EEVblog Electronics Community Forum
Products => Thermal Imaging => Topic started by: MadeForThat on July 14, 2023, 09:14:09 pm
-
Fired up a ratheon camera from the 90's and after getting the chopper wheel freed up and spinning, it makes a nice image! Issue is there is a large gray block in the display, and I am pretty inexperienced to know if this is something on the detector side or the display side of this device, I have attached some images of the boards, I suspect there could be damage in any number of the ribbon cables, but not sure how to know without probing each line of each ribbon. First post, very long time lurker! Thanks for any input
-
Sadly only one image is visible and you need to reduce image size to around 1024 x 768 pixels for optimum forum compatibility :-+
Fraser
-
Wow a horrible display of technical posting! :palm:
-
Board on bottom of device
-
Board on right side
-
Left side
-
Top down
-
Lil microwave transmitter that I removed, so far as I can tell it would have just broadcasted the video feed to a ground station. Replacing doesn't alter image.
-
So when the device is pointed at a candle, in the "dark gray" box in the lower left hand of the display, a few pixles will light at the heart of the flame. It almost seems as if the "gain" on this patch of pixles is set way low. Again, no idea if that is a detector or display issue. Please excuse that I have the camera focused for about 10 feet and it isn't very clear on this flame haha
-
that may be a bad RAM chip, sector failure of the framebuffer
-
Do you think that is something that could be diagnosed? I am assuming that the ram would be related to the display overlay? I really have no clue haha. Wondering if there is a way to independently verify the detector systems are fine. I wouldn't mind running this to an external (or modern) display, assuming there is a place to tie into composite. But I am pretty un-knowledgeable about what that signal even consists of, let alone where to find it, so I may need to disassemble and see if any pins are labled
-
Guess I should have given it a closer look before posting, but there is a "vid" pin haha. Anyone want to weigh in on how to check this feed?
-
This camera was branded under a firefighting company-FRC. Company still exists and after a few emails I am told these cameras were only sold in the 90's and were sourced from "electrophysics." Doing a few searches didn't return much to help me find documentation on this core, or any of their product line. That said, the "bottom board" says Raytheon, so I don't know who was actually making what. Only image I can find of the thing is a fellow in China's PLA using it haha
-
The sales brochures....
(too big to attach)
http://www.fire-tics.co.uk/others/FR-LSS.pdf (http://www.fire-tics.co.uk/others/FR-LSS.pdf)
http://www.fire-tics.co.uk/others/FR-LSP.pdf (http://www.fire-tics.co.uk/others/FR-LSP.pdf)
Oddly for a Raytheon BST, the 'Survivor' brochure claims to provide spot temperature measurement from the sensor. Maybe the brochure is for a later model with say the BAE/Lockheed VOX in there instead ?
An image of the Raytheon processing board would be useful too.
Now to the system....
The Raytheon core was common to Argus2 and Argus3 so teardowns on here and my fire-tics website may be of use.
It output as both BW video analogue and as 10 bit digital data.
Looks like FRC took the digital, added false colour and graphics, again much like Argus3. Finding and viewing the raw Raytheon analogue might be instructive, but I do not know if that is what is on that pin.
The bad news is that that the Raytheon 'ready' rectangle is visible bottom left, suggesting this is a Raytheon core failure, as the 'ready' is overlaid on the dead grey block. If it was the FRC overlay at fault, that would cover the Raytheon graphic
Bill
-
that may be a bad RAM chip, sector failure of the framebuffer
I have seen dead sides, dead columns, dead top half etc, but never a dead rectangle like this on a BST.
The fact a hot object shows suggests a missing chunk of calibration data, rather than sensor fault.
As above, I am suspecting the Raytheon side though, not the overlay/buffer
Bill
-
So this unit certainly does have a temp readout that can be activated in the UI. Seems to be pretty accurate. I went ahead and tested the "vid" pin, and what do you know it works, but the rectangle is still there |O after seeing that I went ahead and tore it all apart. Nothing obviously wrong, some hokey wiring in places though. Got it all put back together and I am noticing a new detail that I didn't have or notice before. When the detector has a hot object in front of it (hand) the rectangle color darkens. Sorry excuse these shots, my only analog input device in the house is an old ctr viewfinder.
Bill W.- are there any options to recalibrate the device. Thinking it would be nice to clear it and calibrate on a cold steel plate or something like that. Maybe that is wishful hahaha.
-
Pics of the Raytheon board
-
Bar darkens, and the plot thickens
-
Setup for testing
-
Well if anyone has any off the wall ideas for things to try, I'm all in. It looks more and more like this is a "let die" kind of moment. Bill, you mention the small bright corner is a "ready" signal for the device? Do the BSTs all have this when operating? Seems odd that it is all the pixels around that indicator, so I wonder if there is some kind of bleed to nearby pixles. Wondering if you have ever gone down the rabbit hole to disable that indicator.
-
The image of the Raytheon board confirms it is a '200A' series 'analog' board. Indeed it must be very early given the bodgery in place to make the crystal run.
There is no recal available for that one that I know of.
Obviously there must be some way to get the calibration into the PCB, but it may be off-camera and loaded in, or via the edge connector.
The 'ready' is part of a set of overlays for [redacted] use, so late on in the processing. It is not bleeding out.
I do not think there was any way to deal with the smaller white 'ready' rectangle, that is why battery indicators were always put in the left !
The only fix I can think of is to get another 200A core, maybe from a camera with a different set of faults. I cannot help - I do not have an export license UK to USA.
-
Thank you for the great information, shame to scrap this in a semi functional state. For my understanding, when you say core, you mean the main Raytheon board as well as the detector assembly?
-
Thank you for the great information, shame to scrap this in a semi functional state. For my understanding, when you say core, you mean the main Raytheon board as well as the detector assembly?
Yes, the detector and the processing board were matched pairs ('the core'), as the calibration specific to the detector is stored in the processing PCB (generally known as SECCA).
The detector board itself is not part of the matching, but was often sold as part of the kit of parts to camera manufacturers.
I would be fairly certain that if say you got hold of an early Argus2 (MSA by EEV) that say had been dropped and smashed the CRT you could simply extract the detector and SECCA.
I have not (yet) tried running an unmatched detector / SECCA to see how bad the image is.
That might be an option as it is the detectors that are controlled, not the SECCA, so I could send you a known good SECCA.
-
I have not (yet) tried running an unmatched detector / SECCA to see how bad the image is.
I tried that in person and it's not bad at all, only a few bad pixels that's not mapped, nothing worse.
Unlike some VOx sensors, which would result unusable picture due to wrong bias/calibration, etc.
-
I guess the question I would pose, if this was your camera, would you bother finding replacement components, or just buy a similar vintage argus/Scott for 200$? I think it's a cool form factor in the box, but not sure it was ever better than the competition in terms of durability. Wondering if there any parts off of this unit that I could sell to you good folks to recoup a few of my bucks haha. You mentioned the detector is the only export controlled item?
-
MadeforThat,
I have just had to call time on repairing a modern UNII-T UTi690B camera and I feel your pain. I hate giving up on a thermal camera and often spend way too much time on a 'patient' trying to revive it. I have had to learn to 'let it go' when it becomes a "time gobbler" ;D
With regard to your BST camera.... it is a very interesting design that I have not seen before. A very boxy format that is clearly designed for utility rather than good looks ! I have several BST technology fire fighting cameras in my collection and E2V/MARCONI ARGUS remain my favourites. The Argus 2 is a bit of a 'Beast' in terms of size but it is well suited to the intended gloved use and very robust. You will not be surprised to hear that some have been sold to film prop companies for use as 'firefighting' cameras or Sci-Fi 'Ray' runs in films ! The Argus 3 was a change of format that you either love or hate. I find myself torn as I like the format, but did not like the large side handles. Well suited to the intended use by fire-fighters, but bulky as heck. I fitted the optional 'slimline' side panels in place or the handles and prefer that configuration. It remains quite large compared to more recent fire fighting cameras, but it can produce nice imagery.
We cannot avoid the elephant in the room though..... BST is an obsolete thermal sensor array technology so is it really worth repairing BST based cameras when modern microbolometer based cameras have become more affordable ? That is a difficult question for me as I respect both technologies. BST does not normally provide a temperature measurement due to the physics of the FPA's operation. The dynamic range of BST is limited so a mechanical IRIS is often employed and the chopper wheel uses a clever membrane to improve performance. The technology was clever and deserved further development. BST offers continuous imaging without the need for FFC flag calibration freeze frames. Those interested in video recording in the thermal domain prefer not to have these freeze frame events present in their recordings. The down side is the mechanical nature of the BST FPA pixel reset using a spinning chopper wheel. With some thought, the chopper wheel could be made quite sophisticated and compact as E2V demonstrated in their military designs. But is the image produced by a BST technology camera any good ? In my experience, the BST based cameras required less clever noise reduction and image processing than early microbolometer cameras. The image in a healthy BST camera system can be very nice indeed. Systems that are showing their age and need some TLC start to show increased noise in the image and the display systems can start to show their age. It needs to be remembered that most BST Fire Fighting cameras provide a 320 x 240 pixel image with a high image refresh rate of anything from 25fps to 60fps.
Once BST development ceased and the focus was moved to Microbolometer development and noise reduction, the World began to forget BST technology. The Microbolometer has come a long way since I first tested it in 1997 in the ground-breaking FLIR PM570 camera. Modern microbolometer thermal cameras come in many formats and quality levels. It is possible to buy a 256 x 192 pixel 25fps microbolometer camera for less than $300 and even less if you can live with a mobile phone dongle format solution. The size and performance of these modern cameras is impressive, given the relatively low cost. It is enough to make people think carefully about whether they but an older BST technology camera or a new microbolometer technology camera from ASIA. The older fire fighting cameras are often beautifully built and very rugged, modern budget microbolometer cameras, less so ! I love the ARGUS 4 and ARGUS MiTic modern Firefighting cameras but you need deep pockets to afford them, even if bought used.
Your thread title reads "Repair or let die". I always try to revive a thermal camera but there is a need to be sensible when dealing with technology that has limited intrinsic value. If I have a very unusual thermal camera in my collection, I will go above and beyond all sensible effort and expenditure to get that camera running again and in nice condition. I am a collector though so can be illogical in my efforts to repair a camera. A general user of thermal cameras needs to decide what a poorly camera is actually worth and ensure that they do not spend more than around 60% of its value and there will be the investment of time to be considered as well. The good news is that there are reasonably priced BST technology cameras on the used market as the cameras are often overlooked by those wanting more modern looking technology.
With this in mind, I would look to repair your camera using a parts donor bought via eBay but I would limit the amount I would spend on the donor unit. A very battered Argus 2 often has a pristine condition electronics package safely protected within the well designed inner waterproof and dustproof casing. If you decide to sell your camera as a project for someone else to repair, I feel sure you could get a reasonable sum for it as it is an unusual model. If you want BST technology thermal imaging then look for a nice condition Argus 2 or Argus 3 but be sensible about the price you pay. The Argus 3 gives you a colour LCD display over the monochrome CRT used in the Argus 2.
Fraser
-
And just as a foot note.... if I had your camera, I would repair it ! :-+
Sady even if I were given the opportunity to buy your camera, the shipping and taxes would likely make it just too expensive for me :(
Fraser
-
Another comment on BST cameras. If they offer a video output to connect an external monitor…. Try connecting a decent monitor to it. You may be surprised at the improvement in image quality. Some early generation Colour LCD displays can be quite poor performers with a grainy appearance. Upgrading a camera to a more modern LCD panel is sometimes possible but much depends upon the signal produced by the video output stage.
Fraser
-
I guess the question I would pose, if this was your camera, would you bother finding replacement components, or just buy a similar vintage argus/Scott for 200$? I think it's a cool form factor in the box, but not sure it was ever better than the competition in terms of durability. Wondering if there any parts off of this unit that I could sell to you good folks to recoup a few of my bucks haha. You mentioned the detector is the only export controlled item?
A replacement SECCA is easy ! PM me
Seems likely to restore decent, if not perfect, image as noted above.
Might even be the best way to recoup costs and buy a good condition Argus.
Yes you can get a BST core camera for around that, but if hoping for a spares donor would need to know it is an 'analogue' 200A type. Argus2 used both so you would need to check serial numbers to be sure - and hope it did not have a replacement core at some time in its' life.
ISG Talisman would be another possibility for salvage.
Bill
-
It needs to be remembered that most BST Fire Fighting cameras provide a 320 x 240 pixel image with a high image refresh rate of anything from 25fps to 60fps.
They are almost all 60Hz 'NTSC'.
There was an option on the later digital BST to switch to 50Hz 'PAL' which windowed the 320x240 thermal image into a 384x288 picture.
( I say NTSC / PAL, the cores were of course monochrome)
Bill
-
Thanks for the replies guys! I guess I will keep my eye out for a core in case I can luck out on another "deal" like this one lol. My other alternative idea for this case would be a pretty sick machined birdhouse, with the old lens hole being the new entrance haha, but restoring the camera to decent glory would be ideal haha! From what I can tell, this core has both power and vid out pins, I am not sure if any of the other boards are needed? See the pins I bent on the image. I have no real need for the spot temp reading, or the automatic iris, or any of the other features. I just want to see hot and cold in B/W contrast lol. Anyone have an idea what the operating voltage of the core would be? I assume 5v, but have no real support for that. My running theory is that the bodgery on the ratheon board is for the spot temp display, but I have no idea haha. I am all for testing with the least number of variables, especially when I don't understand how something functions haha
-
From what I can tell, this core has both power and vid out pins, I am not sure if any of the other boards are needed? See the pins I bent on the image. I have no real need for the spot temp reading, or the automatic iris, or any of the other features. I just want to see hot and cold in B/W contrast lol. Anyone have an idea what the operating voltage of the core would be? I assume 5v, but have no real support for that.
Not 5V, await detailed reply.
Yes you should be able to run as 'volts in - video out' with just the PCB with the detector on it, the 'SECCA' and the chopper motor / sensor etc.
As earlier, I have no idea how they made a BST do on-sensor spot temperature readings. Is it the 'standard' chopper wheel ?
See teardowns here or on my site:
http://www.fire-tics.co.uk/project3/index.htm (http://www.fire-tics.co.uk/project3/index.htm)
and
http://www.fire-tics.co.uk/spares.htm (http://www.fire-tics.co.uk/spares.htm)
Everyone else using BST had to fit a separate spot temperature sensor unit - usually a Raytek.
-
Yeah it didn't look much different from the chopper spare you have available, not sure what I would be looking for on it. Seems the measurement was pretty sensitive, but I didn't check accuracy.
-
Ok, that shows the classic dark halo - it is a standard Raytheon BST with some unknown magic to read spot temperatures when running the standard chopper wheel - 'clear' and 'dispersive' segments.
You can check the volts on the PWR connection to see what is being fed into the Raytheon side. I have not got the full data to hand, but ran these from 8 x NiMH, 8 x AA LR6 and 5 x NiMH.
Keep below 12V especially on an early core as the tantalum caps can be fragile. We had a lot of dead cores from folks wiring them to vehicle supplies.
10V is happiest, as it keeps the current demands down, but it will run down to 4.5V.
One possible dependence on the rest of the camera is if it is running with fixed / calibrated gain / offset ( ? maybe to aid the spot temperature function ? ).
That would mean sensor gain/offest set from one of the other PCB. We did the same in Argus2 & Argus3 and hence the 'set to manual' instructions I give in the tearup pages. An analogue core only has those DC controls which you might need to bypass to go stand-alone.
https://www.eevblog.com/forum/thermal-imaging/argus-3-(bst)-teardown-and-repairs/44/ (https://www.eevblog.com/forum/thermal-imaging/argus-3-(bst)-teardown-and-repairs/44/)
Down from post 44
-
Some thoughts on the 'on screen' temperature measurement. How might you do it with a BST - and why such a problem.
The chopper is the first issue, BST only senses temperature changes hence the chopper.
Second is the need for an iris to stop the lens down when viewing hot objects.
So you need to know where the iris is (looks like this camera might have a stepper motor drive), what the temperature of the iris is, and then what video signal represents a given temperature difference - and finally what that is a temperature difference FROM !
The Argus2 & 3 were set up for a given gain (video height of a 5°C differential)
As above, everyone else (EEV/Marconi, ISG, ISI, Bullard ) simply fitted a separate sensing head from Raytek