• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 582 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.

TPS27S100: High Side driver with accurate Current Sense, Digital Interface

PaulR
Prodigy 145 points
Part Number: TPS27S100
Other Parts Discussed in Thread: INA233, INA260, , INA226

I'm designing a control circuit that requires the load devices be monitored for failures (under current, short, broken wires, etc.), including failure while off (trickle current versus broken leads).

- Some of the load devices to be driven have a failure mode where they draw under 10mA, while normal operation is above 100mA (today). To cover future devices we aim to get a resolution and offset accuracy closer to 1mA (unlikely, but we can aim there).

- Other load devices may draw up to 2A in normal operation

- we need to control both high side and low side power, though low side is typically a simple FET and current monitoring is on High side for better protection in case ground fault.

To cover the variety we are developing a microprocessor based controller that will monitor the devices current, and compare that to stored settings for each device.

I'm considering TPS27S100B, and possibly devices like INA233 or INA260 for more accurate current measurement? (I have also looked at ST VND5T100 and Microchip PAC1720, but considering previous support from them am inclined to stay with TI).

What kind of accuracy and resolution (mA offset and mA steps) can I expect with TPS27S100B? (Please suggest circuits for 2A or even 4A limits).

I like the INA260 for digital interface and internal shunt, but the 5mA offset on measurements is higher than we hoped.

Is there a high side driver with digital current measurement, or a combination of ICs you would recommend?

We have a second project that will need a similar circuit, but also include PWM drive and possibly support loads of 4A to 10A.

Comments/suggestions appreciated.

Paul

  • 0
    Prodigy 145 points

    FYI: To be installed on ships, which have similar power to trucks (24VDC nominal, 28VDC when charging battery, noisy).

  • 0 in reply to PaulR
    TI__Mastermind 34915 points

    Paul,

    For the TPS27S100B at load currents of greater than 1A you are looking at a current sense/monitor accuracy of +/- 3% so this should meet your requirements without the need for an external current monitor. 

    Currently we do not have any high side power switch with a digital current limit device, however in combination with a microcontroller such as an MSP430 or CC2652 that has a decent performing ADC the analog component can usually be designed in with relative ease. In your case it seems that you'd probably want an ADC with at least 12-bits of resolution.

    For the higher load current project that you are considering we do have lower RON parts such as http://www.ti.com/lit/ds/symlink/tps1ha08-q1.pdf that can handle the the higher load current- however at higher voltage levels the short-circuit protection will not work. If you email me directly at timothy.logan@ti.com we can explore other options too. 

  • 0 in reply to Timothy Logan
    Prodigy 145 points

    Hi Timothy,

    The accuracy is needed for evaluating the failure mode of some of our load devices when they draw under 10mA. In the spec I see TPS27S100 accuracy is +-80% near 5mA, that means +-4mA is probably the best I can expect.

    Can this 80% at very low currents be improved? (Circuit changes, other parts, procedures, algorithms, calibration during production, etc.)

    The Low current accuracy is why I was considering adding something like INA233/INA226.

    Thank you for your email address, I'll use it when second project gets closer.

    Especially thank you for your quick response, your support is greatly appreciated, quick is just not something I've gotten from other companies. 

    Paul

  • 0 in reply to PaulR
    TI__Mastermind 34915 points

    Hello Paul,

    I understand your use case now. Unfortunately low load currents like those you are looking at worst accuracy numbers which are not really going to get better across devices in high side switches. I think the best case scenario if you accuracy at a load current that small is to use an external device for the current monitoring such as the INA233. Specifically if you had any questions on how to implement the INA233 in your design I would suggest asking in the amplifiers forum (https://e2e.ti.com/support/amplifiers/f/14)

    If you just care about being able to detect open load events there is functionality built into the TPS27S100 to account for that that involves using a pull-up resistor on the output.

    Could you tell me a little bit about the load are you driving? Will it have any significant inductances/capacitances or is it purely resistive?  Are you PWMing? Depending on your average power dissipation you might be able to get away with a higher RON part.

  • 0 in reply to Timothy Logan
    Prodigy 145 points

    Thank you Timothy,

    Understood. Real life does set real limits :)

    This is for Navigation Lights on ships, different manufacturers lights have different failure models and failure indicators, including "may fail soon". They are getting more creative with newer LED models so I'm just trying to go as far as I can to cover as many future light models as I can guess.

    I'll build some prototypes with TPS27S100B (and with optional INA233/INA226).

    I expect I'll need to do some testing with the different lights to resolve this.

    Paul

  • 0 in reply to PaulR
    TI__Mastermind 34915 points

    No problem! If you are driving bulbs I would recommend reading the following application note:

    http://www.ti.com/lit/an/slvae30a/slvae30a.pdf

    There is a section entirely devoted to driving bulbs and has some useful design insights. 

  • 0 in reply to Timothy Logan
    Prodigy 145 points

    Thank you, I was trying to find that one but I had the title a bit wrong

    Paul