Constant Current NIMH Charger

[gary_ramsgate]

I'm building a slow charger for a 10cell 12V NIMH battery pack.
I've run into a problem.



The circuit is based on an LM117 regulator connected in constant current mode.

With no battery connected I first connected my ammeter across the output of the constant current circuit. I obtained a current reading of : 264mA.  This compares favourably with the calculated value of 266mA.   The voltage between Vout and Adj of the LM117 was 1.25V as expected.

With the battery to be charged connected, I then placed my ammeter in series with the battery. Surprisingly I obtained a current reading of 127mA.   The voltage between Vout and Adj was fluctuating between 0.6 and 0.8V in a kind of sawtooth waveform.

Any suggestions would be appreciated.

As the circuit is set up for constant current, I was expecting to measure a current of approx 266mA with the battery connected.

[Soeren]

Hi,

Is there any long(ish) leads involved?
You're having oscillations - can probably be cured pretty fast with a few caps, but why bother, when you can make a better charger just as easy.

[gary_ramsgate]

Would you like to enlighten me on the better charger.

I have made the circuit on breadboard, so there are some long leads.

Thanks.

[Soeren]

Hi,


Sure, but first I need to know if you're using 10 cells to make up the 12V?

[gary_ramsgate]

Yes, it is a 10 cell, 3AH NIMH battery.

Many thanks.

[Soeren]

Hi,

Here's a circuit then:
http://That.Homepage.dk/PDF/10_Cell_NiMH_Tapered_Charger.pdf

It starts off at around 650mA but tapers off as the battery is charged.
If you want a lower current, R3=2R2 will get you close to the current you selected, but it will be slower (and over time will be harder on the cells).
You can feed it from DC as well - just feed 17..18V into it after the diode bridge (which could then be left out) or up to 20V fed into the AC input (you'll then have input polarity protection like when using AC).

[gary_ramsgate]

Thanks.

[Sandgroper]

Hi guys,

This is my first post here.  I'll be needing a charger for my 'bot in the not too distant future and was looking for information on rechargers for NiMH batteries.  Very informative thread! 

Currently I'm using an 8 cell x 2500 mAh pack (though a 10 cell would be better).  The batteries are rated for 1.4 hrs at 2500 mA for fast charge, or 14 hrs at 250 mA for slow charge.  When fully charged the pack puts out 10.4v.  I usually recharge the pack when it gets down to 8.4v, otherwise things start to go haywire.

I'm no electrical engineer, so I have a couple of questions: 

(1) What rating would the resistors on the Tapered Charger Circuit need to be, assuming a maximum charge rate? 
(2) What would be the optimum charge rate?  A two hour charge time would be great but I imagine this comes at the cost of overheating & reduced battery life. 

[Soeren]

Hi,

(1) What rating would the resistors on the Tapered Charger Circuit need to be, assuming a maximum charge rate? 

R2 = 1k5
VR1 = 470R
Adjust to 11.50V
The AC input should then be 12VAC to keep losses down.

This has nothing to do with charge rate, that's governed by R3. With an appropriate heatsink you could go down to 0R68 (or even 0R47 with ample cooling) for R3.

(2) What would be the optimum charge rate?  A two hour charge time would be great but I imagine this comes at the cost of overheating & reduced battery life. 

If you wanna charge really fast, you shouldn't use a tapered charger. (MAX712 is a very good IC for a fast charger).
The current in this charger tapers off (hence the name) from the initial max. current, when the battery gets near full charge and it gets down to close to nothing.

[Sandgroper]

Thanks for the info, very helpful.  The tapered charge would be probably be better for my application (over night charging at recharge station).

I'd prefer to use DC input for safety.  Is it just a matter of chopping out the rectifiier and connecting the DC input to the top (+) and bottom (-) rail? 

What would be the relevant input voltages / resistor values for DC input? 

Sorry about all the questions but I'm learning as I go, and I want to make sure my project is safe before I breadboard it for testing & fine tuning.

[Soeren]

I'd prefer to use DC input for safety.  Is it just a matter of chopping out the rectifiier and connecting the DC input to the top (+) and bottom (-) rail? 

What would be the relevant input voltages / resistor values for DC input? 

Your battery should charge up to 11.5V. There's a voltage drop of ~0.7V over R3 and the LM317 should drop around 3V, so at TP1, the voltage should be 11.5+.7+3 = 15.2V minimum. up to 16V is fine (if you keep the diodes for polarity protection, you should add around 1.4V to that), but don't go much higher, as any extra voltage just dissipates as heat in the LM317 without doing anything good.

Remember a heatsink on the LM317.


DC isn't any safer than AC - quite the contrary in some instances actually (when the voltage is high enough to make your muscles go into seizure, if you try to shunt a large potential with a body part).