• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 30 subscribers
  • Views 1339 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.

OPT3001's performance under flicker conditions

henning larsen
Intellectual 335 points
Other Parts Discussed in Thread: OPT3001, OPT3001EVM

Some modern LED lamps have significant flicker amplitude at frequencies in the hundreds to the kHz range. How does the OPT3001 perform under such flicker conditions?

The issue is of course that unless the analog signal from the light to current sensor is properly anti-alias filtered before reaching the ADC, the measurement can be very wrong. Note that it may not be of any help to increase the integration time - the aliased-down signal cannot be removed digitally. Nor does it help to integrate in a multiple of the flicker period, because the period is unknown and varying, as it is not necessarily locked to the 50/60Hz grid.

The datasheet has no mention of an anti-alias filter nor a sampling frequency. Can anyone comment on these issues?

  • 0
    TI__Genius 14825 points
    Henning,
    We do see some effect from flicker at lower frequencies, under 1kHz. In my limited testing, I only saw up to about 10 lux of variation with an uncovered device. I also played with the duty cycle a bit and saw some effect there as well, but again, 10 or less. That said, it was me spinning the dial of a function generator, not a precision automated sweep, which seems to be in order to be able to actually say what the observed max variation is. Frequencies above 1khz did seem to average out, from what I recall.

    You can set up the experiment yourself with the opt3001evm as well. I'm out of the office on travel for the next week, but I'll see if I can get someone to set up something to check this out a bit more. Is there a particular range you're looking for?

    Thanks,
    Jason
  • 0 in reply to Jason Bridgmon
    Intellectual 335 points
    Hello Jason,
    Thanks for your answer.
    I have seen LED lamps modulated by 100Hz (rectified grid power) to few kHz. The challenge for light sensing is that the modulation depth can be high, because it is not limited by the LED bandwidth. High modulation depth operation it is not favorable for power efficiency and EMC reasons.
    The metric for sensitivity against source modulation should be in % not in Lux - should'nt it?
    br henning
  • 0 in reply to henning larsen
    TI__Genius 14825 points

    Henning,


    I had my team take some data at 100mS and 800mS integration time.  Legal-type disclaimer: This is one unit, one sweep each integration time, room temperature, one light source, so it's not a guarantee of other unit or future performance, but it shows the performance as we saw it this one time.  We set the frequency, then took 10 samples and checked the min/max of those samples, reported here.


    Let us know if we can help further.

  • 0 in reply to Jason Bridgmon
    Intellectual 335 points
    Thanks a lot for the detailed answer. It looks to work well even in high flicker conditions above the visual frequency range. May I suggest to add this very relevant information to the datasheet or at least an app-note?
    br henning
  • 0 in reply to henning larsen
    TI__Genius 14825 points
    Since we're not testing a statistically significant number of units over various supply and temperature conditions, we probably won't put it in the data sheet, but I will add it to the FAQ in this forum for future reference. Glad we could help!