As the display time extends, the image turns white.

[931436690]

Hi,there.
Thank you very much for the open source material,these are very good projects.

We found a small problem of the miniscope. The setting parameters remain the same, but with the increase of the display time, the displayed image becomes whitish (the image displayed after ten minutes is brighter than the image just after power-on).
The display effect is as follows，(Using ImageJ, the third image is the image ten minutes later subtract the image ten minutes earlier）

My guess is that the heat of the serializer that affects some chips (such as power chip U8_TLV7103318DSER, L4_BLM15HD102SN1D or L5_ADL3225V-470MT-TL000). but I'm not sure.The deserializer chip of DAQ don't get as hot as the serializer chip of miniscope.Probably because the DAQ board dissipates heat better.Moving the affected chip away from the serializer may be an option.
We dissipate heat on the serializer of Miniscope (using a very heavy copper heatsink), and there is basically no difference in brightness. The effect is as follows:

Do you have any suggestions? The whitening problem of the image currently affects the imaging effect.

[ckemere]

Heat is generated by 3 parts of the miniscope circuitry, the excitation LED driver, the image sensor, and the serializer. Decreasing LED brightness and reducing frame rate will decrease the amount of current that they consume and thus the amount of heat they need to dissipate. Heat will increase the noise in the image sensor, which likely explains the "brightening" you're observing. (It can also potentially decrease the excitation brightness, which has the opposite effect.) In the case of cellular-level calcium imaging, field-wide trends like this should be pulled out prior to df/F calculation, but, as with photobleaching (which has the opposite effect), one should take this into account in analysis.

This is a much bigger issue if you're trying to image bulk signals, but in that case, there is also a concern for hemodynamic contamination, which the single-wavelength v4 is not designed account for.

[931436690]

Thanks for your answer; we'll test the way of reducing frame rate. If there is a better solution in the future, please let us know.

[931436690]

Hi, there.
We have another question to consult.
The MiniscopeV4 we purchased from Labmaker had a brightness change within 2 minutes of power-on (the image changed from dark to bright), and the image brightness did not change during the subsequent time (the LED brightness and image frame rate and other parameters were not changed during the entire 40-minute experiment); but The MiniscopeV4 we made ourselves (Thanks to your open source material.), had a brightness change within 2 minutes of power-on (the image changed from dark to bright), and then the image became brighter and brighter (During the entire 40-minute experiment, parameters such as LED brightness remained unchanged). We don't know the differences between ‘the microscope made by Labmaker’ and ‘our self-made microscope’. We guess it is the precision of some devices or the improvement of some chips (we use the circuit board made with 0.5oz inner layer copper thickness and 1oz outer layer copper thickness, this should not be the cause of the problem; we don’t know if the problem can be solved by using high-precision capacitor resistors, such as R8 R9 C33 C34; it seems that the temperature affects the analog voltage).
We hope that the MiniscopeV4 made by ourselves can achieve the same effect as the "microscope made by Labmaker"; do you have any suggestions for this? Thanks a lot for your advice and help.

[karffy]

Indeed the issue is related to higher temperature.

• Higher temperature in imaging sensor leads to higher dark current noise.
• When LED warm up from room temperature to working temperature, normally the intensity increases at the frist few minutes.

On designing the experiment, one may consider warm LED up to thermal equilibrium first to avoid LED related brightening. But the dark current on the sensor should be avoided by cooling, or reduce frame rate as suggested by ckemere.