Flir E4 Thermal imaging camera teardown

[janekivi]

Sorry, forgot add link to my answer, where I explained menu and functions.
I thought this is not so far away, my bad.

[supermikel]

Hi,

i am very happy that i  found  this Site, my E4 arrived yesterday   1,2L  with 2.3  , calibrated 6 days ago, seem to be brand brand new

I have some questions:


1.  where can i download      Set_RNDIS_temporary.fif" and "Set_RNDIS_permanent.fif     ??





Please check if your /FlashFS/system/appcore.d/config.d/conf.cfc is 6608
bytes.


2. How can I do that, Software needed?




3. Is Python 2.7.8 the right Version



Thanks from Germany and forgive my poor english


Michael

[sky@eevblog]

Has anyone encountered 'limited range' E4? Wonder could it be reconfigured via this hack to full range E4 camera?
http://store.flir.com/product/e4-infrared-camera-engineering-special/ex-series-infrared-cameras

[Northdynamic]

Hi,

i am very happy that i  found  this Site, my E4 arrived yesterday   1,2L  with 2.3  , calibrated 6 days ago, seem to be brand brand new

I have some questions:


1.  where can i download      Set_RNDIS_temporary.fif" and "Set_RNDIS_permanent.fif     ??





Please check if your /FlashFS/system/appcore.d/config.d/conf.cfc is 6608
bytes.


2. How can I do that, Software needed?




3. Is Python 2.7.8 the right Version



Thanks from Germany and forgive my poor english


Michael

1) You can find e4hack3c.zip here which includes the RNDIS files:

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg321956/#msg321956

2) You must first apply the RNDIS file to enable a network connection to your camera, I recommend only using the temporary file for v2.3. Follow instructions from the below post to load the RNDIS file.

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg535619/#msg535619

Stop after the step of running the Set_RNDIS_temporary.fif and install FileZilla or any ftp program. When you open your ftp program, under host enter 192.168.0.2 then "flir" for username and "3vlig" for the password. When connected you should be able to browse to

 /FlashFS/system/appcore.d/config.d/conf.cfc

Then just check the file size to see it is 6608 in size. Also this is a good time to make a copy of the original conf.cfc for safety.

3) Yes anything below python 3.x.x should work.

Be careful with a v2.3 E4 until someone creates a new hack procedure. I would only attempt the instructions for the resolution hack. Modification of any other files will likely result in a failure to load the camera software and give you a black screen.

The bad part about this is you can be locked out of FTP access with an invalid password error so you have to re-flash the whole v2.3 firmware over again via Programs(x86)/Flir systems/Flir Tools/bin and execute FLIRInstallNet.exe. In my case the re-flash kept failing until I disabled RNDIS mode first!

Good luck!

[supermikel]

Many thanks to you northdynamic, that will help me a lot 


When i have done like its written here, STEP BY STEP INSTRUCTIONS for Flir E4 (Firmware 2.3.0) for beginners ,  what did i get then? the higher Resolution and the rest stays the same-- Menues, Options, etc.?


Thanks

Michael

[supermikel]

Windows 8 ask how to handle the .fif  file, which program it should use?? Any Tips?

Thanks

Michael

[Scott in Michigan]

Bought an E6.  After using it for a while, I can say this: the capability it provides is great, but overall I'm left a bit underwhelmed.  Translate that to mean - I'm sure glad I didn't buy further down the chain. 

It makes me wonder if an unmodified E4 is of much use at all. 

It seems well built.  Boots up quickly. 

The spot temp feature is useful for surveying wiring and junctions, and seems quite precise.  You can zero in on individual conductors and connections.

The MSX feature is what makes the product most useful. 

Scott

[AleXYZ]

I am happy to report that my E4 with 2.1.0 firmware is now successfully hacked with the v0.3 files.  I now have 320x240 resolution and the induced generated color noise is GONE.  The picture is now stable and noise-less as I hoped.

Two resources helped me that were not mentioned in tmbinc's otherwise excellent instructions:

(1) A video showing the hack procedure for an earlier firmware version (v1.98)


(2) which led me to the missing RNDIS drivers located here:
http://cdn.cloud.flir.se/swdownload/assets/other/flir_device_drivers.exe


I am totally pleased with this brand new camera at a $700 price tag.  Thanks to EVERYONE in the forum who created the hack and made all this information available.

[Hugoneus]

Thank you supermikel, Northdynamic, miguelvp. I appreciate all the help I will try this.

Out of curiosity, after the hack, what does the E8 still offer which the E4 would lack? Is the issue that the firmware version 2.3.0 can't be fully hacked yet?

Thanks,

[mikeselectricstuff]

Out of curiosity, after the hack, what does the E8 still offer which the E4 would lack? Is the issue that the firmware version 2.3.0 can't be fully hacked yet?

With the old pre 2.x hack, the upgraded E4 had features that even the E8 didn't, like manual temp range setting.
AIUI the 2.x hack only improves the resolution, but I'm sure he other stuff will happen in time.
The E8 does have a better resolution spec, and we haven't yet figured out what determines this - it could be they they just add less noise, or could be that they select better sensors.

[Hugoneus]

Out of curiosity, after the hack, what does the E8 still offer which the E4 would lack? Is the issue that the firmware version 2.3.0 can't be fully hacked yet?

With the old pre 2.x hack, the upgraded E4 had features that even the E8 didn't, like manual temp range setting.
AIUI the 2.x hack only improves the resolution, but I'm sure he other stuff will happen in time.
The E8 does have a better resolution spec, and we haven't yet figured out what determines this - it could be they they just add less noise, or could be that they select better sensors.

Thanks Mike. So what is your recommendation, should I down-grade to pre-2.x firmware? (If that is even possible), or is the new firmware worth it? What about the E5, is the hacked version of E5 (with 2.x firmware) essentially becomes an E8 (minus the small resolution improvement)?

[mikeselectricstuff]

Out of curiosity, after the hack, what does the E8 still offer which the E4 would lack? Is the issue that the firmware version 2.3.0 can't be fully hacked yet?

With the old pre 2.x hack, the upgraded E4 had features that even the E8 didn't, like manual temp range setting.
AIUI the 2.x hack only improves the resolution, but I'm sure he other stuff will happen in time.
The E8 does have a better resolution spec, and we haven't yet figured out what determines this - it could be they they just add less noise, or could be that they select better sensors.

Thanks Mike. So what is your recommendation, should I down-grade to pre-2.x firmware? (If that is even possible), or is the new firmware worth it? What about the E5, is the hacked version of E5 (with 2.x firmware) essentially becomes an E8 (minus the small resolution improvement)?

If it was possible to downgrade things would be a whole lot easier (not sure anyone has seriously tried). I suspect the cal data from a later one may not work with earlier FW.
The resolution is the biggest improvement, so unless you can find an old stock 1.2x unit, wait until someone figures out how to hack the other features

[supermikel]

now i know how to handle the fif, but could also get no connection via ftp at 192.168.0.2 to the camera, is it possible that flir changed the IP in the latest version? My Ip scanner also found nothing

[Hugoneus]

If it was possible to downgrade things would be a whole lot easier (not sure anyone has seriously tried). I suspect the cal data from a later one may not work with earlier FW.
The resolution is the biggest improvement, so unless you can find an old stock 1.2x unit, wait until someone figures out how to hack the other features

From Flir's website, the only differences between E5/E6/E8 is the resolution and a minor temperature sensitivity. Does this mean that a hack of the E5's resolution would result in essentially an E8? Since all the extra software features are already available on E5.

[Northdynamic]

now i know how to handle the fif, but could also get no connection via ftp at 192.168.0.2 to the camera, is it possible that flir changed the IP in the latest version? My Ip scanner also found nothing

This thread contains a lot of information but it takes some time to read through and find all the answers. Try reading from this post onwards and you will see some people addressing the ipconfig problems:

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg540027/#msg540027

First I would start by installing the drivers from here (please read page 1 of this post for links and info):

http://cdn.cloud.flir.se/swdownload/assets/other/flir_device_drivers.exe

If the drivers do not help you will have to do some exploring to find your specific issue including any ip conflicts with other network devices seen by your pc. Remember that you have to apply the RNDIS temporary fif file first to activate the network service. Also the temporary RNDIS fif will only be active until the camera is re-booted so it may have to be applied again if you have shut the camera down.

This part of the instructions is intended to make sure you have a connection to your camera,

- Connect your E4 via USB to your PC and turn it on
- Go to "Programs(x86)/Flir systems/Flir Tools/bin and execute FLIRInstallNet.exe (this is part of the FlirTools software installed in step 2)
- Select the RNDIS_xxxxxxx.fif (permanent or temporary) and click RUN FIF button.
  (A timeout error is normal after the command has succeeded because USB is now set to RNDIS mode)
- Wait a moment and open a CMD window, then enter IPCONFIG
- An additional network connection should be listed with IP Address 192.168.0.1 and gateway 192.168.0.2
- Enter "FTP", then "open 192.168.0.2" and a connection to your cam should be established
- Login with user "flir" and password "3vlig"
- Issue a "dir" command and the files and directories of your cam are listed.
- If all that works, type "quit" to exit and close ftp connection.

Just keep reading through the older posts in this thread and you will find the answers. Again, tmbinc has created an excellent resolution patch and compet17 has written great instructions.

[nowlan]

Has anyone encountered 'limited range' E4? Wonder could it be reconfigured via this hack to full range E4 camera?
http://store.flir.com/product/e4-infrared-camera-engineering-special/ex-series-infrared-cameras

I was wondering what model B meant.

This E4 model will produce the same high quality image available on Standard Release cameras under normal operating conditions but the operating temperature range is not the same as the Standard Release E-Series specification detailed in the user manual

Code: [Select]

Temperature Range 0° to 150°C (Standard Range is –20°C to +250°C (–4°F to +482°F))
Accuracy ±5% or 5°C, whichever is greater at 25°C nominal
(Standard Accuracy is ±2°C (±3.6°F) or ±2% of reading, for ambient temperature 10°C to 35°C (+50°F to 95°F) and object temperature above +0°C (+32°F))
150 vs 250 C
5% vs 2%

[rmorgansd]

now i know how to handle the fif, but could also get no connection via ftp at 192.168.0.2 to the camera, is it possible that flir changed the IP in the latest version? My Ip scanner also found nothing

This might be your problem, it was mine when I first tried the hack on my E4......Make sure your computer is in the same LAN network domain range as the camera you're trying to connect to.  It needs to be 192.168.0.[something], where [something] is anything except 0, or 2.   Go to your network settings and set the TCP/IP properties to a static IP address in the same domain, such as 192.168.0.8.  (The '8' is just a number I chose).  Then disable and re-enable the network connection.  You don't need an internet connection on the computer as you do the hack.  To see if it works you can type at the command line.  PING 192.168.0.2  to see if it is now talking.  PING should return a response showing 100% success every second.  To stop the PING hold [CTRL] and press [C].   

[mikeselectricstuff]

If it was possible to downgrade things would be a whole lot easier (not sure anyone has seriously tried). I suspect the cal data from a later one may not work with earlier FW.
The resolution is the biggest improvement, so unless you can find an old stock 1.2x unit, wait until someone figures out how to hack the other features

From Flir's website, the only differences between E5/E6/E8 is the resolution and a minor temperature sensitivity. Does this mean that a hack of the E5's resolution would result in essentially an E8? Since all the extra software features are already available on E5.

There are other differences- E6 & above have picture-in-picture and hot above/cold below palette modes.
A fully hacked 1.2x unit adds manual temperature range (not even available on E8), which is probably one of the more useful add-ons, some extra pallettes and zoom mode (don't recall if E8 has this). 
However like I said the resolution is  95% of the improvement,  the rest is just icing on the cake. Nice to have all the toys but of limited use to most people.
Of course if you can find a way to downgrade a unit to 1.2x then everyone will be even happer (except Flir of course!) 

[Fraser]

For those wondering about the model 'B' versions of the Ex range, there is no real mystery here.

The 'B' model is intended to meet the needs of building surveyors who carry out thermal audits of a structure. This is a fiercely competitive market for FLIR so a cost reduction may increase their hold over such a lucrative customer base. Thermal audits are a significant market in this modern eco friendly world in which we live 

NEC AVIO released the F30 thermal compact camera in a version specified for the building thermal audit community. It had a smaller temperature range but apparently a greater temperature difference resolution as well. Such is important when viewing walls for minor thermal anomalies. NEC named the version F30s and it was also sold by Mikron under the name Mikroshot 'B' for 'Building' version. The camera sold at a relatively competitive price and was designed to fit into a surveyors pocket. Temp range topped out at 100 C but it also has the advantage of manual focus and decent FOV. Thermal range and amazing measurement accuracy are not paramount in building survey work
The lens is a clever design in that it is manual focus at close range and effectively fixed focus at long range so no need to fiddle with focus all the time on a building survey.

I have examples of the Mikron B in my collection but apart from taking one apart I have not had time to play with them. I will have to see what happens when one is pointed at a target hotter then 100 C. Very nice little camera though and perfect for field photography of wildlife as that tends to run cooler than 100 C 

I attach the data sheets for further info.

The FLIR Ex 'B' appears little different in terms of the intended market and the lower price may appeal to those who do not need amazing accuracy or temp range. It should be remembered that where accuracy is concerned anything specified tighter than +-2% is really only for laboratory  experimentation as target emissivity variance destroys absolute  accuracy anyway 

Aurora

[nowlan]

The real question re the model-b is are they crippled in software, or didnt make the grade for true e4.

At first I thought model b was just a revision, eg firmware, L(ock) board.
Almost bought one thinking was a deal.

Guess I need someone to get me a genuine e4 from zoro via proxy.

[Fraser]

There is no reason to believe that the B model is some sort of sub standard 'reject'. FLIR are pretty good with TIC production quality and have no need to sell sub standard microbolometers that fail QA.
Producing a model that is 'hobbled' for specific areas of the market in order to provide a range of cameras with differing capabilities and associated prices is pretty much proven by the E4 upgrade.

It would not be difficult for FLIR to limit the thermal range and accuracy of an Ex in its firmware, to suit the Building survey market needs without endangering standard Ex series sales. Far easier than trying to build TICs from sub standard parts  Even the 80x60 E4 contains a microbolometer that passes FLIR's functional pixel spec.

It is worth considering what an out of spec microbolometer usually fails on. DEAD PIXEL COUNT... simple as that. A VoX element does not tend to have a failure mode that limits temperature range or accuracy. It works or it doesn't simple as that. If a microbolometer die is found to be out of spec it normally ends up as scrap as its reliability is suspect. FLIR offer a 10 year guarantee on the microbolometers in the Ex cameras.

Aurora

[mikeselectricstuff]

There may also be differences in the calibration process. I don't know how many temperature points these units are calibrated over, but  it may well be that B units are at least optimised for the 0-100 deg range.

[Fraser]

@Mike.

Indeed. Calibration can be an expensive part of the production process. I may be out of date as the mass production that is now occurring must surely have some form of automated calibration on the production line ? As you say, a quicker, less precise calibration could be quicker and cheaper.

I was interested to read Mikron's comments on the Mikroshot 'B'. It was stated that though reduced in temperature range it has enhanced temperature differential sensitivity to suit the role for which it was intended. Such could be an advantage for some users over temperature range. If in electronics I see a component running at over 100C I would normally have reason to be concerned 

Aurora 

[Manfred]

I just ran the resolution upgrade on a 1.1L 2.3. e4.  I have lost the spot meter and thermal blending functions but do have improved resolution.  I can revert to the original conf and .dll files and both functions return, is this the expected outcome?  Thanks

[Chanc3]

Does anyone have a T400, T420, B425 they wouldn't mind sharing the config files for?

Trying to add another range to our B400 as the spec sheets say it should have two!