Flir E4 Thermal imaging camera teardown

[Zucca]

I was hunting a good TIC on ebay since months, but

08 Dec@01:30 AM I saw this post and, immediately, I entered in panic mode. I couldn´t sleep anymore.
08 Dec@02:15 AM Finish to read about about 15 pages of this post, but I realized they were written in 48 hours (thank you Aurora to delete some stuff). I understood 65% of it.
08 Dec@02:30 AM I ordered the E4 on Tequipment.NET with 2 day air shipping.
11 Dec@16:30 PM I got my brand new E4, FW 1.19.8. Model E4 1.0. Calibration 13 Nov 2013.

It is not too late guys, don´t think just buy it.

@FLIR: You would never have got my 1000$ without this. Just think about that over Xmas.
@Mike: Just undo it. You are the Chuck Norris of the Electronics Engineer (FYI when Alexander Bell invented the telephone he had 3 missed calls from Chuck Norris).
I´m proud to live in the same planet at the same time with you.

It´s time to blast some AC DC in the air and to get some more resolution...

Update: Done, too easy

[aargee]

Realize this might not be the place for it, but playing around with the flir tools, I haven't been able to get it to record.  (No record button is showing on the app?)  Has anyone had success? And yes its the +

I did find this that might interest folks that don't have Tools+

http://flir.custhelp.com/app/answers/detail/a_id/332

It talks about using VLC to view/record...

.

I think others are using the E4 like a webcam. Mine appears this way to the PC if you connect it via USB without Flir Tools. I assume you're using an E4.

[G0HZU]

Going back to the original subject, you have pointed the camera at a steady uniform surface, then rotated the lens and no differences to be seen?

Yes, I did several quick and lazy tests but my interpretation of 'no difference' might not be the same as everyone. It pretty much looked the same even orange colour but with the usual noise. I have a look through focus tool and all I could see was some random noise on the orange.

I did this on the auto temp scale so maybe if someone tried it on a zoomed manual scale it might be more apparent but I would be falling asleep at the significance of fractions of a degree in non uniformity. This is because I have to look at the overall measurement uncertainty of the entire system plus the uncertainty of the emissivity of the subject and also the calibration drift of the instrument over a few seconds.

If I was worried about uniformity over fractions of a degree across the whole screen area then I don't think I'd be using an Ex series camera. So for my use (and probably most people) it's a non issue with this camera I have here

[daves]

Well, thanks to everybody here. I just received and converted my E4.

I have to say E4 original resolution is pretty low. So low, it is much worse than my old ARGUS camera for half price.
See my argus video hot and cold water - now I can sell this one



After conversion to 320x240 it looks much better and puts old Argus into shame.
I have to do some side by side comparison.

I had problem with drivers on my computer, x64 Win7, camera is not recognized at all. I do not know why.
So I tried my notebook, also x64 Win7. Here it is working well. Conversion was pretty easy, well documented.
Anyone having trouble with drivers ? Windows is saying installation of driver failed. Any suggestion ?

I think original founder of hack deserves some kind of "donate".

[mikeselectricstuff]

Just found another use for the E4 (with  tripod) - checking nozzle aim and evenness of heating when using a hot-air tool to remove a QFN from a crowded PCB without collateral damage - makes it a lot quicker as you can accurately get heat exactly where it's needed.

[Mikesus]

Realize this might not be the place for it, but playing around with the flir tools, I haven't been able to get it to record.  (No record button is showing on the app?)  Has anyone had success? And yes its the +

I did find this that might interest folks that don't have Tools+

http://flir.custhelp.com/app/answers/detail/a_id/332

It talks about using VLC to view/record...

.

I think others are using the E4 like a webcam. Mine appears this way to the PC if you connect it via USB without Flir Tools. I assume you're using an E4.

E4 and I will have to take a look.  I don't remember it showing as a webcam...

[G0HZU]

I tried to think of another reasonable test for the focus tool vs lens error concern:

I hooked the camera up to FLIR tools and streamed the image to my large PC screen
Then I wound out the focus a long way to get it to focus really close to the lens.
I then closed the lens cover and pressed my finger against the lens cover to warm it up.
I then locked the temperature scale
I then watched on the screen as the lens cover cooled off to see if there were any odd pattern effects.
I then rotated the lens a few degrees and tried again. The results were remarkably good. What I did see was occasional vertical bands and some random and very subtle spattery patterns but these were at tiny fractions of a degree and may be a factor of the video compression. They didn't appear to be lens rotation sensitive.

So in this test at least there weren't any obvious/stubborn issues with non uniformity as it cooled down for either lens rotation.

This isn't exactly a conclusive test but I didn't see errors or ugliness that I would call remotely significant at either lens focus angle

[Fraser]

@G0HZU,

This is indeed good news    I have the focus tools so it would have been a shame if that option had 'issues'.

I am currently using the auxilliary ZnSe lens option, as that best suits my needs.

Thank you for doing the tests that you have carried out for us. 

[Fraser]

Just found another use for the E4 (with  tripod) - checking nozzle aim and evenness of heating when using a hot-air tool to remove a QFN from a crowded PCB without collateral damage - makes it a lot quicker as you can accurately get heat exactly where it's needed.

Mike,

Thats an excellent idea. I had not thought of that use before. Thanks 

[jlr134]

Well, thanks to everybody here. I just received and converted my E4.

I think original founder of hack deserves some kind of "donate".

I second that but I have a related question, how is this forum funded because I for one would be interested in donating toward its operational expenses .

[ixfd64]

Well, thanks to everybody here. I just received and converted my E4.

I think original founder of hack deserves some kind of "donate".

I second that but I have a related question, how is this forum funded because I for one would be interested in donating toward its operational expenses .

You can become a Supporter for $25 a year.

[jlr134]

Thanks

[thereza]

Planning on picking one up from amazon if none of them have been patched yet to prevent this hack.   Any reports of it failing?

Thnx,
Reza

[amyk]

I think original founder of hack deserves some kind of "donate".

Buy stuff to send to him for teardown?

[digsys]

   I think original founder of hack deserves some kind of "donate" 

There must be some way to support those who spend a lot of effort, and cost, for our free "upgrades."
Many already donate to EEVBlog, but that doesn't help the others.   Ideas?

[Alphatronique]

since today it a hot day i try my camera outdoor

seem the reading in cold temperature was bit off ..

actual outdoor was only -22 

also did camera put serial number on exif ?
edit : self reply
        yes camera serial number was on the EXIF
        so post original image expose your serial number  camera
        take image put it on flir tool folder and it will show camera SN ..
        have test wit image i take from that tread for make sure info nome from image itself
 

[Fraser]

I have been experiencing inaccuracy at temperatures below 0 Degrees Celsius, but didn't want to post detail until I have determined whether it is user error !  I was seeing -4 Degrees Celsius when a thermocouple was saying +2 Degrees on a house brick and concrete path outside my house. I have yet to investigate what was going on. Could be emissivity but the error seems too large. I changed the refelected temperature setting but it had little effect. More testing needed.

[Alphatronique]

HI Aurora

have take several shot on different surface , snow ,brick  ,sky etc
whit camera a room temp and at outdoor temp (let sit outdor for 10 min)

and same result ,it strange since indoor temp look accurate

and -20 was quite a big error 
will need more test meteo said it was -20 for next 3 day 

[mikeselectricstuff]

I have been experiencing inaccuracy at temperatures below 0 Degrees Celsius, but didn't want to post detail until I have determined whether it is user error !  I was seeing -4 Degrees Celsius when a thermocouple was saying +2 Degrees on a house brick and concrete path outside my house. I have yet to investigate what was going on. Could be emissivity but the error seems too large. I changed the refelected temperature setting but it had little effect. More testing needed.

cold surfaces may have condensation & ice, which will affect emissivity

[Fraser]

I will do some proper tests with test pieces in the fridge and freezer and see how the E4 behaves.

[Alphatronique]

have made quick test indoor / outdoor whit Flir box so it a reproducible surface to everyone here

my thermometer said -17  and image was take away from sunlight

[Fraser]

I will do a test with the PM695 and E4 in parallel to see if they give the same temperature. They are both thermal cameras operating in the same scenario so there should be little difference.

Watch this space.

[tomas123]

uncooled microbolometer: At low temperatures, the accuracy drops dramatically.
Flir defines NETD (noise equivalent temperature difference) at 30°C.
http://flir.custhelp.com/app/answers/detail/a_id/128

I think, the blackbody for calibration of the E4 is also uncooled ;-)
http://www.flirshop.com/product/4-inch-blackbody-source/Tau320-Accessories

--
Don't forget to set the right air temperature and distance!

[unicorn5]

Hi, I"m non technical. Interested in your lens discusion but don't really know what you're talking about.

Inn short a pair of glasses, ie a lens over the existing lens won't help with getting a sharper image, only magnifying or altering focus, but the non-sharpness of images can only be fixed with replacing existing lens, and replacement lens would be a different material to E4 lens and more expensive than E4 it's self. Is that approx correct?

[nixxon]

Here is a IR picture of the full moon taken a while back with a Fluke TiR.

If I remember correctly, there are ca 9 pixels covering the lunar disk, and that should theoretically be enough to get a somewhat accurate reading, disregarding the influence of the atmosphere of earth and of course the reflected temperature of the sun on the moon.

According to my TIR, the temperature is ca 55 ' Celcius.

If the modded E4 has the double FOV (45') and the double resolution, the results should possibly be similar using the E4 with similar IFOV.

Imaging Performance Fluke TiR:
Field of View ....................................................... 23 ° x 17 °
Spatial Resolution (IFOV)................................... 2.5 mRad
Minimum Focus Distance
Thermal Lens.................................................. 15 cm (approx. 6 in)
Visible (Visual) Light Lens .............................. 46 cm (approx. 18 in)
Focus.................................................................. Manual
Image Frequency................................................ 9 Hz refresh rate
Detector Type ..................................................... 160 X 120 Focal Plane Array,
uncooled microbolometer
Infrared Lens Type.............................................. 20 mm EFL, F/0.8 lens