Uni-t UTi260B

[KedasProbe]

Got my new UTi260B camera today works well. (although kept like hanging after selecting language the first time)

When I started to wonder how do I get this overlay text from my screenshots...

This link is very useful to 'remove' (duplicate present) the text overlay of the screenshots and get some extra temp data.
https://github.com/Santi-hr/UNI-T-Thermal-Utilities

Not sure if the person is here on the forum but thanks works very well.

[Fraser]

That poorly UTi690B is haunting me from my “spares & repair” shelf ! I truly hate giving up on a patient and I really want to better understand why that camera is not working with the GC0308 camera modules that I sourced. I have decided to revisit the camera when time permits.

Fraser

[Fraser]

For those who own a UTi260B or UTi690B it may be of interest to hear that when working on the poorly UTi690B, I noted that almost all of the ABS screw towers had been split by the self tapping screw. One that provides a mounting point for the microbolometer heatsink had completely disintegrated into several splintered pieces of ABS plastic. This is not that uncommon an occurrence when designers elect to use self tapping screws into relatively fragile ABS mounting towers/pillars. Sadly in the case of the UTi690B, I found significant splitting and fracturing of the towers and all will require repair. I am thinking of providing details and pictures of how I repair these types of failure in case others need to repair such at some point in the future. If you ever hear something rattling around inside your UTi260B/UTi690B (or other UNI-T cameras for that matter) it is likely a piece of the ABS mounting towers that has broken away.

The solution to this issue is the use of threaded brass inserts moulded into the ABS plastic towers at the time of manufacture. Retro fitting such inserts is not normally possible due to under-size towers intended for M2 self tapping screws. Even brass inserts cab suffer from cracking of the supporting ABS tower, but it is less common in my experience. I judge equipment build quality based upon many factors but it is always good to see threaded brass inserts and not the use of self tapping screws in a products plastics design.

Fraser

[Veteran68]

Thanks, that's a relief then. Unfortunately I looked at but didn't write down the exact version or backup my firmware first (I wasn't aware of the fw backup process until I came here looking for solutions AFTER I bricked it).    I want to say mine might have been something like 1.2.32 but I can't swear to it. I did download someone's stock fw backup that was posted in the backup thread though, so if I can get the UART to work then I'll try that.

The upgrade process did, thankfully, save my calibration data to a file on the SD Card, which I have since backed up. I assume if I apply someone else's firmware image, I'll have to restore this calibration file. I'll cross that bridge when I get to it.

My firmware is 1.2.28 and I can’t find any newer of series “1.” Anything from series “2.” It will not work, at least two of them that I try 2.0.20 and 2.0.22 that’s why I create that simple script in the backup thread you mention, to collect some firmwares here so we make a database since we cannot find nothing on the net.

Any way if you want my firmware I will pass it to you.

I changed only UTi260B_Thermal and device is normal again no any calibration data since I get the same as old, every time you run upgrade menu that .ini file is written to your as card.

Finally I found the time to solder on the UART Dupont header and get logged into the console as root. Copied over the 1.1.23 firmware file that Silicium81 uploaded over in your backup thread, as his was a UTi690B firmware. I don't know if there's a later one for the 690B or not. If you don't mind sharing I'd like to take a look at your 1.2.28 although I expect it's for the 260B model. Supposedly identical, I know, but I'd just prefer to keep mine on the correctly ID'ed firmware image. I wish I remembered what version I had before!

I'm going to follow up with Banggood who asked me to wait until they could hear back from Uni-T. I'll let them know I recovered it, and ask if there's a later 1.x version than 1.1.23 and remind them the 2.x versions they're sending out don't work for everyone.

[Fraser]

I just completed my photography of the broken ABS mounting towers on the UTi690B and it is not good news 

Every single tower mount has either cracked or shattered due to the forces created by the self tapping screws that have been driven into them.
Whilst it is not that unusual to see stress marks or even the odd fine crack amongst a set of tower mounts in equipment, I have not previously seen so much stress related damage to the tower mounts as found in the UTi690B. It is frankly appalling build quality to have so much damage caused by the production process. Some towers have split, but are still likely to do their job, for a while at least, while others have suffered severe cracking in multiple locations and basically shattered. The severely cracked towers are just waiting to fall apart as one did during the initial disassembly of the camera. One of the towers that supports the microbolometer heatsink fell apart into four pieces when I removed the screw that it was supposed to retain. Even the additional buttress reinforcement used around some of the towers have ful, thickness cracks in them !

Look closely at the provided pictures and the full horror of the situation becomes evident. The UTi690B is supposed to be able to survive a 2m drop onto a hard surface. I can see several of the towers disintegrating if this particular camera was dropped. Not great and, in my opinion, unacceptable build quality. UNI-T should have used threaded brass inserts or, if not, they should have done their calculations better on the tower dimensions, screw dimensions and ABS selection.

To the pictures and make your own assessment of this cameras tower mounts........ note that some towers may look OK, but closer inspection reveals at least one full thickness crack extending way down the tower side. Others have fine crazing over them indicating multiple fine fractures within the tower. The towers that have obvious multiple cracks around their tops are severely compromised by multiple long, full thickness cracks down their sides.

[Fraser]

Continued....

[Fraser]

Continued....

[Algoma]

They could vastly improve the strength of those posts with having a glass fiber component in the plastic mix. Always saving a few pennies at the cost of everything else.

[Fraser]

Interestingly, the teardown pictures of a UTi220 Pro show the same design of mounting tower, yet I see less evidence of strain on them from screw insertion. A softer ABS mix maybe ?

I attach pictures of a Hikvision camera and the UTi220 Pro from a previous thread that I created.

https://www.eevblog.com/forum/thermal-imaging/hikvision-ds-2tp31-series-thermal-camera-teardown-307840/

Note the use of brass inserts in the Hikvision camera. A far more professional design that is likely to suffer less from cracking of the mounting towers during its operational life.

Fraser

[Bill W]

Interestingly, the teardown pictures of a UTi220 Pro show the same design of mounting tower, yet I see less evidence of strain on them from screw insertion. A softer ABS mix maybe ?

See image 06865

It claims to be 'PCTPU' rather than ABS.  That suggests some kind of polycarbonante with polyurethane mixed in, and polycarbonate is very hard, brittle etc and susceptible to any adhesive vapours around.
Not to mention the 'Chinese equivalent the same honest' grade, maybe second hand regrounds.

Even if the camera manufacturer tries to do it right, the moulders are a law unto themselves.

[Fraser]

Thanks Bill. Interesting. I was not certain what PC/TPU meant but suspected it was meaning a plastic shell with TPU inlays which are the soft red buffer material. Polycarbonate is strong, as you say but clearly did not like screws forcing their way into it. The “PC” material readily melts when plastic welded with solvent so I can plastic weld all of the cracked towers and come up with a plan for the refitting of fasteners. One option would be to fill the hollow towers with JB Weld hard epoxy, then drill and tap them for machine screws.

As we know, this particular camera is potentially just a spares donor but it makes an interesting project to see how I may repair the damaged casing towers. Such a process may help other owners so it will be worth the effort.

Fraser

[Fraser]

PC/TPU……

https://techarmor.com/blog/all-about-pc-tpu

[Fraser]

I just took a look at a UTi690B teardown by another forum member…..

https://wizeus.de/teardown-of-the-uni-t-pro-uti690b-thermal-camera/

That cameras casing is not showing the damage to the plastic mounting towers that I see in my patient. Could it be that the OEM used different screws that increased the stress on the towers of the camera that I am working on ? As Bill has stated, we could also be dealing with a variance in the plastic quality or composition.

Fraser

[Fraser]

For anyone wondering what products I use to repair damage to plastic casings, I attach a picture 
I also use helicoils and brass thread inserts where applicable. I was fortunate enough to find a heated brass thread insert installer from the USA but a temperature controlled soldering iron can do the job if done carefully.

I have a lot of practical experience repairing plastic casing parts from my years of restoring vintage laptops. Those often have stress fractures in the panels and tower mounts that secure the casing parts and screen hinges. I will not fill this thread with the repair techniques that I will employ for the UTi690B and will maybe create a new casing repair thread.

Fraser

[Veteran68]

I just took a look at a UTi690B teardown by another forum member…..

https://wizeus.de/teardown-of-the-uni-t-pro-uti690b-thermal-camera/

That cameras casing is not showing the damage to the plastic mounting towers that I see in my patient. Could it be that the OEM used different screws that increased the stress on the towers of the camera that I am working on ? As Bill has stated, we could also be dealing with a variance in the plastic quality or composition.

Fraser

As I posted earlier, I had my 690B apart this past week to get access to the TTL pins for UART access to the console, so I could recover from a bad firmware upgrade. While I did not tear it down completely (i.e. did not remove the heatsink), from what I did tear down I did not notice the cracking that you're seeing in those bosses, nor did I have any debris floating around in the case to indicate any prior breakage. I will say one of the self-tapping screws did not seem to want to tighten down to a stop, and I didn't want to keep trying in case it stripped further, but the rest of them tightened down fine.

Perhaps yours was a "Monday morning" or "Friday afternoon" special at the moulding plant. I expect UNI-T's manufacturing cost points don't allow for a high degree of QA attention, so it could be really hit or miss as to what you get.

[Bill W]

Thanks Bill. Interesting. I was not certain what PC/TPU meant but suspected it was meaning a plastic shell with TPU inlays which are the soft red buffer material. Polycarbonate is strong, as you say but clearly did not like screws forcing their way into it. The “PC” material readily melts when plastic welded with solvent so I can plastic weld all of the cracked towers and come up with a plan for the refitting of fasteners. One option would be to fill the hollow towers with JB Weld hard epoxy, then drill and tap them for machine screws.

As we know, this particular camera is potentially just a spares donor but it makes an interesting project to see how I may repair the damaged casing towers. Such a process may help other owners so it will be worth the effort.

Fraser

Yes, TPU could well just refer to the outer (overmoulded ?) 'rubbery' stuff.  In that case pure polycarbonate for the case, with none too surprising results.

That said 'M3 self tapping screw' is a very variable beast, nothing like as well defined as M3 machine screw.  Split towers is a 'screw too big for hole' issue, however it has come about.

[Fraser]

I completed stage 1 of the UTi690B tower repairs this evening. All towers had wire containment applied before being repaired with solvent welding techniques. The more I looked, the more cracks and crazing I found ! Total nightmare. Every single tower had both cracking and crazing that could have lead to complete tower failure. The tower that shattered into 4 pieces has been reassembled and refitted into the casing using plastic welding solvent. Considering the state of that tower when I dismantled the camera, I am very pleased with the result. Modelling plastic welding solvent is great for these tasks as it permeates cracks through capillary action and this includes the areas of crazing. The plastic surfaces then melt together. No adhesive, just plastic merging into plastic. 

If I were repairing this camera on a commercial basis, the failure and repair of the towers alone would likely make this camera B.E.R.

I now have to decide upon the new fastener regime and the options are very limited. Brass inserts would be difficult to use due to the small diameter towers and their already compromised state would make them vulnerable to disintegration during drilling for the insert. Self tapping screws, even if smaller diameter, do not look like a good option here. I am left with filling the towers with high performance JB Weld epoxy and tapping them for M2 machine screws. The M2 screws have a very fine thread which makes the task a challenge but I believe it can be done if care is taken. It will be a time consuming task to fill, cure, drill and tap all of the towers in the camera casing but I can think of no better option at the moment and this approach adds hard epoxy to strengthen the towers. They are hollow to their full depth.

[TobyG]

If looks don't matter, 3D printing new towers and glueing or solvent-melting, depending on the type of plastic, them in place is a viable option that I'd used a couple times in the past, complete with brass thread inserts, that is.


I do repair a bunch of older motorcycle speedometers and Rallye Navigation Equipment, so I do encounter such issues fairly often, too.




That being said, don't beat yourself up over not fixing that thing.
Totally agree with you, with the price point, there simple is a time to call it quits.

Well, I'll be shopping for a higher quality one, now that I know I got a use for it. Not often, but when needed, it's quite a nice thing to have.


Again, many thanks for giving it a shot at all!

[Fraser]

Thanks Toby 

The tower repair is going well. I repaired the cracks with solvent plastic welding and then filled the towers with JB Weld heavy duty epoxy. I have waited for the epoxy to cure for a few days. I will drill each tower to 1.6mm and then tap with a M2 tap set. I purchased a nice assortment of socket head M2 stainless steel machine screws to replace the original 2mm self tapping screws.

I am fortunate to own a selection of excellent Proxxon power tools designed for modelling and these are perfect for working on projects like this. I have a really dinky mini pillar drill with which to drill the small 1.6mm holes so quite an easy task.

I will document the repair to the towers, either here, or in a separate post.

Fraser

[Fraser]

I should also say that I have yet more versions of the GC0308 camera modules coming from China ! This is just to further investigate the issue of the camera not being recognised. They are relatively inexpensive at around £3 each + shipping, so worth investigating.

Fraser

[digant]

Thanks, that's a relief then. Unfortunately I looked at but didn't write down the exact version or backup my firmware first (I wasn't aware of the fw backup process until I came here looking for solutions AFTER I bricked it).    I want to say mine might have been something like 1.2.32 but I can't swear to it. I did download someone's stock fw backup that was posted in the backup thread though, so if I can get the UART to work then I'll try that.

The upgrade process did, thankfully, save my calibration data to a file on the SD Card, which I have since backed up. I assume if I apply someone else's firmware image, I'll have to restore this calibration file. I'll cross that bridge when I get to it.

My firmware is 1.2.28 and I can’t find any newer of series “1.” Anything from series “2.” It will not work, at least two of them that I try 2.0.20 and 2.0.22 that’s why I create that simple script in the backup thread you mention, to collect some firmwares here so we make a database since we cannot find nothing on the net.

Any way if you want my firmware I will pass it to you.

I changed only UTi260B_Thermal and device is normal again no any calibration data since I get the same as old, every time you run upgrade menu that .ini file is written to your as card.

Are we sure the 690B and 260B have the same firmware version numbers? Why would Banggood be providing people this version for 690B devices without asking what version you have to start with? There would typically be a warning in the instructions they provide to check current firmware version. I will reply back to them that I'm hearing that v1 firmware cannot be upgraded with v2 firmware and see what they say.

Let me go through the backup thread and check the versions that I've downloaded, then try to get the UART connection going.

Where can I download the firmware 1.2.28 or last stable version for uti260b? I'm currently on stock 1.1.22 (never updated the fw).

[Kellopeli]

First of of all big thanks for Fraser and other for sharing their experience with this thermal cam.

I got the 690B variant less than a year ago and that sucker died on me.

I paid about 300 European dollhairs, and it got used 3 times. So it ended up costing 100 Euros for each use.
Lesson learnt again - stay away from chintsy home gamer tools. There is a reason why you never see Uni-Ts used by professionals here in north Europe.

After not using it for several months I tried to turn it on. There was just a brief flash of backlight and then it turned black.
I thought it was just an empty battery and quickly hooked it up to a USB-c charger (for Lenovo laptop).
After charging for a while it remained completely dead. I tried Banggood custormer service, but all I got was a "sorry your crap is out of warranty "

So I'll have to try my luck repairing it meself.

Next I stumbled on this thread and was able to gut the camera. I used a suction cup and a little heat to remote the plastic screen protector.
Inside visually everything seemed to be ok. No trace of smoke. I went to ebay and ordered some replacement P-channel mosfets.

Battery:
I also checked the internal battery on the terminals and it is dead, 0 Volts. Hmm.
There seems to be a built in protector circuit in the battery under the plastic wrapping, so I went ahead and carefully removed some plastic wrapping on top of the positive and negative strips connecting the cell to the protector circuit and it measured 3.05 Volts, so the cell was not dead after all. I tried to charge the battery with a regular li-ion charger, but it could not detect the battery. I put some small jumper wire to the exposed strips and connected the chareger directly to the cell. I charged the cell to 3.8 volts, and hoped the protector circuit would wake up. The terminals still show 0 Volts. Maybe the protector circuit died.

Booting up:
So the battery was dead, maybe that was the reason for the unit not booting, so next I hooked a lab power to the battery connector on the main PCB and started experimenting.

3.3V:
First I set the lab power to 3.3V (simulating an almost empty battery) and wanted to see if it powers up. Well, partial success. I get now the Uni-t splash screen and initializing gets stucked at 69%. Current draw is about 350 mA. Power button does not work. after couple of tries the boot gets stuck at 70%.
3.6V:
I raise the voltage a little and the unit gets stuck at 70% for a very long time (several minutes) and then goes to 100% and gets stuck. Intresting. I do a power reset a few times and power button does not work. Unit gets stuck at 100%. Maybe I need to raise the voltage more.
4.1V:
Now the boot seem a little faster and on the first try it booted almost up and did not get stuck at 100%. Instead it got stuck on the main camera view screen with "Initializing". Current draw is about 400mA. Power button has now some effect, it switches off the screen, but current draw remains over 300mA, which is not good. I repeat the same a few times. Power cycling seems to be the only way to stop the unit from drawing high current. Maybe I raise the voltage a little.
4.2V:
This would resemble a full battery. This time the unit boots fully up and starts the thermal camera, wohoo! Now also the power button shuts down the unit properly and current draw goes from 400 mA to zero. I notice the date on the screen is from 2022, so RTC battery probably also died.

What happened:
I am suspecting I had an empty battery, then hooked the unit up to a powerful charger and burned something inside. The unit still somehow works, but requires a higher voltage, and would not work unless the battery was fully charged. I am suspecting the unit has a design flaw or a FW bug causing the unit not to always shut down properly, especially with lower supply voltage, and this may then finalize your battery.


I will continue the repair some day when I have time.

[Silicium81]

Thanks for your comment, it's interesting to know that the battery's BMS is integrated into thelse. These BMS are not very qualitative in general.
I've also had a 690b thermal camera for a year and a half. It works very well if i use it occasionally and recharge it regularly, but no more often than every 4 months and only if the battery starts to run down (2 bars on the charge indicator).

[Veteran68]

Lesson learnt again - stay away from chintsy home gamer tools. There is a reason why you never see Uni-Ts used by professionals here in north Europe.

While I wouldn't consider most Uni-T gear as "professional," I would certainly put it in the mid and sometimes upper range of hobby gear. Far from what I'll call "chintsy" compared to the sea of cheaper no-name brands. For most hobby or even prosumer users, their gear will do fine. For it's price range, the 260B/690B camera is one of the best on the market in terms of features. Alex from Northridge Fix said it compared very favorably to his $6K FLIR camera (although one should probably take his review with healthy dose of salt since he sells the 260B in his store). But when you look at raw specs, it's pretty hard to find a comparable performing camera in this price range. If you want close to the same features in something from a brand like FLIR or Fluke, you're close to adding another zero to the cost.

My 690B has been fantastic. My only issue with it was self-inflicted, attempting to flash a firmware that clearly isn't compatible (although we had to figure that out the hard way). Thankfully their design is such to make it relatively easy to recover when this happens.

I tried Banggood custormer service, but all I got was a "sorry your crap is out of warranty "

That's odd, because mine was almost certainly out of warranty yet they were very helpful with me. Well, at least their attitude was helpful and they were quick to provide me a firmware to try, unfortunately it was the same incompatible version. They said they would take the firmware issue up with Uni-T but never got back to me. After I fixed mine myself, I didn't follow up. But they were far more helpful than I expected them to be.

What happened:
I am suspecting I had an empty battery, then hooked the unit up to a powerful charger and burned something inside.

This is doubtful. It uses a standard USB charging circuit, so a charger isn't going to deliver any more power than the device draws and only up to the negotiated charger limit (unless you're using a non-compliant charger). I haven't tried a USB power meter while charging it (now I'm curious and will do so) but as I recall it comes with a USB-C to USB-A cable so it's not even PD, which means it's charging at standard 5V 1-2A max. I'd say more than likely there was a battery defect or other component defect in yours that led to your failure.

[tchleb]

Which ZnSe lens you can recommend for PCB thermal analysis?