Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[bryana_fl]

I have been operating my second CM15A
(first had problems) since Easter weekend
without being connected to the computer.
It is located in the center of the house in
a closet that houses my alarm system.  All
macros, timers and motion sensors work
fine.  The clock has not drifted at all.

I have:
20 Modules
22 Macros
7 Palm Pads
1 slimline
7 timers
1 universal module (hooked up to Garage
door control)
1 chime
7 motion sensors
I am running version 3.183(if it ain’t
broke..)

PC is Athlon 64 3700 - 1.5 GB Ram
XP Pro Service Pack 2

Have quite a few more modules in a box for
future installation.  I'll get them
installed as well as additional macros
sometime when I have more time.

My setup is nothing fancy (no video, etc.)
but it works for me.

So far so good.

[bryana_fl]

Oh, thought of a couple of other points of
interest:
1)   I have a capacitor installed in my
breaker box to act as a coupler (this was
the source of some of my problems related
to inconsistent operation)
2)   A number of other problems turned
out to be macro related.
a.   Smart macros being used without
flags to assure the macro would not restart
prior to it’s initial completion)
b.   Operator induced logic errors
3)   Location of CM15A.  Originally
located next to:
a.   PC
b.   Wireless network access point/router
c.   Subwoofer for PC speakers
4)   CM15A Plugged into Battery Backup
unit.

[paul k]

I have had mine running for a couple weeks
now with no big problems. Sometimes connected
to the PC (WindowsXP) and sometimes not. But,
I think the USB driver on Windows is flaky.
When I was updating and adding macros a lot
(at 1st), it suddenly would not download
properly (very slow and never finished), and
things got weird. I used a USB sniffer, and
the driver was emitting junk. I stopped and
started the USB driver and I still could not
download. But, the messages were sane. So, I
disconnected the CM15A from power *and* USB
(it seems to draw power from USB - they are
requesting 500mA). It then has worked fine (I
did not remove the battery). I think they
need to look at their USB driver as it may
have bugs - appears to be related to multiple
downloads in one AHP session. I have not had
problems with it connected to the PC or
disconnected since. One big thing to watch -
if using a USB hub, make sure it is a powered
hub. I have a USB current monitor on my
powered hub and they are exceeding their
current allowance (up to 860mA peak). My
guess is a faulty ground plane on the board
(common USB error). This could also cause
people problems with other USB devices
connected to their PCs. So, be careful how it
is connected to the PC.

[paul k]

I have had mine running for a couple weeks
now with no big problems. Sometimes connected
to the PC (WindowsXP) and sometimes not. But,
I think the USB driver on Windows is flaky.
When I was updating and adding macros a lot
(at 1st), it suddenly would not download
properly (very slow and never finished), and
things got weird. I used a USB sniffer, and
the driver was emitting junk. I stopped and
started the USB driver and I still could not
download. But, the messages were sane. So, I
disconnected the CM15A from power *and* USB
(it seems to draw power from USB - they are
requesting 500mA). It then has worked fine (I
did not remove the battery). I think they
need to look at their USB driver as it may
have bugs - appears to be related to multiple
downloads in one AHP session. I have not had
problems with it connected to the PC or
disconnected since. One big thing to watch -
if using a USB hub, make sure it is a powered
hub. I have a USB current monitor on my
powered hub and they are exceeding their
current allowance (up to 860mA peak). My
guess is a faulty ground plane on the board
(common USB error). This could also cause
people problems with other USB devices
connected to their PCs. So, be careful how it
is connected to the PC.

[roger1818]

PaulK:  Interesting.  According to the
schematic
(https://gullfoss2.fcc.gov/prod/oet/forms/bl
obs/retrieve.cgi?
attachment_id=347600&native_or_pdf=pdf) the
USB power pin is not connected to
anything.  Maybe there is a solder short on
yours.  Since they aren’t using it, they
may not be testing that pin.  Strange that
it would request 500mA though.

[paul k]

Roger, I did not get any schematics off that
site - only the report (on radio
transmissions for FCC approval). Were you
able to get the schematics or picture?
In any case, the problem with USB is not
necessarily using the power (although they
should ask for far less if they do not use
it, this is a pain for a hub), but ground
loop and voltage plane interference. They do
not have to use the power, but the ground
must be attached to create a proper reference
for the LVDS signal pair. I wonder if this is
floating and so pulling odd voltage spikes
across the data lines? Are they isolating
this to avoid noise spikes? My hub does not
breakout whether the power draw is across the
data or the Vdd/Gnd, although I assumed it
only used the latter. But, with isolation
devices in place, it may not be able to tell
the difference. It is possible that my CM15A
device is flaky. I may pop it open to see
(trying to decide if I want to return it,
since I think the SmartHome device is a
better deal once I am forced to buy
SmartMacros to do the simplest things - I had
thought the SDK kit would allow me to program
its Flash/EEPROM, not just control from the PC).
Thanks, Paul

[carmine pacifico]

Paul K

Unless you have a defective unit, the CM15a
does not draw any power from the USB port,
the power pin is unconnected.

[roger1818]

Paul:  Schematics are available on the
site.  Try this link
(https://gullfoss2.fcc.gov/cgi-
bin/ws.exe/prod/oet/forms/reports/Search_For
m.hts?
mode=Edit&form=Exhibits&application_id=47670
6&fcc_id=B4SCM15A) and click on the seventh
link (Exhibit B Schematics per 2 1033 b5).
The ground pin is connected and they have
protection diodes from the data lines to
both VCC and GND.  The power pin (pin 2) is
not connected to anything though.

[martin]

My CM15A now looks stable after all the
tweaking I’ve been doing lately (timers
executing on time, all macros responding to
triggers, all Smart Macros conditions
working). In an attempt to help others,
following is what I have:

AHP 3.186 with Smart Macros, CM15A #05C10.
Windows 2000 Pro on a laptop (moved from
Windows XP).
CM15A not connected to USB (when connected,
RF range drops significantly).
After I do major changes in my AHP
configuration, I always clear the interface
memory before downloading.
24 modules, 64 macros and 11 timers, 80.24%
memory left.
I use Smart Macros conditions “Between
Time”, “Flags” and “Module Status”.

I have three motion sensors set-up as follow:

- 1 ActiveEye, detects motion at all time,
Dawn Dusk detection disabled, transceived
through a TM751.
- 1 EagleEye set-up to detect motion at
nights only, transceived through a TM751.
- 1 EagleEye set-up to detect motion at
nights only, transceived through the CM15A.

My TM751s are included in my AHP
configuration and also act as a regular
appliance module (RF sent from CM15A to
trigger them works fine).
There are still some outstanding bugs, like
RF Range and Macros trigger when
corresponding House Code value is not set to
be transceived, but overall it has behaved
satisfactory.

Hope it helps.

[paul k]

Roger,
Looking at the schematic, I think the problem
may be two-fold: there is no R-PD (15KOhm)
from D- to ground as expected (they do have
the R-PU of 1.3K to VReg (they labelled Vusb
for some reason). This can cause ground plane
problems, especially as they have a lot
moderate current things tied to that ground.
I am also not sure why the diodes to D+/D-
off the isolation transfer from the A/C. This
could cause some current draw given the
signal speed of the USB - I assume it was
meant to stabilize the LVDS pair; I am used
to using 12MHz USB and not low-speed, so I am
not sure how much this affects things. But,
it could cause the sending end to work harder
to overcome the diodes and so pull extra
current.

Regards, Paul

[paul k]

Sorry, the 1.3KOhm pull-up (R-PU) is used to
signal high-speed (I was confused at 1st by
the missing R-PD of 15KOhm as perhaps
signalling low-speed). So, at 12MHz, I would
say that the diodes can have an effect on
current draw. But, I still do not see why the
amounts that I have seen. Ground plane drift
is the most likely reason.

[paul k]

Ugh. I meant full-speed, not high-speed. It
has been a long day :-(