Little update at EOSHD with a little help from a1ex and Horshack:
https://www.eoshd.com/news/canon-eos-r5-so-called-overheat-timer-defeated-by-a-single-screw-in-battery-door/#more-24156
LOL, I only pointed Andrew to this link (Horshack's post):
https://www.dpreview.com/forums/thread/4513421#forum-post-64281948Many cameras don't update the values in NVRAM until they're orderly shutdown, ie when the power-switch is turned to off. They keep the values shadowed in SDRAM until then. This is done both for performance and to reduce wear on the NVRAM. You can see evidence of this on Nikon cameras by abruptly pulling the battery and then plugging it back in - any config/exposure changes from the previous session will be lost.
Canon is a little more clever - they have a latch sensor on their battery doors that performs an orderly shutdown whenever the battery door is opened. This prevents an abrupt shutdown. To work around this someone could tape/jig the door sensor so that it's always pushed-in. The experiment would be as follows:
1. Find a way to hold the battery door latch sensor in
2. Power on the camera with the battery door open
3. Record video to the point of thermal cutoff (but not power-off shutdown)
4. Pull the battery
5. Reinsert the battery
6. See what the available video time is
There's a chance Canon periodically updates the NVRAM during the session (or when video recording stops), esp for something as important as thermal management. In that case it may be necessary to pull the battery while the video is actively recording. This may leave the CF/SD card in an indeterminate state so it may require a format when the battery is reinserted.
There are also others who tried this trick in various ways (e.g. in
this slightly NSFW clip, or
this variation with a battery grip), so credits don't belong to me - I didn't discover anything
From what I've seen, the trick works if one removes the battery in the middle of recording, but no longer works if the battery is removed after properly stopping the video. As Canon firmware is pretty much event-based, they probably configure an alarm timer directly on the RTC chip, as soon as video recording is stopped. That could also explain why the camera "forgets" about overheating after removing the clock battery.
A software patch might be relatively easy if one knows where to look, assuming the internals are similar to M50, EOS R and other DIGIC 8 models (I haven't seen the firmware).
However, before attempting to disable these timers, I should get some updated legal advice. If we were to implement this feature, it could raise some eyebrows at Canon
