DS1820, DS18B20 Support

[LarrylLix]

I just got a chance to update here.... SUCCESS!
 
I updated the firmware with the new test version you supplied, no problem. No change in the probe reporting resolution so I power cycled the WC board with still no success. :angry2:
 
Then I hooked up my second probe and I guess I had the power off for about two minutes while I parallelled in the second DS18B20 probe. They both booted up with 0.1 degrees C resolution increments showing in the status page!
 
Thanks to all for their input and CAI for jumping in with firmware to resolve this!

 

[CAI_Support]

If anyone wants to update, please go to here to find update firmware for WC8
 

http://cocoontech.com/forums/topic/26873-wc8-firmware-30218b-released/

 

[LarrylLix]

CAI
 
My 9 bit resolution is back! Somehow my DS18B20 probe is now sending 9 bit resolution temperatures again! (+/-0.5 degree increments.)
 
Weird. I may have interchanged my probe, I have two units, but I believe I had both of them attached previously and they both upgraded to 12 bits of precision transmission.
 
Somehow they have reset themselves and the new firmware isn't upgrading them, as it did after your fix, or the firmware has reset the resolution of the reading setup. I have power cycled the board several times to test firmware initialisation corrections of the probe. No success.
 
Note: This was before the f/w update to ...182b3 f/w which I am currently testing.

 

[CAI_Support]

In the last version we made change instead of re-programming the sensor to 12 bit every time, we only program that particular sensor one time. In that way, we hope the EEPROM will not be worn out.  Now, it seems that sensor does not really remember the 12bit we programmed. So we will change it again to have it based on check result, program the sensor to 12 bit, if it does not remember.

 

[LarrylLix]

Running v03.02.18b3
 
An update with some more weird 1wire stuff.
 
Tonight I disconnected my DS2438 sensor and replaced it with a second DS18B20, losing my humidity sensing. Now I have two DS18B20 probes both with 1m cables attached, one mostly outside (inside a  PVC conduit) and the other inside my shop hanging over the WC8 board metal cabinet. 
 
Restoring the power to the board both probes came up with 0.1 C resolution reporting in the WC8. When I had the mixed probes attached (1m short leads on both) the DS18B20 would only report 0.5C resolution despite power cycling several times.
 
Now I also wonder that this same sequence happened last time when CAI first forced the sensor into 12 bit. I had the power off the board and sensors for several minutes in order to change the wiring. Is it possible these tiny probes have a huge capacitor in them that can keep there settings alive for longer than say...60 seconds?
 
Perhaps these probes have be powered down for more than about 60-120 seconds before they reset or the firmware picks up on the reboot somehow? 
 
Anyway my 12 bit reporting is back????

 

[CAI_Support]

DS18B20 does have large capacitor inside so that even when RESET command pull the bus down for sometime, it can operate in parasite mode.
Without any command sending to it and power off, it will drain down to total reset in few seconds.
 
According to Maxim IC, DS18B20 should not have any running resolution at 0.5C by default. However, we did have two of those running in 0.5 resolution.  One can be re-programmed to 0.1C, the other can not even be programmed to 0.1C.  Maxim IC told us those showing 0.5C resolution were old sensors.  With WC8 3.18b3 firmware, if your sensor can be programmed to 0.1C, it will.  But some sensors may not work, that is beyond software can do.

 

[LarrylLix]

Read again.
 
My probes can be set, and have been set, to 0.1 degrees of resolution. The one probe changed back to only 0.5 degree resolution. After power cycling many times and waiting at least 30 seconds before returning power  many times, the probe refused to return to 0.1 c resolution.
 
After rewiring the probes (removing the DS2438 and adding another DS18B20)  the power was off for more than several minutes and the same 0.5c probe returned to 0.1 degrees C resolution.
 
Try doing this with your offending probes. Leave them powered down for five-ten minutes. My guess is they will come back with 0.1 degrees. If you firmware is adaptive somehow these probes may still report "initialised" after 60-120 seconds without power. Twice this has happened to me that I could not return them to 0.1 until powered off for over five minutes.

 

[CAI_Support]

I think you should bring this matter also to your probe supplier, so that they aware the chips being used are not function correctly.
WC8 logic is very simple, during power up, it will read the configuration register. If that is not 12bit, then write to 12 bit to the configuration register and store in EEPROM.
We do not go back and re-read and re-try during start-up time, otherwise, we could deadlock with a bad sensor.   The 9bit DS18B20 chip we got directly from Maxim IC always programmed into 12 bit in this first try.
 
It sounds like some of your probes will need a few tries to actually store the configuration.  That might be the internal EEPROM in DS18B20 failing.