Review - Thermal Master THOR-002 Thermal Camera - Better than FLIR ? by Fraser

[GreyWolfe111]

New Post created for the 3D Printed Macro inquiry and interest:

https://www.eevblog.com/forum/thermal-imaging/thermal-master-macro-len-3d-printed/

Hi,

I found this forum when looking up info on purchasing my new Thor 002 and found it again when looking up macro accessories and potential of what I could do and  if I could add a macro lens to the Thor 002. For the $250 difference of the 001 and 002 models. I could definitely try and create my own.

ONE thing I should note, is that depending on marketing and how the brochures are printed... they can be misleading on operating hertz. The sensor IS 24hz... marketing sometimes lists this as 60hz. This is the VIEW SCREEN refresh and not the sensor stats.

First I looked at what other thermal cameras had for macros and if I could adapt one of the cheaper ones to fit the Thor or use their lenses in my own creation. After a few weeks of measuring, modeling, prototyping and trials. I have come up with a bit more advanced solution for the Thor002 and printed my own using my Bambu X1C filament printer.

I was able to get a19.05 dia - 60.3mm ZnSe PVD CO2 focus lens off Amazon and currently have a 160mm CVD on order to try and will keep an eye out for a 25mm or 38mm for testing on magnification and focus distance to target. I originally went with the 60mm focal distance so I didn't have to be soo close to focus in. But I think it ended up being less of a zoomed image. I will let you know how the 160 does. If you are patient, you can pick one up at an affordable cost.

Even tho the Focus lens state they allow 99% transmission of thermal radiation the readings appear to be about 4% difference. Germanium lens are listed as losing about 10% transmission thru the crystal so not sure what that would end up being for temperature differences. The Thor 001 version is more due to he cost of the germanium lens material and the firmware differences to account for automatic conversion and handling of the difference in readings. For my purpose, a bit of temperature variance isn't going to change much, as its more the visual indicators that I was going for. Along with the larger viewscreen and higher resolutions.

This version has about 8mm in travel from fully retracted to extended. Any further and the lens would need to be much wider to prevent reduction and tunneling from the sides.(rounding of the side borders like looking thru a tube)  With the current sizing and a bit more modification, the barrel could be changes to accept up to a 25 mm diameter lens. But is currently sized for the 19.05 mm diameter as to what was cheapest to get and the 20mm and 25mm lenses available were significantly more costly.

Depending on the level of interest, I haven't decided to try and print and offer complete lenses, offer the print files for sale, or just make available the files for those wanting to print the lens assembly.

If those that find this could let me know what their thoughts are on this and what they think the value of it would be worth to them. I know that some will prefer to do their own, some may opt for ready to print and others may just want a turn key solution that'll save money over the Thor 001.

I could also just post a link and toss them onto a paid or free 3D file site like Cult3d or Makerworld.
Be sure to check back to see how things go.

Thank you for everyone in your interest if you've made it this far in the post.

 

[GreyWolfe111]

Just an Update:

If you are planning on making a macro yourself and can get a CVD ZnSe lens for close to the same price of the PVD type, go for it.  Not sure if its the focus length, better grade of crystal or the material processing. But the Optical Quality of the image seems much better on the CVD processed lens. Not sure if it shows as much in the thermal image tho. I don't plan of testing this, but it is also suppose to be more durable and a bit more scratch resistance during cleaning.

I don't recall anyone mentioning this. Even when looking up info on choosing a lens. But I did confirm with the second lens that the higher focal length does result in less magnification due to the focal point being further from the target. So if you want higher magnification go for the 25 mm or 38 mm focal length lens. But you'll need to be within an inch or two of the target. So those are probably only useful for detailed small electronic thermal inspection. I am still patiently waiting for a good price on one to try. I'm not in any hurry since it's just to have fun with and not really needed.

The 60.3 that I originally used was around 2-3" or so inches and the new 160mm lens was around  6-10". Even a distance that overlapped the start of the 30cc (11.8") to infinity normal focal length of the Thor 002 without a helper lens, the macro was able to bring it into a better focused and clearer image. But quickly hits it's limit and as would be expected things look worse a bit further on.

For my general use, I think the ZnSe CVD 160mm  will fit my purposes better when using in tighter spaces. But the 60 mm was still fun to play around with.

Still looking a trying to find a suitable lens to make a fisheye on the cheap. But since these lenses are made to focus a beam for lasers. I don't think I'll find one. And the only thing I have found so far are lenses that cost hundreds to a few thousand dollars. Not feasible just for curiosity.

[alexwhittemore]

But since these lenses are made to focus a beam for lasers. I don't think I'll find one.

Correct. But see my post above, ctrl-f "concave." You should be able to do it for about $350. Not fisheye, but at least a wider FOV.

[Marekx89]

Hello, can photo from isotherm mode ? Is as FLIR or ? I see many review on YT, but no one see isotherm mode. Thanks

[GreyWolfe111]

Here are a couple pics from the IsoTherm setting.

One thing important to note, is that it turns off the enhanced resolution imaging. So in this mode it will only show the native resolution of the sensor.
Any recordings or pic captures will show that is in that mode and at what temp settings you selected.

There are Three options. You can select either Higher then set temp, Lower then set temp or set a range between a low and high settings.

The First is a heating duct with temp set higher the 80F'

The second and third are with the 160mm and 60mm macros I created so you can compare with the original M.2 Drive board above and
set to 124F then around higher 140F

And the last is set below 62f to show cold air infiltration thru my front door.

I hope this helps for what you were looking for.

[Marekx89]

Thanks, its only color fron Iron pallete or it change when is change pallette ?

[GreyWolfe111]

You can use all of the colored palettes. The Back hot and White hot are disabled in the Isotherm mode.

[Emiel]

I wanted to share my take on a macro lens holder. I bought a THOR002 back in July mainly because of the initial impressions post by Fraser. It's unfortunate Firmware version 1.2.0 was pulled, I was to late to download it. When I bought the THOR002 I also ordered a ZnSe lens like Fraser in the initial impressions post. Last week I finally took the time to CAD the holder design I had in my head. It simply clamps into the recess of the THOR002. You can find it here. STEP file is included there. I designed it in OnShape. If anyone is interested I'll look into sharing the OnShape file (that would be new to me).

[GreyWolfe111]

This was my revisit and adjustment of the first simpler design I did.
Cleaned a few things up and took the lens retainer ring from the adjustable barrel model I did.



Yours' win for being simple and sweet. But curious at how well it stays in the face during any rugged handling.


You can see the whole lineup of both the Barrel and Fixed pics here.
https://www.eevblog.com/forum/thermal-imaging/thermal-master-macro-len-3d-printed/

But I doubt the adjustable barrel with the spin ring will be more then a novelty.
It does help extend focusing distance a bit while maintaining the same area viewed (the 63mm test was by about 10%).  It's not a multi element lens zoom macro.
But I don't think that really justifies the added cost and complexity of it.

[Aureliano]

how's everything goes now ?

[Marjorie_EE]

Wow, your review is very professional. it gave me a very clear understanding.

[Fraser]

Some welcome news….

https://www.eevblog.com/forum/thermal-imaging/thermal-master-will-be-creating-an-official-account-on-this-forum/

Fraser

[alexwhittemore]

So, I finally got around to doing a thing! As stated earlier (I think), I've noticed my THOR 002 V1.1.2 is reading way off (high) - like, my bedspread was reading 80F in equilibrium with the rest of my room, where a thermometer read 71F. TL;DR, I did a calibrated measurement and, with the measurement parameters all set correctly, the camera does in fact read within 1C. Mine came with an implausibly low ambient temperature setting from the factory.

I had a project from a while back that's a cheap implementation of a blackbody temperature target. It's a PCB with a big thick pad of copper, a ring of resistors on the edge well thermally coupled to that pad, and two TMP117 high-accuracy PCB thermometers, one well-coupled to the center of the pad and one on the corner. It's also got an apparent surface of calibrated/known emissivity - a patch of Acktar Metal Velvet, which specifies e>0.985.



A microcontroller runs a PID controller to drive the temperature at the center of the target using the resistors. The TMP117 has a datasheet accuracy of 0.1C, plus whatever delta there is across the coupling vias (negligible), plus the controller error (known).

As measured, the controller held the target center to 50C within 0.2C, and the corner is pretty consistently 0.7C above that.



Initially, I noticed my THOR002 was only a little off, closer than I'm used to, but I knew I needed to set the emissivity correct for a good test anyway. While browsing Settings>Measurement Parameters to do that, I noticed that the setting for ambient temperature was set implausibly low - something like -8C. I took the opportunity to set that correctly for the current room temp. I couldn't key in 0.985 for emissivity - I had to select the closest stock option of "Skin" at 0.98. The "distance" setting also doesn't go any lower than 0.5m, so I tried to take my measurement from about that distance.

Lo and behold, with settings all correct, the measurement was just about spot on: 49.8/50.0 on center, and 50.3/50.8ish on the corner.



My Thermal Master P3, with all settings dialed in, was also basically bang-on: 50.4/50, 50.8/50.8.



A few minutes later, I'm now measuring 27.4 when the target is bang-on 30C, so perhaps there is still some room for improvement, but again it's technically just about within the error bars.

[Fraser]

I have been testing the two new THOR cameras that Thermal Master sent to me. Today I tested them at a distance of 1.5m from the Blackbody. Accuracy was good. Testing continues but I recommend owners install Firmware 1.2.5. Full details of testing are in my THOR testing thread.

https://www.eevblog.com/forum/thermal-imaging/the-thermal-master-thor-series-testing-thread-by-fraser/

I have also proven that it is possible to revert to older firmware versions without difficulty if a new firmware has issues.

Fraser

[mharding]

I couldn't leave it be before I printed my own copy, so I made some improvements Remix here: https://www.printables.com/model/1478318-improved-macro-lens-adapter-for-thermal-master-tho
Thanks for doing all the hard work!

(Attachment Link)

What plastic do you recommend for printing this macro adaptor?

[Fraser]

Thermal Master appear to have released a new version of their THOR series of cameras. This version contains firmware version 1.2.9(T) and the camera is not compatible with earlier versions of THOR camera. Installing firmware Version 1.2.9 on models that are running version 1.2.5 (and earlier) firmware appears to cause a degradation in measurement accuracy. Reverting to firmware 1.2.5 does not correct the issue and no remedy has currently been released.

Thermal Master have removed the THOR firmware download from their site due to this incompatibility issue. Users have to ask Thermal Master Customer support for any firmware updates so that the correct firmware version may be provided. Do not install firmware V1.2.9 on your early model THOR camera.

I have not tested the new THOR camera version so can make no comment on its behaviour, functionality or measurement accuracy.

What are the latest Firmware versions ?

For those THOR camera owners who own models running Firmware 1.2.5 or earlier, version 1.2.5 is supposed to address the temperature measurement accuracy issue and it is the latest firmware compatible with your camera. Installing Firmware 1.2.9 on these cameras degrades the measurement accuracy and it currently cannot be corrected by the user (re-installing V1.2.5 does not fix things).

The latest THOR cameras are delivered from the factory with Firmware 1.2.9(T) and this is the latest version for these new cameras. If firmware 1.2.5 is accidentally installed, measurement accuracy suffers. Firmware 1.2.9 may be obtained and installed to correct this issue.

Fraser
14 June 2026

[kw1]

Hi, I'm having a strange problem with image analysis using the software.
Sometimes the images get jumbled up when I view them.
Is this due to the camera or the software?

Steps to reproduce the error using the attached images:
I open the 8 images in the offline analysis.
To see the error, I switch from auto to manual for each image.
However, the actual image doesn't always appear; instead, the image from the previous or next frame shows up.
As a result, the temperature readings are incorrect.

Can anyone figure this out?  Thanks


klaus

Firmware 1.29

[kw1]

Hi Fraser,
Am I the only one with this problem?

For me, it would be important to know whether this is a problem with the camera, the firmware, or the software.

thx for your answer