I've been trying to make a reliable NiMH charger for a few decades.
I started by trying to measure the charging voltage this was unsucessful as when
pushing in a lot of current any small junction resistance would cause large changes in
the voltage reading unrelated to the state of charge.
Then it seemed like monitoring the cell temperature was the way to go.
There were several problems with this:
The charger also gets hot (I was using a transformer and LM317s to control the
current both get rather hot. So I went with differential temperture - two thermistors
one next to the cell and one a little way off. It should have worked, but I couldn't
get it to be reliable. Sometimes it would cut off early other times the cells would
almost catch on fire.
Now I am trying to detect the negative dV/dT which is happens at full charge.
It seems to be working so far.
I have attached a typical voltage vs time graph notice that there is a long plateau
where dV/dT is close to zero. To avoid switching off at this point I have also put
in a minimum cell voltage that needs to be achieved (1425 mV).
The y-axis on the graph is mV
I also decided to use one of those USB power supplies (as I have a lot of these from old phones).
The problem then is ypu only have 5V to play with many regulators require a little more over voltage to work. I tried a switched mode but measurement suggested that it was rather inefficient . so I decided to go with a dropping resistor for simplicity. Anyway here is the circuit.
Why isn't the circuit constant current? You can do that with a linear regulator and join the ground pin to the output via a resistor after disconnecting it from ground. the current is the resistor value.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have tried linear regulators (LM317), but there is not enough overvoltage for these to work reliably. As NiMH have a very low internal resistance and the cell voltage does not change too much during charge the actual current through a dropper is fairly constant.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Why isn't the circuit constant current? You can do that with a linear regulator and join the ground pin to the output via a resistor after disconnecting it from ground. the current is the resistor value.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've been trying to make a reliable NiMH charger for a few decades.
I started by trying to measure the charging voltage this was unsucessful as when
pushing in a lot of current any small junction resistance would cause large changes in
the voltage reading unrelated to the state of charge.
Then it seemed like monitoring the cell temperature was the way to go.
There were several problems with this:
The charger also gets hot (I was using a transformer and LM317s to control the
current both get rather hot. So I went with differential temperture - two thermistors
one next to the cell and one a little way off. It should have worked, but I couldn't
get it to be reliable. Sometimes it would cut off early other times the cells would
almost catch on fire.
Now I am trying to detect the negative dV/dT which is happens at full charge.
It seems to be working so far.
I have attached a typical voltage vs time graph notice that there is a long plateau
where dV/dT is close to zero. To avoid switching off at this point I have also put
in a minimum cell voltage that needs to be achieved (1425 mV).
The y-axis on the graph is mV
Last edit: David Stephenson 2020-03-08
I also decided to use one of those USB power supplies (as I have a lot of these from old phones).
The problem then is ypu only have 5V to play with many regulators require a little more over voltage to work. I tried a switched mode but measurement suggested that it was rather inefficient . so I decided to go with a dropping resistor for simplicity. Anyway here is the circuit.
Last edit: David Stephenson 2020-03-08
I use these sort of supplies a lot
https://www.ebay.co.uk/itm/LM2596-DC-DC-Step-down-Adjustable-CC-CV-Power-SuppPTH/402119818226?hash=item5da0357bf2:g:o-wAAOSwZ3Ja868z
This video explains things for readers who know little about chargeing nimh batterys
https://www.youtube.com/watch?v=_ROhH9EkhtU
Why isn't the circuit constant current? You can do that with a linear regulator and join the ground pin to the output via a resistor after disconnecting it from ground. the current is the resistor value.
I have tried linear regulators (LM317), but there is not enough overvoltage for these to work reliably. As NiMH have a very low internal resistance and the cell voltage does not change too much during charge the actual current through a dropper is fairly constant.
Why isn't the circuit constant current? You can do that with a linear regulator and join the ground pin to the output via a resistor after disconnecting it from ground. the current is the resistor value.