One of the entries in the Hackaday Prize Best Product competition is [x-labz]’s pocket thermal imager. It’s more than a prototype, it’s a design conceived to get out into the world and be used by many. Best Product entries are open until July 24th, and with a $30,000 cash prize on the line let’s take a look at some of the things that elevate a project to product status.
Thanks to recent advances in the state of thermal image sensors, a tool that gives you Predator vision is almost a necessity on the modern workbench. The pocket thermal imager will find drafts in your house during winter, will tell you how to cook a steak, figure out what part is shorting out in your latest electronics project, and will tell you how terrible the heated bed is on your 3D printer.
[x-labz]’s thermal camera is based around the FLIR Lepton image sensor, an 80×60 pixel thermal imaging sensor that’s good enough for most uses. This camera is soldered onto a PCB sandwich containing an Atmel SAMD21 microcontroller, full-color OLED display, SD card, and a battery management system.
What we’ve mentioned so far isn’t out of the ordinary for any other entry in the Hackaday Prize. Building something for the Best Product competition is different, though: a lot of thought has to go into the manufacturability and the fit and finish of this device. So far, everything’s looking great for [x-labz]’s camera. There’s a 3D printed case that looks like it could be easily translated into an injection-moldable shell and at least some of the parts of the user interface are unbelievably satisfying. We’re looking forward to seeing the full Bill of Materials and a business plan (a new requirement this year). That’s an area where many hardware designers lack experience; being able to study the examples from Best Product entries will be a welcomed resource.
There’s a world of difference between building a project and building a product, and the entire goal of the Best Product portion of the Hackaday Prize is to reward those people who go the extra mile as aspiring entrepreneurs and show us how that’s done. $50k in cash prizes are set aside for Best Product; $30,000 for the winner as we mentioned before, but there is also $1000 for each of the twenty entries that make it to the finals in this category in addition to some much deserved notoriety from Hackaday’s community of hardware aficionados and early adopters.
20 thoughts on “Best Product Entry: Pocket Thermal Camera”
Sorry, but image on the top of page is definitelly not 80×60 pixel!
It is small spin.
Anyway, good idea, good progress.
The sensor is 80×60, not the display. The trick is to properly overlay the (lower resolution) sensor image over a real image.
The FLIR offerings do the visible overlay, but this project does not (in fact it does not even have a visible light camera). The cover image of the project/article have nothing to do with the lepton imager. The images you can expect will be the same as any other of the dozens of lepton based thermal imagers.
I’ve got a FLIR that does that and I don’t care for it at all. Half the time, the overlay obscures the thermal detail I’m trying to make out. Worse though is the parallax effect, where the two cameras aren’t taking the image from exactly the same angle, so the overlay is almost always off a little. There is a way to manually adjust it, but it’s not worth the trouble. You can turn the overlay off, but it defaults back on when you reboot the device.
FLIR also seems to market their devices as having much higher resolution than they do, citing the visual camera over the thermal one.
Don’t get me wrong, I really like my FLIR (I’ve got a Seek too, which is also nice in other ways.)
With creative optics, they could align the cameras perfectly. Or they could have a distance sensor that adjusts for the parallax effect dynamically, or they could allow me to save the overlay-off setting between power cycles….
I also have a Flir one and love the overlay, haven’t had significant issues with the parallax. I find it useful to be able to see the visible structure as it shows obviously a lot more detail and nevermfound it to obscure the IR image.
Sorry , hardly worth the effort as a potential product. I mean, smartphone based thermal imagers are available for less than this would be on a retail shelf, and perform far better. Nice job as a project…but far from a product.
It’s also pretty straightforward and cheap to build one yourself out of a Raspberry Pi and a Lepton breakout board.
Cheap? How is it cheap? The Lepton breakout board is $35 USD (for just the board) and then you need the actual sensor, which according to DigiKey the cheapest 80×60 9Hz sensor is $175 USD!
So unless you know of a supply of Lepton sensors that fell off a truck, Wizard is right and smartphone based imagers will be cheaper.
Unless the guy who built this thing in the article is stealing them, he also needs to buy the Lepton, so you’re comparing apples and oranges here.
totally agree. Whilst its quite a nice hack/project effort, its not a product that I would buy. I don’t see how it offers any kind of functionality you can’t get in existing commercial devices (which I imagine will be cheaper to buy).
Why does he want to take a thermal picture of his pocket?
I’m sorry but you leave me no choice:
Fits in pocket, but i really don’t know how robust is compared with commercial products
This: We use a lot of thermal imaging equipment of various scales in the lab (including one of Max Ritter’s, referenced elsewhere in the comments) and I’ve been waiting for the knock-off floodgates to open as it appears to have done.
FLIR (the company) has kept the IP of their products tightly held and thus their profits intact (wouldn’t you?). This has slowed development and implementation by several fold and has made hacking/tinkering a pricey proposition. My guess is that the components in the handheld units shown in these ads (that are apparently all from the same source) are being similarly protected to an extent, but the rest – availability of low volumes of bolometers (sensors) and other components – is a matter of time.
As the the cost of the bolometers drops and the community of users grows, things are likely to get a lot more interesting since you can measure any number of kinds of radiant energy with these and they’re used in all kinds of interesting stuff. Backyard microwave astronomy, anyone?
Absolutelyspamming/automation you are easy to recognize… Just stop!
If you consider building / buying something like that, i suggest you take a look at Max Ritters project: https://github.com/maxritter/DIY-Thermocam
From my knowledge Max Ritter was the first person who used the flir sensors to create a standalone thermal cam solution. By version 2it supports Lepton 3(160×120 thermal resolution) and picture overlay using an integrated vga camera.
I have built this using a kit from groupgets and i can definetively recommend it!
higher res OLED and upgrade to the new sensor. It’s an attractive project. I’ve needed one of these for a while for detecting surveillance devices.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)