Optimize touch sensing

[KravitzLab]

Our current method of touch sensing is not 100% reliable at determining which touchpad woke the ESP32. We need to optimize this.

Touch documentation is here:
https://docs.espressif.com/projects/esp-idf/en/stable/esp32s3/api-reference/peripherals/touch_pad.html#_CPPv420touch_pad_get_statusv

[mattgaidica]

I believe I tried touch_pad_get_status() but it tossed a compilation error. I'm less and less impressed with their examples, as they're so bare bones, but this should be ground truth: https://github.com/espressif/arduino-esp32/blob/64b62dc45b008bc578a9918d3b9d537cd07d6ce9/libraries/ESP32/examples/Touch/TouchInterrupt/TouchInterrupt.ino#L14

We tried that (twice I think!) and it was unreliable.

[KravitzLab]

It's annoying! How can waking up on multiple touch pins be such a rare use case? Here is Espressif declining to provide an example of this functionality in Oct of this year - the fact they won't provide an example makes me worried it can't be done? https://esp32.com/viewtopic.php?f=5&t=41970&sid=1f24748610ff561d100d7b6717a8efb4 [cid:aeaf556f-38be-4a38-a9bd-e3122eada450] At some point we might need to add a dedicated touch chip if this is all just too annoying. The other option is to just re-populate the photogates, we left the footprints for them. They are power hungry but our sleep current would still be in the FED3 range (~10mA) with them on, just seems a shame to burn that much power while sleeping. -Lex Alexxai Kravitz, PhD Department of Psychiatry Washington University in St Louis ***@***.******@***.***>

…

________________________________ From: Matt Gaidica, Ph.D. ***@***.***> Sent: Saturday, December 14, 2024 3:06 PM To: KravitzLabDevices/FED4 ***@***.***> Cc: Kravitz, Alexxai ***@***.***>; Author ***@***.***> Subject: Re: [KravitzLabDevices/FED4] Optimize touch sensing (Issue #3) * External Email - Caution * I believe I tried touch_pad_get_status() but it tossed a compilation error. I'm less and less impressed with their examples, as they're so bare bones, but this should be ground truth: https://github.com/espressif/arduino-esp32/blob/64b62dc45b008bc578a9918d3b9d537cd07d6ce9/libraries/ESP32/examples/Touch/TouchInterrupt/TouchInterrupt.ino#L14 We tried that (twice I think!) and it was unreliable. — Reply to this email directly, view it on GitHub<#3 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AD76MGJQFWPDQ5G66T2WYAT2FSMW5AVCNFSM6AAAAABTTYPTJGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNBTGM2DINRYHE>. You are receiving this because you authored the thread.Message ID: ***@***.***>

________________________________ The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

[KravitzLab]

New example here that seems to work better. The important things are:

1. attach T1, T5, and T6 to independent ISRs that keep track of flags for which touchpad was triggered
2. include a 100ms debounce in the ISRs so only one flag can be true at a time

[KravitzLab]

Example moved here - try implementing this on FED4