During testing of a new modification I added, MOSFET power control to bring down the weather station standby current, everything was looking great until I monkeyed around with the 1-Wire rain gauge counter and the windvane. (I completed the 1-wire counter retrofit, BTW, more on that later.) All I wanted to do was add header pin connectors to both device’s cables and run the MetOne weather station code (and as the saying goes: “All I wanted to do was drain the swamp and then I found myself up to my ass in alligators”). Upon reconnecting my newly completed and soldered 1-wire counter module (new post coming forthwith), the debug output was going crazy trying to find any 1-Wire devices. Like…huh?!? This WAS working before.
Here’s what was going on from output from PlatformIO:
Checking for I2C devices…:
Checking for 1-Wire devices…
There are devices present on 1-Wire bus
Searching 1-Wire bus with OWSearch()…
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }
Calculated CRC of device serial is 0
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80 }
Calculated CRC of device serial is 0
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x40 }
Calculated CRC of device serial is 0
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xC0 }
Calculated CRC of device serial is 0
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x20 }
Calculated CRC of device serial is 0
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xA0 }
… (it doesn’t ever stop!)
Great. Now I have to delve back into the 1-Wire search algorithm to find out what the hell is going on. 🙁 So, I loaded up a KNOWN WORKING code and ran that. Same thing!! And I tried another one I had, and it showed the same output. Now my mind is totally blown – what happened?!? Was it the DS2482?? This has me really bummed out – thought I was on the home stretch finally, especially now when (according to Seattle standards, anyway), a monster lightning storm moved through and dumped a TON of rain. And me with non-working rain gauge now. 🙁
As it turns out this was probably good thing that I found this now. The problem was that the 1-Wire bus was shorted to ground (on my breadboard), and I didn’t test for the SD (short detected) bit in the DS2482’s status register. Now I’m going to fix that and update my DS248X 1-wire library. (Interesting that other DS2482 libraries didn’t catch this as well.) 
Considering the kind of rat’s nest of jumpers, maybe it’s a small wonder that it didn’t happen before. Those white wires are connected to ground common to set the DS2482 address. I need more black jumpers apparently to make it more obvious.
So, I added some debugging info showing the bit status of the DS2482 status register to the OWReset() function. This function tests for a “Presence Pulse” from ANY 1-Wire devices on the bus or whether the bus is shorted. Here is some output from my test:
DS2482 Status Register bit definitions:
Checking for 1-Wire devices…
Status Register: 10101010
There are devices present on 1-Wire bus
Searching 1-Wire bus with OWSearch()…
Status Register: 10101010
Found device: { 0x10, 0x67, 0xBB, 0xE0, 0x02, 0x08, 0x00, 0xF2 }
Calculated CRC of device serial is F2
This device is a 1-wire thermometer
Status Register: 10101010
Status Register: 10101010 <– Normal
Found device: { 0x12, 0x05, 0x1C, 0x22, 0x00, 0x00, 0x00, 0x40 }
Calculated CRC of device serial is 40
Status Register: 01001010 <– Normal
Found device: { 0x1D, 0x97, 0x87, 0x01, 0x00, 0x00, 0x00, 0xE2 }
Calculated CRC of device serial is E2
Found 1-wire counter at { 0x1D, 0x97, 0x87, 0x01, 0x00, 0x00, 0x00, 0xE2 }
Status Register: 10101010
Status Register: 10101010 <– Normal
DS2423 returned: 0
Checking for I2C devices…:
Checking for 1-Wire devices…
Status Register: 10100110
There are devices present on 1-Wire bus
Searching 1-Wire bus with OWSearch()…
Status Register: 10100110 <– weird… SD bit (bit 2) set AND Presence Pulse (bit 1) (??)
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }
Calculated CRC of device serial is 0
Status Register: 00000110 <– weird… SD bit (bit 2) set AND Presence Pulse (bit 1) (??)
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80 }
Calculated CRC of device serial is 0
Status Register: 10000110
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x40 }
Calculated CRC of device serial is 0
Status Register: 00000110
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xC0 }
Calculated CRC of device serial is 0
Status Register: 10000110
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x20 }
Calculated CRC of device serial is 0
Status Register: 00000110
Found device: { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xA0 }
Calculated CRC of device serial is 0
In each case when the short is detected (SD bit set) the presence pulse (PPD) is also set. I don’t understand that-it’s not supposed to be set when SD is set. Very odd. Anyway, now to fix the library and the weather station code.