• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 49 subscribers
  • Views 203 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.

Original question:

INA381: Can an RC filter (1ms) be placed between Vout and CMPin

INA381: INA381

cha ens
Prodigy 80 points
Part Number: INA381
Other Parts Discussed in Thread: TINA-TI

Hi,

I marked the former question/issue as resolved bit then i thought of it again and I would like to reopen it.. I replied to the last answer, but may be it went missing somehow.

original post was called: " INA381: Can an RC filter (1ms) be placed between Vout and CMPin"

So i'll open it again:

after presenting the solution to a colleague, he also pointed out that there may be issues with the thresholds for the GPIOs (Vih / Vil may very between parts (same component). Also, this may be affected by temperature..

So I went over the datasheet again and found a section that i have missed:

"8.1.4 Input Filtering

If the INA381 output is connected to a high-impedance input, the device output is the best location to filter, using a simple RC network from VOUT to GND. Filtering at the output attenuates high-frequency disturbances in the common-mode voltage, differential input signal, and INA381 power-supply voltage. "

this brings me back to my original idea.. It seems to even be a recommended practice.

I tried reading the "Closed-Loop Analysis of Load-Induced Amplifier Stability Issues Using ZOUT", it is a bit complicated. I also tried to simulate this using the PSpice for TI but the part is missing from the database.

1) Actually, I'm adding an RC load with no real inductive elements. and as long as R is large enough, i don't see stability / ringing risks. Correct?

2) I thought of RC = 10K * 10nF = 100usec to start with. 

3) or, should i use a 1nF (max output load - No sustained oscillation according to the DS) and a 100K?

this will probably add an hysteresis of ~100mV (according to table 3)

Please advise.

Thank you,

Cha

  • 0
    TI__Mastermind 18780 points

    Cha,

    Section 8.1.4 outlines how to use an LPF filter on the output of the device to filter the output. I believe the disconnect here is that in the original post,  you want to feed this node directly into the CMP IN pin, which may cause issues with the hysteresis. If an RC filter is placed on VOUT and then fed into CMP IN, the filter will have implications on the hysteresis of the device, as pointed out in section 8.1.2. I don't think this is a dealbreaker, but it does need to be taken into account. 

    Thinking about it more, though, hysteresis may not be affected by a cap here, as the cap should be an open at DC, and the hysteresis is determined by a DC current source over the resistance, which should disregard the capacitance once the device comes up. I would advise you to have a look at this parameter and make sure that the added resistance allows you to operate inside your desired specs.

    Regarding simulation, if you wanted to look at this more in depth, you can find a model for the INA381 in TINA-TI here. 

    HD44780U