Are MT10 sensors calibrated?

[PhilipDAth]

@Saralyn, are the MT10 sensors calibrated to meet their specifications (it says accurate to within 0.3 degrees)?

I've been testing it in our company fridge.  First, it said the temperature was 10 degrees celsius (unsafe) - which I could believe.  So I turned the fridge to the lowest temperature and left it overnight.

It reported the temperature dropped to -3.5 degrees celsius and remained that way overnight.

I would have expected the milk in the fridge to freeze at that temperature overnight, yet it didn't.  Maybe it needed longer.

[jbright]

The specifications for the MT10 also show that it is only rated down to 0 degrees celsius. 

Not sure how it would handle or be accurate below that temperature.

Maybe someone from the Meraki product team can chime in on this?

[PhilipDAth]

The more I think about it - the more unlikely I think that the fridge would be below zero degrees - because none of the fluids have frozen.

I'm going to warm the MT10 up in the office and then put it back in the fridge and see what it then reports.

[cmr]

A fridge should be at 3-5C so if the MT10 supports down to 0C then it should report that.  In another discussion it was asked if the MT10 could be used in fridges and freezers and the answer was no, it was not designed for that and there would be new devices to support those scenarios.  I'd think it might be calibrated more for room temperature?

[PhilipDAth]

>I'd think it might be calibrated more for room temperature?

I expect it to be calibrated to the advertised specifications ...

[cmr]

Purely from the spec I'd agree with you, but as it was mentioned that new devices were coming for fridges and freezers, I think the spec may be "optimised" shortly...!  I'm getting an MT10 shortly and will test it in a fridge with a known temperature to see if I get the same errors.

[BlakeRichardson]

Could just be a bad batch @PhilipDAth  Have you tested more than 1 device?

[PhilipDAth]

No.  I just have the one.  I'll give it a bit longer and then raise a support case if it doesn't improve.

[PhilipDAth]

I have the MT10 delivering believable temperature results (warmed it up for a while and then put back into the fridge).

The reported humidity is hard to believe as well.  It is showing 30% swing in a cyclical pattern every 10 minutes.  Most of the time the fridge is closed and not opened.

Do fridges really have humidity variations this great?

[antonis_sp]

Cool that we are discussing fridges here I guess...🆒

While looking for graphs from other sensors in fridges I saw similar results.

So problably the 10min period is the fridge's cooling cycle...

[PhilipDAth]

After another day the reading has settled down for the MT10 in our company fridge.  I also checked the temperature with an infrared gun and got a reading of 2.5 degrees, within half a degree of what the MT10 is reporting.  So this seems reasonably likely to now be correct.

[KrisCarter]

Back to your original question, I'm not sure that we can utilize these sensors in quality environments without evidence of calibration (and scheduled calibration cycle).

[antonis_sp]

I agree. I can only compare the MT10 sensor to other enviroment monitors I have (consumer ones).

It seems that the humidity at least is largely affected by small changes in temperature.

This is with a balcony door briefly opened in a room.

Also as far as accuracy goes, temperature matches other sensors, but four humidity sensors in the same room, all show differnet values (most within +/- 5 of each other).

None of my other sensors though is as quick to register changes as the MT10, which is good I guess?

[BlakeRichardson]

Given there is a pattern with the humidity that matches the temp I would say the humidity changes everytime the compressor turns on.