To start with, that camera is a hobby camera. Turn off the MSX and see if you can identify anything by the palette colors. It’s a big blob. Actually, look closely, you will see that the MSX does not align with the colors.
Electrical IR is the easiest application you can do with IR. But you can’t be looking for Quantitative data as a “go/no go” with that camera. You also are not analyzing the target for how hot it is based on AI data. You need to know what is causing heat and why. Likewise, you not only need to know IR, you need to know everything about what you’re pointing it at, and know what you expect to see when you do. Only 1% of the time are you going to find the device about to melt down as being the problem. What causes the heat? Loose wire connection, loose breaker connection to the buss, defective internal components or the breaker, excessive load, a reflection… With training, you will learn through quantitative observation what each of these are. Then you can go find the problem.
If you see something hot, you need to put down the camera and start looking for the source of the heat. What is the amperage on the circuit. Is it at full load or not. A 20 amp breaker does not run at 20 amps hardly ever. 5 amps can make a lot of heat. What will the temperature be if you up the load to 15 amps? Don’t ever try that!
The temperature you are getting with the camera is an indirect measurement in most cases. Things cool down based on mass, density, conductivity, ambient conditions etc. You will miss the 2,000F Arc where the breaker connects to the buss when it is only 80F by the time it gets to the surface.
The problems go on. The best you can hope for is to use that camera like a glorified flashlight. If you see something in IR, go find out if is anything through normal HI Inspection Standards. Then report it, without putting IR in the report. Unless you can explain psychrometric and thermodynamic reasons for your findings while sitting in a courtroom, don’t do it.
If you can’t afford a couple of thousand dollars on training and get rid of that $400 camera, stick to using it like a glorified flashlight. The course you took qualifies you to turn the camera on and how it works. The camera has no idea what the potential issues are. It just assigns a color for how you adjust the camera, and I bet you did not adjust your camera, right? If not, don’t use the numbers. IR can make things look like a forest fire at a .1 degree differential. They are an apparent temperature and are never correct unless you tune the camera for one target in the scan. These are not a photograph. It is a spreadsheet. Colors are assigned for range and span. You can have a problem hidden in that red spot if you use the wrong settings and color span.