| Author |
|
wasdvd
Groupie
Joined: February 13 2007
Location: United States
Online Status: Offline
Posts: 42
|
| Posted: March 16 2007 at 09:01 | IP Logged
|
|
|
I have recently installed the W800RF32 in an attempt to get more reliable results with the PalmPad and MS14A etc motion detectors. I am trying to use these X10 signals to trigger control of Insteon only devices. No X10 receivers or devices.
The issue is that the codes detected from the W800RF32 by PH are not always correct. If I press the A1 button, sometimes PH event log shows A1, but sometimes it shows D5, or O2 or E1.
I have used rf32.exe to see if the W800RF32 is bad, or if the COM port is inconsistent, but rf32.exe seems to give very fast, reliable results.
Your help will be appreciated.
|
| Back to Top |
|
| |
TonyNo
Moderator Group
Joined: December 05 2001
Location: United States
Online Status: Offline
Posts: 2889
|
| Posted: March 16 2007 at 11:59 | IP Logged
|
|
|
I am seeing that, too. I had chalked it up to a bad PalmPad. Maybe it's something else...
Tony
|
| Back to Top |
|
| |
Deano
Groupie
Joined: February 19 2003
Location: United States
Online Status: Offline
Posts: 75
|
| Posted: March 17 2007 at 10:48 | IP Logged
|
|
|
I think the problem is bad palm pads. I have one that is a pure garbage generator.
|
| Back to Top |
|
| |
wasdvd
Groupie
Joined: February 13 2007
Location: United States
Online Status: Offline
Posts: 42
|
| Posted: March 17 2007 at 17:34 | IP Logged
|
|
|
I had assumd that too, since they are certainly cheap enough. When I used rf32.exe, though, all signals were detected correctly, with the same W800 and the same PalmPads and MS14s.
I have the same issue with MS13,14,18 motion detectors. PH does not seem to interpret the signals correctly, while rf32.exe does. In fact, with TonyNo's help on marcos, this is now my holdup.
|
| Back to Top |
|
| |
Deano
Groupie
Joined: February 19 2003
Location: United States
Online Status: Offline
Posts: 75
|
| Posted: March 18 2007 at 16:51 | IP Logged
|
|
|
I am also using a W800rf32. I don't use palm pads much except for testing but I have a house full of motion sensors and the problem I have with them I thought was speed but I guess it could be garbage. I also miss a lot of photocell status changes.
I have a P2 200 MHz machine which is slow until you compare it with x10 transmission speed.
Often there is a delay of 4 or 5 seconds from the time I see the led blink on the motion sensor and when the macro starts executing. I use Dim 100 instead of off on most of my lights so I can bring them up to 40 or 50 % in the middle of the night, but even though it takes 5 seconds or so to reach 100 I think I should see it start sooner than I do.
I have been timing program segments with ph_gettickcount but I don't know how to determine how long it takes from power home's input routine until a macro starts.
Watching the status screen, I don't see garbage, but I haven't screened the event log for goofy stuff.
I have learned that using immediate instead of post (in the hopes that immediate meant do it now dammit) is the wrong answer and in fact seems to slow every thing down. Apparently immediate doesn't let anything else execute until a macro is complete. Don't know what would happen it you call a macro that initiates a wait with immediate. Might be one of those pull the plug power fail restarts.
Anyway, I'm not familiar with rf32.exe. Does it time stamp and where can I get it.
We must also not forget that that good ole RS232 COM port technology is between the W80032rf and our beloved Power Home.
|
| Back to Top |
|
| |
wasdvd
Groupie
Joined: February 13 2007
Location: United States
Online Status: Offline
Posts: 42
|
| Posted: March 18 2007 at 17:14 | IP Logged
|
|
|
Deano:
<We must also not forget that that good ole RS232 COM port technology is between the W80032rf and our beloved Power Home.>
I thought that the RS232 might be the issue, too, at first. But after using the "demo" program rf32.exe from WGL's site (link below), I think that each signal is capable of being received and decoded correctly. The baud rate is also only 4800.
That is why my guess at this time, is that PH is somehow part of the missing/bad house codes and unit codes. In X10 RF ext, the same issues apply.
On the other hand, on HomeSeer boards, there is discussion of what looks like exactly the same issues with scrambled codes with W800 and HomeSeer.
http://www.wgldesigns.com/downloads/w800rf32_win_decoder.zip
|
| Back to Top |
|
| |
Deano
Groupie
Joined: February 19 2003
Location: United States
Online Status: Offline
Posts: 75
|
| Posted: March 19 2007 at 09:23 | IP Logged
|
|
|
Thanks for the link
http://www.wgldesigns.com/downloads/w800rf32_win_decoder.zip
Would be nice if it would run in parallel with power home
I have 4 palm pads that I can find and a cr12a. When tested with rf32.exe, the one that used to generate garbage is completely dead (I might have fixed it too much), another has buttons that don't work very well, but the remaining 2 as well as the cr12a work just fine and the response is immediate.
While I have not tested all 256 combinations on the three good ones, they seem to work fine with power home as well.
Another observation: If I use palm pads, keychains or stick-a-switches to trigger macros, they seem to respond much faster than when the motion sensors fire. Maybe the led on the motion sensor is fooling me.
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: March 19 2007 at 15:05 | IP Logged
|
|
|
Hmmm...I will dig out my W800RF32A and retest the controller communication routines. If anyone has any test scenarios where they can pretty much make the problem duplicatible (is that a word?), please pass them along.
Dean,
You may want to email me a copy of your database (if you've got most all events being logged. The eventlog only shows resolution in seconds but I can go into the tables and see the timestamp with microseconds resolution. If you're logging all of your events, then I should be able to determine where the slowdown is occuring.
Dave.
|
| Back to Top |
|
| |
dhoward
Admin Group
Joined: June 29 2001
Location: United States
Online Status: Offline
Posts: 4447
|
| Posted: April 01 2007 at 10:37 | IP Logged
|
|
|
Guys,
Ok, Ive broken out the W800RF32A and have started testing. Ive made numerous changes to the code to try and improve efficiency and perhaps fix possible "lost" codes, but nothing really towards interpretation. Ive been sending codes to it and trying to get it to screw up by pressing multiple buttons simultaneously and sending codes as fast as I can push buttons. Everything seems to work just fine. The only time I can get it to mis-interpret codes is when pushing multiple buttons simultaneously. However, I tested the same scenario with RF32.EXE and it does the same. I looked at the raw data and saw that PowerHome was correctly interpreting the data it received.
Soo, Im not sure where to go from here. Ive made vast changes to the W800RF32 code, but I don't think I specifically changed anything in regards to intepretation of the received signals. At this point, we can wait for the next version to come out and see if it performs any better or if anyone has any other additional data to provide, I can see what else I can find out.
Dean,
Got your database (as well as a few others) and will be checking it out shortly.
Dave.
|
| Back to Top |
|
| |
wasdvd
Groupie
Joined: February 13 2007
Location: United States
Online Status: Offline
Posts: 42
|
| Posted: April 01 2007 at 12:56 | IP Logged
|
|
|
Tkanks, Dave.
Still enjoying PH.
|
| Back to Top |
|
| |
Active Topics 
Memberlist 
Search 
Help
Register 
Login
PowerHome Messageboard : PowerHome General
Topic: X10 codes thru W800RF32
