Print

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

[Dave4720]

I own two houses next to each other.  I have already installed an underground conduit to run phone and network to “connect” houses.  I have additional twisted lines not used.

I’d like to bridge the two houses together for X10.

Do I understand correctly that the TW523 pulls X10 codes from the AC line and puts them on the TX terminals, and X10 codes presented to the RX terminals will be transmitted on to the AC line?

If so, can I “convert” AC X10 to baseband X10 with one TW523, pipe it along twisted pair to the other house, then “convert” baseband X10 back to AC X10, therefore “connecting” both houses?

[Brian H]

The data is in raw digital codes. You have to use a computer or microcontroller to make sense of the signals. It also has a Zero Crossing signal used to synchronize the data.
They also do not process the some codes correctly. Dims if memory serve me you get every second one on the output. Extended codes also are not done correctly.

[dave w]

I own two houses next to each other.  I have already installed an underground conduit to run phone and network to “connect” houses.  I have additional twisted lines not used.

I’d like to bridge the two houses together for X10.

Do I understand correctly that the TW523 pulls X10 codes from the AC line and puts them on the TX terminals, and X10 codes presented to the RX terminals will be transmitted on to the AC line?

If so, can I “convert” AC X10 to baseband X10 with one TW523, pipe it along twisted pair to the other house, then “convert” baseband X10 back to AC X10, therefore “connecting” both houses?

If both homes are on the same "pole" transformer you may be able to just have coupler/amps present at each home. Also, you might check with Jeff Volp, but I believe two of his XTB IIR's can be connected togeather through their RJxx connectors. (Jeff are you out there?).

[JeffVolp]

In addition to the issues regarding bright/dim commands and extended codes, there are a couple of other problems with the proposed technique.

The TW523 only reads back the second half of the X10 command.  So the signal relayed to the second house would only contain the second half of each transmission.  While that should still work, it does eliminate the redundancy built into every command to deal with switching transients and other short-term events.

The X10 spec says the TW523 data output should be asserted within 200uS after the zero crossing detect.  However, the data input should be asserted within 50uS after the zero crossing detect.  So, the data pulse fed to the second TW523 could be 150uS late.  If the source of the signal is a controller that relies on the zero crossing, and delay times are near the maximum limit, the extra 150uS will result in the second TW523 not beginning its transmission before the "X10 reception window".  There is sufficient margin built into the protocol that this should still work, but there is little room left for timing errors.

You would not want to couple the zero crossing detector outputs between the two TW523 units.

Jeff

[JeffVolp]

If both homes are on the same "pole" transformer you may be able to just have coupler/amps present at each home. Also, you might check with Jeff Volp, but I believe two of his XTB IIR's can be connected together through their RJxx connectors. (Jeff are you out there?).

Yes, a repeater at both homes might do the job.  ACT also has a repeater that provides a digital link so all units in a multiple repeater installation transmit in sync.  If you have a high-end controller that interfaces through a TW523, you could "Y" the digital connection to directly feed a TW523 in each home.  You would only use the zero crossing detector from one home.  Since the TW523 data output is open-collector, the wire-OR of both outputs would return local X10 signals from each home.  Obviously, if two remote X10 commands overlap, the resulting collision would cause the loss of both commands.

Since the XTB-IIR emulates a TW523 (except it DOES handle concatenated dims and extended commands), it will have the same timing issues as I described earlier for the TW523.

Jeff