REVIEW: Dianyang Technology CA10 Circuit Board Thermal Analyzer - Repair / R&D

[Fraser]

Alternative mounts for the CA-10 Camera head:
 

The CA-10 camera head assembly may be attached to any mount system that provides a standard 1/4" 20 tpi tripod male screw thread. This feature increases the versatility of the system and may assist some users who have special deployment needs.

I provide two examples of mounting systems that I will be using with the CA-10 camera head.

One is a neat little tabletop ball head tripod and the other is a compact articulated arm that is designed for use with disability aids and iPads etc. I was fortunate enough to find two brand new Latitude arms at only £13.50 each but sadly their normal price is nearer to $400    Similar articulated arms are available at far more reasonable prices though. The unit I have is beautifully made and uses a standard Manfrotto Superclamp to which the manufacturer has added their own articulated arm. The length of the arms are perfect for this application and the Manfrotto clamp is excellent quality. The Manfrotto clamp retails at around £25 in the UK.

[Fraser]

For those who have not seen it, this thread may also be of interest........

https://www.eevblog.com/forum/thermal-imaging/use-of-a-thermal-camera-for-pcba-thermal-profiling-and-repair/

Fraser

[Fraser]

Now that I have uploaded the review on the CA-10 I will expand my comments a little to cover aspects of the CA-10 system that I feel are worthy of discussion but not really needing to be included in the formal review. No system is perfect and the CA-10 is no exception, but it is a very good design with only minor niggles and I found no "showstoppers". I will happily use this camera in my daily repair work.

So what shall I discuss first ?

Well there is no doubting the thought that went into the mechanical design of the CA-10 system. The electronics used within its camera head are also of good quality and an area that I would like to expand upon a little. The core that is used in the CA-10 is one of the new offerings coming out of China. For this reason I am interested in its performance when compared to the likes of the FLIR Lepton and Seek Thermal cores. As indicated in my review, I am impressed with the imaging that this core produces and found it more than adequate for the task of PCBA imaging at close range. I was also impressed with the CA-10's temperature measurement accuracy. The one surprise I had was the narrowness of span that the imaging core and software would select when looking at a flat field test plate. The core produces low noise imagery with a span of just 1 Degree C ! I found this hard to believe considering the use of 12um pixels in the core. The images did not lie though. The cores manufacturer has clearly worked hard in the area of image noise reduction and they use the 'thermal enhancer' grids on top of the pixels to improve sensitivity. This is a system used to good effect in the 12um Pixel DRS Talisman cores that also produce excellent imagery. With regard to the noise reduction systems used within the core or its associated software, there is likley a price to pay for such in terms of image definition but I saw no impact on the high frame rate. As can be seen in the images collected from my testing of various PCBA's and the SMT resistor test, there is no doubt that the CA-10 can image the smallest of SMT components with relative ease. The lens is capable of focussing at very short distances and I found that this was a distinct advantage over my other thermal imaging cameras, even when fitted with a supplementary lens. Using a standard lens at such close focus distances is guaranteed to produce some optical effects and the image can suffer from geometric distortion but this is a tool for imaging and measuring PCBA's so such distortion is of little or no consequence. The lens focus system feels solid in the fingers and appears built to last.

At the very small temperature spans that the core elects to use when viewing a flat field, I did note a thermal radient across the flat scene. The far left side of the display was 1 Degree C cooler than the far right side. This level of error falls well within the stated measurement tolerance of the core and CA-10 system and should not be considered a fault. My investigation of the phenomenom suggests that it is related to the FFC flag system design within the very confined environs of the imaging core. This temperature gradient did not cause me concern or interfere with the testing of the PCBA's as the provided images demonstrate.

The CA-10 communicates with the host PC via USB C cable but will happily operate on my USB 2 ports. The PC does some (all?) of the image processing and there is a small amount of delay in the system that needs to be considered when adjusting the cameras focus. I quickly got used to this however and focussing became quick and simple after some practice. I tended to use a combination of course height setting, folowed by fine height setting to frame the PCBA correctly and then fine adjustment of the lens focus. In some cases I used the fine height adjustment as a fine focus control as well. After a little use, the setup and focus of the CA-10 became second nature to me and good imagery was quickly achieved. 

I carried out an environmental temperature test on the CA-10 camera head by placing it in a fridge for an hour at 5C. I was pleased to see that the camera behaved normally as it measured a PCBA and gradually came up to the room ambient temperature of 24C. Those who work in cold labs need not worry about the CA-10 misbehaving 

[jmw]

Wow, thanks for the review - since the company posted about it and said you were doing a review, I've been checking back now and then to see if you had finished. Have you checked in with Dianyang and asked if they might do any specials for EEVBlog members? 

Could you comment more on your experience with the software? How smooth & fast was the UI, did it have bugs/crashes, missing features? When using a system that relies on PC integration, the quality of the software is critical. It would nice if the software side was open/hackable and not tied to Windows, but nothing's perfect. Is the software at least a "good citizen" in the Windows ecosystem (i.e. has a installer, an uninstaller that works, signed executables, doesn't require administrative privileges for normal operation)?

[Fraser]

Ok, to cover your questions in the order asked…….

The UI is smooth and fast with no glitches experienced during testing. The niggle I had was with the temperature Plot screens disappearing if I clicked any other function or anywhere on the desktop that was not within the plot box.

I experienced no crashes with the software during extensive use on a Windows 7 OS but I will need to load it on my Windows 10 laptop to comment on the stability with that OS. I thought I found a bug in that the user cannot take a photo of the 3D screen which I thought odd. DYT confirmed that this was deliberate however as they saw no use for that display beyond real time analysis. As previously stated, there is a small time lag between changes in the thermal scene and those changes occurring on the computers display. This is not uncommon with Dongle type thermal cameras that use a USB link and Host computer for the image processing. The delay was not enough to concern me and I was able to operate the manual focus without issues as you just have to do it slowly to allow the display to catch up with the focus changes.

As for missing features….. this is an area that we can explore further in this thread. DYT are aware that there are features not present in the CA-10 that might be present on the more expensive PCBA analysis thermal cameras. They want to produce an affordable PCBA thermal analysis camera that can be used for any tech but the focus was on real time analysis of mobile phone and similar technology. In that scenario the thermal camera is often used to identify anomalous behaviour in IC’s and discrete components, such as shorted MLCC capacitors of minute size. DYT included the Circuit Design mode to serve users wishing to create plots of temperatures within an ROI. This is why I was disappointed that he plots could not be saved. I used screen captures to save plots and they worked well, but are not ideal. Yes there are functions missing from the DYTSpectrumOwl software that I would like to see added if DYT want the unit to be that best that it can be for R&D applications. I will create a wish list here so that DYT can consider my, and other forum members comments when considering future Software updates. I have already suggested a separate advanced user mode so that the current customer base who desire simplicity of use are not confused by more advanced R&D functionality. It will be for DYT to decide whether they wish to add more advanced functionality to the software package but this thread is the place to let them know your thoughts on the matter. As it stands, the software is a very useful PCBA thermal analysis tool that performs its duties very well and is an invaluable aid to repairing modern electronics. It is not, however, an advanced Research and Development system, more of a basic R&D solution that meets the less demanding needs of some designers and R&D techs. We must not forget that this unit retails at less that $800 and that is very cheap in terms of R&D camera systems.

With regard to open software, I believe that is unlikely to happen. The reason is simple, the DYTSpectrumOwl software has taken a lot of development to get to its current state and DYT are invested in improving it further within their business model. The Software is what turns a generic thermal imaging core into a PCBA analysis tool and DYT are targeting that niche market with the CA-10 system. The DYTSpectrumOwl Software is their advantage over other manufacturers of thermal imaging equipment. That advantage would be lost if the software were to be made open for the user base. I have enquired about the potential for Android, Linux and MAC DYTSpectrumOwl versions and these are unlikely to be released as DYT has found that Windows is dominant in their intended market place and even Android would present significant development challenges due to the variations in users Android configurations. The CA-10 appears to be a ‘Windows only’ solution but for the cost of a cheap used Windows laptop or tablet PC the CA-10 can have its own dedicated host PC for very little additional investment.

The DYTSpectrumOwl software installer is…… interesting… it is partially in Chinese ! That said, I had no issues installing the software on my PC. I will be raising the Chinese text in the installer with DYT as this needs to be corrected. The software seems non invasive as once it has been unpacked into a folder, the user can move that folder anywhere they wish, including another PC and just run the .EXE file for normal operation of DYTSpectrumOwl. No uninstaller is required as the software does not deeply integrate with the OS  It just sits on top of the OS and runs within its own little World  With regard to the software being a “Good Citizen” in the Windows ecosystem…….. in short, the DYTSpectrumOwl is not Signed and Windows does ask the User if they wish to run the software each time it is started. I attach a picture showing the “unknown” Publisher status. This is not an issue for me but others who require signed executables will need to consider their options. The software runs within its own little ‘bubble’ and even saves its images and videos within its own file structure. The photo and video replay functions are handed off to the Resident Windows programs for such file types and so avoids integrating an additional media player function into the DYTSpectrumOwl software. To my eyes, DYT have created a software package that is as non invasive of the Windows structure as possible  I like that approach and, of course, the DYTSpectrumOwl software could be moved to a USB stick and run from there if desired.

More from me later on the features that would be nice to see added to DYTSpectrumOwl if DYT wish to further enhance its capabilities. I have high hopes for this product. It is already an excellent repair aid but it could be so much more as the hardware is of excellent quality and performance. DYT have no interest in turning the CA-10 into a ‘generic’ thermal imaging camera, even though it’s current functionality is adequate for that anyway, but they appear keen to produce a refined niche product at an affordable cost to the Repair and development sectors of the marketplace. I am certainly very pleased with what they have already achieved with the CA-10 and it’s associated Software. DYT will need to be convinced that the demand exists for any additional functionality to be added to the Software as such development takes time and money. We likely all have our own ‘wish lists’ but DYT are a business and need to see a return on their investment by increased sales.

As a final comment… I am so pleased that DYT elected to go down the path of using a domestic Chinese thermal imaging core. This has enabled good communications (same language) between the cores OEM and DYT. The Chinese thermal imaging technology development, driven by the current Pandemic, appears to have resulted in some significant improvements in the available miniature 12um pixel imaging cores performance. Good to see. And let us not forget…. The CA-10 offers internationally shipped 25fps image refresh…. A significant advantage over <9fps when capturing rapid changes in the thermal scene as I witnessed with the Amazon Firestick tests

Fraser.

[Fraser]

I have just been compiling a ‘wish list’ for R&D features that would be nice to have provided by the DYTSpectrumOwl software but I started to feel uneasy about it. DYT have made their ideas behind the CA-10 system clear to me and a key factor was to serve the very popular mobile phone repair industry around the World. That industry is a huge market for DYT to tap into and modern electronics can sometimes be far quicker to repair if the tech has access to a suitable thermal imaging system. By suitable I mean an affordable unit that is easy to deploy and use whilst producing the required imagery. The Circuit Design mode was included to provide a temperature plotting capability that can be useful in some  more involved scenarios. It is not a high end R&D analysis mode however and so lacks features and functionality that I have used in $150K+ thermal camera systems like my FLIR SC4000 and FLIR ResearchIR MAX 4. That is a ~$800 Vs $150K comparison that would be totally inappropriate and unfair to the CA-10 system …. Much like comparing a Ford Focus to an Aston Martin Vanquish car !

It is all too easy to create a huge ‘wish list’ of features without considering the core market that the CA-10 is targeting and the additional complexity with development costs associated with advanced R&D analysis functions that are found in the likes of FLIR ResearchIR. ResearcherIR 4 MAX can cost many thousands of Dollars alone so can we really expect DYT to produce an equal product within the current $800 retail price of the CA-10. If I am honest I do not know but I would hate to create any sort of negative atmosphere around the excellent CA-10 by producing a demanding and unrealistic ‘wish list’ of features that are not currently present in the DYTSpectrumOwl software. This will need some thought in my part. There may be some features that can be easily implemented at little cost and maybe those are the ones that should be highlighted. I can start the list with my desire for Radiometric file saving and the inclusion of Monochrome palettes. I have versions of the DYTSpectrum Owl software that contain a huge list of LUT palettes, yet only 5 have been made available to the user. That might be worth expanding or providing a facility to set a few custom palette selections from the long list provided. Having said that I only missed the monochrome palettes during my tests. I used the Hottest Area function to produce a Monochrome Palette across most of the PCBA but a dedicated Monochrome palette would have been better. Before anyone asks why Monochrome is useful…. The human eyes and brain processes colour and monochrome images differently.

Well that is enough from me. Just some food for thought.

[Fraser]

A fun and useful little ‘hack’ that Bill W told me about ages ago……….

If you source a lens with the same, or almost the same optical characteristics as the Objective fitted in a thermal camera and you mount the additional lens reversed in front of the objective as a supplemental lens …….. you get a thermal microscope  That is to say, if you have a core with 12um pixels the size of pixel on the target item will also be ~12um  The focus distance will be approximately the back focus distance of the second (supplemental) lens so likely less than 10mm in most cases. If you added such a reversed lens to the CA-10 you would gain a true thermal microscope capability with all the advantages of a solid ‘microscope like’ stand. Time for me to experiment with some of my spare lenses maybe 

The supplemental lens will cause a measurement error due to transmission losses etc but thankfully DYT provide a calibration offset option in the setup menu 

Fraser

[Fraser]

Discount on the CA-10 system ?

I did not answer this question earlier, and I am not affiliated with Dianyang Technology, but I believe I can say that the CA10 BoM is not an insignificant cost to produce. The imaging core is a 'bought-in' component and the metal mounting system is complex and looks expensive to produce. I doubt that there is much that can be done on the discount front but it may be worth contacting Dianyang Technology to discuss such matters ?

Fraser

[Fraser]

I just test fitted the CA-10 camera head to the Latitude articulated arm and attach some pictues of the assembly.

The use of an articulated arm would permit the use of the CA-10 camera in otherwise inaccesible areas of equipment. Imagine wanting to image a PC motherboard without removing it from its casing. The articulated arm would reach over the sides of the casing and hold the camera head steady over the required area of the PCBA. As users of optical microscopes in electronics know, it is an advantage to have the option to use a 'long reach' mount for those occasions when such is needed. Working on large PCBA's as found in a Sony PS5 or XBox for instance, can challenge the reach of conventional desk mounting systems. I found the Latitude articulated arm to be both very solid and very steady so it is perfect for the CA-10 head. In a perfect world the CA-10 mounting arm would be removeable from the head to reduce bulk, but that is not the case so we have to work with what we have. The knuckle joint at the head will allow the mounting arm to be rotated 90 degrees vertical so that the articulated arm can lower the head into more confined areas, such as inside a PC casing. For anyone buying the CA-10 system, it would be worth considering the purchase of a well made articulated arm as well. Such will increase its versatility at relatively low cost 

The web site for the Latitude arm is here but the price is eye watering  I have not found it under another brand so suspect that it is a custom Ablenet Inc product.

https://www.ablenetinc.com/latitude-with-super-clamp/

The Latitude articulated arm has a female 1/4" 20TPI mount so I bought a common photographic 1/4" male to 1/4"  male coupler to mount the CA-10 camera head to the arm mounting.

[Fraser]

My new lab PCBA thermal workstation 

The compact stand is just a cheap Monitor stand from Amazon. It is currently discounted from £17.99 to £8.99 and made of Bamboo wood. Fits my requirements perfectly though  It acts as a solid base for the articulated arm and a convenient elevated stand for the CA-10 system. I can even place a laptop on the lower shelf when it is not needed.

https://www.amazon.co.uk/Monitor-Smartphone-Notches-Computer-Projector/dp/B08BHKP2D5/ref=sr_1_3?dchild=1&keywords=monitor+stand+bamboo&qid=1625412536&sr=8-3

Fraser

[Fraser]

As previously mentioned, some of the installation process showed Chinese text which added some fun to the process but it was easy to work out. I have just screen captured the screens that you may see connected with the installation of the DYTSpectrumOwl software. I have used an online translator and attached the translations.

I have informed DYT of this oversight and hopefully it will be corrected soon. Choose the wrong option and you will delete the software installation so random button pressing is not advisable.

I just checked and deleting DYTSpectrumOwl does not delete the CA10 folder or the saved images 

Fraser

[jmw]

Yikes. Unfortunately that makes the CA-10 a no-go in my book, but good to know now as I was initially quite excited about it. It seems to be a common problem that industrial equipment manufacturers do not consider software as a "value-add" and deliver equipment with brilliant mechanical design and excellent finishes, and software that is ... a turd. Privilege escalation prompt on startup, unsigned binaries, and saving files within the program directory structure are signs of sloppy software engineering. I've seen this with Hikvision software and the only way I feel comfortable running it is from within a VM. Malware is no joke. There's no reason a program that takes data from a USB peripheral and displays it on the screen needs administrative privileges, and I expect software that runs this way will stop working in future versions of Windows as Microsoft tightens the screws on platform security. Buyer beware.

[Fraser]

The software deletion process translation.......

Use the Setup file again for uninstall........

[Fraser]

jmw,

Whilst I understand your concern about the way the software may be configured, I do think some common sense measures mitigate the risks to a manageable level.

Solution: Just run the software and CA-10 as a dedicated tool that is not connected to a network or the Internet. Used laptops are cheap and plentiful.

[jmw]

That's a reasonable take; either use a dedicated device or a VM.  If they had an Android version, putting it on some cheap tablet would be a good solution too. Longer term, I'm worried the proprietary software won't be maintained and will break with some future Windows update, but with the rate technology is progressing, something better may be available by then and it may still be worth if it gives you several years of service.

[Fraser]

jmw,

If you are running on a standalone laptop with no network connectivity, there will be no updates to break it 

In the repair world ~$800 will not scare people, it is when a camera costs them $8K that they worry about how long it will take to pay for itself. We have seen with Alex at NorthridgeFIX how quickly some faults can be found using a thermal imaging camera. That time saved frees him up for other paying repairs so I presume the camera is paying its way ? In his case though he is using an expensive FLIR E60 that does not have a very close focus capability. I own an E40 upgraded to E60+ specification and it can be a cumbersome beast with shallow depth of field and not ideal for PCBA inspection.

Sadly there is very little choice in the market where PCBA thermal inspection is concerned. Hence why DYT created the CA-10. I own an ETS320 and it is far from perfect with a fixed focus lens and what is a standard E8 chassis grafted into a different case format. FLIR really dropped the ball on that design  Not cheap either.

[Bill W]

I have enquired about the potential for Android, Linux and MAC DYTSpectrumOwl versions and these are unlikely to be released as DYT has found that Windows is dominant in their intended market place and even Android would present significant development challenges due to the variations in users Android configurations.




With regard to the software being a “Good Citizen” in the Windows ecosystem…….. in short, the DYTSpectrumOwl is not Signed and Windows does ask the User if they wish to run the software each time it is started. I attach a picture showing the “unknown” Publisher status. This is not an issue for me but others who require signed executables will need to consider their options.

Thanks for the review, DYT seem a somewhat better type of Chinese company.

As regards the quoted above:

Might be worth investigating MAC using one of the many "translator/emulator" softwares often used by MAC folks for WIN-only software.

While using a program directory is really quite negligent these days, if the software is moved to a benign section of the PC not policed by Windows (folder of a second hard drive say), does it still trigger the warning on each use ?

Bill

[Fraser]

Bill,

I will check.

I neglected to make clear that the DYTSpectrumOWL software offers user selectable locations for image and video saving. I assume it defaults to the software directory in order that the software is portable and can be moved within the computers directory structure or to other computers.

Dianyang Technology have the very welcome desire to offer the market quality thermal imaging products, with good specifications, at an affordable price. They look for areas of the market that are poorly catered for and try to fill the void with a good product. Having previously communicated with a member of the DYT team it was clear that the company wants to become known for producing quality products and they put the extra effort into the design as a result. Good to see and I hope we see more products from this company.

Fraser

[Fraser]

DYT have confirmed that the software installer for DYTSpectrumOwl was compiled with the wrong language kit. This will be corrected.

Fraser

[JimM]

With regard to open software, I believe that is unlikely to happen. The reason is simple, the DYTSpectrumOwl software has taken a lot of development to get to its current state and DYT are invested in improving it further within their business model.

But wouldn't DYT be able to sell more CA-10 units if they had open software so that a third party could develop software for Linux and Mac?

[Fraser]

I honestly do not know. DYT have their view on the matter and only they can comment on offering open source software. Marketing can be a complex field and I am certainly not qualified to comment 

[Bill W]

But wouldn't DYT be able to sell more CA-10 units if they had open software so that a third party could develop software for Linux and Mac?

I would suggest that there may be too much risk to the main sales.
Yes, they'd win a handful of orders from a few enthusiast cranks, but get a lot of support requests.

They'd risk copycat products using their software, as such an open source would also be open to reskinning for Windows OS.  The use of a common China core is no barrier, their main sales points are mechanics (with associated cost, so someone would do a cheaper, tackier, nastier version in that special bendy Chinese plastic) and their software which no doubt had a significant a cost to develop.

Bill

[Fraser]

Bill_W,

I agree with your comments and love your description of the plastics used on some cheap clone Asia sourced products 

[Fraser]

I am working mobile at the moment and have brought the CA-10 kit with me 

Coming very soon …….. thermal imaging of a 3.5” hard disk drive PCB, demonstrating setting several temperature marker positions and quickly comparing drive PCB’s thermal profile 

Plus maybe some ‘Ant action’ ! 

Watch this space

[Dianyang]

Hi All,

First I would thank for Mr. Fraser's very detail evaluation of our CA-10 PCBA thermal analyzer!

And I would like to answer some questions,

1.   Regarding Apple and Linux, actually we know these 2 OS has a lot of users, but now we don't have enough human resource to develop these version software, and we don’t have Apple computer, we know nothing about Apple OS, maybe later we need to hire one engineer to develop it if CA-10 sell good enough.
2.   Regarding Android equipment, we will consider to develop the Android application in the future.
3.   Regarding white hot and black hot color palette, we are considering to add these 2 color in the future, and maybe more palettes.
4.   Regarding to open the SDK, this is possible but maybe the next generation product. the traditional thermal camera only provides central, max and min etc., general features, we want to let the thermal camera more useful and more easy to use, if you have any requirement and we consider that is possible to develop, we will add it to our new version.
5.   Regarding the bugs like Chinese text etc., we will fix it for the next version.
6.   The image with raw data, we will release this version in the near future, then the users could save all the thermal information in the picture files.
7.   Regarding the mechanical design, actually this cost us 7 months to R&D and produce, the CNC is a very hard job for thermal company, we want to provide a cost-effective product, but not a cheap and bad plastic one, so we changed the design and made samples again and again, and thanks for god, finally we made it.

We will keep investment on this product and will release new version software soon to enhance R&D in the thermal design, also we will release new type products on PCBA analysis, hope you will like it.

If you want to try this product, please go to Amazon url to buy it, https://www.amazon.com/dp/B098JSNSC4?ref=myi_title_dp.

Thanks for your support on our product, and hope to hear more suggestions!

Adel Chai
Key Account Manager
adel@dianytech.com