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.

PGA411-Q1: PGA411-Q1

Part Number: PGA411-Q1

the resolver IC PGA411-Q1  is functional

  • IC PGA411-Q1 is very hot  and  over a time  the  exciter signal has glitches ( with out connecting to exciter coil)
  • when cooled with ab external heat sing and fan the exciter waveform is stable 
  • IC PGA411-Q1 is assembled on a six layer board and land patter as per data sheet 

need your help to find out if a similar problem reported 

and what is cause ?

  • Suresh,

    Are you driving a resolver coil during this test? If so, what is the peak current needed to drive the coil?

    Please look at this thermal application note to see typical temperature rises for the PGA411 and the factors that contribute to it:

    www.ti.com/.../slaa738.pdf

    Thanks,

    -Clancy
  • I believe that this has been resolved offline. Can you confirm if the issue is solved, or if you are still having problems?
  • Clancy

    IC PGA411-Q1 is very hot and over a time the exciter signal has glitches ( with out connecting to exciter coil)

    studied the application note some points i would like to mention

    1) in data sheet section 3.1

    TI recommends selecting the minimal boosted
    voltage required to avoid clipping. Doing so will minimize efficiency losses and unnecessary temperature
    increases in the PGA411.
    (In over case the voltage input is 5V and output is 15V it is in maxim boost mode )

    2) The EVM thermal pad is as per datasheet recommended has itself has 61.2 degree which is not that good is what i feel

    3) In industrial layout ground using top midlayer and bottom layer to remove the heat is not a good sign

    4) having the option of having the switching mosfet outside would have been better
  • the issue has not been resolved
  • No problem Suresh.

    Since the PGA411 is getting hot without a resolver sensor connected, then something is probably wrong with the schematic/PCB, it should not be doing that. A couple of things to check are:

    - Digital IO pins should be pulled up to VIO. So if VIO is 3.3V, then all of the digital inputs should be pulled to 3.3V, not 5V. Digital input pins include AMODE, ECLKSEL, VA0, VA1, FAULT, FAULTRES, etc. Current limiting pull-up resistors should limit the current flow, but if you have anything directly connected to a higher voltage than VIO, then the devcie could be heating up because of that.
    - VEXT should be directly connected to VEXTS.

    About your comments on the app note:
    - The EVM does not have a good thermal design (for instance, there are no vias underneath the thermal pad), and as such, the thermal performance is not very good. The datasheet recommended layout is not what is on the EVM.
    - If the ground planes are not feasible in your system, then you will need to look at all of your system constraints to design a good system. You need to consider the maximum ambient temperature in your system and the maximum power dissipation that you will require from the PGA411.
    - Having the MOSFETs outside would reduce the thermal load on the PGA411, but at the cost of more difficult diagnostic coverage and more PCB space.
    - You can still drive an external power stage with the output of the PGA411. This would reduce the loading required on the PGA411. However, once you fix your board issue, I hope that you will find that it isn't necessary.

    Thanks,

    -Clancy

  •  - Digital IO pins AMODE, , VA0, VA1, FAULT, FAULTRES, etc.   do not have pull up or pull down up it is connected to processor pins witch works on 3.3V 

     - Digital IO pins ECLKSEL was connected to 5V   ( PGA411-Q1  was heating and drawing more current )

    --Digital IO pins ECLKSEL was connected to 3.3V    (  PGA411-Q1 is not heating and draws less current )  and the waveform is stable

     Thanks a lot for your suggestion

    plan to do some more test and would keep up updated 

    - plan to change the exciter voltage VEXT to 15V currently it is working 10V

    - Load the resolver coil 

    regards

    Suresh 

  • Suresh,

    That is excellent news! Let me know if you need more help.