The MH-Z19 in high co2 and humidity

Moderators: Voyager, BertB, grovkillen, Stuntteam, LisaM

Post Reply
Message
Author
JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

The MH-Z19 in high co2 and humidity

#1 Post by JarredAwesome » 18 Oct 2020, 17:59

Hey Everyone,

I’m looking for a sensor to help me determine how much co2 is dissolved in my aquarium. I haven’t used this sensor before, so I am hoping you guys could help me understand how it works.

My first and easiest question is, can I aim this into water to calculate dissolved co2? I’ve read about NdIr co2 that can do that, but I’m assuming it this one can’t. Am I correct?

Assuming it can’t measure it directly, the way I plan on doing it is by putting the sensor in a cup partially submerging it in water. This will allow me to measure the gas in the cup. In theory I can use Henry’s law to calculate the dissolved co2

So here are my questions about this method:

1. Will having constant high levels of co2 mess up the sensors automatic calibration? If it does, what is the best way to reset it? I saw on another post it was suggested letting it run in a low co2 room for 25 hours. Is that the best way to do it without a lab?

2. This will likely be high humidity, will that negatively affect the sensor?

Thanks

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#2 Post by TD-er » 18 Oct 2020, 21:04

What order of magnitude CO2 concentration do you expect to have?
The MH-Z19 is available in various versions. (and some also change the range via commands, but I would not advice it as it has no calibration data for the extended range)
The most common version has a range of 400 ... 2000 ppm, but there's also one for 400 ... 5000 ppm.

I know SenseAir has versions of the S8 that are meant for high CO2 concentrations of upto 50'000 ppm. (quite lethal if you are surrounded in air with such concentrations, but the air you exhale can be as high as roughly 40'000 ppm when leaving your mouth)

I know both the SenseAir S8 and the MH-Z19 is not meant to be used in close contact with water.
But I have no idea what may happen in "close to 100%" relative humidity.
The SenseAir S8 is using a mirror and I can assume condensation will mess up the readings.
So if the internals of the sensor are colder then the air temperature, then you will likely have condensation and thus bad readings.
Not sure how the internals of the MH-Z19 is constructed.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#3 Post by JarredAwesome » 18 Oct 2020, 23:26

My understanding is that the cup might get up to 20,000ppm,

But that’s by my calculations, which might be faulty. I’m very new to this stuff, I’m trying to figure this out. I have a ccs118, but the highest it ever got was 1995ppm

Is there a sensor you guys could recommend?

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#4 Post by TD-er » 19 Oct 2020, 00:31

Are you sure it isn't the CCS811?
That one isn't a true CO2 sensor, but a VOC sensor.
So it does measure some (completely different) other gasses concentrations and then tries to derive a value that could be used as an CO2 like concentration in specific circumstances.
The CCS811 does also need a 2nd sensor to compensate for temp and humidity.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#5 Post by JarredAwesome » 19 Oct 2020, 00:39

Sorry, I did mean the ccs811.

The breakout board I have includes a thermometer. It has a eco2 meter. For my needs, an estimation are good enough.

Is there another sensor you can recommend?

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#6 Post by TD-er » 19 Oct 2020, 15:22

Well like I said, I only know the SenseAir sensors and the MH-Z19
Both don't do well on high humidity environments.

Can't you use some construction with a tube to blow air through the water and collect the bubbles coming out of it in some column above the water?
After measurement you could then release the air and after some time repeat.
This way the air will be a lot less humid.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#7 Post by JarredAwesome » 19 Oct 2020, 15:42

I saw something simular to that used for beer manufacturing. I’m not opposed to the idea, but I’m more worried about how to calculate the results. If the cup’s rim is submerged around the water, the co2 inside the cup becomes proportional to the co2 in the water. If I’m simply collecting the air that raises above it, I’m getting a different resault

In the beer set up they use a moisture trap, particle filter, and hydrophobic cell to prevent moisture. It might be worth looking into.

I did come up with what I think it a better idea then what I originally posted. Instead of submerging the cup, I can simply put a sealed cup on the outside of the tank, and run a tube into the water. That would probably keep the humitiy down.

Can the NDIR read through glass? What if I put the sensor outide the cup, and have it read the air inside of it?

Here is a photo of my prototype. It’s a coke bottle for now, but once I figure out the design. I’ll make something out of glass or acrylic.
The bottle is in the outside of the tank, and the tubing leads from the coke bottle, to the tank.
The bottle is in the outside of the tank, and the tubing leads from the coke bottle, to the tank.
A59AA9BC-CCF5-41AA-A249-9131B420EB42.jpeg (3.1 MiB) Viewed 42450 times
I’m thinking I could mount the sensor on the outside like this:
The bottle is in the outside of the tank, and the tubing leads from the coke bottle, to the tank.
The bottle is in the outside of the tank, and the tubing leads from the coke bottle, to the tank.
A59AA9BC-CCF5-41AA-A249-9131B420EB42.jpeg (3.1 MiB) Viewed 42450 times
Attachments
The sd card case is a stand-in for the sensor
The sd card case is a stand-in for the sensor
0F83D23B-572B-4201-8582-A5E241A77ECD.jpeg (3.35 MiB) Viewed 42450 times

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#8 Post by TD-er » 19 Oct 2020, 18:42

An NDIR sensor is optical, but it cannot be used to measure through glass.
For example the SenseAir sensors use mirrors to increase the path of the light to make a more reliable measurement.

The working principle is to compare the absorption of a specific wavelength.
So you need to know the length of the path of light and the amount of light emitted.
There are more parameters affecting the measured absorption, like direct IR light shining on the sensor, like direct sunlight.

Humidity and air pressure also affect the readings.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#9 Post by JarredAwesome » 19 Oct 2020, 19:00

I could then in theory, make the container out of black acrylic, and just make a small hole for the ir to shine into. As I can silicone around it to make it air tight. Or, use some kind of rubber gasket.

Would that work?

I remember reading the hm-z19 compensates for temperature. Does the ir gather that information as well, or does the sensor have a thermometer in the casing?

Does the sensor automatically adjust for pressure and humidity? If so, does the sensor casing need to be in the same environment, or will having it pointed through a hole work?


Also, assuming my above statements work out, that brings me back to my original question. How to i calibrate it?

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#10 Post by TD-er » 19 Oct 2020, 20:58

The area for detecting the IR absorption is not available from the outside.
So I don't think it is matching your idea of how to implement it.
You can however just place the sensor in a large bottle which is mounted outside the tank.
As long as it remains dry it should work.

Also make sure to have the sensor outside the high CO2 environment every now and then to allow to re-calibrate it to 400 ppm as sensors may become saturated at some time (months).
When left in a closed container for a while, you must make sure to turn off the ABC function of the sensor.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#11 Post by JarredAwesome » 19 Oct 2020, 21:13

I’ll give it a try.

I’ll take precautions to make sure it doesn’t get wet.

How long is a while?

If I leave it running overnight in my living room, is the sufficient enough to keep it at 400ppm?

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#12 Post by TD-er » 19 Oct 2020, 21:19

For > 24h on the MH-Z19, make sure to turn off ABC.
The default ABC period of SenseAir sensors is 7 days.
In the ABC period the CO2 concentration should be as low as possible for at least a few measurement intervals or else the 400 ppm level will be off.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#13 Post by JarredAwesome » 19 Oct 2020, 21:55

TD-er wrote: 19 Oct 2020, 20:58 Also make sure to have the sensor outside the high CO2 environment every now and then to allow to re-calibrate it to 400 ppm as sensors may become saturated at some time (months).
When left in a closed container for a while, you must make sure to turn off the ABC function of the sensor.
This is what I am referring to.

I’m just confused by your response. It’s going to be in the container 24/7, unless I need to recalibrate it.

So do I always leave abc off?
Do I turn on abc when I take it out of the container?

How long do I let it run to ‘unsaturate’ the sensor?

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#14 Post by TD-er » 19 Oct 2020, 23:40

You need to understand the CO2 sensor will "degrade" over time.
Like I said before, the amount of IR absorption is measured, but for that you need to know how much IR was put in.
The light source can (and will) degrade over time.
Also the detector may become dirty and thus measure less IR signal.

Therefore after some time (aging speed may be different during the sensor life time) you need to perform an ABC calibration to get to know the 400 ppm level. (410 ppm is roughly the lowest concentration around the globe, increasing 2.5 ppm per year)

If you will not correct for the aging of the sensor, then your absolute values will be getting off.
This is of course not going to happen in a day or a week.
But keeping the ABC active in a closed container will for sure mess up the readings.
The MH-Z19 with ABC active will be messed up in 24h not seeing proper low CO2 concentrations.
SenseAir sensors with ABC active will be (a bit less) messed up in a week (7 days) not seeing proper low CO2 concentrations.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#15 Post by JarredAwesome » 20 Oct 2020, 01:02

TD-er wrote: 19 Oct 2020, 23:40 You need to understand the CO2 sensor will "degrade" over time.
Like I said before, the amount of IR absorption is measured, but for that you need to know how much IR was put in.
The light source can (and will) degrade over time.
Also the detector may become dirty and thus measure less IR signal.

Therefore after some time (aging speed may be different during the sensor life time) you need to perform an ABC calibration to get to know the 400 ppm level. (410 ppm is roughly the lowest concentration around the globe, increasing 2.5 ppm per year)

If you will not correct for the aging of the sensor, then your absolute values will be getting off.
This is of course not going to happen in a day or a week.
But keeping the ABC active in a closed container will for sure mess up the readings.
The MH-Z19 with ABC active will be messed up in 24h not seeing proper low CO2 concentrations.
SenseAir sensors with ABC active will be (a bit less) messed up in a week (7 days) not seeing proper low CO2 concentrations.

Ok, I think I’m following you. So I should be on if I do the following:

1. Put the entire sensor in the container, but make sure it can’t get wet.
2. Turn off ABC
3. When I see the values are no longer making sense, turn abc back on, and run it in a low co2 environment to recalibrate itself.

Is that correct?

If so, how long do I need to run them for to correct themself?

I plan on using the MHz-19

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#16 Post by TD-er » 20 Oct 2020, 01:12

Since those sensors do not switch back to "400 ppm" in 1 step, I guess you may need to run it for at least 2 times the ABC interval + some.
Another thing you can do is to think of a way to easily swap a unit, so you can let it run "outside" the container for a while on a 2nd node.
Just in case the sensor may be acting up a bit more than expected, or if you need to verify the working of the sensor with a fresh one.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#17 Post by JarredAwesome » 20 Oct 2020, 16:55

I am reading the instructions for this sensor here. Their ABL calibration takes 24hours. So before I put it in my container, I’m going to let it run for 48 hours in my kitchen (I don’t use gas). Then turn off the abc before I put it in the container. I will repeat the process when the results start looking a little unusual.

The instructions also mention a ‘command calibration’, is that relevant to what I am doing?

Also, it also says ‘9.2 When placed in small space, the space should be well ventilated, especially for diffusion window.’

Should I be concerned about this, or is it just because of the ‘saturation’ you mentioned before?


Thank you for all your help, btw. I really appreciate it

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#18 Post by TD-er » 20 Oct 2020, 20:14

The mounting instructions provided by SenseAir suggest to have a gap of at least 1 mm between the diffusion window and a flat surface.
So I guess it will be similar for the MH-Z19 sensor.

Do not try to use the calibration command, as you will not have a known concentration.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#19 Post by JarredAwesome » 23 Oct 2020, 17:25

So I have been running this sensor for about a day and a bit.

Just out on my counter. My co2 levels are showing between 1200-1600ppm. Is this normal?

Is it because it has a burn in time, or because it hasn’t set the abc properly yet?

That seems really high

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#20 Post by TD-er » 23 Oct 2020, 18:42

The MH-Z19 should be calibrated from the factory, so the initial readings should be spot-on.

The so called "burn in", if any, could be that it might need to have the ABC active more often when it is new.
But since it has factory calibration, I would expect it is showing accurate values when new.

Make sure it does not "see" IR light from sunlight.
If you don't live in a city or near industrial area of a busy road, then the CO2 value should rapidly go to below 500 ppm when you have enough ventilation in the room. (e.g. open window)

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#21 Post by JarredAwesome » 23 Oct 2020, 19:33

Ok. I’ll put a opaque cup over it, but leave some space on the bottom for ventilation, to make sure it’s not a light problem. And open a some windows.

However, it’s really high at night when the lights are off, so I don’t think it’s from the sun.

I’ll take a look

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#22 Post by TD-er » 23 Oct 2020, 20:02

Lots of plants in the room?
Plants take in CO2 during the day, but exhale CO2 during the night.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#23 Post by JarredAwesome » 23 Oct 2020, 20:24

There is a tonne of plants in the room.

I thought it might be that, but I also thought that if the plants where making a big difference, the day time levels would be lower.

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#24 Post by TD-er » 23 Oct 2020, 20:32

Well, depends also on the amount of (sun)light in the room and ventilation.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#25 Post by JarredAwesome » 23 Oct 2020, 22:24

Ok, so I just got home and the levels are actually realistic!

Atm it’s at 400ppm, and has been since 2pm

I’m not really sure what happened. When I look at the log, it has been going down little by little since at least yesterday.

The only thing I changed was yesterday. I figured the ABC might be turned off because the numbers where kind of weird. So I turned it on. (The script I’m using only lets me turn it on or off, it doesn’t let me check it’s status)

Could that have been the problem?

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#26 Post by TD-er » 23 Oct 2020, 22:26

Well it could be, but I wonder if it is now the true value.
You have to show charts to interpret it, to see if the ppm values are clipped at 400 ppm, or just 400 ppm when you open windows.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#27 Post by JarredAwesome » 23 Oct 2020, 22:34

I haven’t changed a thing. The windows are still closed. The environment is the same.

I was at work when I posed the question, when I got home it was at 400 ppm

TD-er
Core team member
Posts: 9360
Joined: 01 Sep 2017, 22:13
Location: the Netherlands
Contact:

Re: The MH-Z19 in high co2 and humidity

#28 Post by TD-er » 23 Oct 2020, 22:53

With enabled ABC, you know that in the end the reported concentration is at least once a day 400 ppm.
This is regardless the actual concentration.

For example, what if you have a constant concentration of 1000 ppm, the sensor with factory calibration will report 1000 ppm.
After 24h with ABC enabled, the reported concentration will be 400 ppm.
If you then open a window, the concentration remains 400 ppm.
Now a few people enter the room and the true concentration is 800 ppm, the reported concentration remains 400 ppm as the "true 1000 ppm" is set as being 400 ppm.

The ABC function was still active when the window was opened, so you can assume it is now "correct" when the new base calibration has been applied.
So after it was applied, you can see people entering and leaving the room again.


Now apply this to your situation, where there may not be a good ventilation and lots of plants in a room without lots of sunlight.
This is a room with unknown concentration.
It could be 400 - 1000 ppm during the day, but may also be 1000 - 1600 ppm.
However, with ABC enabled, whatever is measured as lowest value will be set as 400 ppm.
So in both situations, the measured value will range from roughly 400 - 1000 ppm (does not only shift, but also scales, but see this as a simplified example)

If you then have ventilated the room for a while, you may see the reported values "jump up" to something like 800 - 1400 ppm.
Well at least it is clear then you didn't ventilate the room enough, but with ABC on the values will be back to 400 - 1000 ppm the next day.

TL;DR
Just seeing values of 400 ppm with "ABC on" tells you absolutely nothing.

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#29 Post by JarredAwesome » 23 Oct 2020, 23:13

I understand what you’re saying.

I’m going to leave it the way it is for today, so I can see how it jumps up at night. Tomorrow morning I’m going to open the windows and see how much that changes.


One thing was different today actually. I wasn’t home. This is the first day I had to work since I set it up a little under two days ago.

However, it’s in a living room, and it fairly big, when you consider it has no doors in between it, and my hall and my kitchen

JarredAwesome
Normal user
Posts: 27
Joined: 18 Oct 2020, 17:40

Re: The MH-Z19 in high co2 and humidity

#30 Post by JarredAwesome » 25 Oct 2020, 04:30

I opened the window today, and the numbers are closer to what it was when I used it out of the box. So I guess the theory behind ABC is pretty good. I am getting less reading of EXACTLY 400ppm, so I am assuming your statement about it clipping was also right.

I am going to leave it for another day and turn off ABC.

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest