EEVblog
HTML-код
- Опубликовано: 10 сен 2024
- Review of the new FLIR ETS320 Benchtop Thermal Imaging Camera designed for electronics and PCB use.
www.flir.com/ET...
Forum: www.eevblog.com...
EEVblog Main Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
Support the EEVblog through Patreon!
/ eevblog
EEVblog Amazon Store (Dave gets a cut):
astore.amazon.c...
T-Shirts: teespring.com/s...
💗 Likecoin - Coins for Likes: likecoin.pro/@...
It's a nice idea. I use my "enhanced" E4 for monitoring or troubleshooting PCBs (and LEDs) a lot.
rare for me such active discussion on productson a single video from my subscribed youtubers
5:00... I get a lot of flak from people for not using a thermal camera. After seeing the resolution, the functionality, and the price, I stand by my decision that this is ridiculous.
I think the technology has great potential. The same way laptops in the 80s that were 20 lbs with no mobile broadband had potential.... but no. Not buying that POS.
Great video as always Dave!!
Thanks.
There are much cheaper solutions than this. Just a thermal imager for your phone can be all you need.
I'd think that for a stand-mount application, MSX with auto offset detection could be done with an optical proximity sensor
or, have stereo optical cameras symmetrical around the thermal camera, then in software you can find their overlap and the overlap center, and then center that on the center of the thermal image.
It should be rather simple for them to implement a functionality where you say "This is all the same temperature" and it snaps a photo and from that estimate an emissivity for each pixel and use that
Yep, for sure!
Just image the entire powered off PCB and calibrate.
Some of Flir cameras has this feature and it's called Flat Field Correction. Check it in the software.
Nope, FFC is the thing with the shutter where it says "calibrating" and the image freezes for a moment. It puts a shutter of known temperature in front of the sensor and calibrates out every single pixel offset since they drift really a lot. But of course they have that and can build on that to develop the emissivity calibration map easily using a similar mechanism. I think the main reason there isn't already such a thing is that this would be a pretty meaningless feature for all non stationary cameras, so since this seems to be their first dedicated stationary camera it only know makes sense there and they should hurry and implement it (I have been doing this manually in post with still images and it opens your eyes for a whole new world)
And what about cameras that don't have internal shutter and could be calibrated by external "flat field"
Building services bought a general purpose hand held FLIR. We have adopted it in R&D. It brilliant for finding where the extra current is going. amazingly sensitive and saves hours cutting tracks.... Use the tip of a small screwdriver close to the PCB to shield the camera from the source. This more than makes up for low resolution. No problem finding an overheating 0402 😙
How did you solve the minimum focusing distance limitation of a handheld?
So close... maybe next iteration, and maybe by then we'll have 640x480 for the same price-point. In the meantime, back to the freeze spray and IPA testing methods.
Ah yes, the Louis Rossmann method!
Lies! That's the Jessa Method!
IPA is me, freeze spray is jessa.
but this is why I protest the use of these. resolution to price ratio is a joke.
...and Chris Long uses acetone (well, and freezespray too)
YEAH! I learned the frost spray trick from you(Louis)!
I learned it from watching YOU, DAD!
Flir Tools + Licence key after I've bought your expensive hardware now I have to start diking around with license keys too? Really. fail right there.
19:55 create account and login to use? BURN. Seriously.
I hear you. But at the end of the day I have no interest in being the a victim of some companies misguided attempts to prevent piracy. Just as easy as it is to clone the hardware and firmware it's equally easy to hack the desktop software. Making actual paying customers jump through hoops is one of the reasons I refused to follow Eagle into the clutches of Autodesk. The people running the hacked software don't have to put up with this nonsense. Could write a long essay but won't :-)
In the lab, I've had trouble thermal imaging my solar devices because of all the metal reflecting the infrared, but I thermally evaporated about 100nm of C60 fullerenes on top of my device and was able to completely remove the emissivity problem since C60 absorbs pretty much everything. 100nm of C60 coating is still transparent visibly and I was able to see where my shorts were. The heat is still localized too so you don't have to worry about the C60 diffusing heat around it too much either.
I know that not a lot of people have access to organic C60 thermal evaporator systems, but if anyone knows of a insulative water soluble or IPA soluble thin coating that you can spray on top of your PCB or device you can remove all your emissivity problems and then dissolve the coating once your done. The coating has to be thin so that the heat doesn't diffuse to far before it gets to the top surface. You have to make sure the coating isn't transparent to thermal IR. I know PMMA and most polymers are transparent.
Anybody know of a coating?
I think that I will save 2500$ and continue to use wet finger probing :)
If you have to sign in to use the software, what happens if you don't have an internet connection? Or if your connection goes down? If you're in a factory, you have to stop the assembly line?
Seems like waste of money
It is!!
Yeah, even 250$ Flir one has MSX technology
Not sure how FLIR got this virtual monopoly on the market but it seems ripe for disruption. $2.5k for a piece of rubbish like this and they didn't even bother to adapt the software for the use case :/
I'm all for companies making a fair, honest profit but thermal cameras have existed since the 50s ffs!
Indeed. These FLIR people are lazy dullards and seem staffed mostly by marketing people. This product is plainly an engineering embarrassment that should have cost no more than a few hundred dollars. Is FLIR sitting on a heap of bulletproof patents or something?
FLIR could make a budget version of this by taking out the display and CPU, instead giving it USB 3.0 and letting a computer handle that stuff. It would probably be faster too.
The logo is almost as big as the LCD
MSX is probably hard to do at close range without a beamsplitter
It wouldn't be easy I suspect, but given the working distance and limited field of view, if the camera sensor was close enough I'd say it's possible.
But maybe they tried it and it just wasn't workable, who knows.
Maybe it could be a manual thing, have a flip down mirror in front of the thermal lens, take a photo, and then flip it out of the way. Image gets processed and overlayed on the thermal image.
A bit messy, but would work.
Even the ability to manually import and scale an image of the PCB into the software would be useful.
Use the back of the mirror as the calibration shutter? New photo every time it calibrates (or on demand).
Put two optical cameras symmetrically on either side of the thermal sensor. Use the difference between the images to correct the parallax error. Done. Would add something like $5 to the BOM cost but should be the simplest solution.
At the standard 70mm or whatever distance it would be doable. Distortion can be calibrated out with some matrix transforms, though given how unresponsive the unit is, it likely doesn't have enough power to do that in realtime, though 320x240 at 9fps is not an ambitious situation lol. You'd still have mild parallax issues, but that would only affect tall parts on the PCB which would be out of focus anyways. I don't think you could have a beamsplitter that would pass visible and IR (spectrally this unit ranges from 7.5-13um, so the beamsplitter would have to pass ~0.4-13um), but I stand to be corrected. Plus even if that was possible, there would be a lot of light loss and you can tell that there's already heavy noise reduction in place including temporal (judging from the trails in the image). The one feature which they could've implemented on all these models but haven't for whatever reason is some kind of edge detection overlay from the IR data itself. Given how low the contrast can be at times, it may be helpful, particularly without MSX.
Errr... 2500 USD! Are they serious!? Seems expensive but we can assume that some firmware updates will help with the software. Why dont they include a selection of screw on lenses with different focus ratios?
Timothy Giles Since when does the S60 do 320x240? Last I remember, it could only do 60x80.
Opps! My mistake.... Thanks Bob.
Timothy Giles Still a pretty cool phone.
In the software you can make a measurement area and then right-click and choose local parameters and set different emiss-values for this area....useful feature
I have been reviewing this for the last week or two as well, I released a video two days ago on first impressions,
On its own, I agree with Dave, the camera has limited functionality, an SD card to record sequence data while on battery would be nice.
But when coupled with the included software it is a whole different animal and becomes very useful. The height adjustment allows larger object to be viewed but still limited to the 70mm focal distance, but that 70mm could be 100mm above the base for example.
You can also use standard web cam type software to capture the live stream from the camera as it is fully able to support that (But it is just a copy of the LCD screen), it also presents as a USB hard drive when connected to a PC so it is easy to get to the images you may have stored on the camera, even without the FLIR tools.
How about a proper API? You need to be able to draw the polygons and stuff in your software, clear the data, start the camera, then log the data. That would make it useful. You could also drive a relay board and if the temperatures got too high in any area, save a snapshot, and cutoff the power with a relay. Then the techie could get the board and the snapshot and see exactly what got too hot. Is there an API at all with it Dave? I bought a testo 875 back in 2013 or something and when I asked where my API was they told me to get stuffed. Nice.
My 2nd gen Flir one, for 350 bucks has MSX technology. It's a fun toy to bring with you anywhere, happy purchase for me anyway.
As a working electrical engineer that has been looking for a camera like this for years, I think this is a great product. I work for a company that makes mil & aero electronics and let me tell you, finding these features in a thermal camera that is actually intended for electronics (as opposed to industrial equipment, HVAC, etc.) is near impossible. A Fluke camera that can resolve a pixel to less than 10 mil (0.010") cost 10x this one. I think this is actually a great price for what it is. If I did the math right, this one will resolve a pixel to just under 7 mil.
The FOV is fine for 90% of our applications, for the rest we'll just have to move our DUT around to see the whole thing. I wish it had adjustable focus, but the price would skyrocket! I think someone has finally heard us in the electronics world and has come out with the product we need. Hopefully this is only the beginning of a full product line. Before you complain about the price, check out the Flukes then get back to me. The ETS320 is not intended for the hobbyist.
Another comment about the MSX... I haven't tried that feature with FLIR, but I have used the equivalent feature on Fluke's stuff. The problem is it does not work up close. At ranges of 70 mm, you would need the IR sensor and the camera to be on top of each other. If FLIR put the MSX feature in this, you would all be pissed because it would be so misaligned it would be worse than useless.
Maybe I'm the only one here excited about this product. I have actually tried other cameras for imaging our units and they are 4x the price and don't get the resolution I need. I fully understand the limitations, but I will gladly take those in exchange for the resolution and the integrated mount!
I too work in R&D at engineering company and I second your comment. People here are mostly looking from hobbyist point of view and that seems very expensive for what it has to offer. With over 500 engineers in the R&D it is very challenging to requisition any new piece of test equipment. I use other FLIR cameras that are over 10 years old and it costed over $25K each. I was able to get approval of this inexpensive camera for purchase and has been using for about month now and we are very pleased for what it can do for testing circuit boards. FLIR makes wide range of cameras with price going from sub $1K to $1M. I used to work for FLIR in Oregon in 90's and I did notice that marketing department plays very large role in direction of the company. I think FLIR marketing hit the right price point for ETS320 that was intended for R&D and production environment. ETS320 is not intended for hobbyist community. I own $200 FLIR One for personal use and that was right price point for hobbyist.
As always I love your NO Bull Crap opinion ! Honest and to the point ! Great video and as always will be looking forward to the next.
Regarding the licensing, once you have logged in the first time, you dont need to keep logging in, I just verified this with the ETS 320 I have here and it all works just fine even without any network connectivity, so it seems you just have to do it the once.
I love your videos! I am an australian 3d modeller.
There is something about your vids I find fascinating!!
From the thumbnail I thought this will be a review of an overhead projector.
Why did I think of that scene from The Castle?
Haha, great scene.
"...hold your horses. What is it with wogs and cash?"
Feeling like transplanting directly from FLIR E4.
Couldn't they get an emissivity map by taking a thermal picture of the item when it's switched off (and, presumably, at thermal equilibrium)?
Is your "test object" a Fortuna GPSmart by any chance? I still have one somewhere… don't know if it's still working because it had some bug with the ephemeris or almanac data expiring in 2005 and some Russian patched the firmware with newer data expiring in 2010.
Some photographic enlargers had a follow focus cam and lever which kept the image in focus as you moved the head up/down.
Do you think you could use a photodetector from a CDROM drive and some other hardware as an infer red scanner? CD-ROM is supposed to use 780 nm (within the infrared).
OT: Why isn't that ESD mat light blue anymore? I just checked EEVblog #585 do be sure...
It's challenging to find a flir camera with enough resolution and ability to focus to 3 inches for inspection of microelectronics without taking out a second mortgage. It's either low resolution and min focal of 3 feet or pay tens of thousands.
thing is, that thermal sports optics with 640x480 resolution rage in at about 4k€ with a 38mm lens
This product targets an extremely specific market segment. Good luck to the FLIR marketing department...!
Do they make a separate model for lefties?
Ya, an adjustment on both sides would have been nice
This is the "lefties" model. You'd use your main hand for holding a pen or the mouse while working on this and adjust the height with your off-hand. And no, those buttons on it are nothing you manipulate all the time.
looks like you may be able to move it over to the other side but if your connected via the USB cable then it may get in the way, now in a production scenario, you would not be adjusting it much
I just realized, Dave's has the knob on the left, mine is on the right, his must be the "Below the equator" version, mines is the Northern version (Canada)
Either way, it does show you can get it on left or right, or have it moved if you want, time to take mine apart partially to see :)
Looks like a standard off the shelf stand and mount. And the panel is obviously a remote control that has been glued onto the side. Makes sense at $320. Extra zero needs serious justification.
This is at moment the best thermal cam for finding shorts on pcb boards like mobile phones and tablets
They could have made an auto focus or at least a manual focus that that could turn a pot to tell the system how much turns are done if performance was done as one could surely they could add algorithms to correct any variation.
Too me it looks like they just used the guts out of that handheld thermal camera, added a larger LCD, stand alone buttons and put it on some cheap rig.
IIGrayfoxII
For university use this thing definitely lacks a HDMI/... whatever video output for easy use with larger screens. Why not a version completely without a screen and just the bare camera with HDMI/... or a proper more rigid USB port (like USB B)?
Now you can finally Flir the Flir!
Now a VISA etc driver interface would be pretty neat as you could correlate the temperature measurements with voltage/current in the circuit. For example for NI Virtual bench or BenchVue etc. I should go patent that!
if your using win 10 just search for "camera" and you can use that to watch the Flir... the camera app is built into windows
Question, do companies just contact you do review their products, or do you contact them?
They contact me.
They should have looked at some metrology instrements
I think it might be too pricey for what it is. I understand it not having MSX, I'm fairly sure that it would be difficult to get that to work at 70mm, and probably impossible at 20mm. Can the E8 focus that closely if you use the adjustment tool? I assume not as its usual focus was more that half a meter. So they must have used a different lense behind the adjustable one on this unit. If the E8 can focus down to 70mm I'd go for that instead of this, otherwise this does fit a niche.
IIRC the E8 can't get that close. Likely a different lens.
If they add possibility to link emissivity with oval, rectangular selection , the thing may turn to be interesting
It looks and "feels" like a USD 99 worth of product.
I know gold is used as a IR reflector in astronomy applications, so I don't know if it's really possible to calibrate it out.
The sudden drops in the graphs @32:00 probably are a result of the automatic temperature range.
could anyone help, I would like to buy a thermos meter, in about a year / some time in 2018, what would you recommend, fluke do em too, are they as good, I know flir make some of em or something, from a basic, intermediate and a pro (depends upon cost) recommendations, also some advise, even basic, some people use em a lot, therefore know a shed load, its for electrical testing, electrician, I'm a electrician now, engineering made me unhealthy, as I love to eat, being a electrician I am always physical, anywho
hmmm.... I have seek thermal for android phones.
always wondered if there is a way to increase resolution of thermal imaging by using multiple sensors....
The functionality of the PC software makes it worthwhile. The hardware is silly. The wiggling of the unit in the locked position is... ridiculous? Doesn't that defeat the purpose of setting specific masks and spot measurements, if you can throw the unit out of position by touching it? You'd have do a custom mount job to avoid that, which is fine I guess but this is still silly for the price. As for the buttons, it seems that people forgot how to make nice tactile buttons at a reasonable price. It seems that no matter how much something costs, the logic is something like: "does it make contact? at all? eventually? then it's fine." No touch screen on a fixed bench setup? Really? Whether to set the spot measurement or to drag with your finger to create a mask for an area of interest or even just go through the menus? They could've gotten away with the crappy buttons if they had a touch screen lol. Finally, sluggish response times/update rates for ultra basic things like going through some menus is a deep pet peeve of mine. Surely I'm not the only one though. I'd summarize this kind of product as "engineering with no passion" aka "it does what's advertised". I encounter engineers with that mindset more often than I'd like.
I just started watching video, but I already annoyed by the time it takes to boot the device.
20:47 Looks like your Windows installation is set to the wrong region format, so it's displaying the temperature wrongly
No MSX! :o This was the main reason I was considering this. I'm going to move on from this one. Thanks for the review.
I guess FLIR should have got David2 to design the stand :)
I would expect better from a $2000+ lab instrument, not only a focus adjustment ring on the lens, but also tilting camera head for 90 degree angle work. also I would expect the video screen to have a tilt adjustment too & independent mounting bracket from the camera head unit.weight of the instrument stand should not be a problem for a $2000+ lab instrument, so what, if the shipping weight is up because of high quality mounting brackets. FLIR need to do better.
Did someone at FLIR go to the Apple school of marketing and design? Seriously, releasing a new product with missing features so they can "update" it later and sell more.
did he just turn it on WITHOUT TAKING IT APART?
Nice, but way too pricey for me. The prices will come down over time.
Now that and looking heating patterns in 3d printer heat beds...
i would REALLY like the ability to save a snapshot to a flash drive or even an SD card so i can quickly do some manual testing with proof
I liked the video so I gave it a big thumbs up. (:
I wonder if it's really just an early concept product, soon to be upgraded vastly in a newer generation they won't announce yet.
I wish FLIR had a different definition of "affordable." I mean, yes, FLIR devices have gotten a lot less expensive, but if you want any sort of real quality, it's quickly out of the hobbyist price range. Though, the same thing goes for multimeters, oscilloscopes, 3D printers, and lots of other things I wish I could afford to own.
Nevermind the silly 9Hz+ export control business.
Why are FLIR cameras so expensive? Even the cheapest handheld one is still more than 300 bucks - if it were like 70 bucks or so, I'd buy one.
2500 bucks for that piece of shit? Really? This thing has a Chinese knockoff look and feel all over it. A very expensive joke. Someone needs to tell FLIR that April Fools was 25 days ago.
Made in Taiwan, with Chinese parts, and taxed for Swedish social services, I would bet.
No external screen connector?
Clever but it's not up to the job, a few more versions later should fix it.
FLIR 1 --- $250.00
Isn't it just cheaper to get a Flir-C5 (Wifi) and mount it on a cheap stand? At least you'd also get MSX and only half the price!
Wish FLIR would give me one of these
FLIR has always tried to enhance marketing product tiers based the removal/addition of common UI elements and interfaces.
Not sure how effective this would better than a hand held one. Unlike a microscope which is in more use.
It has a closer focus range, and is hackable to even closer.
EEVblog I was referring to the thermal imaging. Why a benchtop when a hand held would be more than enough, I feel.
New for 2017, POS by Flir...get it now for the attractive price of $2500.00+s&h+software. Apparently I got into the wrong business.
We gonna get a two minute teardown on it?
Teardown video shot, will edit tomorrow.
Predator called, he want his Flir back
Did they provide this product for review and/or sponsor this video? Just curious because if so, I am thinking they might be regretting that decision.. lol.
That stand looks like something they pulled off of a $50 microscope.
Need to review the Flir One 3rth edition and Flir One pro.
You are better off mounting a handheld and attaching an external screen.
Dylan T
yep, we made a stand for our hand held FLIR with a block of wood base, chromed tube and a spring clip for next to nothing 😀 and it had adjustable height. The apprentice did a good job of it..
22:39 jpeg is the right choice for an image with such gradients.
Wow. If FLIR thinks putting an otherwise standard FLIR camera on an adjustable/sliding mount justifies the $2500 price tag when the focus window is as limited as it is, it doesn't have wi-fi (which, several years post-advent of the IoT age/wireless _everything_ age, is almost unforgivable, especially considering how cheap and ubiquitous basic wi-fi chipsets have become) *_and it lacks MSX_*, they're out of touch with the intended demographic (presumably) of electronics enthusiasts/engineers.
this seems extremely limited. I think they would have been better off just coming up with some sort of gooseneck and a much better focusing range. For the price they charge this thing should be making me pancakes.....
I wish a company would come out with an affordable thermal sensor for the diy community.....c'mon China I'm looking at you.
The things that are required for it to be a good product were missed. They make a great product line, but this was a failure.
omg that stop ring has a huge amount of play to it. Most hobby machnists would built something much better for much less.
Sharpie: store with tip down :D
I saw a review of it in the morning here on youtube, but yours is much better. nice video. Did they allowed you to tear it down?
Sheesh! It's definitely not priced keenly enough to replace our current mobile-phone solution. I thought it would be ~$500.
Would have been nice for occasional use in RUclips videos, if it weren't for the price.
Smartphone CAT S60 by FLIR visor can do it :) $ 680
Seems like this thing has been intentionally nerfed so that it can only be used for desktop applications, that would explain the fixed focus for example.
Apart from the software,you're almost better of with a flir one and phone and put it on a stand ,saves you 2k
That's one pointless piece of garbage.
They wont be selling many of those
could be better quality for the cost
But can you get raw data from it? Or that shitty software is all you can get?
Functionality aside, from a product design standpoint, this falls so short, the gods of ideation must be cringing...
"don't turn it on take it apart" ????......i i thought you were close to take it apart...&...you ended up with "catch you next time !".......
Given a thumbs down for a half arsed product,.
Lets be real for a minute.... To get your moneys worth you really need to be using this device on a daily basis or your just wasting money
is it still 9Hz.... once you have used a 50 or 60Hz thermal camera you not going to be happy with 9Hz.... also MSX is over rated, its a poor replacement for real thermal resolution....
Pronounced " FLEER"
not flur