• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 64 subscribers
  • Views 692 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.

DRV103: DRV103H IC heating beyond the temperature limit specified when PWM output is turned ON

kirubakar Nagarajan
Prodigy 80 points
Part Number: DRV103

Hi Team,

We have implemented the New DRV103H driver IC for operating the Solenoids (24V DC/1A continuous) in our product to replace the DRV101 (Obsoleted). The design is developed as per datasheet recommended and below are the set parameters for the IC.

Parameters: 

1. Input Control Voltage to IC: 5V DC(0 or 1)

2. Supply voltage: 24V DC

3. IC driving: Low side driven

4. PWM Resistor: 110K Ohm (For 30% Duty cycle)

5. Frequency Resistor: 215K Ohm (For 24KHz)

6. Delay Capacitor: 330nF (For 330uSec initial Delay)

ISSUE OBSERVED DURING TESTING:

1. First we provided the supply voltage (24V) to IC and check if there is any short. 

2. Later, we provided the control input voltage to check the PWM output from out pin.

3. Initially the delay time is observed as per calculation and could witnessed few PWM signals.

4. After few sec (approx 30sec), the IC didnt provide PWM signal and the output pin was in continuous mode

5. Later we removed the daughter PCB from main PCB and repeated the test separately by powering the IC and control input voltage.

6. At that time, we could be able to see the PWM signal continuously, and solenoid also connected across the diode (24V & output 0V)

7. But the IC produce more heat and the heat does not able to dissipate even we provided 4-layer PCB copper pouring for better heat dissipation.

8. The IC exhibits more than +85Deg C in ambient temperature of +25Deg C

9. But as per datasheet, the temperature rise calculated is not matched with the practical data. Also, we loaded the IC only 250mA in PWM mode (1A in continuous mode)

10. If we removed the Solenoid load and tested in open condition, even though the IC generates more heat when PWM output is ON completely.

11. Also, we replaced the IC DRV103U and repeated the same test, the IC produce the distorted PWM waveforms and IC becomes more hotter (>+85Deg C) than the DRV103H

Below are the Gerber images and waveforms provided for your reference. Kindly check the clarification points and suggest the solution. If the DRV103 is not capable, please suggest the new alternate IC for our existing part (DRV101) used in product currently.

1. Initial condition:

Yellow: Status Flag

Green: Output

Blue: Input

Pink: NC

2. Ton is Higher: (theoretical: 28.9uS for 30%DC, 24KHz)

3. No PWM output, Continuous ON

4. Driver IC tested separately and output is working with ringings:

5. Temperature Data:

Gerber:

Top:

Inner layer:1

Inner Layer-2

Bottom

Clarification Required:

1. We like to know whether the DRV103H is capable to support 24V/1.5A(continuous) to load the Solenoid coil.

2. The power Pad can be connected to Ground signal. (In our current design, we isolated the power pad as per datasheet suggested)

3. The copper thickness mentioned in the PCB details need to be increased.

4. Any suggestion on the layout routing.

5.330nF capacitor is connected in the main PCB (The above testing are performed by assembling the delay cap near to the delay pin soldered)

6. Why the IC is heating when PWM signal output is ON.

7. The above IC parameters are set as per the existing DRV101 IC design. Is it required to change any parameters?

8. The Ton+Toff time is correct, but why Ton time is not matched with theoretical data.

9. Why ringing is produced in the PWM signal. Refer the above waveforms marking.

  • 0
    TI__Genius 16145 points

    Hello, 

    Thank you for your very detailed explanation of your implementation and test setup. 

    Please allow me a day to review this information and get back to you. 

    Best Regards, 

    Elizabeth 

  • 0 in reply to Elizabeth Higgins
    TI__Genius 16145 points

    One clarification on your schematic is that with a 330nF capacitor, we would expect a 300ms delay time. 

  • 0 in reply to Elizabeth Higgins
    Prodigy 80 points

    Hi Elizabeth,

    Good day...

    Yes, we will have 300mSec of delay time to have PWM signal

  • 0 in reply to kirubakar Nagarajan
    TI__Genius 16145 points

    Hello, 

    I do not see any issues with your implementation. Could you probe the supply itself to make sure it is staying high after 30 seconds?

    Could you also clarify what resistor pull-up you have on fault and what voltage it is being pulled up to? 5V?

    Allow me to answer your questions I can before more information is provided:

    1. The DRV103H is capable of supporting 24V/1.5A(continuous) 
    2. We recommend to not use the exposed metal pad as a power ground connection as not current should be flowing through the pad. Instead please solder the pad directly to the circuit board. Additionally, with inductive loads we recommend a 22uF capacitor on Vs rather than a 1uF capacitor                                                                                                                  
    3. Allow me to follow up with our internal team about that
    4. thermal vias underneath the power pad will assist in thermal dissipation
    5. I am not sure what you mean by this. Could you provide a picture? Is the cap floating against the pin and red-wired to ground?
    6. Are you saying at this point below the device is heating >85C? Could you elaborate on your measurement technique of acquiring that temperature?

    Best Regards, 

    Elizabeth 

  • 0 in reply to Elizabeth Higgins
    Prodigy 80 points

    Hi Elizabeth,

    Thanks for your response...

    For the point-3, i missed attach the PCB details whereas i provided below for your reference. Kindly check and let me know if there is any issues in Bare PCB.

    Parameter

    Value (No deviation)

    Initial copper

    305g/m2

    Inner Layers

    305g/m2

    Final surface copper

    915g/m2

    PCB Dimension

    11 x 31mm

    Layer

    4

    Finish

    ENIG (RoHS)

    Thickness

    3.2mm

    Drill Size (Min)

    0.3mm

    Point-5

    In actual condition, the Driver IC is assembled in the daughter PCB and the daughter PCB is mounted in the main PCB. Due to space constrain, the Delay Cap is placed in Main PCB. Later while testing, the delay Capacitor is assembled adjacent to the Delay pin of the IC to be as closer.

    Further the delay cap needs to be assembled near to Delay pin as below?

    6. We need a solution to have PWM signal output (With desired set parameters as mentioned below) without any heat dissipation from Driver IC. Now we could be able to see more heat dissipation from IC more than 85Deg C (at 25Deg C) which is not normal scenario. Our product operating temperature is up to 85Deg C.

    So if the IC heats more than 85Deg C at product ambient temp of 85Deg C. The IC will turn ON into thermal shutdown. We required minimal heat dissipation from IC.

    Kindly required TI Support to provide the best solution in high priority...

    Driver IC tested separately, and output is working with ringings pulse falling edge.

  • 0 in reply to kirubakar Nagarajan
    TI__Genius 16145 points

    Hello, 

    Thank you for providing that additional information.

    Again I do not see any issues with your implementation; Could you probe the supply itself to make sure it is staying high after 30 seconds? Could you also probe the output current when the device is unloaded and experiencing an unexpected increase in current?

    Could you also clarify what resistor pull-up you have on fault and what voltage it is being pulled up to? 5V?

    Is this failure occurring across multiple devices and boards?

    With regard to the delay cap, we would recommend proximity to the pin. The further away the capacitor is, the more parasitic inductance will be added. 

    Best Regards, 

    Elizabeth 

  • 0 in reply to Elizabeth Higgins
    Prodigy 80 points

    Hi Elizabeth,

    Sorry for the late response...

    Could you probe the supply itself to make sure it is staying high after 30 seconds?

    Kirubakar: the power supply (24V DC) to the IC was constant till we power it OFF.

    Could you also probe the output current when the device is unloaded and experiencing an unexpected increase in current?

    Kirubakar: No increase in current observed during no load condition. 205mA load current observed in PWM mode while load is connected to the output pin

    Could you also clarify what resistor pull-up you have on fault and what voltage it is being pulled up to? 5V?

    Kirubakar: pull-up resistor at which pin you're asking? Status flag pin? If Yes, the status flag is connected to GPIO and the pin outputs the 5V approx which is connected to the GPIO of the uC.

    Is this failure occurring across multiple devices and boards?

    Kirubakar: Yes, Tested in both DRV103U & DRV103H. Both results to over heat >85Deg C

    With regard to the delay cap, we would recommend proximity to the pin. The further away the capacitor is, the more parasitic inductance will be added. 

    Kirubakar: Delay Cap is connected very closer to the Delay pin of the IC

    Also input supply pin cap also increased to 22uF as per your recommendation but still the IC is heating.

  • 0 in reply to kirubakar Nagarajan
    TI__Genius 13300 points

    Hello Kirubakar,

    Elizabeth will get back to you by 1/4/23 at the latest due to the holiday break. Thanks for your patience.

    Regards,

    Kalin Burnside

  • 0 in reply to Kalin Burnside
    TI__Genius 16145 points

    Hi Kirubakar, 

    I do not see anything wrong with your schematic or layout. 

    Unfortunately due to the age of this device, we really only have the information in the datasheet to go off of. 

    I did come across another e2e thread which mentioned overheating using the DRV103. This issue turned out to be a RC network in series with the flyback diode. Is it possible this is occurring somewhere in your schematic?

    https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1045776/drv103-unexpected-heat-related-pwm-behavior

    With regard to alternate devices, 

    Our group now focuses more on integrated Power Switches(integrated FET, DC devices) and so our driver portfolio can be found in a different place. In Power Switches, DRV103 is the most up to date driver device we offer. You can find more recent offerings of solenoid drivers here:

    https://www.ti.com/motor-drivers/solenoid/overview.html

    We also offer High Power Drivers here:

    https://www.ti.com/power-management/gate-drivers/overview.html 

    If you are interested in any of these devices, please create a new thread with one of these devices tagged so the responsible engineer and group will be able to help you choose the right part.

    Best Regards, 

    Elizabeth