Datacenter (portable) thermal imaging

[Rand__]

We are considering getting a portable thermal camera in the office to check heat distribution in the data centers so I wondered whether anyone had some recommendations or critical criteria to look for.

Been reading up a bit, but still overwhelmed by choices and unknown buzzwords

Budget is probably the usual - as cheap as possible - as expensive as necessary.
Will select a few choices from cheap to medium price with (dis-)advantages and minimum requirements...

Thanks

 

[Iaroslav]

I was looking for something similar last year with a budget of $500-1000. Many parameters matter - thermal range, warm-up time, MSX, but anyway thermal resolution defines the price. Looks like buying Flir E4 and upgrade it to E8 with 320x240 is still very popular. I was even close to buy Flir One phone attachment, but then found nice and compact Flir C3 with 80x60 sensor for $650 and now fully satisfied with it.

 

[Dreece]

As cheap as possible?

Sewer moths! Throw a handful in, they reproduce rapidly, can end up with thousands within a couple of weeks, and they will congregate in their own tribes. Each tribe will select a warm area to rest in, Eventually one should be able to discern heat distribution quite well. Disadvantage is that they don't like to leave.

If the above is a no go , then a more modern and economical solution could be what my sister purchased for her business:
Seek Thermal CompactPRO Smartphone Thermal Camera for Android & iOS - 15Hz

 

[Rand__]

I was looking for something similar last year with a budget of $500-1000. Many parameters matter - thermal range, warm-up time, MSX, but anyway thermal resolution defines the price. Looks like buying Flir E4 and upgrade it to E8 with 320x240 is still very popular. I was even close to buy Flir One phone attachment, but then found nice and compact Flir C3 with 80x60 sensor for $650 and now fully satisfied with it.

Is 80x60 enough? sounds small compared to the others ...

We'll probably want to check rows of racks as well as individual racks and the odd coolant unit on the roof...

Sewer moths!

That sounds very eco-friendly, I like it

I considered the Phone add On too, but no sure we need to go that cheap...

What I wonder - on one hand those things are sensitive up to 0.06°C but accuracy is +/- 2°C (Flir E8) - that doesnt make sense to me...

 

[Rand__]

Anyone who could say something regarding the accuracy? Just a problem of the tech? Any better models out there?

Our Rack sensors are not the most accurate either so worried that in a worst case scenario we might have 3-4° difference from reality which is a tad too much for my liking...
I originally planned with touch temperature probes but o/c it would be way simple to get a single overall picture than needing to measure in 4+ spots ... considering getting one of the combo tools but not sure I can sell that to mgmt...

 

[alex_stief]

What I wonder - on one hand those things are sensitive up to 0.06°C but accuracy is +/- 2°C (Flir E8) - that doesnt make sense to me...

Measurement resolution vs accuracy. It is rather common that the former is better than the latter. It pretty much has to be by definition.
Accuracy vs. Resolution | The Difference Explained

 

[Rand__]

Well yes, but the magnitude of the difference is huge (to the uninitiated eye).
I mean they sell themselves by claiming 0,06° 'resolution' over 0,07° or 0,09°, but without knowing the actual accuracy (which might vary on factor 20 in either direction) that's kind of irrelevant.

I think maybe I am looking at this with a wrong expectation set - I came from measuring thermal or humidity values (roughly ~0.5° accurate) and these things excel in measuring temperature differences within the picture, not its absolute value...

 

[alex_stief]

Yes, that is the niche for thermal imaging. Measuring absolute temperature values with high accuracy is not the strong suit of this method. Especially not at affordable prices with devices operated by untrained personnel.
And measurement accuracy is just the tip of the iceberg here. Once you read up on reflectivity and emissivity, things get even more complicated Different Factors Which Impact Thermal Imaging Quality
Thermal imaging is what it is: a method to visualize temperature distribution. Not more, not less. If you really need high accuracy in your temperature measurements, it is not the best choice.

 

[Dreece]

Just yesterday a client came up with an idea for his horse videos (a breeder), wants to do a thermal video of a horse at rest vs a horse after a heated race. So looks like I may be getting one too now, but will need some form of high-res and fps video capture functionality too, though nothing will happen until this lockdown is over but I'll share what I get when there.

 

[Rand__]

Interesting use case

We are looking for combo devices now - thermal imager for overview with wireless temp probe for specific/more precise surface measurement ...

Edit: Hm not sure those exist - with humidity probe or power yes but not temp. will probably have to ask if I can join up these :/

Edit2:
Testo 872 + testo 605i (Thermo + Hygro)

Or
Visual IR Thermometer
FLIR TG165
Fluke VT04