Custom Shield - Indoor Air Quality Monitor


#201

@mohit: thanks. I’ve choosen a MOS too, FYI I think that I’ll use the NXP BSS138P, found in some MICS reference designs.
I’m finalizing the layout of my board, hopefully soon I’ll be able to put everything in a dedicated thread on this forum. About that, one suggestion - if possible :smile:
I decided to lay down the DHT as in your PCB, but I don’t want to have it hanging out of the board, so I’m considering to use the airgap below the DHT sensor.
Do you have any suggestion/reference about how to “shape” it? Because i googled a bit about that but without any success.
What I have in mind is a “broken” ring, or is it enough just a wide “hole” under the whole sensor footprint?

@Rockvole : the SR6 socket is very nice, thanks to your suggestion I’m also thinking to include it in my design, also for helping with the heat.
But of course it is not so easy to find (as well as the MQ socket, which now is out of stock at seeedstudio, the only place where I found it).
About that, do you know any alternative sources for these 2 sockets?


#202

@cerocca

The idea for the air gap came from the Smart Citizen project. If you look at Google images for “smart citizen” and look carefully at the images which come up you can see there are a couple of variations of the air gap in different board versions.

The MQ sockets seem to be available on aliexpress, just search for “mq socket” although I have never ordered any.

The Figaro sockets I have only found direct from Figaro. Email : figarousa@figarosensor.com


#203

Really nice work you are doing here. I am currently working on similar project and I am also using some of the sensors you used. I am currently working on PCB design and need to know values (resistance, current) for sensors, which are not listed in datasheet, so i would be able to calculate the right power supply i need. Can you provide me sensors information about current, resistace?
I look forward to see your final product.


#204

@mohit - here is a photo of the Figaro SR6 socket which I just got in the post.

They are nicer to use than the SR7 since you don’t need to trim the sensor wires and they fit very easily with no force.
I think at 5X the price they are better for small test rigs.


#205

Those looks really neat. I’m hoping to find an equivalent for the socket that is locally sourced here in China to make manufacturing easier. China is currently shut down for the new year holidays (it literally turns into a ghost town!); be back in business beginning for March.


#206

Hi @mohit
sorry for going back to this post, but would it be possible for you to share how you implemented the “Unity gain buffer on the MQ and TGS lines” ?
I am finishing defining my schematic and I also have - obviously - the same problem, so it would be wonderful to have a reference.

Another note: in the github version of the schematic of your board the TGS is powered at 5V on the Heater and 3v3 on the sensing resistor.
Shouldn’t be both at 5V?

thanks in advance for your feedback!


#207

@mohit It would be genius if you can find something standard that these will fit in. They seem to be an identical form factor to the WSP2110 (grove formaldehyde sensor I have been using) since you can plug the WSP2110 into the SR6 & SR7 sockets.

Perhaps something like a 2-pin double row header but with slightly wider spacing would work.


#208

@mohit - in my V1 shield the DHT22 doesnt work so I plugged a DHT22 grove sensor in the grove socket. I have not managed to get correct readings from it - I tried another and get the same values.

On a breadboard I tried connecting a DHT22 up to a 5V USB header but it didn’t work. As soon as I connected it to the Spark 3.3V line it worked. It would be nice to have the option to use a Grove DHT22 in-case a temperature sensor external to the enclosure is needed. Is there a solution on your Grove shield or am I just doing something wrong ?


#209

@Rockvole, when it was connected to 5V, was the pull-up also connected to 5V? The other thing to remember is to use a 5v tolerant pin. :smile:


#210

@peekay123 - I realized I forgot to change my firmware to the grove DHT22 port pin. My original Grove DHT22 didn’t work but my new one does :blush:

I chose port D2 for the DHT22 on my breadboard because I was trying to keep to the original IAQ shield ports and so my firmware can stay the same. I looked it up and I think that is the only digital pin which is not 5V tolerant.

  • If I chose any other port for my breadboard would it have worked when it was 5V ?

#211

@Rockvole, the DHT22 will work at 5v on a 5v rated pin but in my experience gives different results than 3.3v. However, this is not based a any decent analysis though. :stuck_out_tongue:


#212

@anthonywebb it seems that :
airboxlab -> alima -> foobot

It will be interesting to see how the new one performs. The CTO said he will send me one in a few weeks.
I was worried these guys had given up, but it seems they are still working hard.


#213

@Rockvole please keep us posted on how it performs. I am still really keen on the idea!


#214

@mohit - I have been playing with a USB header and a breadboard with the sensors on. It seems that low quality USB power supplies make the sensors produce poor results - particularly the dust sensor just produces garbage. An Apple power supply fixed the problem - something to keep in mind, I don’t know if there is a way to filter the power supply, but if there is it would save people $24 on the apple PSU.


#215

@Rockvole
Apple sets high standard like always! I have successfully used other USB chargers from makers like Samsung, LG, Amazon, etc. they all work like a charm.

Can you provide me your latest findings regarding the shield, what works/what doesnt? I’m still looking out for sr6/7 socket equivalents. Now that China is back online, things should move faster.

Here is a recent take at the routing:

Next up is labelling and your suggestions. There is going to be a milled out area near the DHT22 as well.

Here are the design files: https://github.com/spark/custom-shields/tree/master/indoor-air-quality-shield

@cerocca
The Core has a max of 3V3 on its analog inputs and hence we needed to run the sense resistor at 3V3. Now that we are using a buffer, it runs at 5V.


#216

while you were asleep
I’ve updated the design with all the labels. Awaiting feedback!


#217

@mohit - thank-you, I will take a look and let you know if I see anything.
Can you number the Grove Ports? - its helpful when discussing where to plug things in.

@peekay123 @epierre - can you take a look if you have time ? I struggle with understanding Eagle.


#218

@mohit - it looks fine to me.


#219

Great, I’m going to spin up these boards and get back to you!


#220

Hi @mohit - any news on the boards ?