Products > Thermal Imaging

FLIR P640 thermal Image distorted vertical lines on screen

(1/5) > >>

Hi All,

Could you please advise what could be causing this issue where i see vertical lines on the screen when the hot subject is focused and the background temperature is lower?
The thermal sensor is not the problem as i swapped the thermal unit/pcb assembly with connecting flex cable and same issue occurs on a different sensor.
If i change the angle the vertical lines are not that pronounced, probably as the object is not in sharp focus, see difference between img 15 and img 17
If someone can explain this it would be amazing.
Thanks all.

Bill W:
The sensors work by having one amplifier / integrator for each column, and it is activated for each row as it is read out.

When you have a hot object in a certain row it results in a lot more energy in the column amplifier and it slightly warms up when reading the hot pixel.
When it comes to do the following row it will be slightly out of calibration compared to how it was on the last 'flag' operation which of course was on a blank scene.

When this is also a colder area the differences show up more as there is less IR energy there.  A very low noise mager like this (30mK) is ironically more susceptible as a 50mK imager would hide this in noise.

The image processing behind a lot of 'shutterless' systems will also result in the same effects although for different reasons.


Whilst I totally agree with your comments on these images, there is a little more to this story. The OP has managed to rebuild a working thermal camera using two faulty units, an SC640 and a P640. I am in communication with the OP via messaging but it will be far better to use this thread to help him.

My issue with the provided images is that these appear atypical of a SC6xx or P6xx camera as the column noise seems excessive in the example images. The SC6xx and P6xx are high performance imaging systems so something is not right here. My PM series cameras are older and less capable, yet do not display this column noise issue. Such is normally dealt with in the image processor sub system. I have asked the OP to detail exactly what parts were combined from the two cameras to create a working unit. We could be seeing either an issue in the video processor sub system or a calibration (NUC) issue. I would have expected dead pixels to appear if the wrong Calibration tables were in use though. It is a bit of a mystery to me at the moment.

If I saw a SC6xx producing these images when new, I would reject it as unacceptable column noise and a non uniform image response.

It is almost as though the cameras normally excellent image processing is not active  :-//

As we know, you have to be careful when transplanting PCB’s in these multi PCB cameras. Some do not play well together if mismatched. FLIR also program each PCB’s serial number into both the PCB subsystems and LiCO main controller. I would expect an error to pop up if the serial number mismatch was a problem though. When FLIR replace PCB’s in these cameras they also carry out a fresh calibration run to ensure all PCB’s work well together.


I believe Ultrapurple owns one of these cameras. He may be able to offer comment on the provided images and compare them to those produced by his camera in similar scenarios. His camera would be a good “baseline” standard to compare against.


Just to get everyone updated, all the components are from the same camera, just transplanted into the donor case.
i did try the other microbolometer from the donor camera and it does work fine, but it exhibits the same visual issues with the vertical bar as the original.
See below first 2 pictures 0066 and 0068 taken at different angles of the same object, you can see one shows the lines and other doesn't.
That confirms Bill's theory that there is a column amplifier issue.
What i am hoping is that it's just a calibration issue which i could correct by recalibrating the unit, i should be able to do that from the service menu which i can access.
I also took some other random photos to show functionality, some with a hot soldering iron for a much larger temperature spam.


[0] Message Index

[#] Next page

There was an error while thanking
Go to full version