IR thermometer saves the day! (story)

Recently all my house lights have been flickering when other lights are turned on! This is pretty nasty and indicates a bad connection somewhere in the 240v supply to the lights, and obviously at the fuse box or between the fuse box and most of the light citcuit wiring.

I opened the fuse box and removed the cover, and carefully wiggled all the wires where they screw into the circuit breakers etc. Darn! no bad connections there. So I worked the light circuit breaker 20 times or so, in case it had corrosion between the contacts. Darn! Light still flickering.

So it was time to start crawling through the roof checking for junction boxes, or pulling out every single light switch and inspecting the wires, and/or every single light fitting in the ceiling... Yep i had been putting this off for a while.

Then I got an idea! I have a cheap IR thermometer something like this one;

**broken link removed**

So I walked through each room and pointed the IR thermometer right at the light switch (about 2" distance), and then at the wall 6 inches to the side of the light switch.

Here are the results, tested at night with about 300W of lights turned on;


It was that easy!

The light in the bathroom wasn't even on at the time but that was where the problem was in the wiring!

I pulled the light switch out and the incoming 240v active wire was sparking visible blue sparks, it had totally sheared at the screw and was simply touching the brass screw with gravity holding the weight of the wire so it sat in the terminal hole.

I was glad I decided to use try the IR thermometer before pulling every light swith out, that room would have been around the middle of my list.

Good catch mate!

Its a good tip for the poor mans version of the expensive heat sensitive camera that my mate uses at work on the massive panel fuse boxes to look for hot spots.

I always wondered what I could use my IR temp sensor for.

Good post and handy to remember..........Al (Domestic electrician for 30 years)

I found a water leak in the ceiling of my sons room using one of those things. It was winter time and snow melt was filling part of the roof, you could tell exactly where the leak pool was because it was dramatically colder than the rest of the ceiling. They're just as functional as a heat camera, they just take a lot more work and a bit of brain power to get the same results =)

I am disspointing in their use as an IC temperature meter though, mainly on aluminum heatsinks have HIGHLY variable emissivity values depending on how oxidized they are.

Just as an example the emissivity of clean non or lightly oxidized aluminum is between .02 and .20 and heavily oxidized goes up to .30 or so I'm guessing anodized aluminum would be considered highly oxidized, but it depends on the type of anodizing done . One alloy A3003 is as high as .40 So math errors dominate the measuring of aluminum. Copper is even worse.

Most IR non-contact thermometers have a fixed emissivity of .95 or so.

Good idea. Another similar trick; take an battery-powered AM radio for a walk. The RFI created by the sparking wire would have been loudest near the switch box.

MikeMl said:

Good idea. Another similar trick; take an battery-powered AM radio for a walk. The RFI created by the sparking wire would have been loudest near the switch box.

Click to expand...

we take a slightly mistuned radio for a walk around the fields to find any arcing from the electric fence due to vegetation etc works a treat!

Thanks for the tip MikeML there are actually 2 "poppy" light switches, i'll replace them soon(ish).

Sceadwian said:

... I am disspointing in their use as an IC temperature meter though, mainly on aluminum heatsinks have HIGHLY variable emissivity values depending on how oxidized they are. ...

Click to expand...

There's an easy fix for that Scaedwian, I do it all the time (actually that's one of the main uses for my IR thermometer) is checking heatsinks and metal things. Get some of those plain-paper (not gloss) adhesve labels, they sell them as "laser printer labels". Then just stick one on your heatsink or any shiny metal, it fixes the emissivity and the IR therm will work perfectly pointed at the label. Afterward any left over adhesive usually comes straight off with alcohol.

I'm confused (easily done), I thought IR thermometers measured the frequency of the emissions to determine temperature. Surely, if it depended on emissivity then moving it closer would result in a higher reading?

Mike.

IR thermometers have a lens system of sorts, so they have a narrow cone of view (~10degrees). Like the "face" metering mode on some cameras. Most of them shine a laser spot where their field of view is centered.

MikeMl said:

IR thermometers have a lens system of sorts, so they have a narrow cone of view (~10degrees). Like the "face" metering mode on some cameras. Most of them shine a laser spot where their field of view is centered.

Click to expand...

The first one I ever saw was at a power station open day, an expensive professional one with a Minolta auto-focus lens on it.

They used it monitoring temperatures at various places, and charting the results - in that way they hoped to detect potential trouble before it became a problem. In a similar way they had acoustic sensors all over the place feeding chart recorders, again looking for changes in the noise to indicate possible developing faults.

For the open day they were demonstrating the chart recorders by sticking a sensor on your chest, and printing out your heart beat, for the thermometer they were measuring peoples temperature - from across the room.

Pommie said:

I'm confused (easily done), I thought IR thermometers measured the frequency of the emissions to determine temperature. Surely, if it depended on emissivity then moving it closer would result in a higher reading?

Mike.

Click to expand...

I think this is an interesting point.

Blackbody calibration sources for infrared pyrometers and thermal imaging systems

Information in the above link seems to indicates the quantity of IR energy emitted is used to determine temperature, not frequency. They do not say this directly but it is there.

Wikipedia agrees

By knowing the amount of infrared energy emitted by the object and its emissivity, the object's temperature can be determined.

Click to expand...

Would IR thermal sensors handle as high of temps as thermocouples would? I need a sensor to monitor the burn pot on a pellet stove and have always been thinking of using a thermocouple. But this might be an easier way to handle the issue, if the IR sensors can handle the heat somehow.

Does anyone have sources on the sensors themselves? I can't seem to find them, just full thermometers.

we use thermal imager a short of device to get infra red enission images from objects, to find out any hot spot in high voltage switch yard. there are hundreds of connections and bolts, but the job is just 15-20min to scan the wohle switchyard. the intersting point in this is the image autoranges from lower to higher temporature and changes its colour from voilet to red. when at 30-40 degree ambiant normally all structures would be seen in red, since its the maximum. but when a sudden hot spot comes to the far field of the imager the range suddenly changes and the far hot point only will be in red.

then for accurate reading a near scanning is done and thermo graphic pictures can be captured for analysis. as i remember its range goes upto 600 degrees C.

I looked into these a while back for exhaust pipe temperature. The sample they sent had only 1/2 the range of the part we ordered and it did not sound like they were in production. Maybe that has changed.

They may go high enough for your stove but at the time they did not for the app I was looking at.

The other thing you need to look at is FOV (Field of View). It is specified in degrees. As you move back from the target the area it looks at increases. I expect you could find a large enough are to view from a safe distance on a stove.

Actually The field of view on almost every non contact IR thermometer I've ever seen is measured with a spot size to distance ratio. Mine is 9 to 1, so at 9 inches the spot it sees is 1 inch. Anywhere from 6 to 1 to 12 to 1 is common. Also emissivity doesn't change with distance it's a property of the surface being measured, thanks for the sticker tip Mr RB, I forget how simple these things can be sometimes =P

Sceadwian said:

Actually The field of view on almost every non contact IR thermometer I've ever seen is measured with a spot size to distance ratio.....

Click to expand...

Perhaps the FOV thing is used more often for the raw sensors which are not as easy to find. https://www.electro-tech-online.com/members/smanches.htmlSmaches asked about them.

As I understand it the emissivity of an object does not change with distance, but the energy level drops of with the square of the distance from the object. As you move the sensor away from a wall the spot increases in size as the distance increases. You see less energy per unit area but have more area due to the larger spot.

I've never looked at raw sensors, just full units, the sensors themselves require a decent amount of calibrated electronics to operate properly, that's why I bought a full unit not a sensor. At distance I think things might get a little flakey, the lens will tend to blur things so you end up with an average, but high temperature point sources with lower temperature ambient materials might lead to some odd readings.

I often use mine for tuning engines.
I just read the temperature of the exhaust manifold inlet ports to find which cylinder is misfiring or running rough. The bad ones always have a much lower temperature.
It works on any engine with two or more cylinders and it also works well for tuning my old Kawasaki motorcycle that has independent carbs for all four cylinders as well. Once the temperature levels for all four exhaust pipes are nearly the same all four carbs are balanced out!

Sceadwian said:

I've never looked at raw sensors, just full units, the sensors themselves require a decent amount of calibrated electronics to operate properly, that's why I bought a full unit not a sensor. At distance I think things might get a little flakey, the lens will tend to blur things so you end up with an average, but high temperature point sources with lower temperature ambient materials might lead to some odd readings.

Click to expand...

The sensors I linked to are temperature calibrated. To say these are raw sensors is somewhat misleading. They also cost between $20 and $40 each.

The MLX90614 is built from 2 chips developed and manufactured by Melexis:
• The Infra Red thermopile detector MLX81101
• The signal conditioning ASSP MLX90302, specially designed to process the output of IR sensor.

Click to expand...

Everything I have read suggest the spot should be at a somewhat uniform temperature. (from memory) I have been thinkiing about how optics would effect the calibration. It may be that as long as you maintain the same spot size you can extend the working distance using optics. I would have to bench test this to be sure.

I think the weirdness I'm talking about would be equivilant to a 'lens flare' with an optical light source. Never really observed it, but it's something that should be kept in mind when you have a possible extremely hot spot that's in the sensors field of view from a distance.

Well we had a job at work where we had to heat a tapered bore coupling via blow torch's to 240C then put the coupling onto a brand new gearbox where the output shaft was tapered to suit the coupling. We went and bought a fluke IR thermometer so we could check the temp of the coupling. We were told at 240C the coupling would sit on the output shaft at the right position so just drop it on.

Well we heated the coupling until the fluke IR said 240C then dropped the coupling on......

The coupling was way too hot and went on 20mm too far.

I just happened to have my multimeter with the temp probe on my desk so I went and got it, the coupling temp was actually close the 350C so thats why the coupling went on too far.

Anyway that job cost over 40K in damages as the coupling had to be cut off, a new one made and a new output shaft for a brand new gearbox imported from Germany.

All this because the company was too stingy to get a contractor with the right gear.

So I found out the hardway IR thermometers are good for non critical measuring but don't rely on them for precise temp measurements.


Cheers Bryan