I have 2000ppm model, this the one that should have best precision for general occupancy application (at home in my case).
The vents can remain closed as long as the air to be monitored pass through the membrane on top of the unit. Those vents are used for application where the air is pumped remotely through the unit. The membrane is permeable to the air and protect the sensor against dust and large particules.
I use z-uno (z-wave) microcontroller that has 3.3V I/O so I had to use a level shifter. The level shifter I purchased (Iduino ST1167) was not right (all resistor the same) so I had to weld resistor in parallel to have proper voltage divider (3.3V/5V).
I tested the T6615 unit with the original connector but removed it and then welded a header so it could just be plugged in the breadboard.
That depends on the specific board model you use, check the datasheet. In my case it was required by the z-uno.
Always best to stay within spec even if it could work outside of it. The overall system remain more robust in all operation cases. If you are willing to pay for a T6615… it’s not the level shifter that will break the bank
OK, I’ll order level shifters. Are there 100% working wiring instructions and code available for ESP32/ESPHome and T6615? I just need a proper CO2 sensor, nothing else.
I do have Airthings Wave Plus sensors but those seem to “calibrate” to false (too low) readings unless I regulary expose them to fresh air. Not sure how to do that, only outside air is fresh enough and there is now -6C…
That explains ESP INPUTS only, not the whole board, are 5V tolerant as per Espressif founder & CEO.
On that Facebook discussion and elsewhere on internet, we can find positive and negative feedback about using input at 5V. Some claim that communication drops sometimes while others saw degradation of hardware.
We need to consider also the other device connected which could not function well if running at different voltage than spec.
The above is applicable to ESP. For Z-uno, it’s different:
Z-Uno can be damaged by 5 V on any of its input/output pins. Use voltage divider for input pins to down voltage to 3.3 V. For example use 1 kΩ and 2 kΩ resistors. Output pins in most cases can be connected directly — most digital sensors interpret 3.3 V as high voltage. If you still need to convert 3.3 V to 5 V output, use transistor. Logic level converter can also be used.
Yes, that’s the one. It appears to be a mosfet-based design, so hopefully quite robust.
When I get it and put it all together I’ll let you know how things work. I’m quite interested in comparing it with my MH-Z19s, especially over the longer term.
I just got the T6615 eval kit and wired a quick test setup. Seems to be working with ESP32. I used a 4.7k resistor in the input pin. Need to check with scope what the voltage level is now.
This T6615 looks like a copy of the MH-Z14A I am using for years. It works with ESP Home using this library.
It works without an issue with D1 Mini using UART without doing anything with the voltage levels. So I can confirm what @chayerd posted above. ESP seems to be 5V tolerant.
Finally got around to testing the T6615, and FWIW just like @johny_mnemonic found, in the first iteration it appears to be working for me connected directly to the GPIO pins. This is an S3 dev kit.
UPDATE: confirmed also working with those bidirectional level converters.
Thx for all the info here…
I ended up with the sdc30 and it was easy to setup…
But it seems that it measures what it wants…
Unplugging it from my room locatio , bringing it outside, lkugging it in and calibrating it outside and then bringing it back inside seems to not work correct…
Staying around 400 outside after calibrating outside but when back in room it easily finds below 400 values…which would be a surprise if i have less co2 inside then outside…
Any ideas?
Nice to see all these DIY approaches on CO2 measuring devices.
For info: at the moment a zigbee device from Heiman “Smart Air Quality Monitor HS3AQ” with a NDIR sensor is sold relative cheap: https://www.heimantech.com/product/index.php?type=detail&id=106
You find them for 40-70 Euro in Europe (I leave the search to you: Look for “Heiman HS3AQ” ). Strangely enough an Ali-express link asks +200 Euros.
I ordered 2 to check them, compare measurement with other devices I have (will open 1 to check the hardware) and will report back
At the moment I have an Airthings View Plus and a Aranet4 (both 200-280 Euro). I like it that both work on batteries. Both integrate nicely with HA.
Airthings View Plus: only via the Airthings cloud. But you can put it everywhere in house as long as the device has wifi.
Aranet4: only Bluetooth/BLE; goes to about 6 meters from your HA bluetooth receiver. So in most homes cannot be moved to every place. Good that you don’t need an external cloud solution (happens that the Aithings cloud is offline).
The Heiman HS3AQ requires USB power source even if it has an internal rechargeble battery (which can run the device for 6 hours the specs say). Stupid, as Zigbee is low power;
Edit: no idea on the quality of the CO2 measurement…; some update: the Heiman HS3AQ without zigbee measurements was in line with the Airthings View Plus out of the box (range 500-800ppm). But as it was without Zigbee I sent it back…
Edit: the Heiman HS3AQ sold on Amazon is without Zigbee… even thought it has same product number as the on Heiman website. One does wonder…
