• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 64 subscribers
  • Views 3096 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ51050 turns OFF and ON during charging

Milan Vandrovec
Prodigy 50 points
Other Parts Discussed in Thread: BQ51050B

I've sent this question to the support, and their advice was to ask here...

We have built a prototype Qi receiver using BQ51050B. At some battery levels (in the middle), the receiver turns OFF every 3-30 seconds, and ON after about 2 seconds.

We have monitored the outgoing communication and it ends with packet 02 04 06, meaning "End Power Transfer" with reason "Over Voltage" (and correct checksum). The documentation for BQ51050B says that this packet is never sent (bq51050b-557127.pdf page 23 Table 1).

Can someone please explain what we need to check or fix in order to avoid "Over Voltage" error?

When we look at the RX coil's voltage, sometimes it goes above 11 Volts just before turning off, other times it seems to stay well below 8 Volts. Also the frequency is around 150kHz most of the time, but rises up to ~200kHz just before disconnect.

The Rfod and Ros resistors have no effect on this behavior (tested Rfod from 100 to 400 Ohm, Ros from 10k to 100k). The coil capacitors have no effect.

It seems that the less inductive the coil is, the better (we have coils 20, 14 and 10 uH). Also partially inserting a ferrite from another coil between the TX and RX coils seems to help a bit, but not always.

Other than the turning OFF and ON, the charging proceeds as designed. We have set Ilim to 400mA, and Iterm to 75 mA.

  • 0
    TI__Genius 12475 points

    Hi Milan,

    What is the status of the Clamp capacitors? Are they atleast >25V voltage rated to cover de-rating?

    Regards,

    Gautham Ramachandran

    Applications Engineer- Energy Management and Transfer

  • 0 in reply to Gautham
    Prodigy 50 points

    Hi Gautham,


    sorry for the late response.

    Both Cclamp are rated for 50 Volts. C3 is rated for 10 Volts.

  • 0 in reply to Milan Vandrovec
    TI__Genius 14890 points

    Milan,

    Can you send a plot showing RECT and BAT? 

    How are you reading the EPT Over Voltage response?  Are you using a Qi-sniffer?

    What transmitter are you using? 

    If you can send a clip of the bq51050B schematic we could take a quick look at that as well.

    The plot you have attached seems to show normal charging, with termination about 60mA at 16:55 or so.  This is followed by a recharge at 19:45 when the battery voltage drops.  You are concerned about the time between 12:15 and 15:15?

    Regards,

    Dick

  • 0 in reply to Dick Stacey
    Prodigy 50 points

    1) Below is an oscilloscope recording in "stable" mode.

    The following are 2 plots of the blackout. It seems about 50/50 chance whether the Vrect signal stays HIGH (and the blackout is short), or LOW (and the blackout is long).


    BAT works the way it is supposed to: in about an hour grows to 4.2V then stays there, with the charging current slowly dropping. Of course the voltage drops a bit when BQ51050B is OFF.

    I'm sorry we cannot provide RECT or BAT plots over a long period of time as we have no suitable equipment for recording.


    2) I've read the packets using oscilloscope between AC2 and COMM2 signals, manually decoding the 0s and 1s:

    A normal Control Error packet 03 E2 E1: (S=start bit, P=parity, E=stop bit)

    The END_POWER_TRANSFER/OVERVOLTAGE packet (02 04 06):


    3) We are using 2 "stock" Qi transmitters. I'll try to find the exact type names.

    It might be true that only one of the two types (but several pieces) are causing the trouble.


    4) I'll try to convert the schematics into a picture but it's a preliminary board so it might be hand-drawn.


    5) Yes, our problem is with the "shaky" part. It actually goes all the way to 0 and then back up. The graph averages these, so it does not look as bad as it is.


    Unfortunately, we have fried both samples, so it may take some time until we can make further measurements.

    Thanks for your assistance so far, and in advance.

  • 0 in reply to Milan Vandrovec
    Prodigy 50 points

    ad 3) The problematic one is a "noname"

    http://www.aliexpress.com/item/Multi-Color-Portable-Universal-Qi-Wireless-Power-Charging-Charger-Pad-For-Mobile-Phone-for-iPhone-6/32579294833.html?spm=2114.01010208.3.62.xoWY5H

    The other is

    http://www.aliexpress.com/item/Qi-Standard-Wireless-MC-02A-Charging-Charger-Transmitter-Pad-For-iPhone-4-4S-5-5S-6/32471193423.html?spm=2114.01010208.3.10.qVws9k . I'm not sure if we observed the bad behavior with this one before. We could not reproduce it yesterday.

    ad 4) The schematic is pretty much the "typical application". I've added the green color for clarification:

  • 0 in reply to Milan Vandrovec
    TI__Genius 14890 points

    Milan,

    There are a few things that jump out at me.

    In normal operation, the RECT voltage should track VBAT until the taper current is about 35% of charge current.

    During stable operation, the communication packets should occur every 250ms or so.  When immediate power levels change, they occur every 30ms or so.

    Your first plot shows normal operation.

    The second plot show packets every 30ms with the RECT voltage increasing.  This tells me the information being sent by the RX is not being decoded properly by the TX.  It seems instead of lowering the power level, the TX is increasing the power level (increasing RECT).  As a result, the RX says to decrease the level - which ends up increasing it.  And that is repeated until the RX shuts down.

    What can you tell me about the interface between the RX and TX (distance, materials)?

    Regards,

    Dick

  • 0 in reply to Dick Stacey
    Prodigy 50 points

    I can't really say if the Transmitter is receiving data. Is there a way to verify this?

    We've noticed that inserting a bit of ferrite between the coils often improves this behavior, but at some charge levels I wasn't able to eliminate the blackouts by moving the receiver, changing distance or adding materials between the two coils.

    The distance to transmitter is about 2.6mm air and 1mm of ABS plastic.

    We also tried just the coil with a few sheets of paper. It seemed to work best with about 2-3 mm of space between the coil and transmitter. When the coil was touching the transmitter, it wasn't very good.

    The coil is TDK WR483265-15F5-G .

  • 0 in reply to Milan Vandrovec
    Intellectual 310 points

    I've had similar problems with a receiver that would only work on a particular type of transmitter. On other transmitters it would turn on and off due to overvoltage. If some loss is acceptable, you can try increasing the series resistance in the LC circuit by adding a small value resistor. Not the most elegant solution, but it solved my problems.