Flir E4 Thermal imaging camera teardown

[tomas123]

I can confirm it:

model E4 1.1
part-no 63901-0101
firmware 1.19.8
appkit 1.0.14
confkit  E4 1.1
osimgkit 16.0.10
prodkit 1.0.0.8
AppCore 22.0.0.5
AppService 22.0.0.1
etc.

[Fraser]

@tomas123,

Many thanks for the conformation. E4 1.1 confirmed 

So 'Confkit' has changed to version 1.1.

Possibly just a production upgrade as this is a new model.

Have any hardware versions changed?....i.e. IRDM 0.0.1.0, POLLUX 0.1.0.0, POLLUX FPGA 8.1.25.0 ?

[tomas123]

IRDM 0.0.1.0, POLLUX 0.1.0.0, POLLUX FPGA 8.1.25.0 ?

... are the same

[Taucher]

Suggestion - let's compare the contents
FlashFS\system\kits.d

Code: [Select]

# confkit.rev
# This file contains revision information for all loaded
# files in a kit
# Do not edit

NAME userconf
VERSION E4 1.0
DATE 12-Jun-2013

# format: FILE <filename> <size> <CRC32>
#     or: FILE <filename> undefined
FILE flashfs\system\appcore.d\config.d\conf.cfg undefined
FILE flashfs\system\appcore.d\factory.d\adjust.rsc 113 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\alignmode.rsc 145 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\extimages.rsc 146 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\hcfalpha.rsc 138 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\intensity.rsc 147 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\palette.rsc 268 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\pathimages.rsc 136 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\power.rsc 341 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\time.rsc 128 <CENSORED>
FILE flashfs\system\appcore.d\factory.d\usb.rsc 114 <CENSORED>
FILE flashfs\system\kits.d\confkit.rev undefined
FILE flashfs\system\service\appcore.d\config.d\conf.cfg undefined
FILE flashfs\system\service\appcore.d\factory.d\bw.rsc 258 <CENSORED>
FILE flashfs\system\service\appcore.d\factory.d\rndis.rsc 84 <CENSORED>
FILE flashfs\system\services.d\config.d\conf.cfg undefined
FILE flashfs\system\services.d\factory.d\no_wzctool.rsc 60 <CENSORED>
FILE flashfs\system\ui.d\config.d\conf.cfg undefined


[Fraser]

Thanks @tomas123.

As your camera is 'enhanced', the new E4 1.1 does look to be an innocent product development version. 

@osteron. Thank you for providing your camera details. Yours was the first report of a new E4 version (1.1). You have assisted in mapping the version history of the E4 model. Cheers 

[mrflibble]

Why censor your supersecret CRC32 that is not a CRC01?  Did I miss the memo?

[osteron]

Code: [Select]

# confkit.rev
# This file contains revision information for all loaded
# files in a kit
# Do not edit

NAME userconf
VERSION E4 1.1
DATE 25-Oct-2013

# format: FILE <filename> <size> <CRC32>
#     or: FILE <filename> undefined
FILE flashfs\system\appcore.d\config.d\conf.cfg undefined
FILE flashfs\system\appcore.d\factory.d\adjust.rsc 113 568770501
FILE flashfs\system\appcore.d\factory.d\alignmode.rsc 145 1145798160
FILE flashfs\system\appcore.d\factory.d\extimages.rsc 146 3959544322
FILE flashfs\system\appcore.d\factory.d\hcfalpha.rsc 138 403022140
FILE flashfs\system\appcore.d\factory.d\intensity.rsc 147 3198357105
FILE flashfs\system\appcore.d\factory.d\palette.rsc 268 767855916
FILE flashfs\system\appcore.d\factory.d\pathimages.rsc 136 2892098096
FILE flashfs\system\appcore.d\factory.d\power.rsc 341 1775527956
FILE flashfs\system\appcore.d\factory.d\time.rsc 128 3305292736
FILE flashfs\system\appcore.d\factory.d\usb.rsc 114 2394224369
FILE flashfs\system\kits.d\confkit.rev undefined
FILE flashfs\system\service\appcore.d\config.d\conf.cfg undefined
FILE flashfs\system\service\appcore.d\factory.d\bw.rsc 258 4267495318
FILE flashfs\system\service\appcore.d\factory.d\rndis.rsc 84 1796347983
FILE flashfs\system\services.d\config.d\conf.cfg undefined
FILE flashfs\system\services.d\factory.d\no_wzctool.rsc 60 2262041015
FILE flashfs\system\ui.d\config.d\conf.cfg undefined


[mrflibble]

The exact same here, except of course the version + date.

Code: [Select]

# confkit.rev
# This file contains revision information for all loaded
# files in a kit
# Do not edit

NAME userconf
VERSION E4 1.0
DATE 12-Jun-2013

# format: FILE <filename> <size> <CRC32>
#     or: FILE <filename> undefined
FILE flashfs\system\appcore.d\config.d\conf.cfg undefined
FILE flashfs\system\appcore.d\factory.d\adjust.rsc 113 568770501
FILE flashfs\system\appcore.d\factory.d\alignmode.rsc 145 1145798160
FILE flashfs\system\appcore.d\factory.d\extimages.rsc 146 3959544322
FILE flashfs\system\appcore.d\factory.d\hcfalpha.rsc 138 403022140
FILE flashfs\system\appcore.d\factory.d\intensity.rsc 147 3198357105
FILE flashfs\system\appcore.d\factory.d\palette.rsc 268 767855916
FILE flashfs\system\appcore.d\factory.d\pathimages.rsc 136 2892098096
FILE flashfs\system\appcore.d\factory.d\power.rsc 341 1775527956
FILE flashfs\system\appcore.d\factory.d\time.rsc 128 3305292736
FILE flashfs\system\appcore.d\factory.d\usb.rsc 114 2394224369
FILE flashfs\system\kits.d\confkit.rev undefined
FILE flashfs\system\service\appcore.d\config.d\conf.cfg undefined
FILE flashfs\system\service\appcore.d\factory.d\bw.rsc 258 4267495318
FILE flashfs\system\service\appcore.d\factory.d\rndis.rsc 84 1796347983
FILE flashfs\system\services.d\config.d\conf.cfg undefined
FILE flashfs\system\services.d\factory.d\no_wzctool.rsc 60 2262041015
FILE flashfs\system\ui.d\config.d\conf.cfg undefined


[Taucher]

The exact same here, except of course the version + date.

Code: [Select]

FILE flashfs\system\appcore.d\config.d\conf.cfg undefined
FILE flashfs\system\kits.d\confkit.rev undefined
FILE flashfs\system\service\appcore.d\config.d\conf.cfg undefined
FILE flashfs\system\services.d\config.d\conf.cfg undefined
FILE flashfs\system\ui.d\config.d\conf.cfg undefined


Yep, same here also - no SN in those CRC obviously...
.. so any change would probably be in one of those "undefined" files

[Taucher]

Why censor your supersecret CRC32 that is not a CRC01?  Did I miss the memo?

To nourish my paranoia ...supported by lazyness to verify if the files are unique or not

[mrflibble]

To nourish my paranoia

And remember: just because you're paranoid doesn't mean they're not after you!!! 

[max-bit]

Can prepare in to save "opportunities" camera "configuration file" in the EEPROM memory which will not be so easily accessible from the "File System"
And then it will be after the event

[schdiewen]

Right, had some fun with FFTs just to see what we can see with regard to banding, vignetting, etc. Turns out that indeed you can see that in the spectral domain.
...

Those are really nice and interesting pictures. And thanks for sharing the numbers as well. After seeing my calibration pic the first time, I thought that the lower vignetting will lead to a lower dynamic range and thus my normalized picture would show more noise than yours. But now i know that our dynamic ranges are comparable, mine is even slightly larger.

So what's the "result", why are those pics that different, you have vignetting and banding, I have spatial noise instead? Perhaps the calibration pictures are already scaled?

Obviously there is quite some variation in the sensor and lens attributes. From trying to interface a thermopile array sensor (Melexis MLX90620) to a µC I remember a rather extensive compensation calculation for each individual pixel, with lots of calibration parameters stored in an internal EEPROM. So I think there are some more interesting numbers buried in the calibration files, which might be helpful in understanding this.

[inline]

So like others probably I'd like an "additional" lens to allow close up PCB inspection.  I've seen some 3d printer files and a few prototypes to provide a "hack" of a lens adjustment tool or additional lense, but is there someone selling these that I can purchase one from via PayPal, etc?

[mtdoc]

So like others probably I'd like an "additional" lens to allow close up PCB inspection.  I've seen some 3d printer files and a few prototypes to provide a "hack" of a lens adjustment tool or additional lense, but is there someone selling these that I can purchase one from via PayPal, etc?

Try a google search for 3-D printing services. There may be one near you or one that will ship to you.  I found a few near me that will do this for both lens adjustment tools and the lens holder for about $15.

[Alphatronique]

Got my from shapeways.com  for 45$ (inc shipping to canada) for lens holder and the adjsut tool
bit more expensive but it professional SLS machine so normally better that make bot ?

[inline]

Do I need both the lens adjustment tool and the lens adapter?

[georges80]

Do I need both the lens adjustment tool and the lens adapter?

Your choice.

The lens adjustment tool allows you to easily set the focus point to 2" or even less. It means every time you switch from far to close you need to adjust the lens.

The lens adapter allow you to leave your primary lens along and just pop in a 2" or 3" etc ZnSe lens when needed. You do have to buy the ZnSe lens so factor that into your total cost.

The ZnSe lens WILL attenuate the signal to some degree since it is an additional optic. I've not noticed much difference in readings at lower temps (say below 40C), but on some measurements in the 90C range I saw of the order of 5C or more lower reading with the ZnSe installed.

I have both

cheers,
george.

[mikeselectricstuff]

Bear in mind that the lens is a plastic thread into a metal thread, and not designed for large numbers of adjustments, so if you are likely to need to change from normal to close-up a lot, an external lens is probably a better option to avoid wear

[Fraser]

For those interested in the typical CO2 laser ZnSe lens spectral response, look here:

http://www.thorlabs.de/newgrouppage9.cfm?objectgroup_id=4313

Its not bad for our purposes but there will be some effect on the cameras calibration as already noted.

[bronson]

Been using my E4 for a few weeks.  Even at 80x60 it's damned useful.  So, fine, I finally went for the hack using a netbook running stock Linux Mint 16.

First, how the heck are you guys getting into the diagnostic screen?  I tried all sorts of combinations of power on, hit one of (Left, Center, or Back), then hold down Play for 5-10 seconds, both before and after the splash screen disappears.  Nothing.  I'm really glad for https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg326705/?topicseen#msg326705 because navigating to the versions screen and holding down Right for 10 sec worked great.

Anyhow, in the diagnostic screen, set USB Mode to RNDIS.  Now plug in the computer.  Mint will spend an unproductive few minutes trying to obtain DHCP from the camera and notifying you about it.  Ignore it.  Open Network Settings, click on the new Wired network (for me, it was the first item in the list), and configure it to be a static ipv4 with address 192.168.250.3.  Default /24 netmask is fine.  No DNS, routes, or gateway needed.  If your computer has multiple wired network interfaces, you might have to do some pluggging/unplugging to figure out which one is the RNDIS camera and which one is the physical port.

EDIT: You only need to set up the network once.  From then on, if the camera is in RNDIS mode, Mint will automatically set it up when the camera appears.  Plug and play.

Now ping 192.168.250.2.  If ping doesn't resolve, unplug the camera for a few seconds and plug it back in.  This worked for me.  No routing or arp commands needed.  If you still can't get anything, check out the linux tricks mentioned in previous posts.

Now that ping can see the camera, you can fire up ftp, ssh, telnet, or whatever you want.  All hail Mike.

Except now I'm a little stuck.  I can't run FLIRInstallNet on Linux, can I?  (ick, no VMs)  How should I put my modified E8.cfg on the camera?  Just FTP it into FlashFS/system/appcore.d/config.d?

I could also zip up a fif file but then how would I install it?

Random notes:

To compile the crc01 executable, 'apt-get install g++', cd to Mike's CRC01_source, and run "g++ *.cpp -o crc01"

Over telnet, "rset .image.flow.maps.combGainDeadMap.pixReplace false" to show dead pixels.  Pass true to hide them again.  Also "restartapp" will reboot the camera without breaking the telnet connection.  If it looks like it's hung, don't worry.  It'll come back.

For the record: E4 1.0, Software 1.19.8

[Taucher]

I'm really glad for https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg326705/?topicseen#msg326705 because navigating to the versions screen and holding down Right for 10 sec worked great.

You're welcome

Except now I'm a little stuck.  I can't run FLIRInstallNet on Linux, can I?  (ick, no VMs)  How should I put my modified E8.cfg on the camera?  Just FTP it into FlashFS/system/appcore.d/config.d?
I could also zip up a fif file but then how would I install it?

The additional config file belongs into \Flashfs\system\appcore.d\config.d\ - try putting the E8.cfg there and powercycle the camera, if the config stays there, then you should be running in E8 mode (unless your CRC01 somehow computed wrongly). Beware of PC style CRLF line-endings while editing!
Btw - I suggest to turn noise generation completely off (AFAIR the example file kept it at low level).

Note: the "flir/3vlig" ftp-login doesn't work if the camera is powered down - just had that issue - but FTP port would be listening *interesting*

[Taucher]

Can it be that Dave's forum broke down on the size of this thread?
The print page gives a "500, Internal Server Error" reply (just http header, no content).
Edit: other threads still seem to "print"

[bronson]

So FlashFS/system/appcore.d/config.d/E8.cfg doesn't work.

But FlashFS/system/appcore.d/config.d/e8.cfg (lowercase e8) does work.  Lovely!

Thanks Taucher.  Now I guess it's time to play with menus...

[Taucher]

FlashFS/system/appcore.d/config.d/e8.cfg (lowercase e8) does work.  Lovely!

@Mike ... I guess it would be nice to include a note that putting the cam in RNDIS mode and FTPing the config will also work - no need for Flir tools