Author Topic: (new pic posted, please help) Cairns Viper, self-heating, some other questions  (Read 4427 times)

0 Members and 1 Guest are viewing this topic.

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
Hi guys.

I just got a Cairns Viper, now I have a few questions about it...

Q1. It close/open the shutter 4 cycles on every power on, is it normal?

Q2. It works well until it heat up, here's the time line in my test:

Start up.
~1 min: First NUC after boot, begin to show slight white pixels.
~8 min: The right side of the image begin to darken to a noticeable level(and growing).
~25 min: While the shadow becoming darker and wider, there begin to have white line on the right edge(FFC want to correct that)(growing thicker with each shutter action).
~41 min: I power cycled the device and the picture is very foggy, until the next NUC.
~56 min: The high-temperature warning icon shows up(see pic 1).
~1h 5 min: The black shadow and white area become much thicker(shadow is also darker), the temperature bar become full(1000 F)(see pic2)
~1h 11 min: Battery died.


There are also some other random behavior when over heating:
a. Sometimes the temperature bar shows 300 F(see pic 3).
b. Sometimes when power on while hot, there's a red symbol next to the battery, in this mode, the shutter will never trigger(see pic 4).

When I take out the "super cool" metal block, it's super hot. Is that a normal thing for old TICs? My friend's BST camera doesn't seem to have this problem.

Q3. What's the red symbol in pic4 means?

Thank you!
« Last Edit: October 28, 2021, 03:54:08 pm by Logan »
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
In a fire camera the temperature warning cam be indicating that the camera is operating in an environment that is too hot fir it to cope with and should be moved to a cooler location. That ambient temperature is measured by a sensor within the camera. If this is the case with your Viper, the over temperature alarm is actually detecting a completely different situation…. A fault within the camera is generating enough thermal energy to overheat the core ! Not a food situation. If the cooling block comes out feeling hot, that seems wrong to me but Bill_W may know more. I understood that the cooling block was intended to act as a heat soak away for the camera whilst the ambient temperature is high in order to prolong operational capability.

You would be wise to open the camera and carry out some testing to determine the source of the internal heating in case it is a fault and could cause further damage. It could be power supply related so check battery temperature and power supply temperature first. I have not worked on a Vioer so sadly cannot help much regarding its internal design. Look into the heating situation before worrying about other fault symptoms as they may stem from the cause of the self heating.

Fraser
 
The following users thanked this post: Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
This thread from 2018 may also be of interest…

https://www.eevblog.com/forum/thermal-imaging/carins-viper/
 
The following users thanked this post: Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
 
The following users thanked this post: Logan

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
Thank you very much Fraser!
So it's not normal? Okay I'll try to fix it...
I did searched the forum and found those threads, but nobody talk about the heating problem.
Look into the heating situation before worrying about other fault symptoms as they may stem from the cause of the self heating.
I know, I just think it's interesting to document it's mad behavior. ;D
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
Microbolomter based Thermal cameras need an ambient and FFC shutter temperature reference in order to correctly report a targets temperature. If the internal ambient temperature rises too high it can mess up the measurement function. Increased internal temperature can also cause uneven heating of the microbolometer via adjacent heat source conduction, convection or radiation. If there is an abnormal heat source near to the microbolometer it has the potential to cause all manner of havoc with the displayed thermal scene. If looking for a heat source adjacent to a microbolometer using the effect seen on the display, remember that some cameras use inverting optics so the microbolometer is mounted upside down so left is right and up is down in the displayed image.

Fraser
« Last Edit: September 12, 2021, 05:12:36 pm by Fraser »
 
The following users thanked this post: Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
Regarding the red symbol next to the battery. That is a classic battery over temperature warning. From memory the Viper uses Lithium Polymer cells and if this is the case, they are not supposed to be exposed to temperatures greater than 60C to avoid self destruction ! Many camera systems monitor the batteries internal temperature during charging and even discharging (in the case of fire fighting cameras) so that the user may be warned if the battery is getting too hot.

As we are potentially dealing with a failed Lithium cell filled battery, I advise great caution whilst charging or using the battery and DO NOT store that battery pack inside your home without it being in a fire resistant container. A lithium battery pack can suffer cell deterioration and there is the potential for cells to overheat for several reasons. Normally the battery contains temperature monitoring, self resetting temperature breaker and a non-resetting (permanent) thermal fuse. Your battery could potentially be getting hot, but not hot enough to trigger the breaker or thermal fuse. Alternatively, something inside the camera is heating the battery either via external heating of the pack or overload due to excess current draw. All worth investigating but please do be careful if you are dealing with elderly Lithoum Polymer cells….. they can be a real fire hazard which is kind of crazy ironic when you think what they are being used in here !  ;D

Fraser
« Last Edit: September 12, 2021, 05:31:48 pm by Fraser »
 
The following users thanked this post: Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
Just looking at the Viper brochure, I see that it does indeed contain a Lithium battery pack. This pack is located next to the cool block and you have already stated that the cool block gets toastie hot. This could be heating the battery pack causing the battery temperature alarm. Your test lasted just over an hour before battery discharge and the original duration specification is only around 90 minutes so it does not sound like a serious over current situation causing battery heating. That said, we cannot discount the battery self heating during use so caution is advised.

Take a look inside the camera and see where the heat has been transferred from that us causing the cool block to get so hot. If the cool block is thermally bonded to the cameras chassis, what else is bonded to that chassis that can generate so much heat ? You could use another thermal camera to monitor the camera when first switched on to see the locality of any internal heat source. Then remove the covers and check again to determine the cause of the internal heating.

Fraser
« Last Edit: September 13, 2021, 10:05:08 am by Fraser »
 
The following users thanked this post: Logan

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
Thank you so much Fraser.
I opened the camera, and since my Seek is broken, I just use my finger to test...
The heat seems come from the processing board, the Xilinx chip is the hottest I can reach, there are some other things like Intel Strong ARM mounted on the other side, but I cannot reach them, because there's a power board stacked close to that side.
The camera works better with case opened, and the battery or microbolometer is not that hot, so I assume they are not the heat source.
By the way, the whole block look like this from side:
|------
|------
The left is the sensor and it's board(please ignore the space, there's only one), while 2 horizontal boards are power board and processing board.
 

Offline Bill W

  • Frequent Contributor
  • **
  • Posts: 826
  • Country: gb
    • Fire TICS
If you can unplug the battery and run from a power supply that will tell us a lot.

A 320 temperature-stable VOX is a pretty hungry beast so expect around 3A at least for up to a minute or so.  May be worse as I think that was an earlier core in the Viper than the Lockheed 205 in Argus3(orange-VOx)

Just guessing but if the thermal control loop was opened (as simple as a thermistor fallen off) the core control might be in a runaway 'I must get colder' mode flat out which could explain several of your observations.  When open it might hold under the 30°C set point so not run away

The thermometer was usually set up to warn of high detector temperature (so only a consequnce of the ambients) and to encourage the user to turn off before damaging it, or to give some warning if the camera turned itself off for protection.  It was generally assumed that a 5 minute warning was enough for the firefighter to get to a safer location similar logic to their air supply.
 
The following users thanked this post: Fraser, Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
As usual Bill, you have great insight  :-+ If the TEC is running hard to cool the microbolometer due to a loop failure, that would certainly generate lots of excess heat within the camera. The reasonably good run time is better than I would have expected though as the Peltier module is power hungry. You could well be onto something though  :-+ As you say the current draw behaviour could tell us a lot.

Fraser
 
The following users thanked this post: Logan

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
Thank you Fraser and Bill.
I actually thought of the TEC feedback failure before, but didn't know how to check it, good point to check the current.
But I may not be able to do that recently, I will report back after test.
By the way, Does anyone have the document for the core? it would be interesting to see.
 

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
Hi, while still waiting for my multimeter to be repaired, I disassembled the camera again and found the wires appears to be connected to the TEC. I ran the camera with TEC unpowered, the camera warned for temperature immediately after boot, and the picture quickly shows sensor artifacts which is inverted compared to the condition with TEC.
So I think our guess is corrent, the TEC keeped cooling the sensor despite it's already colder than normal.
If that's the case, how can I fix it? Thanks.
 

Offline Bill W

  • Frequent Contributor
  • **
  • Posts: 826
  • Country: gb
    • Fire TICS
I suspect some of these thoughts may not be much help, but here goes:

What is the sensor involved, how should it work ?
Is it giving the expected output for ambient + 7 degrees ?
Can it be disconnected from the camera - and is it then OK. Hopefully, otherwise it is in the sensor itself.
If you briefly reconnect the peltier (or power it independently)  does the sensor change in the right way ?

A pure guess is that the sensor die should sit at 30°C, purely as all other sensors did, so again if powered independently adjust for no artefacts and you should be at the setpoint.  That should then tie up with however they did the Peltier control.  Hopefully follow the drive back and find a specialist Peltier IC and cross reference data sheets.  It might however be buried inside software so more difficult to check.

Seems the temperature warning is an 'out of limits' not an 'overheat imminent' as I proposed above, although the user logic is probably the same - turn me off before you get to invoke the warranty.

 
The following users thanked this post: Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
The TEC circuit is closed loop so you will have a temperature sensor mounted inside the microbolometer module, on the die, and this feeds it’s signal to a TEC controller IC that manages the Peltier module that sits on the rear of the microbolometer die, but within the vacuum of the microbolometer module. Neither the temperature sensor, nor the Peltier module may be accessed for repair so let us hope these are serviceable.

The temperature sensor on the microbolometer die would usually be a thermistor but a diode is another possibility. The sensor changes it’s output in response to temperature so it should be possible to monitor the temperature sensor output pin or the pin on the TEC controller that it connects to. TEC controllers should be relatively easy to identify by device number and they usually have pretty heavy duty PCB tracks around them for the Peltier module that draws between 1A and 3A at first start from cold. The TEC and associated Peltier module is used to warm the microbolometer to around 30C and to then hold it stable at this temperature. You should see the signal at the temperature sensor input change as the microbolometer warms up and then a slow sinusoidal variation will be seen as the temperature is maintained at a nominal 30C. The Peltier module drive from the TEC can take several forms so check the datasheet for the IC. Common drive methods for a Peltier temperature stabiliser are variable voltage/current, PWM or on/off switching. On/Off switching is the most crude and least desirable drive method as it can stress the Peltier module. PWM gives excellent control over the Peltier modules heating and cooling behaviour. Varying a DC voltage or current will control a Peltier module but PWM is often preferred.

You need to look at what is happening at the input to the TEC controller and what the TEC is doing at its output in response. The TEC may contain its own internal Peltier element drive circuit or it may use external components / driver IC instead. If an external driver is used, you will need to check its output to see whether it correctly reflects what the TEC wants it to do. You could be in a situation where the TEC drive to the Peltier uses a transistor H bridge that has become shorted so the drive is continuous instead of PWM or variable voltage/current.

Your camera appears to be excessively cooling the microbolometer as that produces excess heat at the Peltier elements heatsink face that is attached to the rear of the microbolometer module. That state is unusual as the TEC normally heats the microbolometer die and does not actively cool it unless the ambient temperature is above ~30C. You need to examine the behaviour of the TEC IC to determine exactly what is happening and why. The likely faults are : die temperature sensor wiring to the TEC chip, TEC output driver failure or TEC IC failure. The TEC will be configured in some way to set an operating temperature and that part of its circuit would be worth checking in case there is a faulty component present. The TEC IC datasheet will likely provide a sample circuit that closely resembles that used in your camera.

Thankfully the failure of the ‘on die’ temperature sensor is unlikely and we can see a change in behaviour when the Peltier module is disconnected, so hopefully the Peltier module is also serviceable.

Hope this helps

Fraser
« Last Edit: September 23, 2021, 09:14:59 pm by Fraser »
 
The following users thanked this post: Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
The datasheet for the well known LTC1923 TEC IC is linked below. This will give you a feel for what a dedicated TEC IC offers and how it is configured to drive a Peltier module. Note the external H bridge drive transistors.

https://www.analog.com/media/en/technical-documentation/data-sheets/1923f.pdf
 
The following users thanked this post: Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
Please provide some nice clear pictures of the cameras PCB’s so that we can see the chipset used and as a future reference  :-+

Fraser
 
The following users thanked this post: Logan

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
Please provide some nice clear pictures of the cameras PCB’s so that we can see the chipset used and as a future reference  :-+
Fraser
Okay, 8 7 in total.
Other 4 battery related photos: https://www.eevblog.com/forum/projects/how-to-make-this-device-think-it-have-a-battery/
Because of "maximum total size 5000KB" forum limit, there will be several posts.
« Last Edit: September 25, 2021, 04:40:10 pm by Logan »
 

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
...
 

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
...
 

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
Picture 7,8 are the back of the sensor board, I cannot take it off to have a full view because the screws are too tight (and my screw driver seems a little bit too big). Full back photo taken, See post below.
Picture 5 shows the lens and shutter assembly, different from the metal one in the forum(and only have 2 screws on the 4 screw holes), they are cutting corners?
Picture 3 shows a 3-pin plug on the right-bottom corner, it's just unplugged as I first open it, why?
Forget it, there are too many unused sockets as well.
« Last Edit: September 25, 2021, 04:41:34 pm by Logan »
 
The following users thanked this post: Fraser, Bill W

Offline Bill W

  • Frequent Contributor
  • **
  • Posts: 826
  • Country: gb
    • Fire TICS
As the designer seems to be a fan of Analog devices, maybe there is an ADN8831/33/34 in there as the Peltier control ?

Bill
 
The following users thanked this post: Logan

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 11452
  • Country: gb
Sadly none of the IC part numbers that I can read are a TEC controller so it is either the one I cannot read on the Microbolometer PCB (picture 5, top left corner square IC) or it is not visible in the images.

Fraser
 
The following users thanked this post: Logan

Offline dertoid

  • Newbie
  • Posts: 4
  • Country: ua
I love my CairnsViper (correct name).
Do you have full kit in a case? You should have spare thermal sponge - piece of metal. When cam starts to heat up you just pull hot one out then swap with spare one.
 
The following users thanked this post: Logan

Offline Logan

  • Regular Contributor
  • *
  • Posts: 175
  • Country: us
Sadly none of the IC part numbers that I can read are a TEC controller so it is either the one I cannot read on the Microbolometer PCB (picture 5, top left corner square IC) or it is not visible in the images.

Fraser
I finally take the full picture of the sensor board back. Now all PCBs (except the LCD one) is here.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf