EEVblog Electronics Community Forum
Products => Thermal Imaging => Topic started by: Fraser on October 05, 2020, 11:29:12 am
-
Some time ago I carried out a complete Teardown of an AGEMA Thermovision THV470 in order to better understand its design. The camera was a spares donor unit due to corroded mirrors so the teardown was not wrecking a working camera ! This spares donor has already provided essential parts to repair a THV450 :)
Before the pictures, some background to the THV470 model is in order to better understand the design.....
AGEMA were a major player in the World if thermal imaging and produced some excellent thermal cameras. AGEMA used to be AGA, another well known name in thermal imaging equipment. Early thermal Imaging cameras were based upon cooled detectors and scanning mirror or prism optical blocks that created a thermal scene using a raster scan system. These mechanical scanning type cameras were born out of necessity as the cooled detector was often a single pixel and not the array type sensors that we see in modern cameras. With just on pixel to work with, the designers had to present the thermal scene to the pixel in a serial mode of operation with the scene broken down into horizontal scanned lines and a complete field created by vertical movement of the scanned line across the thermal scene. A reverse operation using the scan coils of a CRT monitor could recreate the thermal image on a display for the user. With further development the scanner signal became digitized by a ADC and digital signal processing was possible. As stated, the scanning assembly was mechanical in nature and could comprise two spinning Germanium prisms or a combination of tilting and spinning mirrors. Galvanometers and high quality motors were used in these scanning systems and these needed to be carefully synchronised to create the desired raster feed to the detector pixel. These cameras were complex precision electro-mechanical assemblies and this meant that they were relatively bulky, heavy and expensive to manufacture.
Once the thermal scene signal had been captured using the optics and scanning assembly, the detectors signal was passed to amplifiers and then electronic processing stages, either analogue or digital in nature. The detector pixel presented its own challenges to designers however. The detector was made using specialist semiconducting materials but these were effectively blind at ambient temperatures due to their own internal thermal noise. In order to use the detectors for thermal imaging, they had to be cooled to cryogenic temperatures of 77K. This was achieved using Liquid Nitrogen that was contained within the camera in a Dewar that had an integrated detector pixel. At 77K (-196C) the detector pixel was very low noise and was capable of good quality thermal imaging. The drawback to this system was the need for Liquid Nitrogen. Liquid Nitrogen is not the most user friendly substance, it limited the continuous running time without refilling the Dewar and the camera could only be used in a roughly horizontal plane and definitely not invented ! Transporting Liquid Nitrogen also presented logistical challenges for portable use. The use of Liquid Nitrogen was acceptable for laboratory use but a less cumbersome cooling method was needed for portable use. Development work lead to the introduction of the miniature Stirling Cycle mechanical cooler. This is basically a miniature refrigeration heat pump that can achieve a cold finger temperature of 77K or lower. The cooler uses a Helium gas fill to achieve this level of performance. The Stirling Cooler was a godsend to Thermographers working in mobile situations but it sadly had its own drawbacks. The precision cooler had a life of around 2000 hours before it needed a complete rebuild and re-gas. This was an expensive operation as the detector pixel was integrated into the cooler assembly so a simple “cooler swap” was not possible and the camera had to undergo calibration if the cooler/detector was changed. The intrinsic cost of the cooler+detector was also high and this lead to scary service costs for owners of the equipment. The high servicing costs of the sterling Coolers drove development of cheaper cooling alternatives. Some thermal camera manufacturers had opted to use pressurised gas cooling, such as Hughes and their Argon gas based detector cooler. This used the expansion of gas passing through a heat exchanger to create the cooling effect for the detector. It was easier than using Liquid Notrogen, but still not ideal !
Development surrounding the Peltier Element technology discovered that a Peltier ‘stack’ could produce temperatures of -70C reasonably easily. The ‘stack’ was pyramid shaped and comprised several Peltier elements stacked one on top of the other. I will not go into the physics here but you cannot just stack Peltier Elements of the same size and achieve -70C at the detector. The elements must start small and increase in dimensions with each subsequent layer. Such a cooler has limited cooling power at -70C so, like with the Stirling Cooler, the Detector pixel resides inside a vacuum with the cooling source. This lowered the thermal load on the cooler to acceptable levels. Now the sharp minded will note that the TEC only achieves -70C and not -196C so how could this work in a thermal imaging system ? Well new material doping mixes were developed for the detectors to achieve a Band-gap that would operate as a thermal detector at ‘only’ -70C the noise level coming out of the detector was higher than that of a detector operating at -196C (77K) but was acceptable.
It is this Peltier Element Stack based cooling technology that resides within the THV470.
To Be Continued ..........
-
Ok, next installment.... the THV470 camera......
This needs a little history to explain its existence and design.
The THV470 has a brother called the THV450 which looks the same except it does not have a floppy disk drive. That is because it is basically a cost reduced version of the THV470. The cost reductions include the loss of the floppy disk drive and a fixed mount lens rather than a removable lens. There are many areas of the THV450 that are the same as the THV470 however so this teardown will be useful to owners of the THV450 as well.
The THV450 and 470 are both built around a common scanner block that is found in several cameras. Let me expand upon this.
In the beginning there was the THV880 scanner type thermal imaging camera. It used Liquid Nitrogen for its cooling but there was a later upgrade to Stirling Cooler that a user could purchase. The "head" assembly comprised a relatively large and heavy metal 'box' casing that contains a combination of fixed and moving mirrors that pass their output to the cooled detector via some Germanium relay lenses and any filters that are fitted. The signal from the detector is amplified and sent down an umbilical cable to the desktop controller for 'processing' and image recreation. The controller came in several versions to suit the different needs of the user. They ranged from simple analogue systems to relatively sophisticated computer based image analysers such as the BRUTE system. The mid point was a controller that digitised the analogue signal from the detector pixel in order to provide basic measurement and analysis function for the user.
A THV870 scanner head was developed that used the Peltier Element TEC system to cool a 'HOT' detector pixel that could function well at only -70C. This THV870 head looked the same as the THV880 Liquid Nitrogen cooled head and contained the same mechanical scanning engine as the 880 model. The 870 differed at the point where the image signal reached the detector. The 'HOT' detector was integrated into a vacuum package with the Peltier cooler and this was thermally bonded to a forced air cooled heatsink. The Controllers were very similar to those of the THV880 with only minor changes needed to the design.
A drawback to the THV880 and 870 was the lack of portability. In order to make the system portable, a bulky cart was needed to hold the controller, its power source and the monitor. AGEMA addressed this matter by building a Frankenstein THV870 system that integrated the original THV870 scanning head into a common chassis with the electronics normally found in the separate bench mounted controller. The resulting bulky 'all in one' camera solution was turned into a shoulder mounted portable thermal imaging camera with the addition of a large battery belt and electronic viewfinder. Portable it was, Pretty it was not ! Once the side covers of the THV470 were removed, its ancestry was clear to see. A clever piece of engineering but you cannot escape the fact that it is basically a THV870 with the controller PCB's bolted on the rear ! It worked though and it did provide a portable, if still bulky, solution for thermal imaging whilst mobile.
At this point I should state that the AGEMA scanning thermal camera line-up of the period can get a little confusing. The same scanner head can be seen with different model numbers as they formed part of specific 'systems' that AGEMA were marketing. This why the THV487 scanning head has the appearance of a THV870 ! AGEMA were pairing scanning heads with different controller systems and branding the system as a Thermovision THVxxx even though parts of the system were available under their own, different, model numbers. As highlighted by the THV450/470 differences, some systems were designed to be more versatile than others or capable of radiometric measurements when others were not. Lots of marketing at work here ! I should also state that the scanning cameras heads were often available in MWIR and LWIR to suit the users needs. The version is identified on the ID plate and lens annotation.
OK, so now we know the background to the THV450/470 FrankenCamera ;D
I will generate some posts that show the cameras detailed in the above comments.
Fraser
To be Continued........
-
The vernerable Agema Thermovision THV880 scanning thermal imager head unit including a cutaway view of the image path to the Liquid Nitrogen cooled detector.
I also include pictures of the associated THV800 standard controller unit.
Fraser
-
The Peltier TEC cooled Agema Thermovision THV870 camera and controller
(Some images borrowed from : http://wll.kr/bbs/board.php?bo_table=06&wr_id=795 (http://wll.kr/bbs/board.php?bo_table=06&wr_id=795) )
Fraser
-
AGEMA Thermovision 450 shoulder mounted camera with non removable lens and no floppy disk drive.
Fraser
-
And finally the AGEMA Thermovison THV470 shoulder mounted thermal camera.
The pictures of the teardown of this camera model will follow shortly :)
Fraser
-
Ha Ha, whilst searching out the pictures of the THV470 teardown I found that I had already written a guide to the model series and the faults that I discovered in my THV470 and a friends THV450. It is likely a first draft as I lost interest in posting it but I include it here anyway........... There will be repetition so you can skim that part ;D
Agema Thermovision 450 & 470 Thermal imaging cameras.
Specification:
Wavelength : Medium Wave Infra Red (2um-5um)
Cooled or Uncooled: Thermoelectrically cooled
Detector/sensor : MCT 'hot mode' doped SPRITE Detector
Image forming system : Mechanical scanning raster using Lenses and Mirrors
Resolution: 140 lines x 100 sample points per line
Line rate: 4kHz
NETD: 100mK
The Thermovision 450 & 470 are self contained scanner type thermal imaging camera of the shoulder mount format. The camera uses a mechanical mirror based image creation system combined with a SPRITE cooled detector. The detector is cooled, but not by Liquid Nitrogen, Argon gas cooler or a Stirling Cooler. It uses a Peltier element 'stack' operating on the rear of the detector within a vacuum Dewar. The detector is operated at -70C so a long way away from the -196C used in a Liquid Nitrogen or Stirling Cooler based camera. This is achieved by special doping of the SPRITE detector element and careful selection of the Band-Gap.
The Thermovison 870 and its associated 800 controller form a Peltier cooled thermal imaging system that has a separate scanning head driven by a bench mounted controller. The THV450 and 470 are basically a THV870 and its controller combined into a single casing so this teardown may interest owners of the THV870 as well :).
A THV470 appeared on eBay at the right price. This particular example had clearly been stored in conditions conducive to corrosion :( I was hopeful that the camera might respond to some TLC but it was very much a calculated risk.
Initial assessment of the THV470 confirmed that it was not booting and had been stored in damp conditions. The AR coating was separating from the Germanium lenses which is symptomatic of damp induced degradation. The unit appeared to be complete and cared for so had likely fallen out of use and was just placed in storage.
There is a lesson to be learned here though, thermal imaging equipment needs to be stored in appropriate conditions if serious deterioration is to be avoided.
Access to the internal parts in these cameras is made simple by the easily removed side panels. After removing the securing screws the two panels come away from the chassis to reveal the scanning head at the front and the image processing and control electronics at the rear of the camera. The Peltier cooler sits on the rear of the detector at the mid point of the camera design and has forced air cooling of its heatsink. The scanner section of the unit is basically the head assembly of a THV870 that has been cleverly grafted into a longer chassis that accommodates the image processing electronics. The electronics package comprises a mixture of fixed and removable PCB's. A card rack and backplane is used for the image processing and control PCB's whilst two large PCB's sit either side of the scanner mechanics in the scanner head. These PCB's deal with the SRITE detector related signals and the Servo drive functions.
All PCB's are easily accessed and removed and it is clear throughout how much thought has gone into the design of this camera and its stable mates. These units a mechanical-electronic hybrids that required expertise in many areas of design and production within the Swedish R&D team. These scanning cameras came from the heydays of AGEMA thermal imaging camera production and predate the BST and Microbolometer sensor arrays. The wonderful AGEMA Thermovision THV550 came after the THV4xx series and provided a very compact cooled thermal imaging camera that made older AGEMA scanner type cameras look very bulky indeed. The cooled THV550 was a palm camcorder format and used a Stirling Cooler and 320 x 240 pixel staring array operating at -196C. The THV550 effectively marked the beginning of the end for the bulky and heavy scanner type thermal imaging cameras like the THV4xx, 8xx, 9xx series. Industry bought the new compact THV550 and either traded-in the old scanner models or left them languishing in storage. Such was likely the fate of my THV470.
So what did I find inside the poorly THV470 ?
First impressions of the interior of the camera were initially good. It looked very clean. I noted some corrosion around the AA sized Lithium Thionyl Chloride memory battery. Sadly removing the other side panel immediately revealed significant leakage of the Lithium cells electrolyte onto the panel. This was an age related failure. the electrolyte has damaged the power PCB that accommodates the cell and the top area of one of the cameras image processing PCB's that sits beneath the cell in the chassis.
Not great, but by no means fatal.
All easily removable PCB's were extracted and inspected for damage. No serious damage was found. These cameras are literally stuffed with integrated circuits. Very different to modern 'black blob' SOC technology ! The firmware resides on UV erasable EPROM's. Owners of these cameras should make back-up copies of their firmware in case of future age related data corruption.
The scanner section of the camera head uses a pair of moving mirrors and several fixed precision mirrors. These are all highly polished aluminium. As those familiar with aluminium will know, it can be highly polished and works well as a mirror. Sadly it is also vulnerable to corrosion when exposed to a damp atmosphere. I was expecting the worst when I removed the dust covers that protect the mirrors. I was not 'disappointed' by the patient.... all of the mirrors were corroded beyond use :(
The first moving mirror creates the vertical scan movement using a tilt action and is driven by a precision galvanometer. The second moving mirror is a facetted mirror wheel that rotates at high speed to create the horizontal scan movement. This wheel is attached to a precision engineered motor. The mechanics of these cameras must have been very expensive to produce.
Once the poor condition of the mirrors was confirmed, the THV470 effectively became a spares donor as to return the camera to operation would require specialist restoration of the mirrors that placed the unit well beyond economic repair. You will recall that I bought this camera to study its design and I had low expectations regarding getting it serviceable again if faulty. I knew, however, that its parts are of value to others so I could always sell its parts to help others maintain their THV4xx series cameras. I hate to declare a thermal camera beyond help, but on this occasion it was clearly the case. It might go on to save another camera with its parts however.
The story of a friends poorly THV450
The THV470 languished in its case in my garage for a few months before a fellow forum member asked for some help with a THV450 that was booting but not producing a thermal image. After my last dealings with the series I had sworn myself off of them as they do take a lot of time to disassemble and reassemble. They are also a precision mechanical device so there is no simple way to work on the optical scanning section if that has a fault. Remember, no service information is available for these cameras and mechanical alignment of scanning optics can be a real nightmare without the correct documentation and jigs. This was a request from a friend though so I parked my negativity and agreed to look at the THV450, but only at a superficial level for anything obvious. Well that plan soon went out of the window :)
I got stuck in to the THV450 and quickly diagnosed a non functional scanner section but the video processor board and detector were working so there was hope for the unit. The dead scanner section could be caused by a number of faults but a hot favourite would be a failed servo drive board. My THV470 has a servo driver board that is the same as that used in THV450 so an obvious path for testing was to fit that board in the poorly 450 :) The replacement servo board was fitted and then the camera was retested. I was rewarded with the sweet sound of the mirror wheel motor starting so life had returned to the scanner section. The motor did not sound right though and its speed varied all over the place. The tilt mirror was not running and that was a sign that the mirror wheel PLL was not achieving rpm lock. I switched off the camera and noted the almost immediate stopping of the mirror wheel motor. That is not normal in these cameras. The pinning mirror wheel acts like a flywheel and the motor spins down slowly after power is removed. I did wonder whether the THV4xx series employed a motor brake so would need to investigate that as well. The uneven motor speed during the test was the priority. There was the possibility of faults in the motor, motor drive, servo PLL or mirror position sensing system. As an initial check I tested the mirror wheel for free movement....... it offered a weird spongy resistance not typical of a motor or bearing fault. This investigation was getting more and more interesting so I could not resist delving deeper into the scary precision mechanical scanning system. I knew it could turn into a rabbit hole but I hate to be beaten ! I tested the camera again but this time the mirror wheel did not spin-up and only a quiet hum was audible. I had either killed the good servo board or something had changed since the last test was started. I went back to the spongy feel of the mirror wheel when rotated. Maybe something was now jamming the drive to the wheel ? That was an assembly that must be precisely aligned to the other optics in the system and its looked a pig to work on. Oh well, down the rabbit hole we go !
Before attempting surgery on the patient I practiced on my THV470 scanner block to learn about how best to disassemble parts of it in a way that allowed correct alignment upon reassembly. The THV470 was a parts donor and it had already proved invaluable when working on the THV450. Investigation of the mirror drive system in the THV470 revealed age related degradation in the adhesive that was used in a shaft coupling for the mirror wheel and its associated motors anti-vibration mount assembly. Could it be that the mirror wheel coupling had failed in the THV450 causing uneven mirror speed due to slippage ? That would definitely be high on the checks list and the coupling would need testing anyway, even if still complete, as the adhesive may also be degraded. The adhesive had turned to a white powder in the THV470 but thankfully such a failure is repairable as the foam material used in the coupling and motor mount was in excellent condition with no signs of degradation.
The tilt mirror assembly is self contained and easily removed from the chassis. If there was a fault in that sub system I could easily use the motor out of my THV470 after attaching the good mirror plate fro the THV450. I established how best to approach the disassembly of the mirror wheel, its mount, the motor mount and the motor without causing damage to any mirrored surfaces. I also marked the various components positions with a scriber to realign them when reassembling the unit. I was ready to start the work on the THV450.
My first investigation of the scanner head mechanics would be the mirror wheel and its associated drive. Due to my experience with the THV470 I knew that the small hole in the motor mount allowed hex key access to the coupler securing grub screws on the motors shaft. This needs to be released before the motor is separated from the mirror drive shaft. Upon turning the mirror wheel it became clear that the motor shaft was not turning and I could not access the grub screw. It appeared that motor shaft coupling had completely failed. Upon releasing the motor from its two clamps I was able to pull it away from the mirror wheel and lift it free of the camera. As suspected, the motors shaft had only an aluminium ring attached to it. This was a third of the coupling assembly ! An inspection of the tube in which the coupling rotates revealed the other aluminium ring attached to the mirror wheel shaft and some odd looking lump trapped between the coupling ring and the housing tube wall. I was able to remove the trapped 'lump' and it turned out to be a rather mangled shaft coupling foam 'donut' ! It would appear that during the first test of the camera with the working servo board, the mirror wheel motor tired to spin the mirror but failed adhesive in the shaft coupling meant that there was slippage that prevented PLL lock. When power was removed it would appear that the foam donut in the shaft coupling flew out of its correct location and became jammed in the coupling shaft ring and hence the mirror wheel stopped almost immediately. The foam donut 'died' in the event :( Fortunately I know someone with a spare coupling assembly that just needs new adhesive..... me :)
I inspected the motor mount foam vibration isolator and found it partially separated from its metal faces. It would also need to be rebuilt with fresh adhesive. Inspection of all other parts of the scanning assembly showed its mirrors and mechanics to be in good condition.
The repair of the motor mount and coupling components was not a simple "slap on some glue....job done" type of job. I had to select an appropriate adhesive for the task and create alignment jigs to ensure that all mounting and coupling parts were axially aligned. Failure to align these parts would lead to off axis shaft forces and vibration.
Thankfully my mechanical workshop is well equipped and I had suitable mandrels and a decent lathe to help align the parts correctly. The adhesive used for this repair was Evostick contact adhesive used in the fluid jointing mode to enable correct alignment on the jigs. This adhesive is very similar to that used by Agema when they assembled the parts at the factory. Before reassembly of the parts with adhesive, I scraped away all traces of the original adhesive from both the metal and foam parts and then finished them with 1200 grade wet &dry paper to provide a fine surface finish with keying for the adhesive. With great care, the parts were reassembled with the adhesive and left to cure for 4 days. These are load bearing parts so ensuring an excellent adhesive cure is essential.
The mirror wheel, shaft coupling, motor mount and motor were all carefully refitted into the scanning head chassis. They were very carefully aligned with all location marks that I had created before disassembly. The final part to be refitted was the optical mirror wheel position sensor. This was deliberately installed at a slight angle originally so I had to align this component exactly as I had found it to ensure correct sensing. After some fine tuning of the various components final positioning, all screws were tightened and sealed with red varnish.
The camera was then tested. The mirror wheel motor and mirror could be heard accelerating to their normal rotational speed and after a few seconds of "training" the PLL was in lock and the tilt mirror glavanometer servo started driving the tilt mirror.
The camera produced a decent image of my hand after a few adjustments in the user menu for level and span. She lived once again !
The camera was given some tests and pronounced healthy. She is running with the servo board out of my THV470 and I have yet to look at her original servo board to determine the fault with it. I was just happy to have a running THV450 in front of me to be honest ! Her owner is pleased that this camera was not beyond repair. For anyone considering buying such a camera though, do consider the fact that I have worked on such cameras previously and had the benefit of a parts donor on which to learn and source spare parts. These cameras are not an easy repair without a service manual and spare parts availability.
Fraser
-
Comparison of cutaway for THV880 and THV470 cameras :)
I inverted the THV470 image to make a view similar (though not wholly accurate as opposite sides of the camera !)
-
The pictures......
these are relatively unsorted as i was not intending to post them and I am only doing so now at the request of forum member "CAT" who has an interest in these cameras.
Fraser
-
Pictures continued.....
-
Pictures continued.......
-
Pictures continued........
-
Pictures continued.... Getting bored yet :-DD
Note images 2535 onwards are of the THV450 that I repaired.
Fraser
-
Yet more pictures of the THV450 repair... phew this is hard work :-DD
Fraser
-
The repair work on the THV450 that had suffered a Servo board failure and adhesive failure in the rotary Mirror drive/motor mount.
Fraser
-
Pictures of THV450 repair continued......
-
The repair of the motor mount and shaft flexible coupling required accurate alignment of the various component parts. I achieve the required accuracy by using maderels and suitably sized drill bits. A lathe helped me a lot here :) The adhesive used replicated the original used by AGEMA and is a high strength contact adhesive called "Evostick" and it is common in the UK. Superglue or Cyanacrylate adhesive must not be used as the outgassing contaminates the mirrors in the optical section.
Fraser
-
Well that's all folks ! I never intended to post this stuff so it is really just a brain dump and my image archive that I use for working on cameras that I have been inside previously. NOT the most exciting of threads but I know that there are Thermovision Scanning type camera owners who might find the background and pictures of some use.
I did not take detailed pictures of the PCB's as this was a teardown record rather than pictorial record of the PCB's. I can add such pictures if required.
Fraser
-
Thank you so much for the description and the pictures of the teardown!
The corrosion is really extreme, I wonder if it was caused or accelerated by the leaked electrolyte. The Lithium Thionyl Chloride (Li-SOCl² ) battery looks like it was replaced at some time. It appears to be the type with spotwelded wires, maybe the case cracked when the wires were bend to fit on the PCB. Li-SOCl² batteries have often a very "wet" electrolyte without much absorbent material.
I don't want to sound greedy but do you have pictures of the detector assembly?
Pictures of TE cooled detectors seem to be a rarity, the ones attached are from https://vigo.com.pl/ and show how I imagine what it could look like inside the detector.
-
No problem on the detector images front. I will take some as soon as possible. The detector ‘module’ looks like an acorn type thermionic valve with a ceramic ‘skirt’ to which connections are made. An interesting little module :-+
I did take pictures of all the PCB’s and detector but the darned SD card decided to die as I transferred it to my PC :( I should have another go at recording the contents of the PCB’s.
Fraser
-
Detail of the THV450/470 detector module as requested :)
-
Detector module Continued.......
-
The cooled Detector in its vacuum package with connections around the periphery. I am providing lots of images as not many of these are seen on the internet ;)
I include a picture of the Germanium lens that illuminates the detector as it shows the effects of damp or age related degradation on the AR coatings, even deep within the camera !
Fraser
-
The Tilt mirror and its galvanometer. This is the first mirror that the thermal scene energy meets on entry to the scanning block. It provides the the vertical scan movement to create the raster. This Galvanomometer is synchronised to the rotating mirror and only begins to operate once the PLL detects the rotating mirror has reached the correct RPM.
-
The rotating multi faceted mirror that creates the line scans. The mirror is made from aluminium and contains a ball bearing race.
-
The fixed mirror assembly. One fixed mirror is still in the chassis so not shown.
The surface corrosion is clearly visible. These should have a highly reflective mirror polished surface !
It is surprising that there is no moisture control within this camera. No Silica Gel packs, nothing. Aluminium corrodes readily in a damp atmosphere.
Fraser
-
A selection of full resolution images of the cooled detector element for your delight and delectation ;D
-
The principle of the SPRITE detector element. This and the associated pictures may help to make sense of what we can see in the pictures I took of the SPRITE detector.
Taken from a scientific paper on the SPRITE detector....
In the SPRITE an element is fabricated that is around 10 times as long as it is wide. The MCT detector material is biased with a current source. This causes the mobile carriers generated by infrared radiation to drift towards the read-out zone. If the drift speed is matched to the speed at which the scene information is being scanned across the SPRITE element, then all the carriers associated with a particular part of the scene arrive in phase at the read-out region. This improves the signal to noise ratio and greatly simplifies the use of the detector.
-
Wow Fraser - that's a shed-load of photos and I have no doubt they will be used as the 'standard reference' in the future.
I'm always amazed just how methodically you work and how rigorously you document everything as you go. I'm aware that most of the time it's just for your own reference (did the brown wire go here or there?) but I know from my own experience that even getting good quality snapshots is no mean feat.
Thank you for sharing this with the world.
-
It was pretty much a dump of my picture archive and not what I normally like to do with pictures carefully selected for content. There is a lot of duplication so I will go through this threads pictures in slow time and delete the duplications. I provided some duplications, like the sensor as I did not have time to check which was the better focussed or lit. “Cat” wanted to see what the inside of the scanner looked like so I quickly uploaded what I thought might be useful :) I just hope the thread is of use and interest to someone.
Fraser
-
As Ultrapurple has stated, I carefully record the disassembly of equipment I work on. Relying upon memory can be risky when it comes to complex equipment. Knowing the original path that flexible cables took can be invaluable and avoids potential pinching of said cables during reasssembly. Service manuals for thermal cameras, even those from the 1980’s are not generally available (with the odd exception). As such it is a good idea to document working on a particular camera for future reference and to aid re-assembly :-+ I extreme cases, it can be worthwhile to video a complex disassembly operation so that the recording may be played in reverse to reassemble the parts exactly as they came from the factory. I do not usually need such video recordings though. My brain seems to be ‘wired’ for methodical disassembly and re-assembly.
Fraser
-
Hello, I know this is an old thread, and I have not been on the forum for years, but I came across this and had to re-sign-up to contact. Is there any chance anyone kept this thread in a booklet type form (like a pdf) where all could be found (and downloaded) easily? I am very impressed with the information. I would love to see if there are other teardown type information on items like the military AN/PAS-13B or the AN/PAS-23? I am going to start on my own with the PAS-13B because they also have the spinning mirror system like the Agema 470. I am somewhat familiar with the basic design and would be happy to document disassembly with pictures. The motor for the spinning mirror is noisy, not unlike every Agema 470 I have owned, but much quieter. Only by a teardown will I be able to track down if they can be silenced as originally made by Raytheon. Either way I would appreciate any words of wisdom.
-
The spinning mirror motor is almost silent in operation as it is a very high quality unit. It is the spinning mirrors ball bearing race that makes the noise and it is a sealed unit with no easy, low risk, way to rebuild that bearing with fresh lubrication. The THV scanning cameras were always a bit noisy because of the ball bearing race. The mechanical tilt drive of the other mirror is also naturally noisy in operation.
Fraser
-
I have 3 THV450s. All of them are so noisy it is difficult to stay in the same room with them when they spin up. One unit came to me with a box of disposable earplugs as part of the accessories! Strangely enough, the noisiest unit is the only one that works correctly. I am hoping to use your teardown to see if it is possible to correct this and quiet it down. I assumed the noise is coming from the mirror motors as they sound similar to a laser printer mirror motor when they start to spin up. You have referred to the ball bearing race as being the culprit. Is the race you speak of separate from the motors? Have you run into any units that had very loud motors or bearing races?
It is unfortunate the mirror motor race is sealed and at high risk of damage to rebuild. I am not sure how to approach the issue of noise on the THV450 or the PAS-13B units. Possibly I would be lucky and find that cleaning the old race and attempting a small amount of lubrication might help. I will need to pull the motors to decide. I know the PAS-13B motor is audible withing 2-3 inches of the unit's casing. Raytheon sold them as silent to the military, so I suspect something may be wrong.
If you didn't have any other documentation on your THV470 teardown, I will use the thread to attempt it. Years ago, I used to attempt to reverse engineer other equipment I had (including drawing out schematics) to help understand and repair them. I had a CAD program just for creating schematics. If I get anywhere with the THV450s, I will share what I find if you wish. Note that this only applies if this uses 1- or 2-layer circuit boards since it is nearly impossible to trace the runs on multilayer circuit boards. I am not sure that would work on the PAS-13B as Raytheon has been known to erase any identifying characters on the ICs, or have proprietary markings on them.
I have a PAS-23 that I purchased inoperative. I remember someone at some time stating that attempting to disassemble these units will damage them where they can't be repaired. I am wondering if anyone does repairs on these units so I would not have to find out the hard way. Do you have any information on these units?
Any thoughts?
-
The PAS-13B and PAS-23B that I am aware of were microbolometer based but your Raytheon unit sounds different. I shall have to investigate. When I think of early Raytheon systems that are ‘silent’ but make a whirring sound…. I think of the Raytheon BST based technology that uses a spinning chopper wheel. Units like the AGEMA THV 510 uses scanning mirror and ‘purr’ when operating.
The problem with the noisy rotating mirror system in the THV-450 and 470 is not uncommon as the ball bearing race suffers lubrication loss over time. It is a high spindle speed so the bearing starts to sound like someone shaking a tin of bolts ! The challenge is in getting to the ball race. The high quality motor is usually very quiet in operation. The motor spindle drives the mirror wheel shaft via a sound and vibration dampening foam ‘Donut’. This can become unstuck from its metal spindle holders and causes the camera to fail as the spinning mirror is static. The tilt mirror control loop is synchronised to the spinning mirror so failure of the motor coupling to the mirror also prevents tilt mirror operation.
The spinning mirror has a spindle that passes through a high speed ball bearing race. This ball bearing race resides in a recess at the bottom of the mirror wheel. I could see no way to remove the bearing without likely damaging the very fragile polished aluminium mirror faces. It was assemble, so it can be disassembled…. But the risk to the mirror wheel must be considered. The outer section of the ball bearing race is secured to the cameras chassis so the mirror wheel is held exactly in the correct location via its spindle that is secured in the ball race inner ting. Hope this helps.
These cameras often suffer from damage caused by a leaking Lithium Thionyl Chloride battery that supports the RTC and settings memory. The servo drive system that powers and synchronizes the spinning mirror and tilt mirror are also prone to component failures due to age. Failures of the Microprocessor PCB are also quite common.
Basically, the THV450 and 470, if still working, are on borrowed time due to age related degradation in the system. This is why I laugh when I see them on eBay at high prices…working units are basically a ticking ‘time bomb’ waiting to throw up a fault.
Hope this helps
Fraser
-
The relevant pictures of the spinning mirror wheel and associated bearing system.
-
I have found the PAS-13B of which you speak. I only know the later 2006 models. Yours is presumably the cooled version so may well be like the THV-450/470 scanning technology. If such is the case, some scanning units operate the mirrors in a vacuum and this can be an issue if you disassemble the scanning section. I regret I have no information on these military units or what resides within them.
https://web.archive.org/web/20100705003911/http://www.raytheon.com/capabilities/rtnwcm/groups/ncs/documents/content/rtn_ncs_products_tws_pdf.pdf (https://web.archive.org/web/20100705003911/http://www.raytheon.com/capabilities/rtnwcm/groups/ncs/documents/content/rtn_ncs_products_tws_pdf.pdf)
http://www.2ndbn5thmar.com/night/PAS-13.pdf (http://www.2ndbn5thmar.com/night/PAS-13.pdf)
Fraser
-
THANKS so much! I had a thermovision 870 scanner that just recently started giving me issues. it was having problems maintaining speed and would miss the video sync every once in a while until eventually it stopped spinning it's mirror polygon all together. it was that darn foam donut and it's adhesive that had come loose. I probably would've never figured out what exactly the issue was without this thread. I used a little adhesive to reattach the donut coupler to the motor shaft and it works good as new.
-
catjuice,
I am pleased that my post helped you :-+
Fraser
-
Hi Fraser... Thank you so much for all your excellent work on dismantleling this wonderful camera.
We have one that stills partially works and we're looking for someone that could help us.
Our camera does not open the Configuration Menu and as we have tested the rubber side press buttons are not the problem. We suspect is the firmware.
Another problem is the floppy disk that turns on but does nor read or save any file. We tried to replace the floppy drive but it didn't work too.
Could you help us?
We'll be extremely gratefull.
Thank you very mcuh and, again, congratulations for your photos!!!
mario
-
Mario,
Inspect the memory back-up battery that sits under the handle area of the camera. The Lithium cell fails and leaks over time. the electrolyte can travel a long way and onto the PCB's below. If there is leakage you need to fix the damage before doing any further fault finding as the conductive electrolyte can cause all manner of weird issues.
I have a complete set of PCB's and parts for the 470 camera that I will not be needing any time soon. Sadly I am deployed overseas until the end of March so cannot do much for you until I get home. The firmware is replaceable by changing out the EPROMS but I do not have a copy of those EPROMS as I had no need of the code. The PCB's I have do contain a full set of EPROMS but, from memory, the processor PCB has a fault due to battery leakage. Hence why I dismantled the camera.
Fraser
-
Thank you very much Fraser...
We are thinking about the EPROMs here...
Also... Do you have any idea on how to fix the diskette unit driver?
We've exchanged ours for another one that is working properly but it didn't work either.
Any suggestion will help us a lot
Thanks again
mario
-
Sat here in my overseas accommodation I am without my reference material for the 450/470 camera so can be of very little assistance to you. I cannot recall whether the FDD is a standard PC type or one that is specific to the Agema camera. It would be worth checking the original drives part number to see if Google can supply its specs for you. I have met FDD's that have custom connectors and even custom firmware and so cannot be simply replaced with a generic PC compatible version.
As to what could cause a failure in the FDD controller board... I will do a careful inspection of that PCB looking for corrosion. Where EPROM's are involved, sadly there is the possibility of age related data corruption but such is only easily found by comparing to a known good binary file.
Sorry I cannot be of more assistance to you at the moment.
Fraser
-
Thanks Fraser...
I'll check this out and let you know.
Probably will be able to open the camera in the next two weeks. You know... We have other things to do here... :(
Thank you very much again
mario