Print

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

[doug]

I have computers running on a UPS.  I would
like to be able to "boot" the computers
using X10 from AHP (on a different comuter)
as well at the TR551 (telephone
controller), but leaving power to the UPS.

Using the telephone controller on the
battery side of the UPS works fine, but how
can I "jump" the X10 signal through the UPS
from the normal house current side?

Can a phase coupler be used?

Doug

[dave w]

Good question, and it may be totally
dependent on the design of the UPS. In
theory a passive phase coupler should work
if the UPS is simply "not passing" X10" from
input to output. Since you can get X10 to
work at the output of the UPS and
(presumably) on the input side of the UPS, a
phase coupler should work. An alternative is
to plug your "boot" computer into a TM751
and use your CM15A and AHP to send the ON
command to the TM751 by RF instead of PLC.

[roger1818]

Dave’s idea of using a TM751 is probably the
easiest option.

Assuming the PC has an ATX power supply,
another option would be to use a universal
module in momentary mode.  You could then
wire the contacts in parallel with the PC’s
power switch.  Obviously the Universal Module
couldn’t be plugged into the UPS.

If you use Dave’s phase coupler idea, one
other thing you need to consider is that the
UPS probably has a noise trap which will
absorb X10 signals on both ends so you would
probably need to have X10 noise filters
between the UPS and both ends of the phase
coupler (ie. Phase Coupler side A – Filter –
UPS – Filter – Phase Coupler side B).

[Brian H]

I second Rogers thoughts. Most UPS units
have line filters and absorbe X10 signals.
My APC BX1000 killed my X10 signals untill I
used an X10 Type Filter on the AC Input.

[luiz]

Just came across this thread. I've been
experimenting having a TM751 connected to an
APC UPS so I could control some minor loads
(siren for an alarm system directly connected
to the transceiver) even in case of power
failure. Note: the transceiver signal is not
expected to "jump" from the UPS side to the
general house side. But what I found is
puzzling: when it try to access (wireless
remote control KR22A) the transceiver itself
(e.g. K1 with the transceiver house code set
to K), it doesn't work when power is from the
UPS alone (simulating a power failure), only
when the UPS is operating in normal
condition!!! Is there any different between
the power (e.g. not a perfect sin) provided
by the UPS as opposed to being provided by
the house?

[roger1818]

Luiz:  That is interesting.  I wonder if the
UPS is generating a small amount of RF noise
when in battery backup mode, and is
preventing the TM751 from hearing the RF
commands.  Have you tried plugging the TM751
into an extension cord so that you can keep
it away from the UPS?

Are you planning on controlling other X10
modules into the UPS?  If so you will
probably want to have an X10 noise filter
between the UPS and all the X10 modules to
ensure the UPS doesn’t absorb the powerline
signal.

[dave w]

Luiz
Your 01/10/06 post clarifies a few things
and the answer to your final question is a
huge yes.
A UPS uses an inverter to convert a low DC
voltage (battery) to high AC voltage (120V
AC). There are two major designs of
inverters to do this: "True Sine Wave"
and "Modified Sine Wave". The cost of the
UPS indicates which inverter design is used
and 90% of the UPS designed to sustain a
computer for a few "shut down" minutes
during a power outage use the modified sine
design. The modified sine inverter has a
sine wave output that is flattened on top. I
don't know how much this might effect the
X10 signal since X10 signal is at the zero
cross point, at the bottom of the sine wave.

If your inverter has high "cross over"
distortion the distortion will be near the
zero cross point, obliterating the X10
signal. If your UPS is designed to run a
computer for extended periods of time of
power outage, it should have a pretty clean
output.  In other words, you probably can't
get there from here using your current UPS
(a guess only).

[roger1818]

Dave:  The waveform of the UPS shouldn’t
affect the transceivers ability to control
the “Appliance Module” internal to the TM751.
It will affect its ability to transmit that
signal to other modules though.

Don’t forget that any waveform can be
represented by a combination of sign waves at
different frequencies.  By filtering out
those frequencies, you should be able to get
something closer to a “True Sine Wave.”  The
amount of filtering required varies depending
on how close the “Modified Sine Wave” is to a
“True Sine Wave” to begin with.  By putting
an X10 noise filter between the UPS and the
X10 modules you should be able to get a
signal that is clean enough to be useable
(below 50mV or 34dBmV should be good enough
in this case).  Since most X10 filters reduce
the noise by 40dB, if the noise is less than
5V, you should be able to do this with one
filter.  Two filters will get rid if 500V of
noise (obviously you should never have that
much noise).

[luiz]

Well, we have two problems (just summarizing
what Roger posted):
1) The expectation that the TM751 should be
able to control its internal "appliance
module" when it receives a wireless command
for itself no matter what as long as it’s
powered.
2) X10 signals being "corrupted" (sorry,
Computer Science guy here) over the UPS side
of the power line when working on batteries.

My very limited knowledge on the hardware
principles involved prevents me of adding any
meaningful explanation for 2) but I came
across a bunch of other posts where this
behaviour has been observed (not explained
tough).
At this point, my real problem is 1). The
power output of the TM751 itself will control
some small loads which I would like to work
even in case of power failures (part of an
alarm system, if you ask). And Roger, I tried
an extension cord (1 meter) and no changes. I
even tried a second TM751, same thing. In the
exact same configuration, connected at the
exact same outlets, UPS side, power from
battery doesn’t work, power from house works.

[Brian H]

#2 When on battery and the inverter is
running. The X10 Line Signals probably can't
get back through the inverters transformer
or is being swamped out with noise around
the X10 line frequency.

#1 The power supply in a TM751 is line
derived and it could be the modified sine
wave is effecting it. I see you tried an
extension code to move it away from the UPS.
Have you tried it on an X10 Line Filter.
Since you can't get the signal back through
the UPS anyway. Supressing the signal isn't
going to make much difference and it may
stop noise from getting into the TM751 even
though it has no line receiver in it. I did
see another post where the user was using an
off grid battery and modified sine wave
inverter. The waveform overloaded a test
TM751 and burned it out.
I tried one of my TM751s on an older APC UPS
and it switched on and off once when on
battery then it looked like when the relay
triped on or off it sent my UPS into power
down. Maybe from a noise spike that the UPS
though was an overvoltage or the things
battery maybe on its last legs.

[luiz]

Nope, haven't tried a filter. Don't have one.
So far things have working fine without them.
My APC is a brand new BE650BB-CN (650VA). Any
other way you can think of that would
workaround this problem? This is a long shot
but would different loads connected to the
TM751 make any difference?

[luiz]

Update: the RR501 works!

[Noam]

The TM751's internal Appliance module
doens't respond to powerline commands, only
to RF commands. Have you tried turning the
TM751 on manually (with the button on the
front), or with an RF remote control?
The RR501's internal appliance module DOES
respond to powerline commands.

[luiz]

Yes, I tried manually turning on/off using
that button on the front. It didn't work either!

[Brian H]

I will try one on my newer APC BX1000 and
see if it makes a difference. RR501 has
maybe a better power supply. Sure sounds
like the TM751s power supply had a problem
as even the push button was unresponsive.