XTB-II 2-Phase Version

While it would be similar in size to a Leviton repeater, it would still need to get its input from something like the TW523. A outlet on the top would accept the X10 input, just like the present XTB. I thought the cleanest installation would be to mount it adjacent to the panel, and just plug it into a 240V receptacle. Although many of us do it, I believe the code prohibits devices like repeaters from being mounted inside the distribution panel. It was even a little iffy whether the inspector would accept that big X10 filter slipped over the ground lead.

Jeff

Do you have a picture of the case you intent to use?

What about a Power-Brick type case? The kind a powersupply might come in... With a standard power cord input. Then one could use a cord made for 220v, either here in over there... ;-)

Plus the input power connector isn't that deep, hopefully easy to build around, etc. And the case in MY mind's eye already accommodates this by default.

Where would we plug our new RoZetta?

Jeff, like these XTB products of yours, I expect the RoZetta will become a must-have for all HA installations!

A way to get those signals into BOTH or ALL phases will be a God-send.

Thinking and planning ahead?!!

I wanna see the package!

I'll need at least one.

I'll happily Beta test the Kit for you...!

Any idea what the target cost will be for this, both as a full kit and assembled (waiver on file)...?

;-)

Let me know if I can lend a hand.

Sincerely,

Jack :)

Jeff,

I would recommend doing something similar like Advanced Control did for there Coupler/Repeaters. See link to specs. Note they use a standard 4/4 junction box (not supplied). Be advised that a "junction box" has different cover mounting hole alignment then normal 4-11/16 X /4-11/16 electrical boxes.

Let me know when your product is ready for beta. Ill be happy to test for you.

-Dan

The RoZetta would take the place of the TW523 as it accepts the TTL input from the controller and then outputs the desired protocol to the power line. So, unless the XTB did something detrimental to the Insteon or UPB signals it could plug directly into the XTB. The advantage of the RoZetta is that it only puts out the protocol that you specify for a given command. For example, if the controller sends the X10 command A1-AON, the TTL output from the controller goes to RoZetta which looks at the command and and sends the appropriate X10, UPB or Insteon signals. So A1 may be set to send Insteon signals while A2 might send X10.

BTW, Jeff, I'll be happy to beta test and can do my own assembly.

Jack Ed>> Hello All,

Figure about 3" by 5" by 2" high. Probably a Polycase - something like the DC34F or 44F. The exact size would depend on how much the printed circuit board grows. The height will be determined by the transformer.

I'm not familiar with the Power-Brick case.

A single AC jack on the top would accept whatever you wanted to transmit. With more room available, it could also provde over 100 watts.

If it does include a microcontroller, I'll work with Dave to provide a suitable interface.

That is months away, but thanks for the offer.

Since this is apt to be a low quantity item, the PC board will be pretty expensive. Figure about $60 for components. Assembly time will be a bit more than the XTB.

Just to avoid confusion, roZetta will send/receive X-10, Insteon or UPB but _only_ via some type of powerline interface (e.g. 2414S, UPB PIM, CM11A or TW523). It doesn't eliminate the need for an interface and the initial release will require a TW523 or PL513 (or a 12V AC transformer) in order to supply a ZC signal to controllers like JDS, HomeVision, Ocelot, etc.

Jeff will have to answer whether XTB can pass Insteon without distortion. I'll try to test this a bit in a week or two - Jeff sent me a prototype XTB. If it amplifies the Insteon signal, it _may_ exacerbate the Insteon flicker problem. There's also the possibility that it will cause a phase shift in the Insteon signal which could be a problem since Insteon has to be able to clearly distinguish between two distinct phase states of its control signal

- that's what carries the information.

I suspect (but again will defer to Jeff) that XTB & UPB will not get along very well.

Unless there's another skyfall, I will send Bruce the first roZetta late this month.

"BruceR" wrote:

Hey Dave...

How can I get the second one...!?!!

;-)

Jack :)

P.S.

My offer to help with a Logo still stands.

The XTB cannot be used to boost the output of an Insteon transmitter. Testing I did with a couple of Insteon modules produced different results depending on the time of day. Insteon uses a different modulation scheme to obtain the higher bandwidth, and that may be a factor. The XTB was designed to boost X10 signal levels, and it would have been a bonus if it could also boost Insteon transmissions.

I understantd the XTB is being used in several mixed X10 / Insteon installations to combat signal loss caused by multiple Insteon devices. Perhaps those users could report on any incompatibilities they have observed.

A TW523 or CM11A can provide the X10 interface for roZetta. If a 2-phase version of the XTB includes a microcontroller to gate the 1 mS pulses, that microcontroller could provide a X10 line transceiver similar to the TW523.

Jeff,

While I love the integrated couple design, I still would like to see a version that could amplify any X10 signal on the line and not just those from a single unit plugged into it. With signals coming from the CM11A, the ELK panel, and various motion sensors via several receivers, such a design would be much more useful.

Dennis

Jeff Volp wrote:

That function is already covered by several different active coupler/repeaters. SmartHome also makes a plug-in module that works that way, but it doesn't have as much drive as the XTB.

The XTB was really designed to increase the output of a PSC05/TW523. A good passive coupler is still needed when X10 devices are on both phases. Since the optimal location for the XTB is near the distribution panel, offering a version with the coupler built in seemed worth doing.

Jeff

Hello Jeff,

I don't know how the XTB works in detail or even where it can be bought but wouldn't Dennis' situation be remedied if all those units were plugged into the XTB via a power strip? Of course one without any EMI filters.

Other suggestion for a possible XTB-III: The RF transceiver modules contain a receiver that, well, brings on goose pimples for RF guys like me. What if XTB-III contained a "real" receiver with a safe connection socket for a proper antenna? Then it would provide a better RF link and at the same time make sure the amplitudes generated from an RF command would be the same as XTB usually provides.

You have to remember that X10 transmitters are also signal suckers. Both transmitted and received signals on the "shared bus" powerstrip will be attenuated, depending on how many transmitters are used. The configuration you suggest worked fine for one of the beta testers. The number of transmitters that can be shared will be a function how much load is contributed by each one.

Sorry, but the XTB will never include a RF link. But a future XTB may include a direct powerline interface for one of Dave Houstons RF widgets.

Jeff

Hello Jeff,

They wouldn't have to be. If you used a series resonant circuit and a FET as the switch this should not present a serious load while the FET is gated off.

I still think it would be a nice marketability boost.

configuration

The XTB has an output circuit similar to what you describe. But that doesn't help the problem caused by multiple X10 transmitters loading one another on a powerstrip plugged into the XTB X10 input receptacle. The XTB output driver is isolated from its input receptacle at 120KHz.

Jeff

Hello Jeff,

Possibly because the design of the other transmitters is, ahem, sub-optimal? If they'd all tri-state their output stages when not transmitting they should not load down the line. Provided that any output filtering to comply with agency regs is done correctly. A diode T/R switch would be another option but isn't really needed for X10.

My area of work is medical and there we often have a similar task, running pulsers onto an analog bus that is also used for receiving. Microseconds after the pulsers let go they can't even present an ever so slight load because we have to be able to listen all the way down to the noise floor. Else my client's competitors would have them for lunch.