Flir E4 Thermal imaging camera teardown

[Taucher]

Hi Guys, has anyone found a way to keep the temperature scale settings from going back to automatic when powering down the camera ? The auto-scale is nice but most of the time I would like to keep the scale in manual with fixed temperatures so I can compare images taken at different intervals. All of the other menus keep the last feature: Image mode, Measurement, color palette even the new "zoom" feature keeps the last setting.

Thanks !

sure, you just need to add another .conf file to setup the corresponding registry settings as you like.

[svflir]

I am still having problems with my e4 on 1.21 and I fixed my conf_plain to not have typos

I added the crc03, and then re-ran ftool, re-uploaded to the right directory and still have lost thermal msx and don't have higher resolution

Any idea on what to try?  Factory reset?  Something else?

[FrankenPC]

Does the E4 hack drive down the thermal sensitivity to <0.06°C like the E8?

I did a search on the forum and only saw a few posts asking the same question.  I didn't see any responses.

[Gallymimus]

I am still having problems with my e4 on 1.21 and I fixed my conf_plain to not have typos

I added the crc03, and then re-ran ftool, re-uploaded to the right directory and still have lost thermal msx and don't have higher resolution

Any idea on what to try?  Factory reset?  Something else?

I was working on this a few hours ago.  Your CRC03 looks right, maybe you did the ftool incorrectly?  There isn't much to making this work right.  It sounds like it is completely rejecting your config file.

[svflir]

@Gallymimus I will try doing the ftool again, although it seems pretty straightforward

ftool -d to decode, ftool -e to re-encode with the SUID output. 

[svflir]

question about ftool and the 1.21 hack

do I need to have the original conf.cfc from my camera to generate the right suid?

If yes and I blew this away and lost it, is there any way to re-generate it?

I downloaded one of the posted conf.cfc's and used the tool on that, modified and re-uploaded but no luck

many thanks for any help as I have totally lost MSX from my device

[Ivan7enych]

Does the E4 hack drive down the thermal sensitivity to <0.06°C like the E8?

I did a search on the forum and only saw a few posts asking the same question.  I didn't see any responses.

I see, that thermal image show significantly less noise after modification. Less noise = better sensitivity.
To make a numerical comparison, ideally one need to make 2 shots in RAW format, and compare standard deviation values on part of image with uniform color.

I've added 2 image examples from my camera, both with almost the same thermal range (~10degree). Unmodded camera has much worse noise, and I can't imagine any physical explanation of that. Dounsampling the resolution of the image (averaging 4x4 rectangle of pixels into 1 pixel) should make noise 4 time lower. Simple downsampling (taking only one pixel from 4x4 rectangle) should make noise the same, not larger.

May be flir adds some noise  to a good image for low range cameras?

[Gallymimus]

question about ftool and the 1.21 hack

do I need to have the original conf.cfc from my camera to generate the right suid?

If yes and I blew this away and lost it, is there any way to re-generate it?

I downloaded one of the posted conf.cfc's and used the tool on that, modified and re-uploaded but no luck

many thanks for any help as I have totally lost MSX from my device

uh oh, you might be in trouble.  YES you ARE supposed to use the original conf.cfc to get the right SUID.  I'm not sure if it can be recreated.  Hopefully it's just generated algorithmically but I wouldn't know.

[svflir]

oy it was not clear in the instructions to use the original SUID

Who could help me with how to re-generate it?

If I delete it and restore to factory defaults does it regenerate it?

Is there another way to figure out the right SUID for my device or another file I can decode?

[svflir]

thank god there was another conf.cfc encrypted the same way on the device I could get the SUID from!

1.21 hacked

Lessons learned: backup files before messing with them
read all forum posts before trying 1.21 hack

[mikeselectricstuff]

Does the E4 hack drive down the thermal sensitivity to <0.06°C like the E8?

I did a search on the forum and only saw a few posts asking the same question.  I didn't see any responses.

We don't know - the supposedly improved spec on the E8 is a bit of a mystery. Could be down to calibration, sensor selection or just marketing fakery, i.e. all units may actually have the same sensitivity, they just make it look worse on the cheaper models

[daves]

We don't know - the supposedly improved spec on the E8 is a bit of a mystery. Could be down to calibration, sensor selection or just marketing fakery, i.e. all units may actually have the same sensitivity, they just make it look worse on the cheaper models

As I said before - when you take photo of uniform temperature, you will see always about 13°C scale. Scale has 224 levels. 13/224=0.058°C.

[Fraser]

Hmmmm, regarding the thermal differential sensitivity of a thermal camera........

The only way to truly know what the thermal differential sensitivity of the camera would be to test it in a calibration jig designed for the purpose.

In my experience there is a lot of confusion surrounding the "Sensitivity" specification. In simple terms it is best to think of it as visible contrast. The ability to detect a diffference in temperature. The SPAN used on a thermal camera can effect visible difference displayed to the user BUT the camera actually captures the full thermographic range and displays only the segment selected by the user. This is why you can import an image into FLIR Tools and adjust the center temp and span in the software using the cameras RAW thermographic data.

With this in mind, the user selected span cannot influence a cameras true thermal differential sensitivity. That remains fixed no matter what the user does with the span that is displayed via the screen.

FLIR do use an artificial noise generator in the E4. There are various thoughts on why they do this, some considered it a way of improving the images appearance (video processing) whilst others consider it a way of artificially degrading the image on lower end cameras. The Noise generator was originally set to 135mK but was reduced to 5mK by Mike as he correctly identified that the inserted noise had a detrimental effect on the image when so much was inserted. I am very pleased with the 5mK setting that is part of Mike's config file.

Aurora

[tomas123]

I wrote here something about the possibility to change the thermal sensitivity with the parameter targetNoiseMk (the unit is MilliKelvin)

.caps.config.image.targetNoise.enabled bool true
.caps.config.image.targetNoise.targetNoiseMk int32 135

It could actually be noise reduction or shaping - maybe this is a threshold to decide when to apply noise reduction, i.e. adjust filter until noise is below the target value.  "Mk" - mask? Perhaps something like a hysteresis function?

targetNoiseMk is a noise generator in mK (conforming with NETD/ thermal sensitivity in Flir Datasheets of the selected cam)
Formerly I stacked some noisy images with Registax or AviStax  to subtract out the randomly noise and it works great!!

after setting targetNoiseMk to zero, stacking of images don't improve results 

here a sample from a Flir E40 (old cam hacked formerly from 160x120 to 320x240)
in service menu I can measuring the noise

Code: [Select]

rset .caps.config.image.targetNoise.targetNoiseMk  60 (MilliKelvin)

 Temporal noise        MilliKelvin      Digital Units
Pixel Noise      61.28      12.13
Row Noise        14.84      2.94
Column Noise     13.67      2.71

Spatial noise        MilliKelvin      Digital Units
Pixel Noise      25.69      5.09
Row Noise         7.08      1.40
Column Noise      7.82      1.55
Uniformity      198.23     39.25

Total noise        MilliKelvin      Digital Units
Pixel Noise       67.01     13.27
Row Noise          9.71      1.92
Column Noise      10.10      2.00


and now without noise (cam temperature is 25 Grad):

Code: [Select]

rset .caps.config.image.targetNoise.targetNoiseMk  0 (MilliKelvin)

 Temporal noise        MilliKelvin      Digital Units
Pixel Noise      19.85      3.93
Row Noise         6.39      1.26
Column Noise      8.64      1.71

Spatial noise        MilliKelvin      Digital Units
Pixel Noise      18.98      3.76
Row Noise         4.70      0.93
Column Noise      5.54      1.10
Uniformity      210.23      41.63

Total noise        MilliKelvin      Digital Units
Pixel Noise      28.96      5.73
Row Noise         5.76      1.14
Column Noise      7.66      1.52

see the differences (NETD is 0,03 °C @ 25°C)

http://gs.flir.com/surveillance-products/surveillance-technology/imaging-technotes/IR_Technology_Parameters
Uncooled infrared cameras systems are typically a little noisier, in the range of 30 - 120mK. Noise in an image can be spatial or temporal.
Spatial noise is noise across the image at any given point in time. It is perceived as an unchanging fixed pattern on top of the image.
Temporal noise is noise at any point in the image over time. It is perceived as the static that moves in an image.
NETD is typically the measure of both these noise types.

remarkably Flir dont't disable the noise generator at the top cams of a serie (Flir E4 -> E8 , Flir E30 -> E60)

[Fraser]

A useful explanation of thermal sensitivity is to be found here:

 http://irinformir.blogspot.co.uk/2012/05/what-does-sensitivity-netd-mean-when.html

A helpful page here as well:

http://www.infraredcamerasinc.com/Thermography-FAQ/Infrared-icons/Thermal-Camera-Thermal-Sensitivity-NETD.html

[Gallymimus]

thank god there was another conf.cfc encrypted the same way on the device I could get the SUID from!

1.21 hacked

Lessons learned: backup files before messing with them
read all forum posts before trying 1.21 hack

Glad you got it!

[tomas123]

In this thread was repeatedly told that the sensor is heated (and cooled??) to 30°C
(E4 shows an asterisk on display while heating to 30°C).

The sensor will heated only because of the simple calibration?

I think this is bad for the NETD and reduce the sensitivity for low temperatures 

P = s*T⁴ -> http://en.wikipedia.org/wiki/Stefan%E2%80%93Boltzmann_law

sample: air temp = -10°C and sensor temp = +30°C -> (263/303)⁴ = 0.56

[Rainer]

If you mirroring the TIC at glass-windows or any other IR-reflecting surface, you can see and measure the bolometer-temperature.

[Fraser]

All of my microbolometer cameras, including the professional PM series have a microbolometer that is temperature stabilised at +30 Degrees Celcius. I have also confirmed this with my E4 looking at the PM695 microbolometer.

The microbolometers are temperature stabilised on VoX FPA's to establish a set operating temperature point for the purposes of calibration and response prediction. I cannot say why 30 Degrees C was chosen. It may be related to efficiency or intrinsic properties of the microbolometer.

This web page does comment on TEC stabilised microbolometers and non TEC based units.

http://www.sofradir-ec.com/wp-uncooled-detectors-achieve.asp
(corrected the link ooooops. Thanks Tomas123)

For low noise, high sensitivity thermal imaging I used the Stirling Engine Cryo cooled PM550 that had the semiconductor FPA running at something like -190 Degrees Celcius ! The image produced by that camera made the microbolometer based cameras look very noisy indeed, espcially at small temperature spans of 2 Degrees. The down side of such cooling is the need for a mechanical cooler or Liquid Nitrogen. The Stirling Engine cooler is VERY expensive at several thousand US$ and the FPA is integral to it as a 'cold finger' connects the stirling engine to the FPA in a vacuum chamber.

Some thermal cameras used a peltier TEC pile to cool the FPA to -70 Degrees Celcius. The performance was not as good as the Cryo cooled types, and IIRC predated the microbolometer general release. Specific FPA types were needed for opertion at the relatively warm -70C. TEC's can draw a lot of current and the ability to cool to the required temperature is very dependant upon the pile design.

[tomas123]

thanks for clarification
the correct link: http://www.sofradir-ec.com/wp-uncooled-detectors-achieve.asp

here is a Flir technical note
http://www.flir.com/uploadedFiles/Eurasia/Cores_and_Components/Technical_Notes/uncooled%20detectors%20BST.pdf

Exx =  vanadium oxide sensor (VOx)
source http://www.flir.com/cs/emea/en/view/?id=54684

The FLIR E30 is equipped with a state-of-the-art uncooled Vanadium Oxide microbolometer detector

and
ix =  vanadium oxide sensor (VOx)
http://www.systemplus.fr/wp-content/uploads/2013/04/MEMSTrends_2012_10_N12_systemplus.pdf

FLIR microbolometer ISC0601B and the i7:
Each pixel is composed of a thin resistance in vanadium oxide (VOx) covered with an absorber in silicon oxide which absorbs the infrared ray. The temperature of the SiO2 layer is proportional to the temperature of the IR radiation, and the value of the resistance VOx is proportional to the temperature of the absorber.

BUT I don't think the E4 contains a VOX detector (I think it is Amorphous silicon)

[Fraser]

I must admit that I am not certain of the microbolometer technology used in the E4. It does not have the usual vacuum containment and very much looks like an Amorphous Silicon type that can be manufactured more cheaply as per a normal semiconductor die and the vacuum is truly tiny in size and located between the substrate and the front window of the device. The ADC forms part of the die and the TEC forms part of the chips substrate structure. All very compact and mass produceable. This technology originated in the FLIR car thermal night vision work stream as cost reduction was being pursued. I believe I posted some comment on this in Mike's thread on Audi and BMW thermal cameras. There was a document detailing the technology that I uploaded. I will find it and add it to this post. 

UPDATED:

AutoLiv document attached. Chapter 3 details the imaging FPA design.....looks awfully similar to that found in the E4. The lens design and material is also very familiar 

[muvideo]

Is the microbolometer optimized to a specific temperature range, or
probably the same E4 microbolometer is used for higher temperature imaging ?
If it's optimized for low range, is it possible to damage it  imaging a too hot object?

[tomas123]

Flir never said that the E4 is a VOx sensor (you're certainly right that e4 has the cheapest sensor)

I found this 
http://www.infrapuna.ee/files/E4%20Broshure%20English.pdf
Flir ThermaCAM E4

Using the world's best uncooled infrared detector material, vanadium oxide, the E4 delivers unmatched temperature measurement accuracy

[Fraser]

@muvideo,

I have not seen different microbolometers for differing temperature ranges. The normal approach to enabling higher temperature measurement is for the insertion of attenuators and wavelength filters in the optical path to control the thermal energy reaching the microbolometers surface.

The PM695 has options for a 1500C and 2500C. These are filters/attenuators fitted indside the camera on a moving filter holder.

Obviously when a camera has different temperature measurement ranges it will need to have appropriate calibration tables for each. When I select a different range on my PM series, the camera immediately carries out a comlete recalibration routine, not just NUC.

I am not aware of the maximum temperature that the E4 microbolometer can cope with but the Exx appears to be capable of over 600C, and I do not think it needs an attenuator or filter to achieve this.

Modern microbolometers are stated as being "Sun Safe", meaning that they cannot be damaged by being aimed at the Sun (much of the energy is 'out of the lens bandpass'). It would be hard to overload such a sensor to the point of damage, unless you were very determined to do so and directed in-band energy at the sensor at very high levels causing overheating of the pixels and eventual oxidation or melting of such. In the domestic environment or even the workshop I would not expect such conditions to exist.

Flame filters for live furnace inspections are explianed in this FLIR document:

http://www.flir.com/cs/emea/en/view/?id=41800


High temperature filters are not cheap to buy though .... take a look here:

http://www.netzerotools.com/testo-885-2-infrared-camera-high-temp-filter

US$1619 

[DaveWB]

Afternoon all,

Well a massive thanks again to all that helped out. I managed to fumble my way through the resolution fix, with the ftools, not hard but ha to work out how to do it from scratch.

I now have additional menus and also what appears to be 320 x 240 pretty amazing. The manual colour shift is a massive bonus too.

Ive come across a few glitches with my camera though.

The center spot does not work.
When taking an image it only saves the IR, the Jpeg is blank/black.
Alignment distance select option scrolls but when selecting reverts to zero. Yet the focus distance fine tuning works.
MSX option has vanished.

I used the conf.cfc (download from off this thread) downloaded mine from camera, copied all the text from the modified version, so its not something I miss typed.

Although after taking a good look around the file, it seems some are listed as false.

Any idea what to tweak??

I ran into the same problem and I cannot seem to figure it out. Is this the correct information for the config file?

EDIT:Thanks to stefbeer for helping out! Ended up being I forgot a space between # and CRC and also forgot the final empty line at the end.

Code: [Select]

#
# Generated at 2013-10-25 09:37:31
#
.caps entry
.caps.config entry
.caps.config.name text "app E8"
.caps.config.revision text "1.1"
.caps.config.image entry
.caps.config.image.framegrab entry
.caps.config.image.framegrab.fusion entry
.caps.config.image.framegrab.fusion.enabled bool true
.caps.config.image.framegrab.fusion.pip entry
.caps.config.image.framegrab.fusion.pip.enabled bool true
.caps.config.image.framegrab.fusion.hcf entry
.caps.config.image.framegrab.fusion.hcf.enabled bool true
.caps.config.image.services entry
.caps.config.image.services.store entry
.caps.config.image.services.store.enabled bool true
.caps.config.image.services.store.radiometric entry
.caps.config.image.services.store.radiometric.enabled bool true
.caps.config.image.services.store.incompatible entry
.caps.config.image.services.store.incompatible.enabled bool false
.caps.config.image.services.store.incompatible.level int32 0
.caps.config.image.settings entry
.caps.config.image.settings.enabled bool true
.caps.config.image.settings.IRwidth int32 320
.caps.config.image.settings.IRheight int32 240
.caps.config.image.sysimg entry
.caps.config.image.sysimg.alarms entry
.caps.config.image.sysimg.alarms.enabled bool true
.caps.config.image.sysimg.alarms.measfunc entry
.caps.config.image.sysimg.alarms.measfunc.enabled bool true
.caps.config.image.sysimg.alarms.measfunc.maxCount int32 3
.caps.config.image.sysimg.alarms.humidity entry
.caps.config.image.sysimg.alarms.humidity.enabled bool true
.caps.config.image.sysimg.alarms.humidity.maxCount int32 1
.caps.config.image.sysimg.alarms.insulation entry
.caps.config.image.sysimg.alarms.insulation.enabled bool true
.caps.config.image.sysimg.alarms.insulation.maxCount int32 1
.caps.config.image.sysimg.irMarkers entry
.caps.config.image.sysimg.irMarkers.enabled bool true
.caps.config.image.sysimg.irMarkers.spot entry
.caps.config.image.sysimg.irMarkers.spot.enabled bool false
.caps.config.image.sysimg.irMarkers.spot.maxCount int32 0
.caps.config.image.sysimg.irMarkers.arrow entry
.caps.config.image.sysimg.irMarkers.arrow.enabled bool true
.caps.config.image.sysimg.irMarkers.arrow.maxCount int32 4
.caps.config.image.sysimg.irMarkers.box entry
.caps.config.image.sysimg.irMarkers.box.enabled bool false
.caps.config.image.sysimg.irMarkers.box.maxCount int32 0
.caps.config.image.sysimg.measureFuncs entry
.caps.config.image.sysimg.measureFuncs.enabled bool true
.caps.config.image.sysimg.measureFuncs.diff entry
.caps.config.image.sysimg.measureFuncs.diff.enabled bool true
.caps.config.image.sysimg.measureFuncs.diff.maxCount int32 1
.caps.config.image.sysimg.measureFuncs.diff.calcMask int32 65526
.caps.config.image.sysimg.measureFuncs.isotherm entry
.caps.config.image.sysimg.measureFuncs.isotherm.enabled bool true
.caps.config.image.sysimg.measureFuncs.isotherm.calcMask int32 20
.caps.config.image.sysimg.measureFuncs.isotherm.dual bool false
.caps.config.image.sysimg.measureFuncs.isotherm.fixScale bool false
.caps.config.image.sysimg.measureFuncs.isotherm.interval bool true
.caps.config.image.sysimg.measureFuncs.isotherm.invInterval bool false
.caps.config.image.sysimg.measureFuncs.isotherm.maxCount int32 1
.caps.config.image.sysimg.measureFuncs.mbox entry
.caps.config.image.sysimg.measureFuncs.mbox.enabled bool true
.caps.config.image.sysimg.measureFuncs.mbox.calcMask int32 1924
.caps.config.image.sysimg.measureFuncs.mbox.maxCount int32 5
.caps.config.image.sysimg.measureFuncs.mcircle entry
.caps.config.image.sysimg.measureFuncs.mcircle.enabled bool false
.caps.config.image.sysimg.measureFuncs.mcircle.calcMask int32 1924
.caps.config.image.sysimg.measureFuncs.mcircle.maxCount int32 0
.caps.config.image.sysimg.measureFuncs.mline entry
.caps.config.image.sysimg.measureFuncs.mline.enabled bool false
.caps.config.image.sysimg.measureFuncs.mline.calcMask int32 1924
.caps.config.image.sysimg.measureFuncs.mline.maxCount int32 0
.caps.config.image.sysimg.measureFuncs.reftemp entry
.caps.config.image.sysimg.measureFuncs.reftemp.enabled bool true
.caps.config.image.sysimg.measureFuncs.reftemp.calcMask int32 1924
.caps.config.image.sysimg.measureFuncs.reftemp.maxCount int32 1
.caps.config.image.sysimg.measureFuncs.script entry
.caps.config.image.sysimg.measureFuncs.script.enabled false
.caps.config.image.sysimg.measureFuncs.script.maxCount int32 0
.caps.config.image.sysimg.measureFuncs.spot entry
.caps.config.image.sysimg.measureFuncs.spot.enabled bool true
.caps.config.image.sysimg.measureFuncs.spot.calcMask int32 514
.caps.config.image.sysimg.measureFuncs.spot.maxCount int32 5
.caps.config.image.sysimg.visualMarkers entry
.caps.config.image.sysimg.visualMarkers.enabled bool true
.caps.config.image.sysimg.visualMarkers.spot entry
.caps.config.image.sysimg.visualMarkers.spot.enabled bool false
.caps.config.image.sysimg.visualMarkers.spot.maxCount int32 0
.caps.config.image.sysimg.visualMarkers.arrow entry
.caps.config.image.sysimg.visualMarkers.arrow.enabled bool true
.caps.config.image.sysimg.visualMarkers.arrow.maxCount int32 4
.caps.config.image.sysimg.visualMarkers.box entry
.caps.config.image.sysimg.visualMarkers.box.enabled bool false
.caps.config.image.sysimg.visualMarkers.box.maxCount int32 0
.caps.config.image.contadj entry
.caps.config.image.contadj.minSpanFactor entry
.caps.config.image.contadj.minSpanFactor.enabled bool true
.caps.config.image.contadj.minSpanFactor.factorAuto double 2.0
.caps.config.image.contadj.minSpanFactor.factorManual double 2.0
.caps.config.image.targetNoise entry
.caps.config.image.targetNoise.enabled bool true
.caps.config.image.targetNoise.targetNoiseMk int32 5
.caps.config.image.zoom entry
.caps.config.image.zoom.enabled bool true
.caps.config.image.zoom.maxFactor double 8
.caps.config.system entry
.caps.config.system.focus entry
.caps.config.system.focus.laser entry
.caps.config.system.focus.laser.updateFocus entry
.caps.config.system.focus.laser.updateFocus.enabled bool true
.caps.config.ui entry
.caps.config.ui.fusion entry
.caps.config.ui.fusion.PIP entry
.caps.config.ui.fusion.PIP.enabled bool true
.caps.hw entry
.caps.hw.sdcard entry
.caps.hw.sdcard.enabled bool false
[ID HERE(REMOVED)]
[CRC HERE(REMOVED)]