Recharge of Standby battery.?

C

Craftybytes

Senior Member
This query has possibly been covered before but I couldn't find any results covering or near my particular need when I did a forum search.

Is it possible to 'trickle' charge a 3.6v NiMH 80mAH battery (small PCB solder-in unit) from an external 5V supply using only a resistor (to convert to current) - or do I need a higher external supply voltage.?

Am looking at running an 08M on 5V normal (external) and a on-board 3.6V standby battery - need some way to 'recharge' this 3.6V battery.!!

Any help much appreciated.

crafty.
 
M

MFB

Senior Member
If you power the PICAXE from the battery (or via a 3V low drop regulator) it can be charged from the 5 volt rail, but you will also need a diode in series to prevent discharging back when the rail is powered down. The diode should be a low-drop Schottky type.
 
B

BeanieBots

Moderator
This may well flare up into a "battery wars" thread because there are many different opinions out there. Even the "experts" can't agree.

Recent NiMh techology includes chemicals which re-combine the gasses which are produced when a battery is over-charged. It only works for currents which are less than about C/10.
So, yes, you could use just a simple resistor with the caveat that the maximum charging current does not exceed C/10. Personally, I would not exceed C/20 (or lower) in a simple charger.
 
M

moxhamj

New Member
Under C/20 the battery would not even get warm and you could get away with trickle charging long term. 3.6V from 5V is a difference of 1.4V. Lose maybe 0.6V in the diode = 0.8V. If you know the current from 1/20th of the capacity, you can work out the resistor with V=IR.
 
C

Craftybytes

Senior Member
Thanks guys for the quick response.

MFB - yep - thinking to use a BAT85 diode in series with around 1K2 ohm resistor to keep the charge current down below C/10 mark..

BB - will be using a 1K2 ohm resistor as my 'charge' resistor - so at 5V (less diode drop @ ~0.2V) in will give approx 4.0mA of charge for 'trickle' charging purposes.

Might up that 'trickle' current to somewhere between C/10-to-C/5 mark - as the unit will only be connected to the main 5V supply for between 20-30 seconds at a time - 1st after "main" external system powered up - 2nd (& thereafter) once every 5 minutes whilst "main" system is operating.

HTH.

crafty.
 
M

MFB

Senior Member
The drop across a BAT85 diode at a current of C/20 should be less than 400mV. I am not sure your proposed charging duty cycle and current level would fit the accepted definition of ‘trickle’. I have found that even charging in short bursts can cause the battery to fail (they short after months of operation) if subjected to too high a charging rate, but only experience with a specific battery enabled the maximum to be reliably determined.
 
B

BeanieBots

Moderator
The 'short' that MFB refers to is caused by "dendrite" build up.
This is when the chemical crystals are grown slowly (trickle charge only) and get so large that they puncture the insulating material.
This can be reduced/prevented by periodic rapid (1C) charge/discharge cycles.

What is the actual battery capacity?
Don't forget that with such a small avilable volt-drop, the charge current will vary a lot between full and flat conditions.
 
hippy

hippy

Technical Support
Staff member
I did some experimenting with an 08M charging a 1.5V <delete>NiMh</delete> NiCd and that looked promising; basically an R from an output pin, Output High to charge, Input for no load. Each PICAXE output pin can source 20mA-25mA and can be paralled for higher currents. Multipe outputs with differening R's can give differnt charge rates. If it's only 80mAh, C/10 = 8mA should be easy to do. READADC can check for the delta-V ( negative delta-V for NiMh ? ).

One of those projects worth trying out to see how it goes - It was fun and taught me a lot about battery charging.
 
Last edited:
A

alband

Senior Member
I think I understand that last idea. Lets see...
You could then use 7 outputs with different resistors to give you the "numbers" 1,2,4,8,16,32,64 allowing you to make any "number" up to 127.
You could then have three input pins to program the charge rate (up, select and down). Have and LCD or other display. You could even use the button command which is currently being discussed and Hey Presto - Your own programmable trickle charger. :D
 
C

Craftybytes

Senior Member
Thanks guys.

Had not thought of using free 08M outputs as 'charge' control..

Currently in my simple schematic there are 2x unused o/p pins - so maybe with a little extra code one could possibly utilise them to do the 'recharging' of the 3.6V NiMH standby battery - no extra transistor required.

BTW - the gist behind my circuit is that as part of a larger project I need some way to 'memorise' the on/off state of a micro-switch (part of a mini hardware sensor unit) - this 'switch' is connected via a 2x wire link to the main system.

The problem is - this main system may only be powered up for a couple of hours each day or may not even be powered up for several days at a time - and one of its i/p's is to monitor the state of the external micro-switch.

Now this micro-switch can be activated irrespective of whether the main system is powered up or not - and in most cases will only be 'activated' when the main system is NOT powered up - so I need a method to store a "memory" of the 'activated' state - so when the main system is eventually powered up it can retrieve the 'memorised' state of the micro-switch and act upon it.

This 'memory' needs to be able to sustain itself for extended periods - maybe even up to several weeks at a time - before being accessed by the main system.

So I decided that by using an 08M picaxe chip and a few extra components plus a small pcb mount standby battery all loaded on a smd pcb and by using the micro-switch to turn it on/off - the eeprom in the 08M could be used as the 'memory' for the "on" state of the micro-switch I was looking for.

If anyone thinks there is a simpler way to do this - please advise.

crafty.
.
 
B

BeanieBots

Moderator
Battery size??? 80mAhr good for about 3 days.
Weeks at a time! PICAXE draws ~1mA unless you are under-clocking & dissabled BOD etc.

I might have missed something obvious, but why do you need battery backup if you are using EEPROM to store the "state"?
 
C

Craftybytes

Senior Member
BB -
As I stated -

and by using the micro-switch to turn it on/off
Click to expand...
When the micro-switch changes state from off to on - the battery is switched on and is used to actually power the 08M for the fractions of a second it takes to check the switch and write to the eeprom (if need to) - the 08M then goes into ultra-low power mode (disableBOD, etc) till the next reading (once per minute) - or untill the micro-switch changes state back to off - so need the battery..

Hopefully the battery will last a few weeks with such 'small' power usage.

When the connection to the main system is active - the incoming 5V is then used to power the circuit instead of the battery - AND to 'recharge' the battery as well.

HTH.

crafty.
.
 
Last edited:
M

manuka

Senior Member
Quoth Hippy
08M charging a 1.5V NiMh
Click to expand...
NiMH are of course normally only 1.25V- is Rev.Ed onto a new & improved version ?! A great little charging idea however, & worthy of further application. Stan
 
hippy

hippy

Technical Support
Staff member
@ eclectic, post #9 : yes, and yes, as best I can tell. I've updated the tags in the original post linked to so it displays better.

@ manuka et al : My mistake, it was NiCd not NiMh I used.
 
B

BeanieBots

Moderator
Umm.. NiCd nominal voltage is also 1.25v.
Only Zinc and 'regular' alkaline are 1.5v.
Serious gassing if NiCd or NiMh goes above about 1.45v
 
hippy

hippy

Technical Support
Staff member
... okay, change that from a colloquial "1.5V" to whatever is technically correct :)
 
You must log in or register to reply here.
Top