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.

DRV8955: PID control of TEC using 2 microcontroller inputs

Part Number: DRV8955

I need to control 2 thermoelectrics (12V 3.3A Max each) from a 12V power supply. Two Arduino pins would be used as a PID to pulse on/off a single path of an H-bridge until a temp is hit, then switches to reverse path on the H-bridge. So 100% duty cycle is needed. To avoid going fully discrete I was interested in the DRV8955 since it has a charge pump, and I believe it can be controlled with only two inputs. If you can confirm my questions this would be great:

  •  The documentation mentions the DRV8955 having a 5A peak output in parallel mode. Does this still hold true when MODE is connected to ground and half bridges are used in parallel? Also For 100% duty cycle, is the max constant current output still 5A?

 

I have the requirement of using 2 inputs and wanted to confirm my assumptions.

    • Parallel mode- use 1 input to control IN2, then use 2nd pin to control the sleep mode to stop supplying power to TEC.
      • is this recommended? Would this cause any issues with the overall health of the device?
    • Mode to GND w/ 330k- use 1 input to control EN1&2, then use 2nd input along with a Not gate to provide opposite inputs to IN1&2                   
      •    Is this recommended?
  •  Would an issue arise if an input signal is applied to the device before or during the Twake Time?  

 Thank you for your time!

(edit: removed excessive bullet points)

  • Jeremiah,

    We will get a response to you in <24 hours.

    Regards,

    Ryan

  • Hello Jeremiah,


    I will be happy to assist you. I will go through your questions and try to answer them as thoroughly as I can. 

    However before I do, based on your application and your current needs I would actually recommend a different driver. 

    The DRV8432:

    https://www.ti.com/product/DRV8432

    It has been used to drive TEC loads in the past and might be more suitable for your application I recently helped another community member setting up and I believe the following thread might be useful for understanding the device a bit further:

    https://e2e.ti.com/support/motor-drivers-group/motor-drivers/f/motor-drivers-forum/1026357/drv8432-drv8432-drive-4xtec---xlt2418

    However if you wish to use the DRV8955 I have answered your questions to the best of my ability below:

    For the peak output current the following chart shows the maximum peak output current that the device can safely handle in the different modes:

    If mode is connected to ground the maximum peak output current will be 2.5 Amps. For a max peak output current of 5Amps or more you will have to connect the outputs in parallel by using either mode 1 or hi Z. The following table goes into more detail about the different modes and their maximum Itrip.

    Regarding the max constant current outpu, it will heavily depend in your layout. At the end of the day max consistent current will be determined by thermal limitations and will heavily depend on the layout and composition of your board as well as the ambient temperature you are operating at. In order to properly assess if you will be able to drive it at that current you will have to perform thermal calculations, information on how to do this can be found on section 8.2.2.2 of the data sheet. 

    Regarding your need to use 2 inputs I have the following questions:

    Will each TEC have 2 inputs or do you need 2 inputs to drive both of them?

    Will the TECs operate independent from each other or do you want them to mirror each other in behavior?

     

    Regarding the following question:

    Parallel mode- use 1 input to control IN2, then use 2nd pin to control the sleep mode to stop supplying power to TEC.
    is this recommended? Would this cause any issues with the overall health of the device?

    The main problem with this operation is that the device would constantly go into sleep mode, meaning that you will always have to wait for the Twake time to elapse before you can issue any commands to the device which can be as large as 1.2ms. If this is not a concern, theoretically it could work. I will also like to take this opportunity to answer your question about Twake. There should not be an issue if a command is issued in Twake, I believe the device will not process the command. If however an issue were to arise and a latchign fault occurs it could be easily cleared with a pulse to the nSleep pin as detailed in section 7.4.3

    Regarding the following question:

    Mode to GND w/ 330k- use 1 input to control EN1&2, then use 2nd input along with a Not gate to provide opposite inputs to IN1&2
    Is this recommended?

    Initially I see no problem with the logic, however this setup will have some significant drawbacks. You won't have any current control in this setting as there is no way to set up the Itrip system when the device is set up in half bridge control. Testing would need to be conducted to make sure that this is viable. 

    Let me know if I can help with anything else,


    Best,

    Pedro Arango Ramirez