Uni-T UTi690A (120x90 new entry/budget) (UTi690A, UTi120S, UTi712S)

[Paw85]

Was a long time since last posting here (hacked e30bx user) but figured I would share a find of what seems like a good budget thermal imager, with higher resolution than I have found in this price-range before.
Price is currently ~USD 175-190 which for 120x90 resolution is seem quite low, lower than some 80x60 models out there.

https://www.banggood.com/custlink/DKvygTWWDM

The Uni-T UTi690A is not currently listed on Uni-T's website, however one can find an imager named UTi120S with what seem to be identical specification and apperance:
https://www.uni-trend.com/meters/html/product/Environmental/Thermal_Imager/Echo%20Series/UTi120S.html

The store does have a datasheet/manual of it that seem to be from Uni-T, even if not possible to find it at Uni-T's website, at least not yet:
http://myosuploads3.banggood.com/products/20211013/20211013011951UTI690A1.pdf

Uni-T also sell what seem to be a "portable" version called UTi120P with what seem to be the same specifications but just another shape of the device, and those units currently sell for $400 on Uni-T's own aliexpress site.

The UTi690 does not have a regular camera at the front which I suspect is one way they have reduced the cost.
There is however a laser to show where you're aiming it if uncertain, and a led-light if working in poor light conditions.
The unit also have fewer buttons than most Uni-T imagers so adjusting settings I suspect is a bit more time consuming.

Due to the price/performance ratio and looking to be far less fragile than the e30bx I currently own, somewhat waterproof and drop resistant if if to believe the datasheet, I could not resist to pre-ordered one to keep around garage/3Dprinter and as a backup to my other imager.
Will report back with findings when I receive the unit, if it will work well as an entry level or budget thermal imager.

If having any questions regarding the unit just drop a post and I will reply once the unit has been delivered.

Sometimes things are not as good as they seem and there are quite a lot of questionmarks regarding this product:
Why is it named UTi690A when there is what seem to be an identical version named UTi120S?
Why is the unit not posted on Uni-T's website?
How can it cost half of the "portable/flat" version when having identical specifications and even exceeding some like bigger battery and a laser pointer?

Update 2021-11-12: Banggood have now added two videos to the product page, showing another Uni-T camera that is different from the rendered one on the product page... mystery contiues...
Also text added to the product page "The most worth buying thermal imager is comming!!! Original UNI-T Newest thermal imager in 2021 !!! With 25Hz frame rate, 120*90 IR sensor resolution(Image resolution up to 320*240 ) smooth and clear visual performance which is the most cost-effect 25Hz frame rate thermal imager you can find ! Support Chinese and English switching."

[Fraser]

The attached datasheet may be of interest 

https://www.guideir.com/product/detail/id/69.html

[Paw85]

Great to see you're still here and active Fraiser! And thank you for posting the datasheet!
If I find that I can open the imager without risking to damage the IP protection I will give it a try and see if it uses that 120x90 sensor, if they are I wonder how it can be at this price, much cheaper than I have seen on other models that uses it.

[mrisco]

I also order this device a month ago, it has not been not shipped yet.

[Paw85]

I also order this device a month ago, it has not been not shipped yet.

My order was reported to have been shipped yesterday. Did you by any chance also order from Banggood a month ago?
(BG is known to sometimes falsely report shipped items, only to a few weeks later report that it has been shipped once again for real that time, so not sure it can be trusted that it has indeed been shipped.)

[mrisco]

I also got the email to report that it was shipped. Yes, it is from Banggood too.

[Paw85]

Lets hope it's really true that they have been shipped then!

Found another strange thing, on the product page on the store, they have added two videos to the product page today, but not of the correct thermal imager, probably a UTi260B.... the mystery continues...

[Chalcogenide]

I also pulled the trigger and bought one at 169.99$ shipped (found a coupon online). Will report on when it ships.

[mrisco]

The buttons of the device showed in the video of the seller's page appear to be for the UTi120 or similar, or maybe the final device will have all of those buttons. I have the hope that the device can be "hackeable" to enable the real time monitoring using a PC. At the end, the best protection that we, the buyers, have is our union on this forum.

[mrisco]

For add more confusion, the store https://es.aliexpress.com/item/1005003217291124.html is selling the UNI-T UTi120S which is very similar to the UTi690A, but at a lower price.

[snoop33]

Reckon this will be any good for close inspection smd pcb work? I can't find anything about minimum focus distance other than 50cm being optimal (way too much).

[Paw85]

Reckon this will be any good for close inspection smd pcb work? I can't find anything about minimum focus distance other than 50cm being optimal (way too much).

Will try to help answering that question once the thermal imager have been received. (Not used with working with a fixed focus imager, and not at SMD level)
As quite a lot of coworkers that do SMD PCB work were interested in the product, I will try to get their input on if it works well enough out of the box for that task or if it needs an additional lens to focus closer.

Are there anything in particular you'd want to know?

[Chalcogenide]

I am happy to report that my UTi690A just shipped.

[snoop33]

Thanks Paw85. I see it has a tripod screw so I'm hopefully it could be a cheap way to do it, even if it does need another lens. I'll await your feedback!

[connectTek]

I have had the Uti 690B for 2 weeks
It's great, does what I want, don't think you can get anything as good for the price.
BTW, it's 256 X 192 pixels.
https://banggood.app.link/HxDAVCxVjlb

[Paw85]

I have had the Uti 690B for 2 weeks...

There is already another thread discussing the Uti-690B that you can post to: https://www.eevblog.com/forum/thermal-imaging/uni-t-uni690b/
The Uti690B is not the same thermal imager as the Uti690A, it uses a different thermal sensor and an altogether different design/housing. It has more in common with the older uti-260B than the imager discussed in this thread, despite the similar name...

[snoop33]

There is a lens for the C200 InfiRay here https://www.aliexpress.com/item/1005003494967045.html

And the actual lens is also available for $10 ish and the rest is an easy 3d print job. I'm thinking the tripod screw can be mounted onto an arm used for a reflow station hot air gun holder.

In my case I'm sure I'd only use it for SMD work but I don't want to use a phone based one as I know the Chinese will discontinue support for the App etc just as soon as they've released it. To avoid that I need something with a screen. I'm still thinking this way is the best value wise.

[Fraser]

Thingiverse have a lens holder design here.....

https://www.thingiverse.com/thing:4901923

[snoop33]

Now $136. At this price I've pulled the trigger and bought one.

https://www.aliexpress.com/item/1005003311073622.html?

[Paw85]

In my case I'm sure I'd only use it for SMD work but I don't want to use a phone based one as I know the Chinese will discontinue support for the App etc just as soon as they've released it. To avoid that I need something with a screen. I'm still thinking this way is the best value wise.

I had been looking for quite a while for other thermal imagers in this price range and only found another 120x90 but in usb-c format (the UTi120Mobile), but never purchased it due to having previous issues with a flir one gen2 as the app crashed randomly.
Then this thermal imager showed up at lower cost than the usb-c version, a fully standalone unit that hopefully wont have issues with software/hardware comparability and hopefully keep working for longer.
Lets hope it's a good choice we've made!

Now $136. At this price I've pulled the trigger and bought one.

That sound like a bargain, congratulations on the purchase! I'd probably skip waiting on my own and purchase one for the workplace if that deal still would have been available, unfortunately it seems to be back at ~$175.

[snoop33]

Still $136 for me, or $163 with 20% vat (no import charges).

[mrisco]

I am happy to report that my UTi690A just shipped.

We all are waiting for your review 

[Chalcogenide]

It's still in transit and it's unlikely to arrive in less than a week, given the speed of the Italian post system...

[Paw85]

I received my unit today and have tried my best in creating a review from my thoughts here: https://www.eevblog.com/forum/thermal-imaging/uni-t-uti690a-review-(120x90)/

[mrisco]

I saw your review (it is excellent) and comparing against the video from this vendor: https://es.aliexpress.com/item/1005003352759582.html seems to be the same UTi120S device.

[Paw85]

I saw your review (it is excellent) and comparing against the video from this vendor: https://es.aliexpress.com/item/1005003352759582.html seems to be the same UTi120S device.

As suspected it most likely is the same as the UTi120s, at least when it comes to the hardware. When it comes to the firmware there is no way of knowing without testing both, if they differ in features.
(And thank you!)

Hopefully today or tomorrow I can see if it may work for SMD-level detail on PCB without a close up lens.
Already having things in focus from 10cm away feels like having the thermal imager right up against whatever you're looking at.

[snoop33]

Excellent, my 120s will be here in a few days so I will compare and review here also.

I didn't buy the 50mm lens that is on the other site as I wasn't sure if it was the right focal length for the smaller sensor.

[Fraser]

Be aware that my experiences with a Uni-T UTi80P that used the Lepton 2 core showed that it’s normal minimum focus range was around 100mm. This is as per its specification. If the core used in the UTi120s or 690A has similar close range performance, you will see very little improvement when using a 100mm close-up lens. The 50mm close up lens will provide closer range imaging if that is needed. These cameras are fixed focus but some exhibit excellent close range imaging 

Fraser

[Paw85]

I have now tested it as is to see if it works for PCB viewing.
I would say the images are not sharp for normal working distance from a PCB, I was quite easy able to identify components down to a size of ~3mm, but not individual small SMD components.
It does give enough information on where to put proper thermal probes for tests though and should be easy to find where on a board a component have shorted if it heats up, you probably do would want to use an external ZnSe lens if you do PCB work frequently.

If you find a suitable focus length for a ZnSe lens that do work well for the sensor, please do let us know!
I can try helping with designing a 3D-printable lens holder for it if that'd help. It will be a bit of a challange though as the plastic body does not have any suitable back-drafts close to the sensor to be able to snap on to, but should be possible to create a holder that snap around the sides and hold on to it where the LCD is.
I did manage to find some germanium-lenses at work but they were of unknown focal length and none suitable for the imager, however what I did find was that one probably need a lens of ~20mm diameter for it to cover the whole field of view without blocking top/bottom parts.

I've attached two images:
Picture 1: Showing a PCB at ~10cm distance, easy to spot the large components heating up.
Picture 2: Same PCB at an angle from ~2-3cm distance, things get blurry at this distance but could differentiate between objects of larger than 3mm size, but smaller components was not possible to see individually.

Edit: Aaand Fraser beat me to it! Yes, one probably want to get closer than 10cm when looking at PCB:s is my finding as well, my best guess is ~3-5cm is a good working distance for small SMD components, so need a lens to focus that close.
Fraser, do you think going 50mm focal length is a sweetspot for the sensor size?

[Fraser]

Yes, the 50.8mm ZnSe lens would be a good general purpose choice for PCB work. Closer focus distance lenses are less common and likely more expensive. The wide angle lens of the camera dictates the minimum diameter lens for the application. As has been stated, 20mm is a good choice and reasonably priced. For even closer imaging you could buy the 38.1mm focus distance lens. This is less common but available on eBay.

There are two qualities of ZnSe lens used in CO2 laser engravers. The higher quality CVD lenses are justified up when using one with a 320 x 240 pixel microbolometer camera bit the lower resolution of the 690A etc is more forgiving of lens quality so the cheaper PVD lenses may be used I’d desired.

Fraser

[Fraser]

I just checked and a 25.4mm focus distance lens is available on eBay.

I personally would buy a 50mm FL lens for general work and the 25mm FL lens for when dealing with the really small SMD components. At 25mm you will have to consider the ease of access to the PCB if heatsinks are present.

Fraser

[Fraser]

Just a quick warning about doing very close-up thermography……

The Microbolometer operates at approximately 32 Degrees C. In a close up imaging mode of operation the heat energy from the microbolometer die is projected out through the lens system and actually illuminates the DUT. The camera operator and any other local sources of thermal energy will also be see reflected off of high reflectivity areas of a PCB, such as solder joint pts and screening cans etc. Amat a working distance of only 25mm it is possible that he self illumination effect of the camera may cause confusing results that the camera operator needs to recognise. If a part of a PCB is supposed to be unpowered and cool, yet shows a central heat spot at close range, and only at close range, question the truth of the situation as it could be a narcissus effect and you are seeing the reflection of the microbolometers own heat.

As a side note, my cooled thermal cameras project a cold spot out through the optics and their making array is operating at -296 Degrees C ! Close-up thermal imaging can suffer from narcissus effect.

Fraser

[snoop33]

Do you have a link to this lens you speak off?

I deal with 0402 components so I do ideally need to get fairy close but most the time I'm looking for heat from larger components like chips and inductors.

[Paw85]

The lenses I can find are ~€25 with shipping through Amazon. I'm not sure if thats a common or good price, but for 20mm lenses the different focus length cost the same; 25.4mm, 38.1mm and 50.8 mm.
(Ordering from chinese ebay sellers cost a fixed fee of around €10 plus VAT, I cant seem to find it much cheaper than the chinese sellers that got it in Amazon warehouses)

[Fraser]

There is a huge choice on eBay. In this application almost any of the correct diameter and FL will work fine.

Just search eBay for “ZnSe 50.8mm” or “ZnSe 25.4mm” and you will be offered many suppliers, most in China. You need to select the 20mm diameter for this camera due to its wide field of view.

I will do a quick search and highlight a supplier for you but you should check for yourself and select the lens to suit your needs and the amount you wish to spend. For your 0402 components, I recommend the 25.4mm FL lens for use in high density PCB layouts. That is a bit close for a general overview when looking for a hot spot though. 50.8mm or 38.1mm is better for that task
MCW Laser offer the 20mm diameter 25.4mm FL for £16.38

https://www.ebay.co.uk/itm/223541493548?hash=item340c1c7f2c:g:lAQAAOSwTepfQ0Vd

They have 38.1mm and 50.8mm FL lenses for the same price.Cheaper lenses are available though…..

[Paw85]

Just a quick warning about doing very close-up thermography……

That is very good information that most are probably not aware of, I certainly was not.
I did not notice such effect when taking the above images of the PCB and having the imager close to the PCB, it did not seem to reflect heat much at all, and the whole PCB was already above room temperatures, looking at the images probably just around ~30C .
However there was one PCB where testing that reflected huge amount of heat, could easily see my mirror heat image in it, this effect could probably be noticed there.
Thank you for the heads up!

[Fraser]

To see how pronounced, or not, the effect is with your particular camera you can look at yourself in a mirror and see not only your thermal reflection, but also the reflection of energy coming out of the cameras lens ! With these modern lower resolution cameras the microbolometer may not emit as much thermal energy so the effect may not be so much of an issue You can also look at your camera with another thermal camera to get a good view of the energy coming out of the front of the camera via the lens system. It is the close working distance that accentuates this effect as it would not normally be noticed at greater working distances.

Fraser

[snoop33]

Here's one on ali - https://www.aliexpress.com/item/32241074105.html

[Fraser]

Cloudray lenses are fine I have bought a few and had no problems.

Fraser

[Paw85]

I let colleges that work with electronics and PCB have a go with the UTi690A and evaluate if the performance would be useful for their work.
As shown previously the level of detail is not good enough to see small SMD components without an additional lens, but their comments was that the performance was good enough to cover their need of a thermal imager for developing purposes, to check heat transfer on the PCB, find which components to attach thermal sensors to for tests, and find at which area on a board there may be a fault or short circuit. (It would fail to spot a single short-circuit on very small SMD components, but they would narrow down the area of the fault down to ~3-4mm, and from there could find which of the components may be causing the issue with regular measurements of components.)
The images from the UTi690A are certainly no where near as good as from the ~$10k imager they normally use, but it seemed that for everything where they did not need to file a report with high resolution thermal images, this performance was good enough.
And due to the low price but still good enough performance, it was decided that a unit will soon be available at the workplace as well! Unlike the expensive units, anyone would be able to borrow it no questions asked, and if an accident happen would be very little financial impact to replace it.
I specifically asked if they would want a lens for better close up focus, but response was that in the default state was good enough from their point of view (and when very detailed pictures was needed for reports, they had the FLIR units to use).

Still, I'm curious what the performance would be like with a ZnSe lens so may purchase one for private use at least!

[Paw85]

I tried the PC software available at Banggood site that was claimed to be for this thermal imager and even had the UTi690A name in the file, but unfortunately it does not connect properly, so please avoid downloading that.
The UTi690A does not exist on Uni-T's website, and the UTi120S that seems to be the same thermal imager but with a different name does unfortunately not have a software for it listen on the website.
I did however manage to go to Uni-T's website and find software for the UTi120B and that software seem to connect properly to the UTi690A.
You can find the software on this page, under "Docs & Software": https://www.uni-trend.com/meters/html/product/Environmental/Thermal_Imager/Echo%20Series/UTi120B.html

With this software you do get options that are not present in the UTi690A itself if those may be important to you..
You can select to check the temperature at a point, min/max/avg temperatures along a drawn line, inside a drawn rectangle, or in a drawn circle.
A feature that seem a bit annoying is that it automatically adds a Min/Max point as well as a center point measurement to the images even if it was disabled on the unit at the time when taking the picture.

The software does not seem to allow you to import images saved off the device for example on the computer, it can only import files directly from the thermal imager.

The software do allow you to export a very basic PDF-based report of each image.
I did not purchase the thermal imager myself for this type of work doing reports, but I guess it's nice that it is possible even with this imager!



I did also try the freeware "Thermovision_JoeC" and it does seem to import the files produced by the UTi690A just fine (but software appear a bit too advanced for me, not sure how to operate it properly, but for the ones that do, it does seem to work!)

[Paw85]

One thing I did not like about the UTi690A that is shares with many other thermal imagers is a lack of a lens cover.
Today I started working on a 3D-printable lens cover and at the same time a a focus lens holder, developing them in the same item means that I can almost half the number of prototypes needed to be printed to test different features. (They will be separated into two items once done).
Copying a feature of the Flir E30bx that have the lens cover attached to a lanyard so it wont get lost.
Previously I did not think it would be possible to attach a cover at this position where it holds on right now as there is pretty much no backdarft, but printing the item in PLA which is very stiff makes it possible to grip that tiny area to hold it quite secure.

For the focus lens holder, the "support ring" that attaches to hold the lens in place have three cutouts around it where one can use hot-glue to hold it in place, without interfering with the thermal imager.

Are there anything you think could be improved looking at these early prototypes?

[Paw85]

Finished the lens cover today, turned out quite well and tested it on two the UTi690A units to make sure it's a good fit.
Lanyard stays on it really well, and one can jerk the thermal imager quite violently without the cover popping off.
Also had a bit of fun with the vinyl cutter, trying to make use of the different the machine I have when possible.

Development on the focus lens version will be on hold as I dont know anyone with a 20mm ZnSe lens and a UTi690A thermal imager, not do I myself at this moment have a use for one (so cant justify the cost as budget is... super tight right now...)

Will upload it to a STL-sharing site once I find a good alternative for Thingiverse.

[CANUK]

Looking good 

[Chalcogenide]

One thing I did not like about the UTi690A that is shares with many other thermal imagers is a lack of a lens cover.
Today I started working on a 3D-printable lens cover and at the same time a a focus lens holder, developing them in the same item means that I can almost half the number of prototypes needed to be printed to test different features. (They will be separated into two items once done).
Copying a feature of the Flir E30bx that have the lens cover attached to a lanyard so it wont get lost.
Previously I did not think it would be possible to attach a cover at this position where it holds on right now as there is pretty much no backdarft, but printing the item in PLA which is very stiff makes it possible to grip that tiny area to hold it quite secure.

For the focus lens holder, the "support ring" that attaches to hold the lens in place have three cutouts around it where one can use hot-glue to hold it in place, without interfering with the thermal imager.

Are there anything you think could be improved looking at these early prototypes?

I have bought a 20 mm ZnSe lens (50.8mm focus, as kindly suggested by Fraser) and I was looking at designing and 3D printing a holder. If you want, I could definitely try to print your design and see if it works!

[Paw85]

I have bought a 20 mm ZnSe lens (50.8mm focus, as kindly suggested by Fraser) and I was looking at designing and 3D printing a holder. If you want, I could definitely try to print your design and see if it works!

Thats great to hear, then I will resume the development on the lens holder as well, it may take a couple of days as I'm away from home a lot but will post a reply when having it ready to test.

[Paw85]

It would appear that UNI-T have launched yet another re-name of this unit, now UTi712S with specification and appearance that seem identical to UTi690A and UTI120S.
https://www.uni-trend.com/meters/html/product/Environmental/Thermal_Imager/UTi%20Echo%20Series/UTi712S.html

Does UNI-T usually do these things with their products? (seems to be happening to the UTi260B/UTi690B too?)

[mrisco]

But the UTI-120S was removed from the official ECHO series on their website.

I just discover that there are two Echo series webpages:

the old one: https://www.uni-trend.com/meters/html/product/Environmental/Thermal_Imager/Echo%20Series/
and the official: https://www.uni-trend.com/meters/html/product/Environmental/Thermal_Imager/UTi%20Echo%20Series/

Now the 120S is at only $130 and free shipping here: https://es.aliexpress.com/item/1005003683936847.html

[Chalcogenide]

In the end today, being half sick after the booster vaccine, I felt like giving a go at designing the lens holder. I have come up with a rough design, but I don't seem to gain a whole lot in terms of resolution. My knowledge in optics is fairly limited, so does anybody know the optimal distance at which to place the additional 50.8 mm lens in front of the camera lens? Should it go as far back as possible or is there an optimal distance?

[Paw85]

I think one have to experiment a bit, perhaps by holding it in front of the thermal imager and see when things change?
From what I read in the forum posts regarding ZnSe lenses, most seem to be mounting them as close as possible to core, probably for a reason?
Have you already tried mounting it both ways and you see no difference?

I managed to find a lens within my budget so have placed an order of a 50.8mm one too, will take a week to arrive though but will help with experimenting once it arrives.

[lordium]

I recent bought the UTI120S. Mainly for redoing some doors and windows now that temperature is dropping again.
I also bought a lens (20mm, 38.1mm) because why not. The lens works fine and allows me to hold the camera further away from pcb to get same picture as without.
This is helpful if you have mixed small and large components close to each other. But if you have a relatively flat pcb, the lens is not needed since you can get the camera very close.
The lens holder is something I made quickly using fusion 360 and a 3D printer. It is held in place by 2 magnets to either side of the lens. Works just fine, but if you try to swing it like Thors hammer, it will come loose.

Attach a few pictures for reference.

[Paw85]

I recent bought the UTI120S. Mainly for redoing some doors and windows now that temperature is dropping again.
I also bought a lens (20mm, 38.1mm) because why not. The lens works fine and allows me to hold the camera further away from pcb to get same picture as without.

Great to see more users with this fine unit, and great idea with the magnets to attach it, thats really clever!

I am a little confused about the results that you and Chalcogenide are getting with the ZnSe lenses, reading other forum posts, it would seem that one should be able to hold the thermal imager much closer while retaining focus, not further away?
Will help with this experimentation once I receive my lens, just a week to go...

[lordium]

Here are 2 more images. Both take from a height of about 4cm above the pcb using a vise as a holding jig. one with a 38.1mm lens, and the other without.

(don't mind the temperatures, they are not stable as pcb on/off intermittent, just look at scale of the holes)

[Fraser]

A small word of caution when using magnets near to the imaging core…….

The Flat Field Correction solenoid uses magnetism to operate. A high energy magnet placed near to that solenoid can interfere with its operation. Likely not an issue in this case, but worth considering when locating rare Earth magnets anywhere near to the cameras core and lens assembly.

Fraser

[lordium]

A small word of caution when using magnets near to the imaging core…….

The Flat Field Correction solenoid uses magnetism to operate. A high energy magnet placed near to that solenoid can interfere with its operation. Likely not an issue in this case, but worth considering when locating rare Earth magnets anywhere near to the cameras core and lens assembly.

Fraser

Yes the magnets I use are not very strong. I do not notice anything strange/different when they are close or not. Same quality image both cases.

So I step it up and attach 20 more magnets, now I can see a difference in operation, there is the sound like relay being toggled (like when calibrating) and image goes dark, but does resume normal operation when calibration is pressed again.

So Frasers warning is very valid indeed, don't use too strong magnets. small one will work just fine though (for at least MY UTI120S)

[Chalcogenide]

Here is my take on the lens holder for this camera. Everything fits by friction and snapping in place, no magnets or glue required:

https://www.thingiverse.com/thing:5181663

I hope someone might find this useful.

[Fraser]

Nice design  I see that it is a bit more than ‘just’ a friction fit Those side clamping pieces look excellent for holding the assembly onto the camera 

Fraser

[Paw85]

The lens cover I made can now be found here: https://www.prusaprinters.org/prints/110716-uni-t-uti690a-uti120s-uti712s-lens-cover-thermal-i
Had time to find a viable alternative to Thingiverse, didnt know that Prusa had a STL hosting site that requires no login to download files, and with a search engine that seem to be working well!

Waiting for arrival of 38 and 50 mm ZnSe lenses to test out what distance from the front works well, and will publish a lens holder once tested and see which lens works best for this core.
Meanwhile created a "manual holder" for a 20mm ZnSe lens to make it easier to test out lens distance on the fly, If interested, you can download it here: https://www.prusaprinters.org/prints/110811-znse-20-x-2-mm-manual-lens-holder

[Fraser]

Paw85,

Your ‘manual lens holder” is a most excellent idea  That will help me when experimenting with ZnSe lenses. I want to build a mini optical bench for experimenting with thermal lenses and your little manual holder makes me think I should print the parts for the optical bench. I think there are already designs on the Internet that I can adapt 

Thank you for the time you spent designing these parts for the forums use.

Fraser

[Paw85]

Received the ZnSe CVD lenses today, a couple of days earlier than planned, luckily was done with my lens holder yesterday so all was set for testing to begin!
I took images at identical distance per set (used a tripod).

My findings is that while the 38.1 mm ZnSe gives better detailed images over all than 50.8mm when in focus, it requires one to be very close to gain the benefit.
At ~50mm working distance, it performs worse than the 50.8mm lens as expected. But when moving close, the 50.8 mm does not seem to be performing much worse than the 38.1mm one.

My personal opinion is that for the UTi690A, the ZnSe 50.8 mm is the best performing lens to choose, one can even spot the solder joints on components that I believe to be 0402 size?
Even with 120x90 resolution of this thermal imager, with the help of a ZnSe lens PCB work should be OK!

Would like to hear what is your opinion from looking at the sample images?


Will publish the lens holder tomorrow, spent way too many hours over-engineering it, but offers protection for the ZnSe lens from both the front and back when not in use which I find important (when having the cover hanging in the lanyard).

And for fun before I might be selling my hacked E30bx (to finance other hobbies in these times...), figured I'd show a comparison when shooting board wide of the test subject.
The E30bx does seem suffer a non flat focal plane when focusing this close, but if one is doing professional PCB repair, this amazing thermal imager is clearly a benefit, it easily shows solder joints of 0402 components when taking a picture of the full PCB at once, it's a beast for sure!

EDIT: Before going to bed, attached a quick photo of the ZnSe lens cover as a sneakpeak if someone's curious.

[Paw85]

Just finished uploading the ZnSe lens holder/cover I made for the UTi690A, you can find the STL files here: https://www.prusaprinters.org/prints/114142-znse-lens-cover-uni-t-uti690a-uti120s-uti712s
This ZnSe holder/cover is just 1mm thicker than the cover for the thermal imager previously posted.

The most important feature I focused on is that the lens is covered both from the front and the back when not in use, as the whole item is basically a cover for the thermal imager when not using the ZnSe lens.
The sliding cover "clicks" in place at both open and closed position, and the circlip is designed such that it puts a force to hold the ZnSe lens in place when assembled, so no rattling from a loose lens.

I also recorded a video to show how it works, how to assemble the item and what to think about when printing it.
I do apologize for my spoken English language not being very good, but I hope you can understand what I'm saying.



Also attached some pictures of the ZnSe lens cover down below.

[CANUK]

Awesome job! 

[mrisco]

@737mechanic
The UTI690A and UTI690B are totally different devices.

[Halcyon]

@737mechanic
The UTI690A and UTI690B are totally different devices.

Looks like he was another spammer in the making. New user, posting the same thing everywhere. His posts have been removed and the user banned.

[Fraser]

@Halcyon,

I was not sure if the posts were SPAM or just someone excited about their new purchase. The sample pictures were actually interesting to see. Yes they posted in several threads, but those threads were all related to Uni-T cameras, so it seemed a reasonable act ? No link to a specific seller of the product IIRC. You have greater insight however so must have seen something suspicious

Maybe the SPAM link was to follow in the next message ?

Fraser

[snoop33]

I'm playing with mine, using a 38.1mm looking at SMD's around the 0402. I'm unimpressed so far, difficult to see what I'm looking at.

I have the got the lens the right way, flat side closest to the sensor? 

[Fraser]

The low resolution of this camera does make PCB analysis more challenging but it may still be used in this role. You need to consider the “pixel size on target” with your FOV and distance from the PCB components. This will tell you the minimum size of a component that will fully illuminate a single pixel on the microbolometer. Normally you will want a few pixels on the component to provide a useful image. Imagine a SMD resistor with a single pixel imaging it Vs four or six pixels all seeing it. The more pixels ‘on target’ the more detail you can pull out regarding shape and thermal energy distribution. The sample images provided by PAW85 show a camera that is capable of PCB thermal imaging but the components need to be generating heat to be seen clearly.

You will need to learn to interpret the thermal signatures of components. If there is a hot 0402 resistor in the scene, you should still see it against the cooler surroundings. A PCB that is not generating much heat in its components will look somewhat thermally flat.

With higher resolution cameras you are able to image the solder points at the end of components or IC leads due to the high reflectivity that often reflects your own body heat into the camera. These shiny areas also exhibit low emissivity when not reflecting heat energy from their surroundings (so appear cooler than the PCB). Between the solder points you may see nothing if the component is not generating thermal energy. With a lower resolution camera you do lose that level of detail and may only see components that are generating heat. This is the disadvantage of buying a cheaper, and lower resolution, camera. The UTi260B is my current recommendation for affordable PCB inspection using an “all-in-one” camera rather than a phone dongle etc. An upgraded FLIR E4 and ETS320 are other options, but are more expensive.

For more detailed PCB thermal analysis, I recommend the DYT CA-10 and CA-20 as excellent value for what they offer the user.

Fraser

[Paw85]

I'm playing with mine, using a 38.1mm looking at SMD's around the 0402. I'm unimpressed so far, difficult to see what I'm looking at.
I have the got the lens the right way, flat side closest to the sensor? 

What one can see when looking at small SMD components is usually the solder joints, as they are metal they reflect the surrounding temperatures.
While yes this is difficult to see, if it has the same temperature as the surrounding PCB, from what I know, it's likely not an issue with the component.

If you look at the last image of the Exxbx vs UTi690A comparison, you'll see that the same is true for that high resolution thermal imagers, you only see the solder joints of the 0402 components.
(Marked a few of them on the image below).
It may take a while to learn; If there is a short circuit component that is generating heat then you should be able to see the heat signature between the solder joints, even with the UTi690A.

Dont get me wrong, the UTi690A is in no way optimized for thermal imaging of PCB.
If using it daily for PCB work then you may want a higher resolution sensor for sure, but for occasional use at least I find it good enough, but that may also be that I've used thermal imagers for a few years now?
If running a buisiness or doing these types of work daily then there are specialized thermal imager developed for PCB work as Fraser mentioned, those will outperform the UTi690A for PCB work in every way, but you're also paying a lot more for it, and they are not a standalone unit that you can use everywhere.

For the low budget range the UTi690A is to my knowledge the thermal imager with the highest resolution, most in this price range are 32x32 pixel sensors, and a very few at 80x60 pixels.

For lens orientation, flat side against the sensor did seem to work best for my images when I tried.
Perhaps you can upload a few images with information about the working distance you used?

[Johnny B Good]

@Paw85

 Thank you for that last very helpful posting (as well as all your previous postings). After reading through the whole topic thread, I thought that you'd nicely summed up the pros and cons of this latest UNI-T offering which I'd only become aware of from a Banggood marketing email today. It was my search for more information, over and above all the glowing customer reviews (yours included as I've just discovered ), that led me to this topic thread.

 They're currently offering it at a knockdown price of only $169.99 (£127.78 inclusive of VAT and free postage to the UK). I'm rather tempted to 'impulse buy' it, despite it being 30% more than I paid for an OWON XDM1041 4 1/2 digit bench multimeter just a couple of weeks almost a month ago.

 The price of these thermal imaging cameras has always struck me as being rather exorbitant for the seemingly paltry image resolutions they provide but since this high pricing applies across the board for any thermal imaging camera, I've long since come to accept that thermal imaging "Is bloody expensive"

 After checking out eBay prices for this model, Banggood seem to be undercutting "the competition" by anywhere from 47 to 26%. Not only that, Banggood's cheaper 8x8 pixel devices with only one 168.75th of the UNI-T's areal resolution are still over one third the price of this unit.   On the face of it, this does seem to be somewhat of a bargain buy in its class.

 Unless I receive any advice regarding cheaper same resolution alternatives or higher resolution options for a relatively modest increase in price point, it looks like I'll be giving in to my impulse in a few days time.

[Paw85]

The price of these thermal imaging cameras has always struck me as being rather exorbitant for the seemingly paltry image resolutions they provide but since this high pricing applies across the board for any thermal imaging camera, I've long since come to accept that thermal imaging "Is bloody expensive"

I hope that the information posted have been useful. One should probably view thermal imagers image resolution differently than that of regular cameras;
You're actually collecting a temperature reading of objects at every single pixel of the area that sensor can "see", it's like having a huge amount of temperature sensors that you can use to collect data with from a distance, it's amazing really and if that data is interpreted to colored pixels on a screen that turns out to be beautiful, then that should just be a bonus

I believe that at least my conclusion is;
If you're fine with the 120x90 pixel resolution, then the UTi690A is a great unit for the price. (And often cheaper than lower resolutions units).
If you need higher resolution, then save up double cost and go for the UTi-260B as it also should be a great unit for its price. (Have not tried this one myself but it's recommended everywhere so it's probably a good unit).

I usually look at larger components like stepper drivers, cables and motors in 3Dprinters and other machinery and I find the UTi-690A works fine for that use, and in the very rare occasions that I need to look at tiny SMD components then with the ZnSe lens the performance was good enough for my use. Please note; Not great or good, but just good enough to even spot tiny 0402 components if they should heat up from short circuit or similar.
If doing a lot of PCB SMD work then saving up for the UTi-260B is advisable for sure with the higher resolution.

I cant say if 120x90 pixels will be enough for your use cases, but if it is, then the UTi-690A is a great buy

[Johnny B Good]

 Thanks for that insightful advice. I'm currently limited to a single pixel resolution (i.e. an IR thermometer ) when trying to get some sort of handle on powerfet temperatures and such like. That 690A represents a considerable upgrade to my ever growing armoury of electronics test gear.

 Although it would be nice to have more pixels, I figure I'd be better off enjoying the thermal imaging experience at the level being offered by this throw away priced unit before investing in a considerably more expensive device when I'll have a better idea of just exactly what I need.

 The 690A (with additional lens attachments) might well serve my needs for several years before the compulsion to upgrade overwhelms my decades long habit of financial jurisprudence (lifetime long habits are hard to shake off).

 Well... I've just placed my order with Banggood.   With shipping and tariff insurance it came to a total of £130.88, money that I'm sure is well spent. It seems  it's "Estimated to ship before February 19, 2022" (from China) so it could well be another three weeks or so before I can offer any feedback.

[snoop33]

I'm just coming back to this, 3d printing your cover now.

There's an interesting cover design for the higher res model in the other thread but the design has more to grip on to.

[snoop33]

I have found the slider to be very difficult to print, this is on an Ender 3. Those overhangs require supports which are difficult to remove and destroyed the rise to the left when I tried to remove them. I don't think the design is viable.

[Paw85]

I have found the slider to be very difficult to print, this is on an Ender 3. Those overhangs require supports which are difficult to remove and destroyed the rise to the left when I tried to remove them. I don't think the design is viable.

I hope you don't take this the wrong way, but that does sound like a 3Dprinter/slicer issue.
The design is not the easiest thing in the world to print, but I've printed it without issue on 3 printers myself during development, and there are at least 6 other people that have printed it successfully without issue that I know of, one even used PET-G that is a lot more challenging when it comes to support sticking to prints.

If your supports are difficult to remove perhaps try and see if you can minimize or solve that issue and it will likely help with all your following 3Dprinting that use support as well.
I did that myself a year ago and it have helped a lot on all prints with support!

You can send me a private message with an image of the print with issue if you want and I can try my best to help, but there are guides online that are probably better than me at explaining what may help, english is not my first language.
Lets try not to flood the thread with 3Dprinter discussions, you can also contact me on the prusa website, that's a perfect place to discuss issues with 3Dprinting things.

You're always welcome to make another design that is easier to print, the more options people have the better!

[Johnny B Good]

@Paw85

Update on my UTi-690A order status from Banggood:- I've just received an email advising me of delayed delivery due to excess demand and given the choice of waiting longer or cancel and request a full refund. I chose the latter option since I had  started suffering "buyer's remorse" as a result of taking a closer look at the UTi690B you'd recommended and saw that it also offered a visible light camera, allowing optical and thermal image fusion to help clarify what you can see.

 I've added this:-

 https://www.banggood.com/UNI-T-UNi690B-256+192-Pixel-Infrared-Thermal-Imager-15~550C-Industrial-Thermal-Imaging-Camera-Handheld-USB-Infrared-Thermometer-p-1755789.html?cur_warehouse=CN&rmmds=search

 to my shopping cart for the time being since I figured I may as well be hung for a sheep as for a lamb (you're actually getting 4 times the number of IR pixels for a 'mere doubling' of the cost).

 I've got another week before the offer expires so I thought I'd ask your opinion of this Banggood bargain before I 'pull the trigger'. I'd checked out pricing on AliExpress but their pricing versus model numbers is confusing to say the least and the only certain price point on an actual UTi590B I found was 268 quid, seemingly a close match to Banggood's pricing until you remember that you have to add VAT onto their pre-VAT prices, making it 20% dearer than Banggood's vat inclusive price.

[AlexP.]

The CZ warehouse offer is okay but it was a tad cheaper when I bought mine, but not by much.

If you can combine with one of their discount coupons and/or VIP advantages, the price can become even sweeter sometimes.

[Johnny B Good]

@AlexP.

 Thanks for mentioning the possibility of applying a discount coupon. I managed to locate a £39 off coupon that reduced the special offer price of £268.95 (plus £7.45 for shipping and tariff insurance) down to a total of just £265.94 all in (the coupon only discounts the full, non special price of £298.09). Not quite as big a discount as I thought I'd be getting but another £10.46 saving is not to be sneezed at. 

 I'd been viewing youtube reviews when I saw your reply. After looking at an 8 month old Defpom review where he mentioned the possibility of a discount coupon, I just DDGed for for "UNI-T UTi690A discount codes" and managed to find a current coupon which, to my surprise, was accepted at the checkout.

 Banggood say they are packing my order (really?) yet state an expected shipping time of 22nd March (do they really mean that or is it a mis-translation for 'expected delivery date by 22nd March'?)

 Never mind, this time it looks like it may well be another month or so before I'll be able to report my own experiences of this thermal imaging camera bargain.

[Johnny B Good]

 Well, my UT690B finally arrived this sunny Sunday afternoon whilst I was out on the bay window roof, erecting my GNSS RTK antenna (bought just before Christmas!).

 I've only now gotten round to unboxing it, some 3 hours later after completing the antenna installation and tidying everything up (and admiring the 50+ dB c/n signals in u-centre ).

 The camera seems to be working ok but I've only given it the briefest of tests - I have other projects vying for my attention right now (Rubidium reference oscillator demanding a better quality GPS reception in order to shorten the syntonising process, hence the long overdue antenna upgrade).

 I'll post a fuller report later, as and when I can find the time.

[snoop33]

It's not, when designing parts to be 3d printed you have to design them to be 3d printed.

The tiny overhangs where gaps will be infilled with supports which can't then be easily removed. Needs a rethink.

[Halcyon]

It's not, when designing parts to be 3d printed you have to design them to be 3d printed.

The tiny overhangs where gaps will be infilled with supports which can't then be easily removed. Needs a rethink.

How is this related to the thread?

[Johnny B Good]

It's addressing an issue of printing shorter focal distance lens holder adapters here (and even further upthread as I recall)

https://www.eevblog.com/forum/thermal-imaging/uni-t-uti690a-(120x90-entrybudget)/msg3912533/#msg3912533