Electric / Battery Charger

With four NiMH battery packs now totalling over 10 Amp Hours in the boat and one smaller pack in the transmitter, the issue of battery charging has become a most significant one, Who wants to wait around for days while these are all charging. I regularly read the electric R/C forums on various websites and while there are undoubtedly some great chargers out there, they are expensive and it seems some are almost too complicated for users. It seems that the cleverer the chargers get, the more online arguments there are, and potentially the more dangerous they become, there are people blowing up chargers and various types of batteries all over the world!  Also false peak detection termination and uncharged batteries are leaving planes and heli's falling out of the sky etc.

There is a good commercial charger here for around US$100:  Batteries Wholesale

I have bought (June 2005) a new Mega Power Infinity-660SR charger from www.eflypower.com  (US$154) This is a sophisticated unit capable of charging or discharging 1 to 26 cells of NiMH or NiCad (all three of my 8 cell packs in series) and 1-10 cells Li-Ion or Li-Poly at up to 5A charge or discharge current. I haven't used it all that much yet but it looks good so far, I want to start playing with Li-Poly batteries and I needed a way to charge them. This is meant for connection to a car battery with crocodile clips so the first thing to do was put a pair of Deans Ultra plugs in the inlet lead so I could power it from a 12VDC (5A plus)  power supply and still retain the ability to connect it to the car battery and also put a Deans plug on the output lead to connect to my battery packs.

NiMH Battery Charging

The charging efficiency of a nickel metal hydride cell is around 60%, meaning that you must put around 150% charge into the cell or battery for every 100% you get out. The faster you charge the cell  the worse this charging efficiency becomes due to heating.

Many NiCad chargers use the characteristic voltage bump that can indicate end of charge in a NiCad cell but this voltage bump is much less pronounced in NiMH cells and it is also very temperature dependent. Some NiMH cells also exhibit voltage bumps in their charge curve early in the charging cycle, particularly when cold. NiMH cells are liable to be damaged if overcharged when the charge rate is over C/10 and since the voltage bump is not always easy to recognize by some chargers an overcharged or undercharged battery seems to be highly likely. For this reason using voltage bump as a termination method for NiMH cells does not seem to be generally recommended.

As the battery cells reach a fully charged state, oxygen starts to form at the electrodes, and in normal operation is recombined by a catalyst material within the cell. After the cell is fully charged continued current flow will start new chemical reactions. In NiMH batteries this consists of generating and recombining larger quantities of oxygen. This process heats the cell. This fairly sudden increase in temperature rise, which can be easily measured with a temperature sensor, can be used to terminate the charging cycle . This seems to be becoming the safest way to detect end of charge during a fast charge. Cell temperatures will stay somewhat around the ambient surrounding temperature for most of the charging cycle and then start to rapidly rise as they approach and pass the fully charged state - of course this approach relies on all cells in a pack reaching a charged state at essentially the same time.

June 2005 - I have done a fair bit of NiMH charging by now and I decided to test my favorite constant current / temperature rise method to see how close it gets the cells to a fully charged state, so I fast charged a 3000mAH pack at 1.2 Amps (C/2.5) until the temperature started to rise fairly quickly and then dropped the current back to a C/40 charge and measured the mAH's needed to reach an ultimate cell voltage of 1.40V and it only required another 60mAH so you could probably say with reasonable confidence that the cell was charged to within 60mAH of 3000mAH (or say 2940mAH) by the fast charge temperature rise method, that's not bad.

Trickle Charging

The simplest way to charge a NiMH (or NiCad) battery pack is to trickle charge at 10% of the rated capacity (per hour) or C/10 (cell capacity divided by 10). So a 3000 mAH battery would need to be charged at 300 mA for 15 + hours. This method requires no sensing and ensures a fully charged state is achieved and it is also commonly used for many other cell types. Most, if not all, NiMH cells have an oxygen recycling catalyst which prevents damage to the battery during an overcharge situation but this oxygen recycling cannot keep up the oxygen  re-combination if the charge rate is significantly high. To maintain long and healthy battery cell life a timer should be used to prevent overcharging from continuing past 15 or so hours, and this assumes a fully discharged cell. All new cells should be charged this way at least once on their initial charge, particularly if they are made up into a series connected battery pack as it is the only sure way to know that all cells in the pack are fully charged. This is called 'forming' the battery and if this is not done, when the pack is subsequently discharged, the cell with the lowest stored charge will most likely completely discharge and then reverse its polarity while the other cells in the pack continue to pump current through it and it may be damaged irrepairably. Even though continued trickle charging at C/10 may not cause cell gas venting, it will warm the cells by a few degrees Celsius and this temperature rise is quite easily detected. 

It seems that it is not widely recommended to constantly trickle charge NiMH cells past the fully charged state, instead the idea of pulse or burp charging on a very long duty cycle seems to be gaining popularity, so following a full charging cycle, monitor the cell voltage until it drops to 1.30 Volts and then apply a C/10 current until the cell voltage rises to 1.40 Volts (at 25 degrees Celsius) and then reapply this C/10 current once the cell voltage drops back to 1.30 Volts and this pulse or burp charge may be continued indefinitely.

I have used this type of basic constant current source circuit many times over the years and it is extremely reliable, the LM317 (1.5A) and LM350 (3A) devices are very rugged with internal temperature and current limiting and are rated up to 40 and 35 volts input respectively. 

The example shown above with a 3.9 Ohm resistor will regulate the current to somewhere between 307 and 333mA or approximately 1/10 C (or C/10) for a 3000mAH Battery. The regulator drop out voltage at 300mA can be as high as 1.75 Volts so if you use a 12V power supply where it's output voltage is at the low end of its tolerance (say 11.75V) it may not fully charge an 8 cell pack which needs at least 1.4V x 8 or 11.2 volts minimum. 

The 3.9 Ohm resistor needs to be rated at 500mW or more and should be wire wound or metal film. 

For higher currents use a lower value resistor using the formula in the diagram  I = E divided by R   or if expressed to find the resistor value R = E divided by I .

 i.e. for 1000mA (1/3 C) the resistor would need to be R = 1.25 (V) divided by 1 (A) which equals 1.25 Ohms. The resistor wattage can be calculated by the formula E x E divided by R or 1.25 x 1.25 (= 1.5625) divided by 1.25 = 1.25 Watts.

A higher voltage power supply can be used of course (I use a 40V power supply and charge all 4 batteries in series) and the power dissipation in the regulator can be worked out as E x I where E is the voltage across the regulator (power supply voltage minus total battery voltages) and I is the current through it in Amps. Make sure you screw the regulator to a reasonable heatsink and use some heat conducting compound in between and also note that the heatsink will be connected to the regulator's VOUT pin unless you insulate it.

What do we need for Fast Charging

If the temperature sensing method is used to detect a fully charged state, NiMH batteries can be charged at rates up to at least 1C or 100% of the battery capacity in amp-hours for a nominal period of 1.5 hours. The jury seems to be still out on how far above 1C we can go and it will be brand dependant and will surely be reassessed over time as the battery wear out mechanisms and chemistry becomes better understood. When terminating the charging cycle based on temperature rise, the recommended rate of rise should be set at about 1 to 2 degrees Celsius per minute and due care must be taken that this temperature rise reflects all the cells in the pack. If the starting temperature is above 40 degrees Celsius or below zero degrees Celsius either don't start charging at all or start charging with a C/10 charge. If the discharged battery voltage is less than 1.0 Volt/cell start charging with a C/10 charge. If the discharged battery voltage is above 1.30 Volts/cell start with a C/10 charge then start fast charging at a rapid rate (1C) until the cell temperature reaches 35 - 40 degrees Celsius, or the rate of temperature rise indicates full charge. After terminating the fast charge, resume slow charging at C/10 for a few hours to ensure all cells are brought up to a full charge. If the cell voltage climbs to 1.5Volts/cell terminate the charge. Do not allow the fast charge cycle to exceed 1.5 hours and do not allow the trickle charge to exceed 15 hours.

A number of people recommend discharging a partially charged NiMH battery until the cell voltage is 1.0 Volt/cell, so that it can then be fast charged, given that the cells may already be partly or maybe even almost fully charged this seems dumb in the extreme! This approach was developed for NiCad batteries to overcome their partial charging memory effect and my many experiments using cell temperature rise to indicate a fully charged state in NiMH cells seems to indicate in an almost identical manner regardless of the initial charge state of the cells.

Thought for the day, an ATX  PC power supply has a +12Volt output at around 20 to 30 Amps and is a very in-expensive power supply with world wide approvals that is safe and already fitted with a power inlet and often a power switch as well. A short wire between the thin green wire and any black wire on the ATX motherboard plug will turn the power supply on.

Linear Technology www.linear.com has a new Lithium-Ion / Polymer charging IC p/no:  LTC4061-4.4 designed for precision single cell charging at up to 1A which looks like an interesting device but only for single cells unless you put some more clockwork around it.

Stay tuned.