Internal Light Leakage
Internal Light Leakage
I've recently received two red shifted v4.5 miniscopes for jRGECO1a and assembled them based on the wiki instructions. I haven't glued the PCB yet and wanted to first check the recordings and in both miniscopes and I see internal light bleed appear in the top right of the FOV once the LED goes above 50% (gain at high) and the top 5th of the FOV becomes completely saturated at 100% LED power. At normal gain and LED power the FOV looks fine.
Let me know if you've seen this issue before and what is the best recourse. The tape I'm testing isn't exactly black electrical tape, but I've tried it on many black surfaces and have seen the same results.
(I'd like to upload images of the FOV and miniscope to help describe this better, but WM says I don't have permission to upload) Hale Tobin (talk) 14:34, 13 October 2026 (PDT)
- Thanks Hale. I've uploaded the images you sent.
- @Federico can add more, but before the PCB is glued in place, you typically need to apply some pressure with your finger during testing to prevent light leakage. I'd try that first to see if it resolves the issue. If so, gluing it in place should be the permanent fix.
- You really shouldn't need to use such high gain or LED power during an actual experiment. If the image looks fine at normal gain and LED power, you should be good to go. Samara Hecht (talk) 15:32, 13 October 2026 (PDT)
Samara Hecht (talk) 15:32, 13 October 2026 (PDT)
- Adding to what Samara said: the pattern you're describing (bleed in one corner that only appears at high LED power and high gain) is very consistent with stray LED light leaking through the gap between the sensor PCB and the housing before it's glued. A quick check is to press the PCB down firmly with your finger while recording at 100% LED. If the saturated region shrinks or disappears, the gluing step should fix it permanently.
- Keep in mind, too, that the thin coated filters used in the miniscope don't block excitationn light perfectly, and the leakage grows as LED power goes up. Some bleed at 100% LED with high gain is expected, so testing at those settings will exaggerate the problem compared to what you'd see at normal working settings. Federico (talk) 00:14, 20 October 2026 (PDT)