• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 41 subscribers
  • Views 333 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.

TIDA-010941: Sensitivity for the reference design TIDA-010941?

Per Ellefsen
Prodigy 70 points
Part Number: TIDA-010941

As I understand, the reference design has been tested according to UL217 and is compliant with the sensitivity requirements from this standard.

I am working with a project where the sensitivity requirement is 0,001dB/m, averaged over a period of 5 seconds.

Will the reference design fulfill this requirement? As seen from the design guide and the test-report, the obscuration is measured in "%/ft".

I am not familiar with the "%/ft" to "dB/m" convertion.

If the reference design doesn't fulfill this requirement, are there any changes we may do to improve the sensitivity?

BR Per

  • 0
    TI__Prodigy 640 points

    Thank you for the inquiry.  I am not familiar with the dB/m unit for alarm threshold.  We will research this and get back to you as soon as possible.

    In the meantime, what is the application requiring the 0.001dB/m sensitivity?  Is this requirement coming from a particular standard?

    Regards,

    David

  • 0 in reply to David Stout
    Prodigy 70 points

    Thank you for your answer David, the requirement is coming from our customer. The product we are working with is for industrial use,

    where the smoke detector is mounted in the end of a long pipe. Will check out this requirement some more. Looking forward to hearing from you!

    BR Per

  • 0 in reply to Per Ellefsen
    TI__Prodigy 640 points

    Hi Per,

    The dB/m unit is Optical Density per meter expressed in dB which is expressed as OD(dB) = -10*log (T), where T is the transmittivity of light through the air measured at a given distance.  Or written a different way, T = 10^(-OD(dB)/10).  In order to change from dB/m to dB/ft, the dB/m value is simply divided by 3.28 ft/m.

    Obscuration in %/ft or $/m depending on the unit of measure for distance is expressed as Ou = (1-T)*100.  This establishes the relation between Obscuration and Optical Density in dB.

    Using the above, converting 0.001dB/m to %/ft is as follows:  Ou(%/ft) = (1 - 10^(-.001/10/3.28))*100 = 0.007%/ft

    Based on your description of the application, this will be for an aspirating smoke detector or some sort.  These work by pulling in air through pipes that are distributed though the sensing area which are typically quite large.  This type of detector has very stringent sensitivity requirements (as described above).  These detectors generally use a sealed and calibrated laser-based module for the optical design and are powered from DC line voltages instead of batteries.  These are two major differences compared to the application this reference design is targeting.

    For this reference design, the variation in threshold at 1%/ft was 5 to 20 LSB at the ADC.  Assuming a linear improvement in variation, a 0.007%/ft sensitivity would need at most 0.035 to 0.14 LSB.  At present the noise floor of this system with 300 pulses is 0.13 LSB.  Currently, the system noise is dominated by the TIA amplifier stage followed by the ADC quantization noise.

    Changing to the laser based enclosed optical module would provide a larger source signal with lower noise and near perfect non-interference from ambient light.  Changing the amplifier to a precision version with ultra-low noise (also the larger source signal would mean smaller TIA gain resistance and therefore lower noise contribution) should be able to make the ADC noise floor the dominant source.  For a 5s average, we could increase the sample rate to something like 300 pulses every 200ms which would give 25 points to average.  In theory that would provide a 4-5x noise reduction.  These are examples of changes that would need to be made to this reference design to approach the required sensitivity level.  At this time we have no way of knowing how close these changes would be to meeting the requirement or if additional modifications would be needed.  Note also we have not measured any of these circuit changes described and is only based on the knowledge of the existing design and the theory of operation.

    Best Regards,

    David

  • 0 in reply to David Stout
    Prodigy 70 points

    Hi David,

    I also looked into the "Obscuration" and "Optical Density" definitions, and found the same formulas as you did.

    I understand that this sensitivity requirement of Optical Density at 0.001dB/m is quite hard.

    Where do I find information about "the laser based enclosed optical module" you mention?

    If I understand this correct, using this "laser based optical module" could be a solution,

    and another solution could be to change to a better op.amp, and then also increase the sampling rate (this would of course also increase the current consumption). Correct?

  • 0 in reply to David Stout
    Prodigy 70 points

    Hi again David,

    Read some more in your Design Guide, and by "the laser based enclosed optical module" you probably refer to the "laser-based PM sensor" which is used as as a reference during testing??

    If this is the case, we probably will go for a solution where we try to improve sensitivity level of your ref. design as you describe in your answer.

    Is it possible to buy an evaluation board of your design?

    BR Per