• State Resolved
  • Replies 5 replies
  • Subscribers 30 subscribers
  • Views 104 views
  • Users 0 members are here
Support feedback
Options
Options
Related

DRV110: Adjustable keep time design

Zach Zwanzig
Prodigy 10 points
Part Number: DRV110

Tool/software:

Hi all!

I'm working on a project design right now to drive a 1.5mH load at up to 13 A. I chose the DRV110 due to it being a hit/hold chip with adjustable current settings. I am using the SLIC024 calculation tool provided as part of the DRV110 product page (https://www.ti.com/tool/download/SLIC024) to change the current levels dynamically using a DAC, but I realized there is no such advice given on how to adjust the keep time. The datasheet and the calculator both show you can change the capacitance to change the time, but it doesn't seem like there is any way to change that value without changing the capacitor itself.

With the way the keep pin works (sourcing 1uA to a capacitor and switching Vref values when the keep pin reaches 100mV), we had the thought to try and include a variable current sink to draw a portion of the current away from the charging capacitor. We utilized a Widlar current mirror sourcing from a DAC output, but the MOSFETs being used had too much thermal instability (a small temperature swing would cause the charge time to jump drastically), and thermal stability is a requirement for the design. We also tried implementing an op-amp current sink (like the one seen here Programmable low-side current sink circuit (Rev. A)), and while that solved the thermal issues, it introduced its own forms of instability.

Are there any suggestions on how we might be able to adjust the keep time dynamically? Ideally it could be controllable via a DAC output, similar to how the hit and hold currents can be set with a DAC output. If there is no method of changing the keep time, are there any other driver chips that might have this functionality? 

  • 0
    TI__Expert 4770 points

    Hi Zach, 

    Thank you for your question. 

    Please let me investigate about it and get back to you by tomorrow. 

    Best regards,

    Mojtaba,

  • 0
    TI__Expert 4770 points

    Hi Zach, 

    Thank you for reaching out to us.

    it is mentioned in datasheet that " The keep time, tKEEP, is set externally by connecting a capacitor to the KEEP pin. A constant current is sourced from the KEEP pin that is driven into an external capacitor resulting in a linear voltage ramp. When the KEEP pin voltage reaches 100 mV, the current regulation reference voltage, VREF, is switched from VPEAK to VHOLD. The internal current source is switched off, and the capacitor is grounded for discharge." However, we do not have any reference design for utilzing DAC for changing KEEP value dynamically to provide 100 mV for the input of comparator. 

    Zach Zwanzig said:
    are there any other driver chips that might have this functionality? 

    You can use TI Brushed DC drivers with current regulation, and current sense feature and develop your own firmware based on your application. 

    BDC drivers with current regulation feature. 

    Brushed DC (BDC) motor drivers product selection | TI.com

    Best regards, 

    Mojtaba.

  • 0 in reply to Mojtaba Afshar
    Prodigy 10 points

    Motjaba,

    Thank you for taking the time to look into this. Unfortunately, due to the requirements of our design, we still believe this chip is best suited for our needs. The hit/hold functionality of the DRV110, its ability to automatically regulate the PWM according to the DAC signal, and the ability to (once settings have been determined/set) run with only a pulse on the enable pin (rather than having to sacrifice microcontroller cycles to regulate speed/current/etc) is too valuable to lose, and I don't see any of the brushed drivers that have these features.

    We are going to test using an analog switch with a series of differently sized capacitors to see if we can properly regulate the keep time using that but worry about the leakage current through the mux. If you have any other suggestions we could try, I'm all ears and willing to test more.

    Thanks!

  • 0 in reply to Zach Zwanzig
    TI__Expert 4770 points

    Hi Zach, 

    Accept my apology for late reply. let me inquire more about it internally for any possible solution and get back to you. 

    Best regards,

    Mojtaba

  • 0
    Prodigy 10 points

    For anyone who stumbles across this down the line looking for a similar use case, we ended up using an analog mux, having the common port tied to the tKeep pin on the driver and a small capacitor, then having a series of differently sized capacitors we could select from on the other side of the mux. We chose a mux with a very low leakage current, and this allowed for consistent and repeatable results, even when running high current loads.