In my opinion your burning leaves smell would be picked up by the Grove Dust sensor. I think the sensor which has the best chance of picking up the Garbage smell (out of the sensors we are using in this IAQ shield) would be the Figaro TGS2602 hydrogen sulfide sensor.
I have not yet played with any of the shields precise components yet, but I have a dust sensor and it does go high from smoke. I have a formaldehyde detector and I know when that gets high readings. I had a professional CO detector a few years ago.
I still hope that Spark will produce a run of these shields, but I expect it will come down to the number of people expressing interest in them. Also, I dont think the run will be imminent, I think Spark are busy with office moves and the recent funding
So it turns out I was wrong about the Alima/Airboxlab and it does have a CC3000 in it - so in future it should allow wifi passwords entered by smartphone and automatic firmware updates.
The beta testers are having to update their devices over USB with the latest firmware to fix the CC3000 issues right now. My device was bricked due to that update - so Im waiting on another device being sent to me.
Hi @kennethlimcp - it looks like Mohit is busy, so I thought I would ask you since you reviewed the board. Do you know if there is a way to cut power to the shield sensors from the spark core ? I donāt think there is - but Iām still at the āHello Worldā stage with hardware and firmware.
Thanks.
I had a suggestion for a MICS-5525 replacement from a friend who is experimenting with a lot of air quality sensors.
He suggested the MICS-4514 since this sensor is designed to detect pollution from car exhausts - this would be an excellent addition to the shield. It is the sensor chosen for the Smart Citizen Project :
@mohit - do you think we could use this as a replacement ? I cant find any supplier where it can be purchased.
I just started with a Sparks core so bare with me if Iām incorrect. I have a question regarding the reading of the mics sensors, as theyāre connected to +5V, the MICS signal might very well be above 3.3V depending on the pot. However, the analogs on the Spark core are limited to 3.3V?! not sure if Iām mistakenā¦
Thereās no level shifting. If you know that your gas concentration levels stay in certain ranges, you might very well turn the pot to a level, such that the output doesnāt exceed 3.3V. At widely varying concentrations you would have to readjust the pot for every level with some kind of feedback loop.
But I would still like to know how the shield was designed as my personal preference is to make sure thereās never a chance for the sensor to output more than 3.3v and damage the IO pin.
I have the same concern. Probably it should be redesign/extended such that itās limited by the hardware implementation rather than by turning a āknobā.
@Perrot, you could use a fixed gain inverting op-amp followed by a unity gain inverting buffer, both powered by the Sparkās 3.3v supply. Many op-amps will tolerate inputs higher than their power supply. Under those conditions, the output of the op-amp will simply clip to 3.3v. I believe Texas Instruments also has programmable gain op-amps that can be setup via I2C. You would then use one of those with some logic to create an auto-gain for the best range. Bottom line, create an analog front end to āmassageā you signal prior to inputting to the Spark.
Sounds reasonable. Iāll be testing a MICS 4514 with 3 resistors (refined readings at different ppms) instead of the pot and use the op amps to have a reasonable gain (<3.3/5) protecting the spark core
Hi @Perrot
Thanks for spotting that - its unlikely I would have noticed a problem since there is no MICS chip on my board.
Did you manage to find a supplier for the MICS-4514 ? The sgx cdistore dont seem to be interested in responding to requests for single chips.
