High Gain/Low Gain

[Jonathan Hodges]

Hello everyone. Many FLIRs have the capability to swap calibration ranges from -40C-150C to 100C-650C by changing from high gain to low gain. Does anyone know if the Lepton supports this functionality?

[Max Ritter]

Very interesting question ! There is a register "LEP_CID_RAD_GLOBAL_GAIN" in the radiometry part of the Lepton, but I am not sure how/if it affects anything on the Lepton2.

[Jonathan Hodges]

Looking through LEPTON_RAD.h on the purethermal github, I see lines 90-91 have a comment for High Gain and lines 136-137 have a comment for Low Gain. This makes me believe that the Lepton indeed supports two calibration ranges.

I will play around with it some more, but this is pretty far outside my experience. I'm currently using the Arduino Due setup if anyone has ideas.

[Max Ritter]

Yeah, also noticed that! But I looked through all the functions over and over, and was not able to find any matching one. 

As the calibration stuff does not work on the Lepton2.0 anyway, maybe the Low/High gain switch will be accessible with the release of the Lepton 2.5

[Jonathan Hodges]

I think the functions are in LEPTON_RAD.c lines 266-415. However I have not been able to find where they are being called.

[Max Ritter]

I tried some of them, and they are not helpful on the Lepton2.0, b/c the hardware revision simply does not support them. They will be implemented with the next HW revision that does add all the calibration stuff then.

For example, all the RBFO registers (which should deliver the Planck constants for flux to temp conversion) only deliver the standard values, which are R: 395653, B: 1428000, F: 1000 & O: 156000.

You can go ahead and enable radiometry mode with register 0x25 for SET and check it with 0x24 for READ, the raw values will change their offset then. This should compensate the ambient temperature (see section 8.4.2 of the datasheet), but it's not very accurate. I find myself more comfortable using an external temperature sensor and comparing those value with RAW value 8192 to calculate an offset (figure 10 of section 8.4.2: 8192 corresponds to the ambient/camera temp).

RAD_GLOBAL_GAIN & RAD_GLOBAL_OFFSET deliver changing values when radiometry mode is enabled. I suspected the gain value would be Planck constant R2 and the offset would be O, so I spent some time to put that into the conversion formula. But it's not making any sense, so I stopped there.

[Jonathan Hodges]

I realize it has been awhile since I posted this, but I have been busy with other projects. I contacted FLIR and they said low gain/high gain support will not be added until the next lepton version. However, they recommended using a neutral density IR filter to increase the temperature range. I decided to use this filter from Thor labs to see what temperature range I could measure (http://www.thorlabs.com/thorproduct.cfm?partnumber=NDIR10B). I've attached calibration curves generated with and without the filter. Heating was performed using radiation heating panels set to emit a known heat flux and surface temperature was measured simultaneously with a Lepton and a FLIR A655sc. Note, these are estimates as ambient conditions and internal camera temperatures will change the calibration curve, but it should give a general feel for the capable temperature range.

[Jonathan Hodges]

And I realized that I forgot to actually attach the curves. Here they are.