NiMh Charger - 4x AAA in Series
JBWolf
Posts: 405
Hello,
I have been having a little trouble finding suitable information to create a schematic & build a battery charger for my needs.
Batteries: 1.2v AAA Energizer NiMH "Recharge Power Plus", 700Mah
Usage: All four batteries in series to accomplish roughly 5vdc
Separation: Each battery is 18" from the next consecutive battery forming a 6' ring for even distribution of balance.
Primary limitation regarding charging methods is that the batteries are permanently installed in series, contained within rigid tubing and cannot be removed for charging via conventional charger.
I consider this to be similar to a 'battery pack'. Temp sensing will have to be accomplished via thermistor or similar component.
A standard 110vac outlet will be available for the initial power source for charging... There are no restrictions on the charger size or weight.
---
What I need to accomplish is a method to charge the batteries safely, an indicator to visibly identify when charging has been completed is required.
I plan to use a 4-terminal headphone style adapter to plug in the charger to the 'battery pack'.... two connectors for the anode & cathode terminals, two for a thermistor connected to one of the batteries.
For the past few days I have been researching as much as possible in hopes of figuring out how to do this safely and cost effectively.
I have read extensively about the %C charge rates, time & temp change of charging NiMH batteries.
The BQ2000, BQ2002 or BQ24400 IC's from TI look to be a suitable management controller to create a suitable charger... the BQ2000 & BQ24400 are the only ones which the datasheet contains an example schematic and possibly enough information to modify for my needs... however I am a bit unclear on my needs and the proper modifications that should be made.
Along with the particular energizer batteries I purchased, I also purchased an energizer 'smart charger' with fuel indicator made for them with the intention of using it to get base readings on a fully charged battery.
There are small differences between each battery, but not by much.... here are some readings recorded 12 hours after completing a full charge.
Batt1: 1.354v @ initial peak 8.13A, stabilized at 5.92A after 20sec
Batt2: 1.355v @ initial peak 8.23A, stabilized at 6.65A after 20sec
Batt3: 1.351v @ initial peak 8.34A, stabilized at 5.62A after 20sec
Batt4: 1.350v @ initial peak 8.23A, stabilized at 5.65A after 20sec
Oddly, repeating the tests immediately after taking those... I got a higher stabilized amperage on each one at around 7A.
Combined voltage in series measured at 5.11v even with repeated tests
So for target max voltage/amperage I am thinking of using 5v/33A.
Now the sources I am starting the schematic design from are the BQ24400 datasheet and the BQ2000 Datasheet ... however, the BQ2002 is the only part specifically listed as a purposed 'fast charger'... but from what I have read, fast charging is simply at a 1/C or 2/C rate which all three components are capable of.
The BQ2000 datasheet provides the best examples of circuit design.
I'd like a recommendation as to which IC component would be best for my needs.
I also could use some pointers as to max charge time, how to figure maximum temperature for an NTC thermistor (also which thermistor to use since there are so many types/values), setting peak voltage and also whether I should bypass trickle charge.
The max temp and thermistor stuff is especially stumping me.
Sorry this is kinda a big one, hoping someone has experience with this and can help out
I have been having a little trouble finding suitable information to create a schematic & build a battery charger for my needs.
Batteries: 1.2v AAA Energizer NiMH "Recharge Power Plus", 700Mah
Usage: All four batteries in series to accomplish roughly 5vdc
Separation: Each battery is 18" from the next consecutive battery forming a 6' ring for even distribution of balance.
Primary limitation regarding charging methods is that the batteries are permanently installed in series, contained within rigid tubing and cannot be removed for charging via conventional charger.
I consider this to be similar to a 'battery pack'. Temp sensing will have to be accomplished via thermistor or similar component.
A standard 110vac outlet will be available for the initial power source for charging... There are no restrictions on the charger size or weight.
---
What I need to accomplish is a method to charge the batteries safely, an indicator to visibly identify when charging has been completed is required.
I plan to use a 4-terminal headphone style adapter to plug in the charger to the 'battery pack'.... two connectors for the anode & cathode terminals, two for a thermistor connected to one of the batteries.
For the past few days I have been researching as much as possible in hopes of figuring out how to do this safely and cost effectively.
I have read extensively about the %C charge rates, time & temp change of charging NiMH batteries.
The BQ2000, BQ2002 or BQ24400 IC's from TI look to be a suitable management controller to create a suitable charger... the BQ2000 & BQ24400 are the only ones which the datasheet contains an example schematic and possibly enough information to modify for my needs... however I am a bit unclear on my needs and the proper modifications that should be made.
Along with the particular energizer batteries I purchased, I also purchased an energizer 'smart charger' with fuel indicator made for them with the intention of using it to get base readings on a fully charged battery.
There are small differences between each battery, but not by much.... here are some readings recorded 12 hours after completing a full charge.
Batt1: 1.354v @ initial peak 8.13A, stabilized at 5.92A after 20sec
Batt2: 1.355v @ initial peak 8.23A, stabilized at 6.65A after 20sec
Batt3: 1.351v @ initial peak 8.34A, stabilized at 5.62A after 20sec
Batt4: 1.350v @ initial peak 8.23A, stabilized at 5.65A after 20sec
Oddly, repeating the tests immediately after taking those... I got a higher stabilized amperage on each one at around 7A.
Combined voltage in series measured at 5.11v even with repeated tests
So for target max voltage/amperage I am thinking of using 5v/33A.
Now the sources I am starting the schematic design from are the BQ24400 datasheet and the BQ2000 Datasheet ... however, the BQ2002 is the only part specifically listed as a purposed 'fast charger'... but from what I have read, fast charging is simply at a 1/C or 2/C rate which all three components are capable of.
The BQ2000 datasheet provides the best examples of circuit design.
I'd like a recommendation as to which IC component would be best for my needs.
I also could use some pointers as to max charge time, how to figure maximum temperature for an NTC thermistor (also which thermistor to use since there are so many types/values), setting peak voltage and also whether I should bypass trickle charge.
The max temp and thermistor stuff is especially stumping me.
Sorry this is kinda a big one, hoping someone has experience with this and can help out
Comments
The problems are not just in the charging phase, but it the discharging phase having one cell discharge excessively and causing a reverse polarity state in an adjacent cell with complete damage.
I have one AAA charger for NiCad and it charges 4 cells with 2 pairs in parallel. From what I can see when shopping a large selection at the local electronics stores... that seems to be the preferred charging configuration for AAA cells regardless of NiMH or NiCad chemistry. So the industry seems to avoid charging long chains fo cells in series..
One can build something from scratch by starting with the the proper range of Voltages for charging, but 'Fastcharging' is always going to be an opportunity for disaster. At least, you should have thermal sensing of each cell to abort the fast charge if there is excessive heat building in any cell.
These cells have internal components.. the anode or cathode that have an extremely thin film coating that makes them efficient. That film is easily damaged by excessive heat or too high a charge or discharge. You project just may be too challenging for the internal construction.
Wire size is directly related to amperage. Do you realize how large your wiring will have to be for 33amps? It is the same as for 33 amp household wiring. Insulation is directly related to voltage.
Charging is nearing completion when there is a sudden rise in the charging voltage toward a per-determined cutoff voltage that is below the absolute safe charge voltage of the cell. But the NiMh have a very flat belly in the middle of the curve.. so the faster you charge, the easier it is to overshoot the cutting off the voltage rise before it does real damage.
If the cells are in series, won't you only need 7 to 8 amps?
I am not using 33A, but that is what will be detected by the IC during checking phases. If a cell in the series dies, the series connection would be broken and cease to charge on these particular charger IC's, or generate heat due to resistance which would also cease the charging cycle.
The arrangement in conventional chargers for AAA batteries from what I have seen are primarily for convenience... try holding four AAA batteries end to end and you'll get an idea why it is easier to charge in parallel. In particular, the IC datasheets actually address the issue of incredibly hard to detect plateau voltage changes of a single sell which is why 2 batteries are commonly used as a minimum in chargers.
There are an incredible number of examples where batteries in series are charged... power drills, rc airplane packs, even other batteries such as an A23 which is nothing more than button cells in series (though the A23 itself is not commonly sold as rechargeable, there are rechargeables).
The best examples are the rc airplane packs... they are also the closest to what I am trying to accomplish. Yes there is a Thermistor inside to detect overtemp condition due to cell degradation which causes the charging IC to terminate before reaching the hard-set peak voltage... this has become a standard as cell degradation is inevitable.
I absolutely must have 5vdc and it absolutely must fit inside a 10mm diameter package... these are absolutes I must have, no negotiations.
the AAA is a 10440 batt = 10mm diam x 44.0mm long which makes it the only viable choice I have found as the amperage on a AAAA is too low and I have not found any to exist in rechargeable form.
If you know of a single rechargeable =<10mm diameter battery that can discharge 5v @ >5a, please let me know! I have considered using a single AAA cell or two in parallel and boosting voltage from 1.2v to 5v, however the amperage would not be sufficient and die very quickly, plus this would require adding even more electronics that must fit in a 10mm package which introduces more weight/cost/heat/efficiency loss. It was difficult enough creating a board for the propeller small enough to fit in there, the QFN chip did the job with nothing to spare on the sides.
The datasheets for the BQ series IC chargers do specifically describe usage intended for 4 batteries in series and refers to them as a 'pack'.
So I hope I have convinced you that making a charger for 4x AAA batteries in series is not only viable, but accomplish-able "safely".
We can observe an example of this on page 3 in thie datasheet for the BQ2000: http://www.ti.com/lit/an/slua064b/slua064b.pdf or on page 1 for the BQ2002: http://www.ti.com/lit/ds/symlink/bq2002d.pdf
The first thing I'd like to work out is what I am most confused on... the thermistor usage.
an example is provided in the datasheets, however... is it possible to use 4 thermistors (one on each battery) by connecting them in series? or would an abundance of temp increase in only one not cause enough of a change across the thermistor network.
for the temp sensorrs . the standard is a 10 K
and If you want a quick and dirty Over temp you can use a quad Opamp as a comparator and then a final OR gate to combine them so that If any cell over heats the output of a opamp is HIGH ) the pack stops charging .
I just made a 2.5 AH NiMH pack last night! for my Vid light . It has a DC DC boost to let me do a small float charge to the batt off ANY input from 4-28 V
and it has a 10 turn pot on it sooo I use it with a 10 K termistor in series with the feedack resistor so that if the pack gets warm the DC DC pumps out less Voltage !!!!!!!!!
I dont know the PID curve but with a crude test It WILL keep the pack safe !
the thing drops to about a 30 mA If the batts thermistor hits 120 F ..
Sounds good to me . or at least better then any other way I can do in a night build
ON THE TRAIN so Ill add more If I get more bars ..
Peter
Those current figures suggest you measured the short circuit current of your batteries - now you'll have to replace them
as that sort of abuse will have reduced their capacity and lifetime considerably.
_Never_ short circuit any type of battery, its a recipe for ruined batteries or burnt out wiring. You measure the voltage
of a cell to get a rough indication of charge and you measure the current in series with the charger to see what the
charging current is. An ammeter is a short-circuit to all practical purposes.
Charging 4 NiMH AAA's in series can be done with a constant voltage charger of about 5.4V limited to 250mA or so,
but you can check the manufacturer's datasheet for the exact batteries to see what charging rates they recommend.
A couple of points...
1. AAA cells aren't 10mm diameter, they're 10.5mm
2. AAAA cells ARE available.
3. You'll probably melt the whole d@mn thing if you try to pull 5A from an AAA cell, and you'll get that power for less than 8 minutes.
4. Cells of the same dimensions as the AAAA, but slightly shorter can be found in neat 7.2V packages with the same shape as 9V PP3 batteries...
Here's a few docs that may be of interest:
http://www.gpina.com/pdf/miniature.pdf
http://www.candlepowerforums.com/vb/showthread.php?52477-Typical-Internal-Resistance-of-NiMH
Discusses the internal resistance of cells, how to measure it and how it changes.
Thanks for the useful info guys... I'll look through this
I'm especially interested in prospects involving a AAAA.
I put one of the AAA's in my micrometer, diameter = 10.15mm (I believe the .15mm comes from the plastic wrap which can be safely removed, however is not necessary).
The reason diameter is important is because I am limited to a 5/8" ID tube with AAA's where a AAAA would allow for use in a 1/4" ID (must allow room for LED strip)