Poor Man's Thermal Imaging

I use a temp. probe on my multimeter very fast response. I saw in some catalog of stickers calibrated to certain temperatures that will indicate a hot spot when you place them on a suspected area that could get hot.

Most all scales/printers in all major supermarkets use thermal labels.... You could probably bum a couple or few labels or buy a slice of cheese from the deli and use the label to see if it reacts to your heat source ( I am not sure... but I believe it darkens to full " black " at about 150 degrees F.... )

Did you find http://www.sensorprod.com/thermex.php?

Under certain conditions, I have used a 3M product called Fluorinert FC-77. If you immerse a PCB with a short circuit into the fluid, you can get localized boiling of the fluid, which will allow you to locate the short quickly. The material has a relatively high dielctric constant so it may not work on high frequency circuits. For locating shorts in DC systems it works great.

I recently read that the Apple IPod Nano has the ability to capture infrared images- maybe not sufficient for a serious application, but maybe sufficient for a quick look for hot spots? I don't know if it would work for that, but at $200, it is a whole lot cheaper than a real IR imager- and if it doesn't work for this application, you could always use it for listening to music...

My wife got a sample of a product that was a heat sensitive color changing plastic sheet, don't remember the company but they also market a light sensitive color changing sheet

One of my secretaries, a few years ago had a heat sensitive Tee shirt. If you made her blush her boobs would glow red! (I'm not kidding)

More on topic, get a reel of thermal printer paper, i.e from an old fax machine.

I've got some rolls of it right here beside me but it's a bit far to send them th Canada.

Chas

http://www.tempil.com/index.asp

thank you gentlemen, that's something I can work around with.

This forum 'rocks', as my daughter would say..

Spencer's gifts (Usually in the Malls) has had the thermal liquid crystal sheets about 4"x6". Problem is it will need to be flat and allow some time for the heat to transfer if good physical contact can not be made.

Good Luck!

Have any of you heard about the conversion of an old digital camera to an IR camera by placing exposed film over the CCD? I have toyed with the idea, just out of curiosity. There are a few websites that give instructions, but I am not sure how good it would be at finding hot spots.

Actually this won't do it BUT you are close.

What you can do is open the camera and someplace between the lens and the CCD is a IR filter to cut off the CCD's IR response. Do a search on your model of camera and removing of the IR filter that blocks the IR part of the spectrum.

This will ruin the camera for normal imaging. The color balance will be horrible. looking somewhat halfway between a regular looking photo and a true IR photo from a very expensive camera.

You can push the heat sensing effect further for very little cost by adding a IR pass only filter to the camera to help block the shorter wavelengths.

Joe

Idle-Tyme The Original Rolling Ball Clock

http://www.idle-tyme.com

Joe, before I 'ruin'(convert) my digital camera, is it based on your personal experiments?

I Can't find the main article I had for this, But here is a step by step how to from one person,

http://geektechnique.org/index.php?id=254

and some images made from them,

http://www.wrotniak.net/photo/infrared/

Infra-red cameras are expensive. If it was as simple as switching a filter or two to convert a digital camera to infra-red, everyone would do it. If your interest is in locating hot spots on an otherwise cool circuit board the converted camera may barely do the trick. To do thermal mapping you need expensive lenses that will pass long wave lengths, sensors that work at these wave lengths, and emmisivity compensation.

Yes it is NOT as good as a real one. but spen 20 bucks vs 2000, it's still better than all the tapes and probes etc. it will still show the hot spots, you won't know how hot, (hence part of the reason the expensive ones are expensive because they tell you what the temp is) but a hotter spot will show up brighter, ya just don't know how hot it is.

It's easy to say that, and it's easy to believe it. But that doesn't make it so. Different emmisivities can and do lead to completely wrong readings. A cool part can read hot if it has high emmisivity and a hot part can read cool. There are methods for coating circuit boards with talc or other powders to give all components similar emissivity, but these methods add time and cost. Then you must remove the stuff.

You could try a non contact IR thermometer, some of them are pretty good and are under $100 US.

In Poland they have bear on which a label has magical text

In Poland a bottle of bear with a label on which apears "you can drink" if the temperature lowers to drinkable. You can order these bottles and use the label for your tests. Whats wrong with you? Cant you just move your hand close to the PCB to detect hot spots? Are you one of those that look for excuses for lack of results? Perhaps $100billion IR tlescope place on the moon could help you to find these hot spots? Do they tollerate your excuses?

Stan

Temp-dots are heat sensitive stick-ons used in Operating Room to monitor patient temperture, they come 50 to sheet for about $7.50... also available calibrated for higher temps

I have a feeling that many of the suggestions are made by folks who have never actually performed this measurement. The problem with surface sensors such a liquid crysals, TC probes, etc. is that they all report their own temperature. In order to measure the temperature of a suspected hot spot, one needs to get the sensor temperature somewhere close the temperature of the hot spot. The only way you can hope to do that is if the sensor makes good physical contact with the hot spot. Close is not good enough. If you are using a piece of thermal paper or liquid crystals, you must have the sensor actually laying on the hot spot. This is not always easy considering the topological features of a PC board. If you use a thermocouple probe, the probe mus be in physical contact with the hot spot and enough heat must be transferred to the probe to raise its temperature. That's not always easy. Not to mention that both these measurements can alter the temperature of the hot spot because they extract heat. Radiation measurements can be fast and foolproof, but require some care.

thanks for clarification, I'm looking for thermal camera, I guess wont be building one..

I have a Sony video camera, its a few years old now (3) but it has a "Low Light" feature that switches into IR and it also uses an IR LED to flood the (sometimes unsuspecting!) subjects in pitch darkness.

It would be very easy to cover up the IR LED and then only heat sources would be filmed.....sorry videoed......

That would be a cheap alternative if you have yourself or know someone with such a camcorder.....I am sure that many Camcorder makes offer something similar........it does not have to be a Sony.

The results are a sort of Green/Black image.....

yes, this was pointed out to me and I used mine to check if some remote controls were working. A useful thing to know.

But in this application would it have to be used in total darkness so the sensor wasn't flooded with visible light?

Haven't tried it, but I would guess so......

But why would that be a problem? Just turn all the lights out or wait till sundown......

Not all IR is created equal, so don't waste your time with trying to convert a CCD or CMOS camera.

A video camera can see IR and removing optical filters and adding optical filters will help this. This IR however is called Near IR and is basically near the visible spectrum at about 700nm-1400nm.

Thermal imaging is done in the long IR which starts around 3000nm, where your video cameras won't work.

Some other issues here are that glass and many other materials that would be used as a lens and the glass built in over the sensor block thermal IR.

Real thermal cameras don't use a CCD or CMOS sensor. Most modern FLIR cameras use a sensor called a microbolometer as well as a few other kinds of sensors.

For finding hot spots in a circuit board I would try a IR thermometer, I have done this with some success. This idea of finding hot spots on a circuit or cold spots in a room spawned my idea to create an image from this. So far I haven't had time to do this and no one has really stepped up to help.

In the comments of my post someone had the idea to use liquid crystals held very close to the threshold to detect heat and use this to create an image. I have no idea if it would work but its worth a try.

Thermal imaging has finally gotten down to the hobbyist's budget...

http://www.thermal.com/