Re: [misterhouse-users] Stargate Thermostats

[Martin D. <mdo...@po...>]

Didn't think I would get a chance to try that change until later.   That 
modification worked like a charm,  Now it's a exact mapping from SG to 
MH.  Only thing I wonder is if anyone else beside you and I use SG and MH 
together would they potentially get confused that SG vars, SG flags start 
at zero?

as for the first example confusing you.  I had spell check on but I guess 
content check slips a little  late at night :)

Should read:

> > SG     MH (items) result set
> > 1     (down)   1 (down)       SG & MH zone 2 set (gr)
> > 2     (gr)    2  (gr)         SG & MH zone 3 set (up)
> > 3     (up)    3 (up)          nothing set (debug show trying to set 
> zone 4)

Hope that is clearer.


Since my SG clock seems to loose time, took 6 months to loose 10 
minutes,  I also tried to use the set_time from stargate.pm.  I got the:

Stargate bad write on set_time: 38   as the error.   I looked at the 
homebase version and the only difference was this line:

stargate.pm   = $main::config_parms{time_zone}

Homebase.pm = abs($main::config_parms{time_zone})

I made that change to stargate and it worked like it should.

Very minor question:  Noticed that both stargate and homebase have a 
comment that isdst should be defined in the mh.private.ini for use in 
set_time .  However the value is assigned on the next line with 
$isdst="00"; . and a config_parm for dst or isdst  is not part of the 
sprintf function.  (at line 1059 in my editor)  let me know if I'm missing 
something or if i should do as I did and ignore the comment.


Thanks,
Marty




At 10:00 PM 1/20/2003 -0800, you wrote:

>Martin,
>
> > I've just gotten my thermostats in stargate working with MH.  They work
>and
> > I can do  setbacks in MH instead of stargate and make them hopefully a lot
> > smarter.  I was wondering if anyone else has the same problem i did.
>
>What is your exact .mht definitions, we support both the SG485RCSTHRM (if
>you have a comport tapped into the Stargate485 bus you can use this) as well
>as the STARGATETHERM (this interfaces via the Stargate COM1 as with
>vars/flags/etc).  I wrote both, and as I recall both are 1 based since the
>Stargate refers to them in a 1 based fashion.  I'm going to assume your
>refering to the STARGATETHERM objects and NOT the 485 modules.
>
>FYI, here are my zone defs
>SG485RCSTHRM,   1,                      DownstairsThermostat,
>Thermostats
>SG485RCSTHRM,   2,                      UpstairsThermostat,
>Thermostats
>SG485RCSTHRM,   3,                      MasterBedroomThermostat,
>Thermostats
>
>#STARGATETHERM, 1,                      DownstairsThermostat,
>Thermostats
>#STARGATETHERM, 2,                      UpstairsThermostat,
>Thermostats
>#STARGATETHERM, 3,                      MasterBedroomThermostat,
>Thermostats
>
>As you can see, I'm using the 485 bus interface, so it is possible I did
>something stupid in the code for the other type.
>
> > First try
> > SG MH (items) result set
> > 1 1 MH zone 1 set
> > 2 2 MH zone 2 set
> > 3 3 nothing set
> >
> > Second try
> > SG MH(items) result set
> > 1 0 nothing set (tries to access zone 0)
> > 2 1 MH zone 1 set
> > 3 2 MH zone 2 set
> >
> > Third try
> > SG MH (items) result set
> > 2 1 MH zone 1 set
> > 3 2 MH zone 2 set
> > 4 3 MH zone 3 set
>
>The first example confuses me, your saying that 1 mapped to 1, 2 mapped to
>2, but 3 didn't work?  As such I don't see how the thirdtry (2 mapped to 2,
>3 mapped to 2, etc) fixed anything?  Presuming we are off by one (very
>likely).  Can you try changing:
>
>     my $output = "##%5e" . sprintf("%02x%02x%02x\r", $self->{address},
>$command, $data);
>
>to
>
>     my $output = "##%5e" . sprintf("%02x%02x%02x\r", $self->{address}-1,
>$command, $data);
>
>And then use 1 based defs in your .mht file.  Let me know if that does the
>trick and I'll get the SG code updated (I just restructured alot of it, and
>I can't send it to you as you'll neeed other changes that Bruce is merging
>in)
>
>Bill
>
>
>
>
>
>-------------------------------------------------------
>This SF.NET email is sponsored by:
>SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
>http://www.vasoftware.com
>________________________________________________________
>To unsubscribe from this list, go to: 
>http://sourceforge.net/mail/?group_id=1365