Announcement

Collapse
No announcement yet.

"simulating" a battery

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • "simulating" a battery

    Se here's an interesting situation.... I have a QStarx 818x...love the reception, very reliable signal...however the battery crapped out (apparently it cooked itself while inside the dash panel under the windshield). Without the battery, it will not stay on. It will simply turn on and when the juice runs out, it shuts off until the battery charges enough for it to turn on again.

    So is there a way to 'simulate' that there is a battery? I noticed there are 3 terminals on the battery, and the battery is 3.7v (I think, maybe 3.6v). It takes the 5v input from the USB and uses that to charge the battery. How could I take that 5v and make the GPS receiver work without the battery in?

    I tried just jumpering the 5v to the battery terminals but it didn't work. Also tried a variable resistor in line to adjust the voltage to 3.7 but it seemed to fluctuate a lot...my guess is that 3rd terminal is some sort of 'feedback' and the charging circuit is trying to compensate to the missing feedback.

    Ideas? I can post pics of the circuit board if needed.

    Thanks,

  • #2
    A capacitor may do the trick. Hopefully charging it from zero won't be an (over-current) issue.

    The 3rd wire is probably a voltage feedback to indicated battery charge state which has to be simulated (maybe a resistive divider) but you have to find its relation ship for simulation.
    I'd expect if you found that, a charger alone should work (excluding "peak" current surges that overtax the charger).

    Comment


    • #3
      Would opening up the battery help any on that 3rd terminal?

      I hadn't thought of the Cap...I may rig something up with a cap + variable resistor for the current and see if anything comes of that.

      I'm itchin' to hardwire the thing so there's no battery needed, if I were to replace this battery, it will have been the 3rd one in less than 2 years.

      Comment


      • #4
        Opening the battery - maybe... but you might just find a chip.

        Can you find the specs on the battery - maybe its a "typical" Li-Ion (3.6V is a common phone battery) - it sounds wrong for a LiPo type?
        Or find what voltage that 3rd terminal provides for a full battery, then merely the USB into the device, and a resistive divider (2 resistors or a trimpot) across the USB 5V to the 3rd terminal.

        (Warning - it's been a while since I looked at 3-terminal batteries....)

        Comment


        • #5
          I'll dig that info up tomorrow and see what I can find. Thanks for the tips, I'll post back tomorrow with what I find out. I know it's a Li-Ion and not a Li-Po battery. My other GPS receiver may use a similar (if not the same) battery...haven't checked and haven't thought to until now. I'll charge it fully and take some readings from it.

          Comment


          • #6
            removing the battery completely will mean the gps receiver will have to do a full lock every startup (probably 10minutes +)

            But anyways, if you have a friend with an oscilliscope or a friend that goes to university for electrical or computer engineering, I would have them probe the 3rd mystery port.

            It could be a digital IO pin, not a simple analogue output voltage.
            Fusion Brain Version 6 Released!
            1.9in x 2.9in -- 47mm x 73mm
            30 Digital Outputs -- Directly drive a relay
            15 Analogue Inputs -- Read sensors like temperature, light, distance, acceleration, and more
            Buy now in the MP3Car.com Store

            Comment


            • #7
              Most likely on a single cell Li-Ion pack, the 3rd pin is a thermal sensor and often times is labeled TS. Check that's there's no voltage between (-) and TS. Read ohms, set it in the fridge for 10 mins and check again. It should slowly rise back to the original reading.

              I wouldn't connect it directly to 5V since it's looking for 3.7V and has a charge circuit. It would be better to use an adjustable linear regulator and diode (compensate for the diode voltage drop) and connect it to constant power.

              Comment


              • #8
                I've seen that before with the thermal sensor, but the last pack I took apart from a laptop had a 1 wire serial interface. The battey was spewing out serial data from that thrid pin. There was an initial start character that was the same every time, then about 3 or 4 bytes of data, then the line went dead for a second, and it started again...

                So different things can happen. I would definately suggest to the OP to try what you said and check if it is a simple thermal output. But nowadays, even some batteries require MCU communication...
                Fusion Brain Version 6 Released!
                1.9in x 2.9in -- 47mm x 73mm
                30 Digital Outputs -- Directly drive a relay
                15 Analogue Inputs -- Read sensors like temperature, light, distance, acceleration, and more
                Buy now in the MP3Car.com Store

                Comment


                • #9
                  Oh yes, a multi-cell Li-Po setup needs to do a lot of work. There's a thermal sensor on each cell or at least 1 shared between each 2 cells, plus a wire between each one to monitor individual cell voltage. If just 1 cell is overcharged or drains too much it can create a hazzard.

                  Comment


                  • #10
                    ok - took it all apart and took some readings.

                    There are two pads on the board. One is labeled Vbus, the other vBat. When plugging the receiver into USB WITHOUT the battery I get the readings labeled in the picture below.



                    Now, when the battery is plugged and charging, the T terminal reads 0v, vBus =5v, vBat = 3.67v

                    Would a simple resistor + cap setup bridging the - and T? This is where I'm lost

                    possibly a relay triggered by a cap that shorts the T to ground?


                    :EDIT: I forgot to mention...reading the resistance from T to ground I get 67k ohms. Would simply jumpering T to ground via 67k resistor be a possibility?

                    Comment


                    • #11
                      It will only charge the battery when it has vUsb, and use the battery for power without. T is only used when charging, so you shouldn't need T connected when it's using the battery for power which is the only time you need it. If it's open, the only thing that may happen is the charger won't function properly. 5V may also damage it, so be carefull.

                      Comment


                      • #12
                        So if T isn't really necessary...how would I go about simulating the battery being installed so the unit turns on without the battery? I tried measuring the resistance between the - and + terminals and there wasn't any (or it was so small that my meter wouldn't pick it up on the lowest setting).

                        from the datasheets I looked up on the mtk chipset, input voltage is from 3.3v-5v -/+ 5%

                        Comment


                        • #13
                          I wouldn't think that it uses the thermister to detect the battery, but a resistor would simulate it. 67k would be around abient, but anything around that should work, like 65 - 70 I'd think. It probably has a 100K range.

                          Comment


                          • #14
                            Well, connecting T to ground via 67k resistor doesn't do anything. The power light goes green...but flashes very fast...I'd guess at a rate of around 5-10hz. It won't turn on though. I'm kind of at a loss...I thought for sure this was possible.

                            Comment


                            • #15
                              Try a 7.1k resistor (from the battery's 3rd connector to -ve).
                              Not sure how accurate that needs to be - a 56k paralleled with an 8.2k is 7.15k (<1% error).

                              As I understand it, the battery is equivalent to the common Nokia BL-5C (Li-ion; $10 upwards) and merely has a thermistor temp signal.

                              Comment

                              Working...
                              X