cyclon SLAs are a bit different and have a slightly different recommended voltages.

Burglar alarms I've seen have a factory set float voltage and don't include temperature compensation, but the batteries seem to survive well enough. The chargers chucked in with the x million candle power halogen lanterns are truly awful, along the lines of a 9V unregulated supply, about 14V O/C, with a resistor in the lamp to limit the current. It's no wonder that there are so many reviews complaining that they are useless and fail after a month or two

The other thing about lead acid batteries is that they degrade if less than fully charged and so need to be left on an accurately set float charger, or regularly topped up, and they suffer with deep discharge. These conditions don't fit with the pattern of use most people expect with a rechargeable torch, or motorised wheel barrow, much less with the chargers they chuck in with them.

There's a world of information on charging lead acid batteries on the WWW.

I believe Maplin sell a multi voltage lead acid charger for SLAs, but I wouldn't trust it myself.

 
Posts: 6972
Joined: Sun Aug 14, 2011 11:02 am
Location: Co. Limerick

Re: Battery charger design.

Post by Michael Watterson » Wed Apr 16, 2014 11:47 pm

Good summary.


Return to Dials, Logo's & Batteries



Who is online

Users browsing this forum: No registered users and 1 guest