Is this your own layout or is it on the DC2038A demo board? If your own layout, can you send it to me along with your schematic? Feel free to send me a friend request and send it privately.
It may be that the LTC4162 is actually getting through the battery detection test, but stops charging immediately. Usually, passing the battery detection test is not an issue with a good battery if the CELLS pins are configured correctly. If the battery is in pack-protect, it may take a few detection cycles to pull it out depending on how the pack-protect is implemented.
1. I'm not sure what you mean, can you clarify?
2. Yes, it is approximately 1.6ms for the ADC readings. See the Measurement Subsystem section in the datasheet.
Thx to help me,
For the first question:
- Does the value I put in Iin_limit_target or input_undervoltage setting (or others) are linked to suspend_charger or other "error" registers (no_rt, vin_uvcl or even battery detection test) ?
For the second:
- I spoke about data refresh rate of "debeug" register, like vin_gt_vbat when we plug or unplug supply power ?
Moreover sometimes when I unplug the battery ( and keep supply power ), the register en_chg = 1 ?
Thank you to help me, I am lost with the component behavior
1. No, the register values themselves will not change those other registers. However, the results of setting the registers may impact them.
2. You can consider it to be near instantaneous - it will take longer for you to read the register than it will take for the register to update.
For your last question, refer to the Operation Without a Battery section of the datasheet:
Thank you for all this informations
But if the battery is in charging cycle and we unplug battery, how long do we need to wait before charger knows the battery is missing ? (30s like when battery detection test fail) ? (Or never cause I don't use tcvtimer) ?
with a collegue, we detect this afternoon, the charger is on only if we run program before plug battery.
Is this behavior is "normal" or what could be the explaination ?
It is normal sometimes the charger is desable (en_chg = 0) but no "debug" register is set ?
And a last thing, what are the condition for no_rt == 1 ?
On the datasheet we have "The RT pin impedance detection circuit will typically indicate a missing RT resistor", I have this resistor Or "battery charger is not enabled such as when there is no input power available." , I have also have that
So I Don't understand when no_rt =1
Thank you for your time
I took a look at your schematic. INFET is facing the wrong direction, so try flipping that and see if that helps. I'm not sure that it will. I also don't recommend using resistance in series with the INFET gate.
When the battery is unplugged, the charger will sit in CV mode until it terminates. If you are not using the timer, I assume you are using C/x termination? If that's the case, the charger will terminate when there is less than C/x current into the battery node. If there is a load on your battery node, you may never terminate. If there is not, this can be a good way to detect a missing battery.
alsimon said:with a collegue, we detect this afternoon, the charger is on only if we run program before plug battery.
I don't know your program, so I can't be sure why. However, you shouldn't need to change anything if the LTC4162 is configured correctly (correct chemistry variant, correct CELLS pin states, etc).
alsimon said:It is normal sometimes the charger is desable (en_chg = 0) but no "debug" register is set ?
What do you mean by "debug" register?
alsimon said:And a last thing, what are the condition for no_rt == 1 ?
It basically means that the charger is not charging. It certainly doesn't often mean that there is no RT resistor installed, so it can be misleading. I would ignore it for debugging purposes.
Thank you for layout recommandation
I mean by "debug" registers, all registers except measure ones (including threshold and setting).The debug registers are status, state, ...
About my program, I set threshold and parameters like input_undervoltage or voltage_charge_setting, ... Then I read all mesures, thresholds reach, and all "debug" registers periodically (state, status, vin_gt_vbat, en_chg).
If the battery is already pluged when µC woke up, the battery didn't charge until I plug and unplung it 2 or 3 times. Do you have any explaination ?
I do not have an explanation, is it repeatable? Is the battery at a voltage where it is expected to receive full charge current?
It is repeatable, but only with an ampermeter in series.
And no, my battery is < 2.85V, I don't think, it is the max voltage configuration.
Without the amperemeter, the battery charge less than 1s, don't keep the charge and "burst" like that randomly in time.
I suspect the protection pack. What do you think ?
Do you thinks the amperemeter can change some electical parameter. And lead to a correct configuration for the LTC4162 and charge.
I suspect I am at the limit of detection, so some time the battery enter in condition and most of the time it is not without the amperemeter.
With the amperemeter, we just need to plug in right order and wait less then 10s to charge the battery
And I read Chemestry register and it indicate my 1 so LTC4162-L42
Can you clarify: does the battery detection have problems when the ammeter is present or does it have problems when the ammeter is not present?
When the ammeter is present in serie, the battery detection = 0, and my battery charge (in the good side of my "detection limit") . And every thing is ok, from LTC4162-L.
When the ammeter is absent the battery don't charge, the battery detection is unstuck to 0.
In general all register = 0, except VIN_GT_VBAT, NO_RT, VIN_GT_4P2V, INTVCC_GT_2p8V.
Sadly some times device set and reset random registers like VIN_GT_VBAT = 0, with power supply plugged. Or battery missing fault with the battery plugged.
How ammetre can change something in my circuit ?