suda-morris
commented
7 months ago @rpaulbeam Thanks for reporting! I don't have a MAX31820 before my hand, It would be very appreciated if you could submit a PR in this repo for the fix.
Open rpaulbeam opened 7 months ago
suda-morris
commented
7 months ago @rpaulbeam Thanks for reporting! I don't have a MAX31820 before my hand, It would be very appreciated if you could submit a PR in this repo for the fix.
rpaulbeam
commented
7 months ago The MAX3180 is a compatible temperature sensor and it works just like the ds18b20. The problem with the search routine is a discrepancy on bit 0, which can happen with different kinds of devices. I'm not experienced with esp-idf so it took me a while to move the component from a managed component to one I can modify. I will fix over the weekend and post as suggested.
Get Outlook for iOShttps://aka.ms/o0ukef
From: suda-morris @.> Sent: Saturday, January 20, 2024 4:48:00 AM To: espressif/esp-idf @.> Cc: Paul Beam @.>; Mention @.> Subject: Re: [espressif/esp-idf] OneWire Search not working for mixed devices (IDFGH-11938) (Issue #13017)
@rpaulbeamhttps://github.com/rpaulbeam Thanks for reporting! I don't have a MAX31820 before my hand, It would be very appreciated if you could submit a PR in this repohttps://github.com/espressif/idf-extra-components/tree/master/onewire_bus for the fix.
— Reply to this email directly, view it on GitHubhttps://github.com/espressif/esp-idf/issues/13017#issuecomment-1902046032, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ARIKJB3FKKLDTUTQG32RD6LYPOHFBAVCNFSM6AAAAABCCUHWSSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMBSGA2DMMBTGI. You are receiving this because you were mentioned.Message ID: @.***>
suda-morris
commented
7 months ago
Answers checklist.
General issue report
I'm using esp-idf 5.1.2 with ESP32-S3-DevkitC in a Windows/VS Code environment. I have 2 onewire devices (DS2408 and MAX31820) on my bus, which is just a breadboard. Using the Onewire example, it can detect either of the devices, but not both. Plugging the same "network" into a Linux SBC, both devices are detected, so I believe the wiring is correct. The ID of the DS2408 (470000003FB84629) supersedes the 31820 (0100000ED5745128), so only it is detected if both are attached.
Looking at onewire_device_iter_get_next() with the JTAG debugger, I can tell that search_direction is never 0 -- last_zero = rom_bit_index is never executed even though a discrepancy is detected. I think the issue is different device families. The problem may be that last_dispepancy is initialized as 0 and rom_bit_index is initialized as 0, so a discrepancy of the first bit will be detected as the 2nd occurrance of this discrepancy? A different implementation (esp32 bitbang onewire) initializes last_discrepancy as 64. The Analog Devices App Note initializes last discrepancy as 0, but the first bit is 1 and not 0.