• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 30 subscribers
  • Views 403 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM26: current sinking capability of OS_lowactive output

martin demke
Prodigy 10 points
Other Parts Discussed in Thread: LM26, LM26LV

We use LM26-CIM5-TPA in our design and we must have a time constant at the OS_lowactive pin, as this signal is used diode ored for several functions.

We pull up this output by 100k ohms to 4.2V, capacitve load should/must be 33nF.

That means the LM26 must sink the load of the 33nF directly, when going low active. I found on LM26lv, that this product can source 1.6mA@3.3V. That is enough for the application.

The only question is: can my device lm26-TPA be destroyed because of (Ic=C*dU/dt) ? I just can find on the datasheet a current limit of 5mA into any pin.

But if the internal FET sinks some mA in any case, that would be no problem, as the low active state only has to sink this current dynamically and not in DC operation.

Thanks Martin

  • 0
    TI__Expert 4905 points

    Hi Martin this my second reply to this. I even attached a drwing to the first. go figure...

    OK so the 5mA current rating is an abs max and is applies when you are exceeding the power rail mostly. You are correct though that no max current for the output is specified so if you want to be ultra conservative and reliable you could add a simple resistor as shown in the attached schematic and limit the current flowing into the output to 5mA. You will only see about 40 to 50 mV across the 1k ohm as shown if the pullup is 100k and the power supply voltage is 4.2V. So the logic high level should still be within the limits you are looking at, that is if there is not other pullup other than the 100k. If ther is another pullup then this is not a viable option.

    Also everythime the output switches without the 1k as suggested the current will be limited as the Ron of the output device is in the order of 300 ohms or so. But then the current will be three times higher and you will get some self heating of the device. thus a temperature error on the VTEMP output.

    I hope this helps!

    Take care,

    Emmy Denton

    Precision Systems Applications

    Silicon Valley Analog (formerly National Semiconductor)

    Texas Instruments