Just realized I probably should have posted this in here, so I apologize for the double post...

So, I have a situation. I have a new prototype system that uses a 13.2 Volt NiMh battery pack, and I need to figure out a quick and dirty charging system for temporary use (we are developing a more universal buck/boost charging board for the product in the near future). For now, we just need to charge the system up during a demonstration and evaluation phase, where it will typically be allowed an overnight charge between uses.

We have a readily available standard 18V unregulated DC power supply being used for the previous generation of this product, and would like to use that supply if possible. That power supply is rated at 18V 533mA, and shows about 24.2V output measured with no load.

Any suggestions on a good quick and dirty charging circuit? I've been looking at so many variations (LM317 based, some specific charging chip based, etc) that I'm going crazy, and I think I may be overlooking the best and most obvious choices.

Appreciate any help on this... Might save me from a nervous breakdown!

 

Multiply the AH capacity of the battery by .095. Configure the LM317 as a constant current circuit. Divide 1.25/(AH x .095). This is your resistor value.
I have built many this way and they will never overcharge the batteries. I just took a lunch break from building another one for some AA batteries.

 

You could also use a simple series resistor to limit the maximum charge current to .095 x AH, but that would vary some as the battery charges or the power supply voltage changes. Likely a better solution is using the LM317 in a constant-current configuration as k7elp60 suggested.

 

And it may not be optimum but plenty of folks have gotten away with an LM317 set to constant voltage. The built-in overheat, over-current protections in the LM317 cover up a lot of badness. If the battery cannot tolerate bulk charging at 1A - which the LM317 may be capable of until it overheats - then you need current limiting.

 

Use a constant current charger as mentioned in the earlier posts they are safer, a circuit like this is the simplest, divide the mAh capacity of the battery by 10 and use that as the current to charge the battery.

so if your battery is say 4000mAh then use 400mA to charge it for 1hour MAX!

To set the resistor value divide 1.25V by the current in amps, Example 1.25/0.4A = 3.125 ohms 1Watt

 

Use a constant current charger as mentioned in the earlier posts they are safer, a circuit like this is the simplest, divide the mAh capacity of the battery by 10 and use that as the current to charge the battery.

so if your battery is say 4000mAh then use 400mA to charge it for 1hour MAX!

To set the resistor value divide 1.25V by the current in amps, Example 1.25/0.4A = 3.125 ohms 1Watt

Your circuit is exactly what I use. I add a diode in series with the + output lead. Just a safety factor in-case the power goes off. The reason I use .095 instead of 0.1 is back in days of Ni-Cad batteries I learned that as the batteries get near charge the temperature of the cells increases and there is a slight - bump in the terminal voltage. Reducing the charge current reduces some of the heat and the charger does not need to be timed. In other words the charger can be left on indefinatly.
This info was passed to me by a former supervisor who had many years in the battery pack rebuilding and battery charging business.

 

Thanks for the guidance, guys! The LM317 solutions kept swinging around as I was doing my research, and it's good to have some real feedback on how/if they work. So simple and elegant, too!

This should definitely give me what I need to get through this prototype / alpha testing phase. You guys are life savers!