Daniel,

The origin of 85 was that the original sensors were 8 bit and only went 
to 85c and thus all 1s (FF)was an error. When the extended the range to 
125c and the bits extended, the error value was already set and produces 
the problem you have. Hind sight says they should have used the extra 
byte up front and sent an out or range number, but that doesn't help now.

I know of no way to tell the difference between an error and an 85c 
reading, others may be smarter about this.

In my experience there is a fair amount of low bit noise in the readings 
of the 18b20s at 12 bit, so there would not naturally be a long string 
of 85c readings in a row if the sensor is working. So I just discard all 
85c readings. Reading at 8 bits may be more problematic.

Another trick that I do to avoid unreasonable readings is to use a 
median filter. With this, you take an odd number (I use 3) of the latest 
reading and pick the median value (sort it and pick the value in the 
middle.) So if you get a single crazy reading, it has no effect on the 
filter output. This again has the net effect of slowing down the rate of 
update (in my case, one read cycle.)

I use simultaneous readings and read from uncached to make sure I get 
the latest reading while making the conversion time tolerable. With my 
moderately slow response heating and cooling system, these decisions are 
completely reasonable. I have about 40 sensors running right now. YMMV.

jerry

On 07/13/2013 08:11 AM, Daniel MacKay wrote:
> I notice my DS18S20 and DS18B20 sensors, if there's a network failure or.. 
> perhaps a hardware failure from overheating the sensor when I'm soldering 
> wires onto the leads, reports "85" as the temperature.
>
> As far as I can tell the "crc8" value is correct in these instances.
>
> a) why does it do this?
> b) this really is not a helpful number. if it was -999 then I could say, "ah 
> that's an error value" but 85 is a reasonable temperature for the stuff I'm 
> doing
> c) is there any way to tell that this is an error 85 as opposed to the sensor 
> actually being at 85°C ?
>
> Many thanks.
> ------------------------------------------------------------------------------
> See everything from the browser to the database with AppDynamics
> Get end-to-end visibility with application monitoring from AppDynamics
> Isolate bottlenecks and diagnose root cause in seconds.
> Start your free trial of AppDynamics Pro today!
> http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
> _______________________________________________
> Owfs-developers mailing list
> Owfs-developers@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/owfs-developers


------------------------------------------------------------------------------
See everything from the browser to the database with AppDynamics
Get end-to-end visibility with application monitoring from AppDynamics
Isolate bottlenecks and diagnose root cause in seconds.
Start your free trial of AppDynamics Pro today!
http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
_______________________________________________
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers

Reply via email to