Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Device stops sending data after a while
Thanks for the detailed analyse... i need some time to check in detail what can be happening
what i want to understand is - does this happen with this power supply only - that is - with others it behaves well, or in general it works like this - especially the CO2 reading not updating much ?

and also - did you add this property or not ?
prop_jset "event.setupEnd"##nop 0##
It seems to be specific to this power supply: using a different USB wall-wart power supply and running it directly off my laptop USB port it works fine, including the CO₂ updating regularly.

Even with this USB hub powered from the mains but with the USB uplink lead plugged into the laptop and with the other devices mentioned plugged in it works fine (CO₂ updated regularly).

No, I didn't add the event.setupEnd property, just the LED colour properties detailed in Option 2 plus my attempt to have the LED turn red on Wifi disconnection.
I did some measurements as well.
In general if the power supply outputs more than 5.25v, the CO2 sensor does not work. I tried addinga schottky diode between the 5v of the board and 5v of the device, to drop it by 0.2 v, and it seems it started to work.
if you happen to have one of those or simillar, you can give it a try. if not i will send you tomorrow one, and will post some photos how to connect it.
Also the fact that the LED is not changing colors - this happens when the voltage is above 5.3v, then apparently this LED is not working well. I had added some diode to drop the voltage only for the LED, but it drops just 0.1v, so sometimes it is not enough
the sensor operates well at as low as 4.7v, so with the Diode included - it should be mostly fine.
I didn't really expect that that one is so picky to voltages, as all others i use work just fine. Apparently for the next revision, i will need to do something about this
(11-16-2016, 05:35 PM)Ed Davies Wrote: .

i just sent you a package with one diode, jumper wire and a Capacitor.
I will post later some photos how to attach the diode. 
The capacitor is necessary only if the power supply is very noisy, which i have only seen in cheap 1$ iphone style chinese power supplys and has to be soldered on the board
Thanks Vladimir. Capacitor, diode and lead arrived the other day (bit of a mystery prize as I hadn't looked on here for a few days!). I'll look forward to seeing how to connect it - it might be a day or two before I can give it a try though.
Any more on this?
ah.. i will send some photos tomorrow morning..i i did them but then i was too lazy copying them from the phone to the pc (got an iphone recently, and everything is so much complex than with android and google Sad(( ).
Hello Vladimir,

maybe could you send me the photos too how to add the diode and capacitor, or distribute here in the forum?

@ Ed,
maybe you could help me too here.

Thanks in advance,
Best regards,
(01-22-2017, 06:39 PM)joshi04 Wrote: Hello Vladimir,

maybe could you send me the photos too how to add the diode and capacitor, or distribute here in the forum?

@ Ed,
maybe you could help me too here.

Thanks in advance,
Best regards,
They were in the other thread Smile
Hi Guys,

i've done lots of improvements to the firmware in the last weeks. Some of them were related to the WiFi configuration, others to stability and performance.
Up to now the device was spending 10 seconds each iteration to scan where is hte i2c bus and what devices are there
In addition each 30 seconds it was trying to connect to the router and then killing the connection w/o sending anything most of the time.
in addition, once it boots and takes an IP via DHCP, then it doesn't get it again each 30 seconds, but it directly reuses it.
now it boots for 1-2 seconds and scans for hardware changes only after factory reset. i hope this will improve the stability, or if your devices are already working fine - keep it
but i would be glad if you could give it a try, just update to the latest firmware

most of this changes are needed, because the next part of my project is to have also a battery-operated board, and there each second of active runtime can decrease total runtime by a month Smile

Regards, Vladimir

Forum Jump:

Users browsing this thread: 1 Guest(s)