A couple more 1-wire questions

[pete@malibubeach.com]

I'm getting very close to not only getting it working but also understanding the technology (which, having been in the technology business for 35+ years is something I seem to always try to do). But I have a few more questions.

First, I have three Hobby-boards.com modules: a temperature only, a temperature/humidity and a temperature/humidity/sunlight. For the moment, Ineamed the HS devices with the sensor type and the last 4 digits of the sensor serail number, so I know what Iam looking at. All three units are located within 3 or 4 inches of each other in an open area (i.e. I would expect the readongs to be the same.

So here are the questions:
1. First, the onle I understand: some of the measurements show up with two different names. I believe this is caused because the 1-wire xap program had seen them before I gave them a name and this is propogated through Homeseer. I believe that changing the 'accept' to 'reject' and/or excluding them in the 1-wire setup will solve this.

2. I get five temperatures: one from the DS18S20 in each of the three mdules and one from the DS2438 in the temp/hum and temp/hum/light modules. Would you expect the temperatures to be so varied? I thought they were speced as 0.5 degrees C, yet I see a variance of over 4 deg F (or 2+ deg C). If I eliminate the one I've called 'Temp-5010' which is the temp only unit and it is the only one not in the Hobby-Boards plastic case, perhaps based on heat generated by (maybe) the voltage regulator. That would not be a good thing in my mind. Perhaps I'll take them out of their cases and/or remove the jumper so everything runs on parasitic power and see what that does.

3. I was trying to figure out the relationship between the three voltages Michael's program reports and the DS2438 hardware. I think I've got it but perhaps, Michael, you could confirm: V0 is the voltage register with the A/D configured to convert VDD; V1 is the voltage register with the A/D configured to convert VAD (which comes from the humidity chip); V2 is the current sense register, which is really the voltage across the resister in series with the sola diode. Also, it appears that the 'sunlight' number reported is V2 times 400; and the relativve humidity reported is somehow inversely proprtional to V1. Not sure why I care except in my attempt to figure out the next one.

4. One of the humidity sensors is reporting a constant humidity of 100% (and V1 is a very low number of .09). The other correclty reports 47% (my Davis Vantage currently says 45%, so this is well within the expected error range for both) and V1 is 2.21. I am assuming that the one reporting 100% is defective and I should probably send it back to Hobby-boards for a replacement. Is there any other possible explanation?

5. Some of these have never been sent to Homeseer. Is that because they are only sent on change, so if Homeseer wasn't configured to accept them when it started and they never change (like the 100% humidity), they are never updated?

6. Finally, is there a unit available that is better packaged for wall installation than the hobby-boards units? Specifically, it would be nice if the RJ45's came out the back so the cables could go into the wall without being visible.

Thanks for any help or advice!

Pete

[Michael McSharry]

1. First, the onle I understand: some of the measurements show up with two different names. I believe this is caused because the 1-wire xap program had seen them before I gave them a name and this is propogated through Homeseer. I believe that changing the 'accept' to 'reject' and/or excluding them in the 1-wire setup will solve this.

Essentially correct. I implemented xapmcs1Wire to include the name as part of the address and mcsXap to use the address to identify the item. If you change the name at xapmcs1wire then when it sends it will be a new sensor from HS's perspective and will need to be accepted as a new device. You can change the name via mcsXap and then HS will know it by whatever you want and it will be different than what xapmcs1wire is sending so you may become confused later when you forget what you did.

2. I get five temperatures: one from the DS18S20 in each of the three mdules and one from the DS2438 in the temp/hum and temp/hum/light modules. Would you expect the temperatures to be so varied? I thought they were speced as 0.5 degrees C, yet I see a variance of over 4 deg F (or 2+ deg C). If I eliminate the one I've called 'Temp-5010' which is the temp only unit and it is the only one not in the Hobby-Boards plastic case, perhaps based on heat generated by (maybe) the voltage regulator. That would not be a good thing in my mind. Perhaps I'll take them out of their cases and/or remove the jumper so everything runs on parasitic power and see what that does.

4 degress is quite a bit. The calibration feature of xapmcs1Wire can be used to account for a bias. This assumes the difference is only a bias.

3. I was trying to figure out the relationship between the three voltages Michael's program reports and the DS2438 hardware. I think I've got it but perhaps, Michael, you could confirm: V0 is the voltage register with the A/D configured to convert VDD; V1 is the voltage register with the A/D configured to convert VAD (which comes from the humidity chip); V2 is the current sense register, which is really the voltage across the resister in series with the sola diode. Also, it appears that the 'sunlight' number reported is V2 times 400; and the relativve humidity reported is somehow inversely proprtional to V1. Not sure why I care except in my attempt to figure out the next one.

I would need to go back to look at schematics to assess your questions with assurance. I use encapsulated functions for most of the DS2438 interface. I played games with faking the current resistor to get a solar in the range that was similiar to temperature so I could graph it easier.

4. One of the humidity sensors is reporting a constant humidity of 100% (and V1 is a very low number of .09). The other correclty reports 47% (my Davis Vantage currently says 45%, so this is well within the expected error range for both) and V1 is 2.21. I am assuming that the one reporting 100% is defective and I should probably send it back to Hobby-boards for a replacement. Is there any other possible explanation?

I think you are correct. Of course consult with Eric.

5. Some of these have never been sent to Homeseer. Is that because they are only sent on change, so if Homeseer wasn't configured to accept them when it started and they never change (like the 100% humidity), they are never updated?

That is correct. xAP has a provision in the spec for Basic Status and Control that the device needs to respond the current value of any of its points when queried. It only reports events on change of value. The bottom of xapmcs1Wire has a button to send current values. It could also be queried for other xAP locations.

6. Finally, is there a unit available that is better packaged for wall installation than the hobby-boards units? Specifically, it would be nice if the RJ45's came out the back so the cables could go into the wall without being visible.

I'm not familiar with various assembled units. I purchased the raw sensors from maxim-ic and made my own mounts (crimp onto a RJ11 plug). My humidity devices are installed in areas that are not visible.