• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 63 subscribers
  • Views 338 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

BQ24650: Part gets burnt sometimes with no apparent reason

Tomer Bruchiel
Intellectual 640 points
Part Number: BQ24650

Hi Team,

We designed a solar charger for a LiFePO battery.

The charger is part of a board that has many other functionalities including driving 3 BLDC motors.

After the initial (and extensive) testing of the board, we manufactures 150 of them for a larger scale testing before going to mass production.

We installed them in the field (Outdoors solar plant) and we saw that about 10 of them are not charging.

We inspected the boards and found out that the BQ24650 has been damaged.

Further testing shows that the resistor Rsr (the one between SRN and SRP) is burnt as well. Note - it was burnt to open not to short.

It seems to me that some extensive current was flowing there and that caused the issue.

I replaced the resistor with a working one but the issue remains - the BQ24650 was damaged and will not charge.

We need some help please understanding what might cause such a behavior before going to mass production.

The parameters of the system are as followed:

5 cell LiFePO battery (16V to 18V).

Max charge current is set to 4A

Solar panel is 50W 28V

Rsr is 10mOhm, 1W resistor.

MPP is set to 21.5V

Attached are the schematics and measurements comparison done between a working board, and a malfunctioning one. For the malfunctioning board we replaced Rsr to see if it solves the issue but it didn't.

Cases 1,2,3 - Working board

Cases 4,5 Malfunctioning board, Original Rsr - failed to open (infinite resistance)

Cases 6,7 Malfunctioning board with new Rsr.

Case 1 2 3 4 5 6 7
Board Working board without battery Working board With battery and current limit to 1A Working board With battery and no current limit Malfunctioning board without battery Malfunctioning board With battery, with and without current limit to 1A Malfunctioning board without battery - new Rsr Malfunctioning board with battery - new Rsr
Supply Voltage 24 24 24 24 24 24 24
VCC 24 21.5 24 24 24 24 2.3
REGEN 5.97 5.97 5.97 5.98 5.98 5.98 5.98
BTSBT 22 22.5 23 5.9 5.9 5.9 19.2
HIDRV 18.2 21.2 21.3 0 0 0 16.35
PH 17.3 16.6 17 0 0 0 16.35
LODRV 0 1.2 1.4 0 0 0 0
VFB 2 1.97 2 0 1.95 0 1.95
STAT1 2.2 0 0 0 0 0 0
STAT2 1.4 1.4 1.4 1.4 1.4 1.4 1.4
VREF 3.3 3.3 3.3 3.3 3.3 3.3 3.3
EXT_TS 2.07 2.07 1.95 2.04 2.04 1.95 1.69
TS 2.07 2.07 1.95 2.04 2.04 1.95 1.69
MPPSET 1.3 1.2 1.3 1.31 1.3 1.31 1.31
SRN 17.35 16.6 16.89 0 16.4 0 16.38
SRP 17.35 16.63 16.94 0 0 0 16.38
CURRENT 0A 1A 2.95A 0 0 0 0

Thanks,

Tomer

  • 0
    TI__Mastermind 24075 points

    Hi Tomer,

        Do you have a test procedure to evaluate performance before sending out into the field? SRN and SRP are sense lines, and do not carry current through them. The total charge current definitely passes through RSR, but if it is indeed 1W rated, then it should be able to handle the power dissipation across the resistor.

    If you have an EVM available, are you able to solder on the BQ24650 that are not working on your board, to the EVM to evaluate if it can work on the EVM?

  • 0 in reply to Kedar Manishankar
    Intellectual 640 points

    Hi Kedar,

    I don't have an EVM but I can solder it to another board that is working (we have around 100 of those).

    I don't see how it will help as the measurments shows that something is wrong but I can do it.

    I have another theory - if there will be a short circuit between SRN and SRP, PH theoreticly will produce infinate current - could that happen and harm the part and Rsr?

    Thanks,

    Tomer

  • 0 in reply to Tomer Bruchiel
    TI__Mastermind 24075 points

    Hi Tomer,

       This part has battery short protection as well as a secondary cycle by cycle charge overcurrent protection:

    The charger has a secondary cycle-to-cycle over-current protection. It monitors the charge current and prevents the current from exceeding 200% of the programmed charge current. The high-side gate drive turns off when overcurrent is detected and automatically resumes when the current falls below the overcurrent threshold. 

  • 0 in reply to Kedar Manishankar
    Intellectual 640 points

    Hi Kedar,

     

    We tried to replace the parts but it is very difficult to remove and place this part by hand so we couldn't do it.

    The part is almost certainly damaged as I replace the Rsr resistor and it didn't help.

    Also, shorting SRN and SRP on a working board didn't do anything except by increasing the current by 1 amp.

    Any ideas on how to proceed?

    We now have about a 100 working boards in the field - all are verified. We are monitoring them - any idea to conduct tests on a daily basis?

     

    Thanks,

    Tomer

     

  • 0 in reply to Tomer Bruchiel
    TI__Mastermind 24075 points

    Hi Tomer,

      The conditions you have are in the recommended operating range so I dont see anything out of the ordinary that would cause this damage. Usually damage is caused by an over voltage or over current condition.

    Is it possible to capture when the damaging event occurs?

  • 0 in reply to Kedar Manishankar
    Intellectual 640 points

    Hi Kedar,

    Thanks for the confirmation.

    We are trying to catch this. I will update once we succeed

    Thanks again,

    Tomer

  • 0 in reply to Tomer Bruchiel
    TI__Mastermind 24075 points

    Hi Tomer, 

       I will keep this thread open till then.