Post Go back to editing

LTC1760 SafetySignal timing characteristics

Hello,

I am designing a system with two removable smart batteries and I am concerned about preventing arcing when hot-swapping one of the batteries. I gather from the datasheet that LTC1760 cuts off the battery's output switch when its thermistor connection goes OVER-RANGE or disconnected entirely, but how much time does it take until the switch is cut? The time for the switch itself to act is given in the datasheet as a few microseconds, but I could not find the timing characteristics of SafetySignal measurement. I need to know this in order to decide whether it will be sufficient to put SMBus and SafetySignal connections on shorter pins in the battery connector (I am eyeing TE 18276x4) and rely on the few milliseconds' delay between the shorter and the longer pins losing contact, or if I need a limit switch of some kind to have LTC1760 cut off the output FET in order to prevent arcing when the battery is being removed.

Thank you for your attention.
Anton

Parents
  • Preventative. I'm asking this question before committing to the first schematics and board designs for charger/manager and battery. I plan to have around 4x3000mAh in each battery and sustained discharge current can reach 15A if only one of the two batteries is being discharged. The PowerPath fast algorithm should prevent the system from losing power, but possibility of mechanically interrupting this high a current worries me.

    it is mostly the insertion of the battery that will cause arcing and high currents

    Hm. Doesn't LTC1760 only open the battery's switches after it observes the SafetySignal in a good range?

  • Hi Anton,

    I commend your taking preventative measures here. At 15A discharge currents, it is a reasonable concern.

    Unfortunately the LTC1760 doesn't do much to help you here because the turn-off is not fast. If you are powered by a battery alone (I assume that is the case here considering we are discussing the battery discharge suddenly stopping) then the charger is not monitoring the safety signal. So, it is not actually detecting battery presence. Here's a plot of what happens during battery removal:

    YELLOW = GB1O, BLUE = battery node, PINK = VOUT

    I have a 1A load on VOUT, otherwise it slowly droops down. You can see from this that the FET is not cutting off at all.

    You can manually turn off the path if you'd like. The attached guide gives an example of how to do so.

    PDF

    Regards,

    Zack

Reply
  • Hi Anton,

    I commend your taking preventative measures here. At 15A discharge currents, it is a reasonable concern.

    Unfortunately the LTC1760 doesn't do much to help you here because the turn-off is not fast. If you are powered by a battery alone (I assume that is the case here considering we are discussing the battery discharge suddenly stopping) then the charger is not monitoring the safety signal. So, it is not actually detecting battery presence. Here's a plot of what happens during battery removal:

    YELLOW = GB1O, BLUE = battery node, PINK = VOUT

    I have a 1A load on VOUT, otherwise it slowly droops down. You can see from this that the FET is not cutting off at all.

    You can manually turn off the path if you'd like. The attached guide gives an example of how to do so.

    PDF

    Regards,

    Zack

Children
  • Hello Zack,

    Thank you for the explanation. I was envisioning a situation where there is no DC power and both batteries are discharging, but one is near empty and the user removes it to swap in a fresh one. 15A may be a tad too conservative an estimate for discharge current from one battery, but better safe than sorry. I see that I will have to implement some sort of per-battery switch cutting off the battery either on the LTC1760 side, or on the battery side. Probably I will make virtue out of necessity and combine it with whatever mechanical latch will be locking the battery in place.

    Best wishes,
    Anton