I am putting together a NiCd charger as seen in this project:
For R4 and R5 I don't have any 33k or 15k resistors on hand, I do have some 10k's, some 4.7k, and some 2.2k. I was wondering if I needed to use the exact values he used in his project for the voltage divider, or if I can use different values. I did see in the code that if the analog reading is less than 70, it assumes no battery is attached.
I'm also modifying it a little bit, instead of using the red LED, R3, and the TIP127 for constant current, I'm using an LM317 and a MOSFET to switch it on and off.
Any advice would be greatly appreciated. Basically, the little 400mAh charger that came with my remote control tank is a POS. I would like to put together a faster charger until I can upgrade to LiPo.
There are two key elements of selecting the best values for the voltage divider.
First it's the ratio of the two resistors ( 15k / 33k = .45 ) that determines voltage stepdown constant. The program for the device 'knows' what voltage level to use to make it's decision to start or stop a charge cycle. In theory any two resistors that calculate out to have the same ratio can be used. If you can't obtain the same ratio then you must have the means to modify the program to factor in the new voltage values.
Second the voltage divider resistors are wired full time across the battery, so there will always be some current being drawn from the battery even if you have the device turned off. In that case you want to use as high a resistance pair to minimize unwanted battery discharge. A simple on/off switch in series with the batteries + terminal would solve this.
Edit: The below statement is what would have to be changed if your resistor divider ratio is different
const int multi = 1614; // multi coefficent for obtaining millivolts
BackWoodsBrewer:
Thank you very much! How was the 1614 value calculated, or, how do I know what value to use for the resistors I have?
Well it's math, you will have to read the code to see the equations they use that factor in and back calculate it see what equeventent battery voltage values they are using in the switching decisions. Not something I'm willing to do at this time
Me, I would just substitute a 50K ohm pot for the two resistor divider network and just adjust the pot until the charger switches at the proper voltage levels. But then again I'm of a hardware background.
Thank you, I hadn't even considered a pot. I guess I could use a multimeter to dial it in to a .45 ratio. I'll try to make more sense of the code, but if I'm not badly mistaken all it really does is average the measurements and shut off when there is a .02 voltage drop, which indicates a 100% charge on NiCd (actually the -.02 drop occurs just a tad past an overcharge, but it seems to be one of the most common methods for smart chargers to recharge them other than temperature monitoring, or a combo of both).
BackWoodsBrewer:
Thank you, I hadn't even considered a pot. I guess I could use a multimeter to dial it in to a .45 ratio. I'll try to make more sense of the code, but if I'm not badly mistaken all it really does is average the measurements and shut off when there is a .02 voltage drop, which indicates a 100% charge on NiCd (actually the -.02 drop occurs just a tad past an overcharge, but it seems to be one of the most common methods for smart chargers to recharge them other than temperature monitoring, or a combo of both).
Well to be able to detect a .02 vdc 'drop' it has to have to proper factor so as to directly relate to the real world voltage of the battery terminals it's monitoring. It's just a scaling thing. Again I would just use a pot and verify proper operation with a good quality mulitmeter measuring the actual battery voltage.
So called 'peak chargers' have been avalible for a couple decades at least as they were/are the preferred method used by the radio control hobbyist. The same charging method works with Nimh as well as ni-cads. When you convert to lipo a totally different charger will be required.
Thanks. I just placed an order for a 5000mAh LiPo, balance charger, and a low voltage alarm so I'm not too concerned with the little 1700mAh NiCd that came with it anymore, though it will be fun to experiment. Actually I think the NiCd is damaged, because the tank came with a cold solder joint on the battery switch and the wire had fell off during shipping. I didn't realize this until I tried running it and had to look under the hood. I charged it last night until it read 8.5v but it only ran for about 3-4 minutes, seems like it has shorter life every time I recharge it :~
