Flir E4 Thermal imaging camera teardown

[Fraser]

Interesting experiment.

I purchased a used reflector type telescope from the local Charity shop for just such experimentation but have had no time to play lately.

I recommend that you do not remove your E4 lens as dust on the micro-bolometer is a real PITA to sort out.

You may be interested to hear that Inframetrics used to offer a thermal camera telescope based on the compact reflector lens structure often found in physically short, high magnification SLR telescope lenses. I will see if I can find a picture of it.

Sadly SLR reflector lenses are of no use as they use a reflector that may or may not be surface metalised and glass lenses are used in the design to shape the image for the camera body. Pity.

UPDATE: Pictures of Inframetrics X10 reflector lens added. Expect to pay in the order of $1500 for such a lens, and that is very cheap considering its original price !

[mikeselectricstuff]

Nice little lens on Ebay UK :


http://www.ebay.co.uk/itm/200mm-F1-0-Germanium-Thermal-Imager-Lens-Rank-Taylor-IRTAL-Super-Fast-Objective-/271429492535?pt=UK_Lenses_Filters_Lenses&hash=item3f32755f37

[muvideo]

Interesting experiment.

I purchased a used reflector type telescope from the local Charity shop for just such experimentation but have had no time to play lately.

I recommend that you do not remove your E4 lens as dust on the micro-bolometer is a real PITA to sort out.

You may be interested to hear that Inframetrics used to offer a thermal camera telescope based on the compact reflector lens structure often found in physically short, high magnification SLR telescope lenses. I will see if I can find a picture of it.

Sadly SLR reflector lenses are of no use as they use a reflector that may or may not be surface metalised and glass lenses are used in the design to shape the image for the camera body. Pity.

UPDATE: Pictures of Inframetrics X10 reflector lens added. Expect to pay in the order of $1500 for such a lens, and that is very cheap considering its original price !

Nice Images, thank you!!

You are right that SLR lenses can be of no use, since usually
have refractive elements in the path.
I was thinking to cheap Newtonian telescopes, that have
reflective only optics. The downside of these optical arrangements
is that are bulky, and not well corrected for field flatness, coma or
spherical aberrations.  But long wavelengths are more forgiving
than visual, and a small sensor like the one in E4 will see a flat field.
It remains to be seen the behaviour of these mirrors on thermal IR.
I updated my previous post with a quick test, using an "ocular" lens.
I'm confident that direct imaging (without E4 lens in the way) will
give good results, but the dust on the bolometer is a big danger.

I definitely need to find a cheap thermal imager for experiments,
my E4 is needed to be in perfect shape for all other things.
Unfortunately my alerts on ebay for firefighting cameras arent
showing much hits.

Mike: I made a 100GPB offer, who knows

[treefiddy]

Marphy , your app just worked for a 1.22.0 full hack and a 1.19.8 menu hack.  Beautiful work.

BTW, are y'all still compiling the list of serials/fw/calibration dates?  Thanks guys.

[DaveWB]

Marphy , your app just worked for a 1.22.0 full hack and a 1.19.8 menu hack.  Beautiful work.

BTW, are y'all still compiling the list of serials/fw/calibration dates?  Thanks guys.

Yes it can be found here https://www.eevblog.com/forum/reviews/flir-e4-the-useful-information-thread/msg332428/#msg332428

[lewis]

Nice little lens on Ebay UK :

Nice!

I've come across that seller before and he is hugely overpriced, some of the stuff he has has been on for literally years.

[Sk_rmouche]

Hi from France !

First post here, please excuse my mistakes in english  just got my modified E4, I've been using before a Flir i7 and the extra functions offered by this mod are quite useful. No need to go with the Flir Tools software to deal with advanced measurements tools and other palettes. Thanks to Mike with his incredible work !

I've compared the E4 with the i7 and it seems that Flir made a lot of improvements with the image noise.

I made this to compare :

i7 :




E4 :




Like the E4' images are softened ... ?
The enlarged FOV is quite a good news for me, I work as a HVAC technician and we almost never have space to check the equipment, I already love my E4 for that !

[Fraser]

Hi Sk-rmouche,

Welcome to the Forum.

The wide FOV of the E4 has attracted much positive comment amongst professional thermal camera users. Many suffered the same 'lack of space' challenge as yourself when using a TIC. The only option open to them on other FLIR camera models was a VERY expensive additional 45 Degree lens. Such an optional lens actually costs more than an E4 !

I am used to using industrial FLIR cameras such as the FLIR PM series. These come as standard with a 24 Degree FOV, and to widen it to 45 Degrees requires a large auxilliary lens and lots of cash. I am fortunate enough to own both the 45 Degree and 12 Degree optional auxilliary lenses but I would have hated to have had to pay the retail price on them when new.

On the down side, the 45 Degree standard FOV can be a little wide for some applications and so a telephoto lens is needed to narrow the FOV. As the E4 is not designed to take auxilliary lenses, some creativity is needed when it comes to adding one, as in the case of the ZnSe close-up lenses. With the arrival of the electronic Zoom function in the new 'taucher' menu at least it is possible to halve the FOV by selecting X2 Zoom(at the cost of losing optical resolution), and this often meets my needs when I cannot get close enough to a target when operating mobile.

All in all, I am very pleased that the E4 has the wider FOV. It makes it a very versatile unit for most users.

[Sk_rmouche]

I agree with you, optional lenses are highly priced for me ( and certainly many others ), I have X2, X4 and X8 electronic zoom, and both X4 and X8 are quite unusable, unless accuracy and sharpness aren't required !

Maybe it can be useful to occasionally detect heat elevation or moving part at something like 10-15 meters... But for the price of the E4, it's more than I could afford with an official E8, and clearly a nice update for my i7.

By the way, I saw on the web that i7 sensor is actually 320*240 pixels, but stuck @ 140*140 for rev 2 model, there's nothing new about that right ?

I'd love to play with InSb cooled sensor, but you don't want to look the price :p

[Fraser]

I have worked with cooled thermal cameras, and they do provide superb sensitivity and a very clean image. When I first saw the FLIR PM570 demonstrated, I thought it a very noisy image and was not impressed. I was used to a superb PM550 with its striirling cooled detector.

The thermal cameras that require cooling sadly have their drawbacks, as I am sure you already know. The ones that require liquid Nitrogen for cooling are a PITA to run, and the ones that use bottled Argon gas discharge cooling are similarly so. The Peltier cooled cameras are 'OK' but not as good as other cooled options. The ultimate for me was a stirling cooler that provided the required low temperatures of around -196 Degrees Celcius, BUT they are mechanical coolers that require re-gassing and rebuilding all too often. Such maintenance can be very expensive. FLIR used to charge around GBP5000 per service. For this reasson I have avoided buying cameras that have stirling coolers as their accumulated run time, time since last service, and condition of cooler is often not known. These cameras become very expensive paperweights if you need a cooler rebuild, or even a re-gas.

I have a lovely, 'as new' Agema 880 industrial thermal camera.....it is Liquid Nitrogen cooled, but as I have no access to Liquid Nitrogen, it is just a memento of the early days of thermal imaging, when I used just such a camera to good effect with its associated BRUTE computer and Husky computer. My poor 880 sits in its huge metal case, along with the controller, monitor and Husky, in my garage. One day I will decide what to do with it. It is actually more useful and valuable as a source of Germanium lens elements as the camera has large, high quality lenses that work in the Long Wave band.

[Sk_rmouche]

Yeah, one way to go if I had some extra €€€ is to buy used cryocooler on eBay, to make LN2 at home, and then use it when I want crystal clear image, but anyway I don't have cooled thermal camera.

The Flir T640 has a nice 640*480 bolometer, and I think it could be sufficient for a lot of purposes. Does bolometer operate with less noise as it's cooled ? I know that the E4 use TEC element to heat up the bolometer around 30°C, just to avoid drifting, don't remember if Mike actually shows it in the tearing video ...

[Fraser]

The microbolometer is an inherantly noisy thermal detector but manufacturers are using various techniques to 'filter' some of the noise from the images these days. I have not met a cooled microbolometer yet. Most microbolometers wer temperature stabilised but FLIR recently advised me that the E4 does not have a thermo-electric cooler (or heater) to stabilise teh sensor. They are working on calibration maps that extrapolate microbolometer data corrections based on the ambient temperature measured on the chassis. This is why Mikes camera went crazy when he sparayed freezer spray on that chassis sensor.

The FLIR E4 is built to a price, and a low one at that. I can forgive the removal of temperature stabilisation as it meets the needs of the intended market without such. I use my PM series cameras for work where more temperature accuracy is needed.

The cooled sensors are usually semiconductor based and these need the cooling in order to reduce the level of their internal noise that would otherwise mask the energy that they are receiving from the lens. Similar techniques are used on professional radio telescopes with super low noise front end amplifiers etc. Liquid Nitrogen temperatures make seniconductoors nice and quiet. My Thermocam 880 at room temperature, without the Liquid Nitrogen, does not produce any form of thermal image, just rando noise.

I investigated buying some Liquid Nitrogen fron BOC in the UK but the Dewars are very expensive here, and there is the ever present risk of killing yourself through hypoxia if a major spill occurs ! I'll stick to microbolometers and BST sensors for now.... nice and safe.

[mucek]

Most microbolometers wer temperature stabilised but FLIR recently advised me that the E4 does not have a thermo-electric cooler (or heater) to stabilise teh sensor. They are working on calibration maps that extrapolate microbolometer data corrections based on the ambient temperature measured on the chassis.

As Ex and Exx cameras obviously have the same sensor, but different (specified) thermal sensitivity (E30: < 0.1 °C, E40: < 0.07 °C, E50/60: < 0.05 °C - source: http://www.flir.com/cs/emea/en/view/?id=41959) do you think, that they also have different correction maps? If so - does i.e. hacked E30 to E60 still have "old" thermal sensitivity, but higher (image) resolution, or does the hack also "unlocks" maps for higher thermal sensitivity? (probably yes, as the firmware is the same for all series)

Regards,
Gregor

p.s. A newbie question: would it be possible to have some "ND like" filter for thermal camera in order to measure (actually view only - don't need to measure absolute temperatures) higher temperatures? (with original camaera only up to 160 degrees)

[Sk_rmouche]

but FLIR recently advised me that the E4 does not have a thermo-electric cooler (or heater) to stabilise teh sensor. They are working on calibration maps that extrapolate microbolometer data corrections based on the ambient temperature measured on the chassis. This is why Mikes camera went crazy when he sparayed freezer spray on that chassis sensor.

Oh okay, my mistake  I was believing that Flir is actually using TEC element to control the module temperature, if it goes above 30°C, TEC is cooling, and under 30°C, reversing polarity and TEC in heat mode.

But then, why is there a ~ sign before the temperature, around 30 seconds after power on ? There's not such thing with my i7. I should compare that on a fixed heating resistance 

[Fraser]

 Sk_rmouche,

I thought along the same lines as you and believed that the E4 would have TEC temperature stabilisation. I stated that belief here in the forum and Mike rightly commented that there did not appear to be enough room behind the micro-bolometer wafer for a TEC. I decided to ask FLIR CS for their comments on the cameras temperature stabilisation method. Their response surprised me. They stated that the micro-bolometer naturally sits at around 30 Degrees C and this is why the temperature calibration tables are centred on that  temperature. No TEC is used in the Ex series.  I can't say that I was exactly happy to hear the news, but this is a budget camera after all...... what about the E8 that costs a lot more though......no temperature stabilisation is not great when working at narrow temperature spans as I do.

Regarding the Tilda (~) symbol, it is believed that this is shown for a period until the electronics and micro-bolometer rise to their nominal operating temperature. Only then will the calibration and temperature correction tables be valid, due to the lack of the TEC.

Take a look at this post and those that follow: (#112 onwards)

https://www.eevblog.com/forum/reviews/flir-e4-the-useful-information-thread/msg412095/#msg412095

[Sk_rmouche]

Thanks for the link, that's helpful. So lenses of Ex series aren't made of Germanium 

I should open my i7 to see if it's the same way about TEC stabilization or not.

[Fraser]

Not an answer but maybe of interest to you:

http://www.systemplus.fr/wp-content/uploads/2013/04/MEMSTrends_2012_10_N12_systemplus.pdf

http://www.i-micronews.com/upload/Rapports/Yole_FLIR_ISC0601B_Micro-bolometer_Sample.pdf

Note that the Ex series use a completely different micro-bolometer format to that of the ix series, and there is no hermetically sealed hard vacuum module, as the vacuum is integrated into the micro-bolometer wafer layer structure. The E4 is using micro-bolometer technology developed as part of as cost reduction process for automotive TIC applications. I believe I covered this and the lens details in my "E4 Useful information" thread. Worth a read ?

[Iphone_hack]

Hi guys
First time posting
I have been looking at this E4 hack for few weeks
Worry that I missed the opportunity to get a E4 with performance of E8
So today I went to order one from my local store. $1000 Canadian
Might take 2 weeks but I can't wait, keep my finger cross that at least it is 1.21 or 1.22 where can still be hacked.
I was waiting for Flir one but still not out yet and only VGA resolution.
My worries is anyone have any problems after the hack?
And can it be reversed to original factory setting in case of warranty issue?
Thanks

[Gallymimus]

Hi guys
First time posting
I have been looking at this E4 hack for few weeks
Worry that I missed the opportunity to get a E4 with performance of E8
So today I went to order one from my local store. $1000 Canadian
Might take 2 weeks but I can't wait, keep my finger cross that at least it is 1.21 or 1.22 where can still be hacked.
I was waiting for Flir one but still not out yet and only VGA resolution.
My worries is anyone have any problems after the hack?
And can it be reversed to original factory setting in case of warranty issue?
Thanks

Good question for the Q/A newbie thread:
https://www.eevblog.com/forum/testgear/flir-e4-teardown-qa-and-newbie-questions/

[Kay]

Hi,

anyone knows if the E4 units that ships from Flirdirect as of now, May 2014,can still be hacked and what firmware they run?
Thanks for any input.

[echen1024]

Hi,

anyone knows if the E4 units that ships from Flirdirect as of now, May 2014,can still be hacked and what firmware they run?
Thanks for any input.

It is prolly still 1.21/1.22, so yes, it can be hacked.

[Fraser]

FLIR handed this community a wonderful gift by using a 320x240 micro-bolometer in all its Ex series camera, and then not protecting the software configuration hobbling adequately. Sadly the newer low resolution cameras, such as the FLIR One will likely use a dedicated low resolution micro-bolometer, or one that is hardware crippled to prevent users upgrading them. It is my belief that FLIR have learnt a valuable lesson here, and will not repeat the mistake on new hardware. The FLIR Ex series (and some other earlier cameras) will likely be the beginning and the end of a superb opportunity for everyday users to own a high budget TIC capability on a relatively low budget. Thank you FLIR 

Anyone still sitting on the fence wondering whether the E4 is a good buy should think very carefully about the above. He/She who hesitates is lost 

[tjb1]

I already almost missed the chance when I was on the fence and they put out 1.22 luckily the forum came through with the mod for that and I finally jumped off the fence   I do agree, do not pass this chance up.  It may never come up again.

[mikeselectricstuff]

I wouldn't  be surprised if at some point the E4 gets replaced by something with a Lepton core

[ixfd64]

The hacked E4's on eBay are probably what prompted FLIR to patch the original hack. However, I still don't think FLIR has much to gain by making new hacks impossible. It's not like the eBay buyers would have bought an E8 otherwise.