Shop OBEX P1 Docs P2 Docs Learn Events
GP2Y1010AU0F Dust Sensor — Parallax Forums

GP2Y1010AU0F Dust Sensor

paturner4paturner4 Posts: 4
edited 2012-11-14 09:49 in BASIC Stamp
I am working on a project that requires monitoring dust in air following though a tube. The Sharp
“GP2Y1010AU0F Dust Sensor” should work. I would like to use a BS2 which I have used for many projects, but the problem is that the Dust Senor requires a pulse cycle of 10ms with a pulse width of .32ms. My understanding is that the shortest BS2 pulseout width is 1ms. Is there a way around this or can I use a 555 timer to generate a pulse width of .32ms? Any ideas?

GP2Y1010AU0F Dust Sensor see: http://www.sharpsma.com/webfm_send/1488 (see page 4 and 5)

Thanks, Pat

Comments

  • PJAllenPJAllen Banned Posts: 5,065
    edited 2012-11-04 17:41
    paturner4 wrote: »
    My understanding is that the shortest BS2 pulseout width is 1ms.

    The BS2 can PULSOUT in 2 microsecond increments. For 320 usec that'd be PULSOUT pin, 160

    Does that device require a constant rep rate or just a one-shot?
  • paturner4paturner4 Posts: 4
    edited 2012-11-04 18:04
    Thanks vary much for your input.
    The instructions by one user was to pulse the sensor input for .32ms, check the output Voltage (.9 to 3.4 V) and then pause for total cycle time of 10ms.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2012-11-04 18:18
    "Check the output voltage" presents a challenge.
    You'll need an ADC IC or you can use RCTIME with the circuit in the following link
    http://emesystems.com/BS2rct.htm#SmallV
    The transistor used there isn't "widely" available, an ADC-based approach would be more easily implemented.
  • paturner4paturner4 Posts: 4
    edited 2012-11-04 18:43
    I was thinking of using RCTIME, But as you side “ADC-based approach would be more easily implemented”. I have never used a ADC, can help find the information I would need?
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2012-11-04 18:54
    Here's the device -
    http://www.parallax.com/StoreSearchResults/tabid/768/txtSearch/ADC/List/0/SortField/4/ProductID/245/Default.aspx

    Find some solid information on using it -
    http://www.parallax.com/Portals/0/Downloads/docs/books/edu/Web-BasicAnalogDigital-v1.4.pdf
    starting on its page 43
  • paturner4paturner4 Posts: 4
    edited 2012-11-04 19:07
    PJ,
    I will give the ADC a try
    Thanks again for all your help.
    Pat
  • ercoerco Posts: 20,257
    edited 2012-11-05 09:16
    This hard-to-locate PDF (bookmark-worthy, and/or save the PDF before it's deleted) is brief and will get you up & running with an ADC0831 and Sharp analog sensor in a hurry: http://www.parallax.com/dl/docs/prod/acc/SharpGP2D12Snrs.pdf
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2012-11-07 09:25
    The signal response accompanies the 320µs excitation pulse.
    Screen shot 2012-11-07 at 9.07.22 AM.png

    The analog to digital converter has to sample the response, either by holding the peak value, or by integrating the response above baseline. It can be done with a Stamp, but it does require external op-amp circuitry in addition to an analog to digital converter.

    You might also take a look at the Shinyei particle sensor, available here. There is a nice animation at the bottom of the Shinyei page that shows how it works. It depends on flow through the chamber and light scattered from particles as they move through a focused beam. Therefore the signal is moved up away from baseband and the sensor is more sensitive than the Sharp sensor and less affected by reflections and ambient light. The flow in the Shinyei is normally created by heat rising from a resistor, but you might be able to shunt some of your flow instead. There has been a lot of discussion of this sensor, links in this thread.
    Screen shot 2012-11-07 at 9.07.22 AM.png
    608 x 218 - 19K
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2012-11-07 10:35
    I was under the impression (mistaken) that it got a pulse and then locked an output voltage. Ow!
    Couldn't it just be taken and left HIGH, read it with the ADC and then send it LOW or does that sample time result error if on too long?
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2012-11-07 11:38
    The data sheet leaves a lot open to speculation. It comes down to their test circuit in figure 1:
    Screen shot 2012-11-07 at 11.18.40 AM.png

    Resistor values inside the device are not listed. A chart says that the "LED terminal current" is 10--20mA, but I believe that is referring to the drive current labeled ILED in the diagram, not to the actual drive to the LED. Typically, the IRLED in smoke detectors is driven at 200mA or more in short pulses. I suspect that is the case here, and the pulse of current comes from the external 220µF capacitor. The current is limited by the internal resistor, but we don't know the value. It would be easy enough to figure out with some test equipment.

    If the LED stays on beyond the allotted 320µs, the current will drop off due to the 150Ω resistor in series with the 5V supply and the signal will follow suit. Furthermore, the LED is not rated for continuous drive at high current and leaving it on would shorten its life. The data sheet says as much in the Notes section.

    We don't know the response of the receiver circuit either. It evidently has a pole that leads them to suggest 0.28ms as the optimal sampling time. It may also have a zero that causes the response to fall off after a time. Would have to be subjected to a bit of reverse engineering.
    Screen shot 2012-11-07 at 11.18.40 AM.png
    835 x 300 - 31K
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2012-11-07 11:39
    They do suggest that the sensor can distinguish dust from smoke "by pulse pattern of output voltage". That would involve the frequent sampling (they suggest 10ms between pulses) and analysis of variance. Dust exhibits larger variance than smoke. Home smoke detectors normally sample at a much lower rate, more like one sample per 2 seconds, unless a suspicious condition is detected.
  • ercoerco Posts: 20,257
    edited 2012-11-12 10:14
    Interesting thread. Here's a marginally related question for Tracy, PJ, and everyone else smarter than me:

    What is the only synthetic element to have found its way into the household?

    Try before you click the hidden white link below...


    http://en.wikipedia.org/wiki/Americium
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2012-11-12 22:54
    Erco, happy birthday!

    That would be americium, 241Am, found in ionization smoke detectors, 1µC, about one-quarter of a microgram. It decays to neptunium, and I thought that might also be synthetic, but Wiki says it does occur naturally in miniscule amounts in natural uranium deposits. You might say the decay product of a synthetic element is itself synthetic. Would that apply to the alpha particles (helium nucleii)? 37000 decays per second.
  • ercoerco Posts: 20,257
    edited 2012-11-14 09:49
    Ah yes, everything decays. Me too, a bit with each passing birthday. Ashes to ashes, dust to dust...

    Which eerily loops right back to the OP's dust sensor question. Everything's connected! Was Kevin Bacon mentioned yet?
Sign In or Register to comment.