Flir E4 Thermal imaging camera teardown

[OrBy]

Interesting unit! The export agreement has got some really specific wording on it as well. Small lens on the front, looks like it's running faster then 9Hz as too.

[ixfd64]

There is a RUMOUR that the camera contains a FLIR core but I could not confirm this.

That's what teardowns are for.

[heavybarrel]

Just throwing in my two cents here but.. It might be possible that the people over at Flir are in a panic for a different reason?
We need to remember that these engineers and designers are not stupid people! I have a suspicion that when they caught wind
of the lovely resolution patch and extra bonus menu items, they started to dig into what they have been putting out as far as the firmware goes and had an "UH OH!" moment. I believe they scrambled to fix a "boy i hope nobody figured out how to exploit this yet". Logic dictates that the camera was crippled by the firmware, so the hardware can do much much more. I would bet the farm on the frame rate being able to be exploited on earlier versions and the latest countermeasures were in fact not to eliminate the ability to patch the resolution but to lock down the features that could in fact put them in front of a judge. Fear of a lawsuit can make a company do very odd things. Like targeting ebay listings, really.... Its just a broom sweeping at some of the crumbs to intimidate the rest of them. I spoke to the legal department at my place of employment (i work for an engineering firm that does automation, software, and hardware) and the answer i got was "They don't have a leg to stand on." and "They must not understand what the law defines as property." and finally, "Software including firmware can only be copyrighted in it's current state and must go through that process each time it is modified." In other words, once you change it, it is not covered anymore. The automation industry would go out of business if that wasn't the case.

[mikeselectricstuff]

I would bet the farm on the frame rate being able to be exploited on earlier versions

I highly doubt this - we have no direct evidence, but as the FPGA is doing all of the processing, including frame avaraging to allow the use of the small lens, there is no reason for it to support higher framerate throughput. However it's not inconceivable that there could have been a test mode in there.

 
 "Software including firmware can only be copyrighted in it's current state and must go through that process each time it is modified." In other words, once you change it, it is not covered anymore. The automation industry would go out of business if that wasn't the case.

Nonsense. What process ? Copyright exists as soon as something is created.

[heavybarrel]

Just telling you what our legal guys said. They do it every day, i don't. As far as copyrights go i am no expert. I do know that in the world of graphics there is no such thing. You can sell superman t shirts all day long as long as it doesn't say the words "Super Man" on it. I guess the name is the only thing protected. I do know our legal guys are top notch! We deal with all of the giants of the manufacturing industry. From mining equipment manufacturers to nuclear power plants. I would take their advice if i was going to face a judge. Again that was just my two cents, no intent on initiating an argument. 

Edit:

Might just be a U.S. thing, not sure but i found this.

http://www.copyright.gov/help/faq/faq-register.html

And this.

§ 411 . Registration and civil infringement actions11

(a) Except for an action brought for a violation of the rights of the author under section 106A(a), and subject to the provisions of subsection (b), no civil action for infringement of the copyright in any United States work shall be instituted until preregistration or registration of the copyright claim has been made in accordance with this title. In any case, however, where the deposit, application, and fee required for registration have been delivered to the Copyright Office in proper form and registration has been refused, the applicant is entitled to institute a civil action for infringement if notice thereof, with a copy of the complaint, is served on the Register of Copyrights. The Register may, at his or her option, become a party to the action with respect to the issue of registrability of the copyright claim by entering an appearance within sixty days after such service, but the Register’s failure to become a party shall not deprive the court of jurisdiction to determine that issue.

Note: I didn't read 106A(a) 

[Galaxyrise]

As far as copyrights go i am no expert. I do know that in the world of graphics there is no such thing. You can sell superman t shirts all day long as long as it doesn't say the words "Super Man" on it. I guess the name is the only thing protected.

There are certainly such things in the graphics world!  For your example, there are numerous trademarks on Superman, including the logo and his likeness for clothing.

[heavybarrel]

Trademarks and copyrights are not the same. That's all i have on that one. 

[ixfd64]

On the subject of laws, are restrictions on use after purchase legally binding?

For example, FLIR sells some educational kits with deep discounts but requires that they be used for non-commercial purposes only. Are buyers legally obligated to comply with this?

[PedroDaGr8]

On the subject of laws, are restrictions on use after purchase legally binding?

For example, FLIR sells some educational kits with deep discounts but requires that they be used for non-commercial purposes only. Are buyers legally obligated to comply with this?

They can be based on the EULA.  This would not apply to sellers because they are not end users.

[PedroDaGr8]

On the subject of laws, are restrictions on use after purchase legally binding?

For example, FLIR sells some educational kits with deep discounts but requires that they be used for non-commercial purposes only. Are buyers legally obligated to comply with this?

They can be legally obligated based on the EULA.  This would not apply to sellers because they are not end users. I think flir is on very shaky legal ground for pulling sales that even mention it can be hacked. They could really open themselves up to a DMCA-abuse lawsuit.

[ixfd64]

Still, it looks like FLIR has achieved their goal with scare tactics. It's been several weeks since 2.x.0 was released, and there seems to have been no serious attempts to hack it so far.

[mikeselectricstuff]

Still, it looks like FLIR has achieved their goal with scare tactics. It's been several weeks since 2.x.0 was released, and there seems to have been no serious attempts to hack it so far.

I don't think it has anything to do with whatever Flir may or may not have done -  the simple fact is the boat has mostly sailed.
Anyone interested in buying & hacking a camera will already have done so, and anyone looking now will try to get hold of a unit with older firmware, of which there are still a few around on ebay and probably also in distribution - 1.22 rwas eported on a unit from Amazon on Jul 11th.
(It would be interesting to know if Flir pulled old stock back like they did with 1.22, though we've had a report that an old unit updated by Flir to 2.3 is still hackable so maybe little point.)

The simple fact is that there are far, far fewer people with any need to hack the current firmware than was the case for the previous releases.  Maybe once 2.3 has been around a while someone will have a serious crack at it, though by then the E4 may be using Lepton cores.
 

[mikeselectricstuff]

Just stumbled on some internal pics of the E60 on the FCC approvals site:

https://apps.fcc.gov/oetcf/eas/reports/ViewExhibitReport.cfm?mode=Exhibits&RequestTimeout=500&calledFromFrame=N&application_id=523320&fcc_id=ZLV-FLIRE49001
 

[OrBy]

I was making some pano's today and then running them through tomas123's flir.php script and was getting nothing but failures for some unknown reason.
I managed to track it back to the newer FLIR Tools 4.1 seems to be putting values differently into the corrected radiometric jpg and convert is failing to be able to extract them.
Downgrading to a older 4.0 version of FLIR Tools fixed the problem and I can make them again.
I tested this with non-panoramic files and they were affected as well.

Just a FYI in case anyone else runs into issues!

Oh - and I got ZnSe/20mm/FL: 4"/101.8mm lens this week and got my local maker space (assentworks.ca) to print off that holder file that was published in ABS on their best printer. Here is some close up goodness.

[pubo]

Just for whoever that is still interested to purchase one, I've just purchased one from tequipment.net via PriceUSA on the 8th of July and received today. Total cost is AUD 1224. The unit I received is as follow.

Model: E4 1.1L
Serial: 63924xxx
Part Number: 63901-0101
Software: 1.22.0

Hope that helps.

[Echo88]

Hi,

does anyone know which format the uvc-stream of the Ex-series has? Is it an MPEG-Stream consisting of JPG-frames (with EXIF-rawthermalimage) or just the live-stream of the display without EXIF-stuff?

Greetings,

Echo

[vmp]

I know its off topic but how i should configure my flir e4 to stream video to "vlc" or "flir ir camera player"
rtsp://192.168.0.2 don't work for me. Camera is online beacus i can connect to via filezilla.

[ThermalGuru]

I'll try to buy a E4 1.22 before the supply dries up completely. Plan to use it for PCB work.

Has anyone tried to make a new casing for it and move the internal parts over to make for a more easy fit on a tripod?
I'm also thinking in terms of designing in an easily accessible focus adjustment using the standard optics and a new form factor (such as Flir A40M). I can design up the casing in my CAD and then have it 3D printed with Objet or SLA for a nice finish at reasonable cost.

• Please let me know thoughts on pro/cons doing this and interest.
  
• Will the MSX function be offset when changing the focus on the lens?
  
• If anyone has made CAD files of the E4 or accessories for it, please share as it can speed up the design (most time will go into making the internal parts into 3D CAD)
  

Cheers

[mikeselectricstuff]

I'll try to buy a E4 1.22 before the supply dries up completely. Plan to use it for PCB work.

Has anyone tried to make a new casing for it and move the internal parts over to make for a more easy fit on a tripod?
I'm also thinking in terms of designing in an easily accessible focus adjustment using the standard optics and a new form factor (such as Flir A40M). I can design up the casing in my CAD and then have it 3D printed with Objet or SLA for a nice finish at reasonable cost.

• Please let me know thoughts on pro/cons doing this and interest.
  
• Will the MSX function be offset when changing the focus on the lens?
  
• If anyone has made CAD files of the E4 or accessories for it, please share as it can speed up the design (most time will go into making the internal parts into 3D CAD)
  

Cheers

MSX won't work close up, but you don'treally need it as when close up it will be obvious what you're looking at.
Not sure it's worth the effort re-casing - it's not too hard to rig up a tripod mount of some sort.

If you want something for nothing else but PCB work, Therm-App might be a better option, as it has adjustable lens and more mounting options

[Echo88]

I disassambled my E4 and stuffed it together with my 2 stereo-webcams in a bigger aluminium-case, its not that complicated. The 4 screws which hold the pcb in the magnesiumframe are separated by about 50mm each. The On/Off-Button is emulated by a flexprint-button and the battery is connected via a selfbuild adapter. The other buttons are controlled via Telnet.

[RAWebb]

There are some tripod mounts, auxiliary close-up lens holders, and focus ring tools for the Ex cameras over at Thingiverse. Nothing, that I'm aware of, for remounting the internal components.

[Sentient BeanCurd]

Hi Folks,

New owner of a FLIR E4 v 1.22.0 here , and I'm trying to perform the amazing upgrade hack.

However, I cannot seem to connect to the camera using Filezilla as described in this post.

I am receiving a "Connection failed with "ECONNREFUSED" -Connection refused by server. message most of the time, and "Could not connect to the server" other times.

I have:
1. Turned off Windows Firewall and McAfee
2. Confirmed IP address is 169.254.146.11 in ipconfig and used that IP address
3. Tried using the 192.168.0.2 IP address as described in this thread
4. Tried the connection on 3 different Windows 8 and 7 machines...same errors
5. Of course installed the RNDIS drivers from FLIR
6. Setup the camera in RNDIS only mode in the instrument USB hidden menu

So...I' wondering if FLIR have somehow changed the RNDIS drivers? I can communicate with the device using FLIR tools, and I can stream video to an IP capture device once the device is reset (not in RNDIS mode) so I know the E4 is working.

Any help mucho appreciated...something new must have happened recently, since the Filezilla connection is so effortlessly acheived in these posts.

[vmp]

RNDIS drivers i used to connect via filezilla.
http://cdn.cloud.flir.se/swdownload/assets/other/flir_device_drivers.exe

[RAWebb]

It's been observed that it is sometimes a bit of a dance to get the FLIR and PC to link up over RNDIS.

What I've gotten to work (fortunately, it hasn't been needed recently) is some combination of (1) disconnecting/reconnecting the USB, (2) leaving the USB connected and power cycle the FLIR, or (3) change the FLIR's USB mode from bare RNDIS to RNDIS+[one of the other modes]. Eventually, the PC and camera have always managed to find one another. Just keep checking ipconfig for the right subnet.

[Sentient BeanCurd]

Hi,

Thanks for your suggestions. I've tried them all, and the "best" I can do is confirm the IP address of the FLIR E4 can be pinged, and get an "ECONNREFUSED" error in Filezilla and FTP error in the TIConfig program..

I've attached a few screenshots in case anyone would care to take a look.

It seems to me that the camera is connected with a valid IP address and responding to pings but not accepting the user password?