Go Down

### Topic: help me understand the operation of these circuit (Read 2521 times)previous topic - next topic

#### fatality

#15
##### Apr 05, 2012, 03:36 pm
referring to the datasheet, http://www.ti.com/lit/ds/symlink/lm117.pdf under the typical applcation note,theres an example of 50mA Constant Current Battery Charger. i think this is the basic circuit that peter hayles used in his circuit to construct the charger right? correct me if im wrong but i think my test circuit is correct. its just that the output voltage of the regulator are a bit higher than the voltage of the battery. its gives about 8.8v but the battery voltage is 8.4v so is it safe for me to charge the battery with this voltage?

#### BillO

#16
##### Apr 05, 2012, 03:49 pmLast Edit: Apr 05, 2012, 04:20 pm by BillO Reason: 1
Yes, it is pretty much what he did, but it is not what is shown in your test circuit.  Can you not see that?

The resistor MUST be in series with the load.  Yours is not.

Have a look at the attached schematic.  Do you see where the 24 ohm resistor is?

The 1K resistor is there to limit current through the transistor and is not a major contributor the the current limiting through the battery.

Can you see the difference between this and your schematic?  In your schematic, the 3.9 ohm (wrong value) resistor is NOT in series with the load, so will not do anything to limit current through the load.

The low current you are seeing is because your input voltage is too low.  12V is NOT enough.

Facts just don't care if you ignore them.

#### fatality

#17
##### Apr 05, 2012, 11:09 pm
my wrong there,i draw it wrong in orcad but yes,in the real test circuit the 33ohm resistor is in series with the load. and i remove r2 and r3 from the circuit to make sure all the current goes to the battery and to remove all other variables. when i turn on the adapter and measure the voltage of the battery,it goes up to 9.5v and then i turn off the adapter because the battery is rated at 8.4v only (dont have any goggles around).
what is wrong?
or i shouldnt remove r2 and r3 from the circuit?
this is the test circuit but without r2 and r3 and the value of r1 is 33ohm

*will try to build the circuit as you suggested this weekend and hook it up to my arduino and re measure everything

#### BillO

#18
##### Apr 06, 2012, 12:07 am
No, there is nothing wrong.  Under charge the cells will float up to around 1.5V per cell.   That would make your battery go as high as 10.5v or more.  You should stop charging once it gets to that voltage (or maybe a bit less ~ 10.4V).

But do not use the 3.9 ohm resistor.  Use 24 ohms.

And do not use 12V in, use 16V, otherwise the regulator cannot work properly.
Facts just don't care if you ignore them.

#### fatality

#19
##### Apr 06, 2012, 08:24 am
thanks for the info. im thinking of using the delta v method to stop the charging process or should i just stop at ~10.4v?
i only have 12v adapter around to power up the regulator. should i make a bridge circuit so that i can power it up directly from the wall outlet?
thinking about adding bridge circuit and another regulator to set the input voltage to ~16v

#### BillO

#20
##### Apr 06, 2012, 01:32 pmLast Edit: Apr 06, 2012, 01:44 pm by BillO Reason: 1
Well, there is always the chance that the adapter is putting out more than 12v.  Most of them do, and many of those only use half wave rectification or do not have a filter capacitor.  Put a big (220uF - 1000uF) electrolytic capacitor across the adapter and measure this 'filtered' voltage.  Be sure to observe the polarity and the voltage rating of the capacitor.  The capacitor should be rated at more than 16v.

The delta V method varies depending on the internal resistance of the battery.  It is most effective for large, low impedance batteries like those used in tools.  Smaller, high impedance batteries, like those used in cordless phones and other low power electronics tend to float higher while charging and the delta V is not as sharp.  Especially if the adapter you are using as a source has a low voltage.  Remember, the diode and regulator are going to drop the input voltage by  a minimum of 2.5 volts (and another 1.25V across the resistor, if you get things running), leaving you with only about 9.5 if indeed that adapter puts out only 12V.  This is not enough to effectively charge your battery.

So, do the test I mentioned above and let us know what the resulting 'filtered' DC voltage of your adapter is.

You really do need at least 16V or this circuit will not work.  If you can't get 16V, then I would recommend just using a resistor to limit current and cutting off the charge when the battery gets to 10.4V.  The exact value of that resistor will not be known until we know the exact filtered output of your adapter.

Edit:  By the way, I do not recommend you using a bridge, or anything else, to rectify the power coming out o the wall.  What is the AC supply voltage in Malaysia?  240?  Please don't try to power this thing off 240V directly.  Very bad things will happen.
Facts just don't care if you ignore them.

#### fatality

#21
##### Apr 09, 2012, 02:25 am
sorry for the late reply. quite busy with other projects. anyway, measured the filtered voltage of the adapter using a 1000uF 16v electrolytic capacitor by placing it between the output of the adapter..the value is 12.4v. the same as the unfiltered voltage. its hard to get a 16v adapter here. most of them are either 12v or selectable adapter up to 12v

#### BillO

#22
##### Apr 10, 2012, 03:16 am
Then the regulator is wasted.  Just replace the regulator and 29 ohm resistor with a transistor and 50 or 60 ohm resistor.  Turn off the charge (turn on Q2) when the battery gets to 10.4V

See attached schematic.
Facts just don't care if you ignore them.

#### fatality

#23
##### Apr 10, 2012, 04:10 am
thanks.it works like a charm
anyway,im also planning on implementing a discharger with my system so i found this constant current sink
im planning to drain the battery at 100mA and then sending the data through serial comm which will later be analyzed by labview to produce a graph and the actual mAh of the battery.
I got the labview program running OK but cant find those mtp3055. Can anyone suggest other logic level n channel mosfet that is more common and can be applied to this constant current sink circuit?

Go Up

Please enter a valid email to subscribe