• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 1349 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.

TPS1H100-Q1: Inductive-Load Switching-Off Clamp and device protection

Oliver Filsch
Intellectual 430 points
Part Number: TPS1H100-Q1
Other Parts Discussed in Thread: TPS1H000-Q1

Hello,

in an actual application this device is used to drive a DC contactor (Gigavac HX241) with 24V coil (dual coil design with economizer, the first bigger coil is internally disconnected after pick-up).

This contactor has an integrated coil suppression. External coil suppression is not recommended and is not implemented, because this would slow the release time wich could reduce life cycles and lead to contact welding.

  • The contactor's back EMV is 55V according datasheet for the 24V coil version.

  • The TPS1H100-Q1 has a drain-source clamp of appr. VDS,clamp=58V at room temperature according Fig. 14 in datasheet.

Actual measurement with Vin=24V shows on Ch3 the coil voltage referred to GND when switching the coil off.

So the voltage across Drain-Source of TPS1H100-Q1 is 36V + 24V = 60V (@ 25°C).

The waveform shows that the voltage is clamped by the TPS1H100-Q1, right?

How can I estimate if the IC can handle the switch-off energy, or if it needs external drain-source protection?

Unfortunately, there is no information about the coil's inductance(s).

However, the hold current after pick-up ist 90mA @ 24V continuous.

Regards, Oliver

  • 0
    TI__Mastermind 34965 points

    Hello Oliver,

    The key parts that you need to know here on whether the integrated VDS clamp can handle the inductive turn-off are:

    1. The steady state load current going through the inductor (Vsupply / Rload)
    2. The inductance value of the coil

    Once you know these two numbers you can simply refer to the following graph in the datasheet:

    Without the inductance value of the coil it is not possible to determine if the integrated VDS clamp is capable of handling the turn-off. I would recommend contacting the manufacturer of the coil for this information.

  • 0 in reply to Timothy Logan
    Intellectual 430 points

    Hello Timothy,

    thanks for your reply. I'm in contact with the relay manufacturer regarding the inductance value of the coil, but until now I did not receive the information.

    I measured the inductance in the relay off state: appr. 600mH.

    The inductance in the relay on state might be much higher, for example omron ly datasheet lists doubled inductance values with armature on.

    • The diagram is valid for VVS=13.5V ("TI provides the upper limit of single-pulse energy that devices can tolerate under the test condition: VVS = 13.5 V, inductance from 0.1 mH to 400 mH…")
      Can it also be applied for other supply voltages, like 24V in this case?

    • Since the diagram stops at max. 400mH, what would be your suggestions if we had, lets say 1200mH of inductance? The current in the application (90...100mA @ 24V) is much lower as the min. value in the curve.

    Regards, Oliver

  • 0 in reply to Oliver Filsch
    TI__Mastermind 34965 points

    Oliver,

    It can be applied to the operating range of the device, however at 600mH which is out of the range of what we measured I can say that you will need an external clamping solution like a freewheeling or flyback configuration. The attached presentation (slide 9) does a good job of explaining how to do this with an external diode. 

    2019_FAE summit_Driving inductive loads using power switches _cameron.pptx

  • 0 in reply to Timothy Logan
    Intellectual 430 points

    Hello Timothy,

    thank you very much for the attached presentation, it helps a lot for understanding.

    The presentation uses following formula to estimate the energy in the switch during demagnetization.

    If we assume Vclamp=55V, Io=0.09A, Vbat=24V, L=1.2H, then Ed=8.6mJ, wich is far below 70mJ (max. inductive load switch-off energy of TPS1H100-Q1).

    If I understand you correctly, this calculation is still not applicable, since the inductance is out of the range of what was measured at TI?

    If so, I will use external zener clamp across the load.

  • 0 in reply to Oliver Filsch
    TI__Mastermind 34965 points

    Oliver,

    Is the steady state load current of this device only 90mA? This seems very small for the TPS1H100-Q1 which is rated for up to 4A continuous load current.  If your load current is 90mA I would also recommend the TPS1H000-Q1 as it will be a better price point.

    If you are only running 90mA of load current when the inductor is fully charged then the internal VDS clamp would work in this case. 

  • 0 in reply to Timothy Logan
    Intellectual 430 points

    Hello,

    the used contactor switches on with up to 2A using two coils, then after approx. 100ms one coil is electronically removed by the contactor-internal drive circuit.

    You're right, the remaining coil draws approx. 90mA DC in steady state  @ 24V DC.

    Another reason the TPS1H100-Q1 was chosen is that we use the current sense output for diagnostics.

    Regards, Oliver

  • 0 in reply to Oliver Filsch
    TI__Mastermind 34965 points

    Oliver,

    If at the point of turn-off there is only a charged inductor of 90mA you should be fine without an external clamping circuit. We do not have anything specified in the datasheet as normally the device is driving a higher steady state current. A better idea here might be to just put a freewheeling diode in parallel with your inductor to ground. The voltage drop of the diode is going to be significantly less than the clamping voltage of the integrated VDS clamp and it will take longer for the charged energy to dissipate, however at 90mA it really should not take that long. This would be the safe approach as you mentioned we have not tested anything within your loading conditions. While I am sure it would work- it is not guaranteed by datasheet parameters.