Flir E4 Thermal imaging camera teardown

[Psi]

I would not be at all surprised to find that there's someone in Sales over at Flir telling management that "Those guys over in that hacking forum just cost us over a quarter-Million USD in profits in the last 2 weeks alone".   

hehe, yeah

[Mark_O]

Buy the E4 and then you can inspect the prospective houses and be more informed.

No kidding!  You can spend $400 for a home inspection, and still not have them spot hidden, internal things like pest infestations, or water intrusion, etc.  A walkthrough with an E4 could net you $thousands$ in pricing accommodations, OR tell you it's time to walk away.

[mikeselectricstuff]

Mike,

The drift in pixels is normal and my PM695's do exactly the same. It is the downside of the micro-bolometer technology, and always has been. That is why cryo cooled cameras are so nice to use.....clean images, no pixel drift to be seen. The calibration routine should become less frequent as the detectors temperature stabilises, but this can take a while as there is hysteresis in the system and some initial over/undershoot on the Peltier module.

The comment on TICs becoming cheaper is interesting. One area that is buying into thermal camera technology is the car industry. Its use is safety of pedestrians/animals and predictive braking systems. We could see a large increase in TIC deployment in cars over the coming years. It may follow the model of other safety systems such as air bags and lane dicipline cameras. That can only be good news for TIC fans as more production into a larger market should bring innovative technology and lower prices ? In the UK we did not have 'expensive' air conditioning as standard in our cars. These days it comes as standard on many models sold in the UK and is not the expensive option that it was. It is also more refined, in the form of climate control 

I read of a new 80x80 mini TIC called the Atom80 that has just been released. It is designed for the less demanding TIC needs of the marketplace. No idea of cost though. Sofradir is a big player in TICs. It looks like a 'camera on a chip' type solution    Could be fun to play with, but I doubt its cheap at the moment.

http://www.sofradir-ec.com/landing_atom80.asp


UPDATE:  To save you registering on the site, I attach the ATOM80 datasheet here.

Info on the sensor, from their subsidiary ULIS
http://www.ulis-ir.com/index.php?infrared-detector=MICRO80P-044
Bizarre statement in the brochure claiming 2 years on 2xAA cells, and on the same page <25mW  - clearly some cockup there...
Wouldn't surprise me if we see TICs based on this soon.

[dr.diesel]

No kidding!  You can spend $400 for a home inspection, and still not have them spot hidden, internal things like pest infestations, or water intrusion, etc.  A walkthrough with an E4 could net you $thousands$ in pricing accommodations, OR tell you it's time to walk away.

Likely the average home owner would just think your trying to scam them out of money, if doing it yourself.

But yes, would make you a more informed buyer at minimum.

[dr.diesel]

A manufacturer wants its customer base to 'trust' it, and its products. This thread kind of introduced an element of doubt that FLIR will not wish to see extended into its more expensive products. I would not wish to be the salesman visiting a factory to demo a TIC and getting bombarded with questions about hobbling and hacking of firmware 

Why would being hackable lose any customer trust?  99.999% of customers will never even know and 1/2 of the people reading this thread only bought one cause they could hack it, IMO they've lost nothing and only benefited by selling a few extra E4s.

[Psi]

A manufacturer wants its customer base to 'trust' it, and its products. This thread kind of introduced an element of doubt that FLIR will not wish to see extended into its more expensive products. I would not wish to be the salesman visiting a factory to demo a TIC and getting bombarded with questions about hobbling and hacking of firmware 

Why would being hackable lose any customer trust?  99.999% of customers will never even know and 1/2 of the people reading this thread only bought one cause they could hack it, IMO they've lost nothing and only benefited by selling a few extra E4s.

Software crippled hardware rubs people the wrong way.

[dr.diesel]

Software crippled hardware rubs people the wrong way.

Perhaps, ignorance is proliferating, at least in the US.

[BravoV]

Even the E4 isn't a cheap product - but I can guarantee that nobody here was going to buy an E8 ... and from what I count there were already 50+ E4 sold...

That's true.  However, I would not be at all surprised to find that there's someone in Sales over at Flir telling management that "Those guys over in that hacking forum just cost us over a quarter-Million USD in profits in the last 2 weeks alone".   

The fact that that is completely false doesn't change the situation that different folks see things in different ways, and for some, that will be their perception.

And if the sales results/numbers for non E4 units ended up so bad that it deviates huge amount from the annual 2013 sales ... target performance of the whole "Sales & Marketing Dept.", someone up there at FLIR will be so pissed off, that he/she will likely bring the blame on the Tech & Design Dept. 

[dr.diesel]

dr-diesel

I have 'interviewed' FLIR salespersons on several occasions when they have been hawking a new product. To date I have had no reason to think that they hobbled units in firmware. The word is well and truly 'out' in the circles in which I move....industrial & Government techs are not dumb. We will ask FLIR for the best deal, and now be aware that they have employed hobbling to differentiate between models that intrinsically cost the same to produce. I have not seen such publicly proven before. Would you 'trust' a salesman who now stated that the E8 is an intrinsically higher spec product than the E4........before this hack, he/she could have done so without question. Doubt breeds awkward questions 

I fully understand the economics of software crippling, almost everyone does it, it's nothing to be ashamed of.  From Scopes, to computer processors, to TICs, to you name it.

When I buy equipment for customers I buy based on their requirements and never suggest hacks, of any kind.

[mikeselectricstuff]

Looking at the pricing and features, compared to the Ex0 range and the competition I can't see how they ever expected to sell any E8's.

[Petrlib]

I did some screenshots (VLC -> direct show capture) showing the camera's BETA1 menu/features - examples how it looks here (might vary depending on your caps settings) .. there are some minor issues like when selecting hot/cold spot then the palette gets set to gray (might be related to order of menu items).
I've also noticed that my cam loses lang-settings when cold starting... but no deal-breakers

Attachment: zips contain everything (following postings due to size limits)
Images: peek preview/examples for the zip contents

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg326024/#msg326024 (ADDMENU-BETA1.zip)

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg325934/#msg325934 (extracting RCC)
https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg325975/#msg325975 (repacking RCC)

Will be possible to add some parts of this BETA to Mick hack / new colors pal., condensation, isulation mode /?
Or could you help us - how to add this  to present system??

I like this BETA / watching on downloaded pictures - dont know how to add /

Thank you

[mikeselectricstuff]

I did some screenshots (VLC -> direct show capture) showing the camera's BETA1 menu/features - examples how it looks here (might vary depending on your caps settings) .. there are some minor issues like when selecting hot/cold spot then the palette gets set to gray (might be related to order of menu items).
I've also noticed that my cam loses lang-settings when cold starting... but no deal-breakers

Attachment: zips contain everything (following postings due to size limits)
Images: peek preview/examples for the zip contents

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg326024/#msg326024 (ADDMENU-BETA1.zip)

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg325934/#msg325934 (extracting RCC)
https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg325975/#msg325975 (repacking RCC)

Will be possible to add some parts of this BETA to Mick hack / new colors pal., condensation, isulation mode /?
Or could you help us - how to add this  to present system??

I like this BETA / watching on downloaded pictures - dont know how to add /

Thank you

At some point I was going to do a seperate  add/remove script for this, but wanted to wait a little first to see if there are any issues - I can make it lock up fairly consistently by flicking between palette options, but haven't seen any issue sin more sensible operation.
All you need to do is create a .zip (renamed to .fif)  with the file tree and a camera.cmd to add the files, and another to replace with the originals to un-hack.
A potential issue is that this hack may be more sensitive to different firmware versions than the E8 one.   

[dr.diesel]

I'm sure some people see it that way, it's unfortunate.  What I see if a manufacture that lowered overall costs (or maybe raised stock holder profits and screwed us.  ha ha) by intelligent engineering/manufacturing/integration.

In the end though, I bet your correct and they see it as an attack on their profits, even if it sold more E4s than lost E8s, in terms of actual income.

[nitroxide]

Mike,

The drift in pixels is normal and my PM695's do exactly the same. It is the downside of the micro-bolometer technology, and always has been. That is why cryo cooled cameras are so nice to use.....clean images, no pixel drift to be seen. The calibration routine should become less frequent as the detectors temperature stabilises, but this can take a while as there is hysteresis in the system and some initial over/undershoot on the Peltier module.

The comment on TICs becoming cheaper is interesting. One area that is buying into thermal camera technology is the car industry. Its use is safety of pedestrians/animals and predictive braking systems. We could see a large increase in TIC deployment in cars over the coming years. It may follow the model of other safety systems such as air bags and lane dicipline cameras. That can only be good news for TIC fans as more production into a larger market should bring innovative technology and lower prices ? In the UK we did not have 'expensive' air conditioning as standard in our cars. These days it comes as standard on many models sold in the UK and is not the expensive option that it was. It is also more refined, in the form of climate control 

I read of a new 80x80 mini TIC called the Atom80 that has just been released. It is designed for the less demanding TIC needs of the marketplace. No idea of cost though. Sofradir is a big player in TICs. It looks like a 'camera on a chip' type solution    Could be fun to play with, but I doubt its cheap at the moment.

http://www.sofradir-ec.com/landing_atom80.asp


UPDATE:  To save you registering on the site, I attach the ATOM80 datasheet here.

Could you, or any other that's also interested in the price of Atom80 try make an inquiry at sofradir-ec? I've sent one a few weeks ago and still got no answer...

[nitroxide]

Yeah, I've noticed that. Anyway, they should be able to provide at least a price range if not an actual price estimation. Though, I'll be surprised if it would be under 1K...

[tnt]

Note that AFAIK there is no evidence that :
 - Lenses and sensors are not "binned" depending on their quality
 - The more expensive models could be going through more extensive testing / calibration to ensure their tighter specs

[olsenn]

Making a piece of hardware, especially a test tool, is very costly and time consuming. There's not only the R&D to design the bloody thing, but there is also prototype testing, release candidate testing, compliance and requirement testing, safety approvals, EMC evaluations... It probably took them a few years to just design the one Ex camera. If they had to design an E4, E5, E6, and E8 camera all separately, with all different hardware, then it just wouldn't be feasible.

The big kicker for FLIR (or any company for that matter) is that they want to squeeze the most out of every buyer. They don't want to sell there camera to a person for $1k if you could of sold it to them at $6k. However, they also don't want to not sell a product at all that they could have sold for $1k. This is self evident, but the only way to bring prices down and get CEO's off their fat asses, is to force them to sell for less in order to stay competitive.

[georges80]

Interesting read that implies that people will be upset paying more for un-crippled hardware (e8 versus e4 pricing).

Yet it is very common in the software industry.

You download a large application, you pay a certain license fee and get an unlock code. You want more features, you pay a higher license fee, get a different unlock code and that opens more functionality. It's the same software. People live with this business model (ever priced some of the Synopsys tools??).  Just like you have to pay for FLIR Tools+ features.

Would people be happier thinking the E4, E5.. E8 hardware is free? What you are paying for is the software license fee and it costs more to get the higher resolution.

cheers,
george.

[Taucher]

when menu opened, SD card and wireless icons appear alongside battery  neither is useful

yes, they suprized me as well - seems like as soon as the .rcc file contains the images, they will be shown
I didn't bother (yet) removing them as it might come handy for a further hack

Is Temp scale->unlock useful? seems the same functionality can be had by switching to auto or manual modes

It does the same as the auto mode, true - but it's accessible with one click less
That item would be useful if one of the more advanced modes would work - like setting lower or upper limit and letting the system scale.

Rainbow high contrast palette seems to be same as white hot

rainHC.pal uppercase bug (in Exx series it's UC, in Ex it works just if lowercase hc) - you already found it

I've had it lock up a couple of times while scrolling around palette menu  - after to cold restart by long button press I got an "application error" popup window onscreen during boot. Removing battery seems to fix.
Could this be a missing pallette file  or menu error for Rainbow High contrast?
What I currently have on the E4 is arctic,iron,lava,rainbow and rainHC,.
Also noticed that if I select RainbowHC and exit (when it doesn't hang), on re-entry to colour menu it shows palette as Interval, so I think something may be wrong with the RainbowHC option

dito - it needs it's palette file in lowercase
another specialty seems to be related to delta measurements - seems like those two modes create files according to the stored presets (which didn't exist in Ex but Exx)

I think your latest zip also omits the rainHC.pal file - I have it on my E4 but wasn't on the original files I backed up a while ago so not sure where it came from - maybe an earlier version of the menu hack zip?

just checked the zip - rainhc.pal in lowercase included in Beta1 while Alpha up to 3 did include an uppercase version... wasn't windowsCE supposed to be case insensitive? (I suppose some Java Script inside the menu files isn't)

[Lukas]

Any recommendations where to buy the E4 in Germany? Many resellers have it for 995€+VAT, whereas it's 995$ (=745€) in the US Does Tequipment ship to Germany?

[ixfd64]

Any recommendations where to buy the E4 in Germany? Many resellers have it for 995€+VAT, whereas it's 995$ (=745€) in the US Does Tequipment ship to Germany?

I don't think they do. However, you could probably ask a trusted U.S. forum member to buy one for you, and then given them the money and cover the shipping costs. You can also ask them to ship the camera separately from the rest of the package if you're worried about nosy customs officials.

[Taucher]

I did some screenshots (VLC -> direct show capture) showing the camera's BETA1 menu/features - examples how it looks here (might vary depending on your caps settings) .. there are some minor issues like when selecting hot/cold spot then the palette gets set to gray (might be related to order of menu items).
I've also noticed that my cam loses lang-settings when cold starting... but no deal-breakers

Attachment: zips contain everything (following postings due to size limits)
Images: peek preview/examples for the zip contents

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg326024/#msg326024 (ADDMENU-BETA1.zip)

https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg325934/#msg325934 (extracting RCC)
https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg325975/#msg325975 (repacking RCC)

Will be possible to add some parts of this BETA to Mick hack / new colors pal., condensation, isulation mode /?
Or could you help us - how to add this  to present system??
I like this BETA / watching on downloaded pictures - dont know how to add /
Thank you

the insulation/condensation modes seem to be a bit pointless gimmicky to me (but nice colors) as really useful modes would take into account room humidity, relative temperature etc... in principle every owner with a "below" mode can already perform the same thing - just color anything thats either colder than 9,3°C(condensation) or colder than 18°C (insulation)

open the ADDMENU-BETA1.zip from the referenced posting - there's a short readme

BETA
-------------------------------------------------------------------
Disclaimer:
   Only for educational purposes,
   Use on own risk,
   Know what you're doing
   ... keep backups
-------------------------------------------------------------------

This archive contains a mashup of published Ex and Exx
files which perform the following changes, compared to a
regular Ex series cam:

- Change Bootup logo (9Hz version is not export restricted)
- Make hidden menu accessible (menu: Device Settings)
- Make FLIR logo in lower left corner 100% transparent
- Add quick-access to USB-Mode menu (also Device Settings)
- Enable PIP in two sizes
- Add extra options at the measurement-menu:
  - Center/Hot/Cold combined mode
  - Delta center-hotspot
  - Delta center-user defined temperature
  Note: not tested with advanced measurements enabled (tends to be crashy)
- Add extra palettes
  - Rainbow high contrast
  - White hot (=gray)
  - Black hot (=gray inverted)
  - Arctic
  - Lava
  - Condensation (color anything below 9,3°)
  - Insulation (color anything below 18°C)
  - Interval (user defined range - use cursors to set)
- Add "manual" range to temperature scale options
  - has 3 submodes: Min, Max, Both (use cursors)

Installation:
You need to have the Ex camera in FTP-accessible mode,
then simply overwrite the corresponding filestructure.
(Filezilla suggested).

Then cold boot your camera or use restartapp via telnet.


Troubleshooting:
Some features can require .caps modifications
(not documented here)

ok, maybe the installation part needs a readup across this thread

- apply mike hack
- put camera into IP mode (RNDIS) (hold "right" button on version screen for >10s = hidden menu)
- access camera over IP (default gateway IP of "infrared" interface = camera - usually 192.168.0.2 ... but had seen own_ip+1 scheme too)
- use filezilla to backup anything readable - esp. the FlashBFS folder
- overwrite the FlashBFS folder with the attached one, do not remove the old one as the new zip just adds or overwrites certain stuff
- turn off camera, remove battery, repower

have fun, report back

[Taucher]

Mike,

I saw the ULIS site...very interesting stuff there, Note that the micro-bolometer does not need a Thermo-Electric Cooler so they have managed to get rid of the Peltier module and maintain temperature stability ?

maybe they secretly replaced the bolometers with miniature thermopiles? *who knows*

A thermopile would be happy to deliver either positive or negative voltage - always in relation to it's own temperature vs. the incident temperature (IR is the predominant way of heat-exchange - so a lack of inbound IR means that the sensor radiates heat away... )

[G0HZU]

Has anyone worked out how the 'manual' focus operates yet?

I made a tool to rotate the lens surround yesterday but I'm uncertain which way to turn it to get closer focus. I'm nervous that something will come loose or run off a thread etc.

Also, does the lens move relative to the bit that rotates? i.e. does the little lens 'poke/protrude' and the surround stay in the same place (if that makes sense)

Or does the whole assy move in and out together?

I'm concerned about the lens poking forwards and hitting the tool because this first one isn't hollow and there isn't much clearance.
The good news is that the tool fits really snugly and it's easy to rotate it smoothly. But I've only moved it maybe a quarter turn each way.

Someone already asked how many turns are needed and nobody replied. So can I ask the same question? How many turns will I need to turn the lens for maybe 6 inch focus? ... and which way do I turn it?

[Mark_O]

What I am saying is that this hack does create doubt in customers minds as to whether they are seeing value for money from FLIR.

...but I wouldn't expect them to welcome the news that they firmware hobble their high end products to create low end versions. It just sits poorly with the customer base. No one likes to think they have been overcharged. Nothing more than that.  ...

I am just an end user, so maybe I am wrong about all this ?

I don't think you're wrong about end-user perceptions, that may be held by some customers.  In fact, I'm certain you're correct.  There's a group of folks who think that anyone engaging in such practices is ripping off their customers, plain and simple.  In their view, since it's physically possible to create the hardware for an economy model E4, that's substantially (if not completely) the same as the E8, there should be NO model variations.  Just an E8, for $995.

Of course, the fact that such a thing would be impossible is of no concern to them.  They want what they want.  The thing is that the ability to sell one level of the product at a higher price, helps recoup substantial R&D costs.  And being able to sell another variation, with fewer capabilities enabled, let's them sell at higher volumes, which brings down manufacturing costs.  But please, don't bother them with realities of the development/manufacturing process.  Or offering customers different options, at different price points.  That's just "devious".

It did not sit well with the customer base and is still quoted to this day as an example of devious sales tactics and the potential consequences.

In my opinion, Flir (and other companies exercising similar practices) are doing nothing devious at all.  The same thing is going on in many area of tech development.  You need look no farther than other topics in the EEVblog Forums, for oscilloscopes sold for ranges of 5:1, that are essentially the same device.

The fact is that if Flir could sell only a single model of their E series, it would cost nowhere near $995.  Nor would it be 6-grand.  The price would likely (just a WAG) be in the $3-4k neighborhood.   I guess everybody would like that better.  That would be much more "fair".     After all, everyone's needs are the same, and people who don't need the higher performance or special capabilities should be forced to pay for it anyway.