• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 515 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.

AMC6821: AMC6821 PWM output stop

Andrew Du
Intellectual 1840 points
Part Number: AMC6821

Hi, Team:

1. My customer use AMC6821 to drive a 4 wire Fan and monitor the RPM every 5s.

They found some time the FAN stopped; if they find the FAN stop after 30s, they will write a 100% PWM and make the FAN recovery.

For the chip power up initialization, they just set the FDC1: FDC0 = 00, write the DEC to 0x22, keep other register no change.

When read the RPM, They set the TACH_EN: TACH_MODE = 11;

Because such issue happen in their end customer, they cannot dump the register. But they confirm that the temperature read back is correctly. 

Do you have any experience or debug solution for such issue?

2. One month ago, the customer report that some system fan stop and they found the PWN_EN =0; but they feedback host didn't do any operation on the chip. Any reason make the chip change the PWN_EN from 1 to 0 by itself?

  • 0
    TI__Expert 6475 points
    Andrew,

    I have a few questions:

    1. Is this a new design? Have the customer used this device before in other projects?
    2. You mentioned some time the fan stopped. What is the frequency of this occurrence. Like 3/10 of the end equipment has this issue or do all of them have this issue every now and then.
    3. Do you have a schematic or block diagram of the system?
    4. Does the fan recovery procedure works 100% of the time?
    5. When the fan stops, does the PWM output stops also? I want to rule out the fan/driver itself as the issue.
    6. I haven't heard of this issue before but then again I haven't supported this part for that long either.
    7. I'm not sure why the PWN_EN would change from 1 to 0 by itself right now. I need to look into the datasheet for more clues.

    -Kelvin
  • 0 in reply to Kelvin Le THS
    TI__Intellectual 1840 points

    Hi, Kelvin:

    The feedback as below:

    1. this is not new design, the custoer already use this device for several projects.

    2. not all of the product have this issue. about 4/120pcs have this issue. but this four fail euipment is in different city.

    3. See below schematic.

    4. Yes, it is 100% recovery.

    5. They didn't duplicate such issue in their lab, but from their field log information, the PWM_EN = 1 and never = 0.

    6. Do you know if we have EEPROM or flash in this chip?

  • 0 in reply to Andrew Du
    TI__Expert 6475 points

    Andrew,

    1. They never had any issues with the previous projects? Did they use a different manufacturing flow for this project?
    2. Are they planning on debugging these failed units in the lab? The need to be able to recreate the fail condition in the lab. Why would the 116 units work fine but not the other 4?
    3. Are you saying that the field log (PWM_EN = 1) contradicts what the customers observed (PWM_EN = 0) on the PWN_EN bit?
    4. No EEPROM or flash that I know of.

    BTW, I don't understand what you mean by "write the DEC to 0x22".  0x22 is an 8-bit register.  What is "DEC"? You mean "DCY"?
    -Kelvin