FLIR Exx & Txxx 2000 Celcius Temperature range doubler - T197993 / T199235

[Fraser]

I recently purchased an unusual accessory for my FLIR Exx series camera and thought others might be interested in it.

I often search for FLIR supplementary lenses but on this occasion I found a supplemental temperature range doubling "lens" for my FLIR Exx series camera. I had not seen one of these adapters before so it immediately caught my eye. It looks like a normal Exx series supplemental lens but it is a very different beast inside that barrel.

What I had found was the FLIR High Temperature Adapter for Exx and Txxx series cameras. It is available under two part numbers, T197993 and T199235 as is common with FLIR accessories that serve more than a single product line. Bothe part numbers use the same hardware and lens mount.

The adapters specifications are as follows:

Compatability: FLIR Exx and Txxx series cameras (Calibration required)
HFOV: 25 Degrees
Min Focus: 0.4m
Focal Length: 18mm
F Number: 1.3
Transmission: 50% (2um to 16um)
Temperature range : 200C to 1200C (2000C)

Now these specifications are a little misleading to those without the adapter in their hand. At first glance the sight of HFOV, Min Focus, Focal Length and F number might suggest that this is some form of lens. This is not the case. It is a flat piece of Germanium, like a window, but it is coated to form a reflective-absorptive Neutral Density FILTER that has a transmission characteristic from 2um to 16um of 50%. Only half the energy that enters the filter, passes out of it and into the cameras optics.
This ND Filter is rated for operation up to 2000C but, as a range doubler, it would require a camera with a 1000C Max Range to achieve this and thermal protection might be required for the camera and operator. On my E40(E60+) the adapter increases the maximum temperature measurement of the camera to 1200C.

OK so this is an ND filter with a 50% transmission figure..... you can slap it on any thermal camera and it will double the cameras temperature range right ? ...... ummm sadly no, life is never that simple 

This adapter halves the energy that enters it in order to permit a thermal camera to view and measure temperatures twice that normally available with the standard built in ranges. but, and it is a big BUT, the Planck curve needs to be considered. Using the ND filter the Planck curve of higher temperatures is different to that in the cameras normal calibrated measurement range. See the attached Planck Curve for a number of temperatures. Just fitting the filter to a camera and doubling all measured readings would result in an error. This error could be calculated by the user familiar with Planck curve tables but that is a royal pain the in the derriere ! It is far better to have a camera compensate for the use of the ND filter and to be calibrated to provide accurate measurements. This is the path that FLIR have taken. The Exx and Txxx cameras for which this adapter was created contain support for it and the required compensation tables to provide accurate temperature measurements from +200C up to +1200C. FLIR state that the camera should be calibrated with the adapter and this is understandable. The filter option is not even visible on some cameras and it needs to be enabled before the filter is used. It may also be necessary to carry out a two point calibration on the camera whilst the filter is in the optical path as a reference against which to apply the corrections to measurements.

So it is not just a simple case of slapping one of these high temperature adapters onto your camera and getting accurate measurements up to 2000C  I am OK on this front as I have Exx cameras that support the ND filter adapter and contain the required Service Menu with full 'Lens' calibration routines  I also own the required accurate Blackbody calibration references to calibrate the camera with the ND filter attached 

So how much does one of these high temperature adapters cost from a UK FLIR Agent ? Expect to pay around £2K plus the cost of calibration of the camera by FLIR  Not exactly cheap then  I looked at the cost of an equivalent 2um-16um 50% transmission ND filter from one of the specialist providers and this large diameter ND Filter would cost around $700 +TAX and is just a bare filter with no mount ! Eeek !

Those of you who take a close look at the pictures I provide will notice something strange about the FLIR High Temperature adapters appearance. Yes its casing looks normal enough, but look at the filters mounting....... it is angled ! This is not a fault due to transit damage, nope this is deliberate on the part of the designer. This angling of the filter makes for a complex filter holder design. So why is the filter angled ? In truth, I am not certain but I have met angled Germanium windows in thermal cameras before. I am thinking it may be to do with combating Narcissus effects or deflecting radiated energy away from the target rather than back at it "on-axis". remember, we are potentially dealing with high energy levels at close range when using this high temperature adapter. I am also aware that angling a filter can change its characteristics and presume this has been taken into account by the designer. I am nor expert on reflective-absorptive optics !

Well that is about it for this introduction to the FLIR High Temperature Adapter type T197993/T199235. I have never before seen such an adapter so they are likely not that common. I am pleased to have added it to my Exx camera kit though  A FLIR Exx 10mm wide angle lens will also be arriving soon to keep it company in the case  I have been very lucky finding the adapter and the wide angle lens in the last couple of weeks

Fraser

[DaJMasta]

Sunglasses for your thermal camera?  Neat 

[ixfd64]

According to the E60 specs, the maximum supported temperature is 650° C: https://flir.com/support/products/e60

There is no mention of any extended ranges. Does the software still display temperatures up to 1,200° C despite what the specs say?

[Fraser]

The E60 MK1 User Manual does not mention the extended range option, but the MK2 User manual does.
I checked the MK2 firmware and FlashFS contains the support for the High Temperature Filter in the calibration configuration. Upon checking in the same place on my MK1 camera, I found the same entry.

See page 22 (section 6) of the E60 MK2 manual:

https://www.flir.com/globalassets/imported-assets/document/flir-exx-bx-series-bx-user-manual.pdf

The HT Filters data sheet (provided above) also states Exx MK2 series compatibility 

The filter option is enabled in the firmware in a calibration configuration file that I found. I am assuming that this will make the calibration routine for the filter available to the Service Menu. It may require enabling in the EEPROM via the Service menu as well. I will not know until I look into it further. I have a second E40 that needs to be upgraded so I will play with that unit and test my thoughts on the enablement and calibration of the HT filter option

The HT Filter appears to have been made available as an accessory in 2015. It was released well after the end of MK1 production. I suspect the firmware of the Exx MK1 camera contains the HT Filter support as a result of either a planned HT Filter accessory release, planned support for customers future needs or availability of such a filter for another camera series that uses the same basic firmware build.

Fraser

[bap2703]

What I understand is that there is no antireflection coating, as the material look plain grey.
That's a cheap way to get the desired attenuation.
But that means there's a lot of reflections and you don't want them to be "on axis", otherwise that reflected light might bounce somewhere else and come back on your sensor.
A common way to avoid that phenomenon is to tilt one surface.
It also works against interferences in the window.

Look for "wedged window" in any optical shop.

[Fraser]

Bap2703,

That is interesting information. My memory of tilted Germanium Windows was in connection with scanner type thermal cameras and from my recent investigations into their use, it turns out that scanner type cameras suffer from narcissus effects and tilting the window deals with this issue effectively.

Interesting that you mention no AR coatings. I am no expert in this area but have learnt that the ND filters are reflective and absorptive. If totally absorptive the year build up in the filter would be a very bad idea, especially with a temperature sensitive Germanium substrate. I understand the Germanium is actually coated with a very thin layer of Nickel or a metal alloy to produce the desired transmission figure. As you say, no mention of any AR coatings.

For anyone interested in these ND Filters.......

Edmund Optics page detailing their ND filters for MWIR and LWIR.

https://www.edmundoptics.com/f/infrared-ir-neutral-density-nd-filters/13765/

Newport page:

https://www.newport.com/f/infrared-neutral-density-filters

Thorlabs page:

https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=6007&gclid=CjwKCAiAyeTxBRBvEiwAuM8dnZmF_Tgu_Hk2d8dcpuXk5guSnmGBM8jmlbZsx7GYG8beo8tND1aU9RoCFVQQAvD_BwE


Fraser

[Fraser]

This site is good, with a nice range of ND filters offered.

https://www.andovercorp.com/products/neutral-density-filters/infrared-nd-filters/

Now Andover Corp do offer AR coatings on one side of the ND filter on some select filters that they offer. They make specific mention of internal surface reflections as a potential issue. As has helpfully been said by Bap2703 the angling of the filter would help in this scenario.

Fraser

[Fraser]

 Also worthy of note,

The 0.3 ND filters (50% transmission) are shown in the various manufacturers transmission vs wavelength plots as having a transmission characteristic of 46%. Hmmm that sounds familiar..... yep that is about the nominal transmission figure of an uncoated Germanium window. I now realise that Bap2703 was saying this ! For lower transmission filters the nickel or alloy metal coating is added to provide more accurate characteristics, but when you want a 50% transmission figure, it actually exceeds the capabilities of pure Germanium that is not AR coated. 46% is the best that you can do easily.

Interesting stuff because I once thought of buying a plain non AR coated Germanium window to use as a ND filter for range extension 

I attach two plots of IR ND filters showing the 0.3 (50% transmission filter) with its true 46% characteristic. I also attach a plot for pure uncoated a Germanium that also shows a transmission figure of 46%.

Fraser

[Fraser]

For $40 you can buy a 1” diameter Germanium window that has no AR coatings. That would seem to be a good candidate for a DIY thermal camera ND filter. All my previous comments on the camera calibration and Planck curve still apply of course  For some users, measurement accuracy may not be that important. If so it is a cheap “range extender”. Sadly such a parallel surface ND filter will not work as well as a wedge type, even if angled, but it may be ‘good enough’

https://www.ebay.com/itm/1-inch-diameter-HAND-POLISHED-Germanium-Metal-Disk-99-999-5g-Element-Ge-Sample/232258898858?hash=item3613b58baa%3Ag%3An7YAAOSwal5YDrWh&LH_ItemCondition=3

Fraser

[bap2703]

In case a calibration is required anyway, why not just reduce the aperture of the lens ?
For example by 3d printing a part that would fit as close as possible to the lens.
You can easily halve the light coming from the scene. However the out of focus aperture will radiate and create a somewhat homogeneous "glow" on the image.

[Fraser]

bap2703,

Indeed, that is effectively what was done within the 3 element lens assembly of BST core fire fighting cameras to increase their maximum temperature capability with limited dynamic range. The iris sits behind the front lens element though. There are many ways to reduce the cameras sensitivity. I even used a standard Tiffin photographic glass single coated Haze filter to thermal image gas jets and it worked well. I was not interested in measuring the temperature of the gas jet plume though. ND filters are presumable a reliable and simple means of controlling transmission for calibrated use. As you say, a fixed iris might also work. I have seen that several of my FLIR cameras contain a calibrated aperture that is brought into the optical path when in the higher temperature ranges. Those fixed apertures are carefully positioned within the optical block however. placing one directly in front of a large thermal camera lens might be more challenging in avoiding unwelcome effects however ?

Fraser

[Bill W]

The aperture simply needs to be somewhere afocal, ie not reducing one part of the field of view more than another.  As such covering part of the front element is perfect.

I recall the Bullard BST camera (T4) has a 'Manual Thermal Throttle' function, otherwise known as a manual iris.  When the inevitable complaints came from sales as to why we ( = Argus 2) did not have one so they could compete in tenders, they were surprised to be told we had a fully automatic thermal throttle. 

Bill

[mahony]

Agree to Bill W - a nice side effect: you significantly increase the depth of field too. This may also be of interest for some application.
Sharp edges on the aperture and centering on the lens do help with image quality. Cheap and easy ...

[Fraser]

Thanks chaps 

Very helpful information.

I wonder why FLIR chose the ND filter route if a simple thin calibrated aperture plate would have worked ? As I said at the beginning of this thread, I have not seen many ND filters used in thermography which is why this accessory stood out for me. I have an IRISYS thermal camera that has what is effectively an ND filter attachment to extend its temperature range. IRISYS used a plastic membrane however.

Fraser