Xin Xout in Europe

[RFsolution]

No X-10 users from Europe ??? please help

[ardhuru]

Hi,

I had replied to your earlier post on the same subject ( Xin & Xout Europe (50hz) vs US (60hz) ).

1) Is there any difference in timing or parameters (define) for 60hz or 50Hz operation ? (Europe 50Hz)

No; as long as your transmitter and receiver are on the same electrical phase it would make no difference. If they are on different phases, there would technically be a timing issue. However, in the real world as long as you have signal couplers (0,1 uF caps) accross the phases, it works fine, This is on account of the relatively wide 'listen' window deined by the X10 protocol for receivers.

2) I found in the latest on-line documentation a define: DEFINE XOUTXLAT_ OFF 1 but dont find any additional explication.

I have done dome development work on X10, but not used the XIN / XOUT, so here I'm guessing (vaguely remembering something I'd heard). DEFINE XOUTXLAT_ OFF puts PBP in a mode where it reproduced the raw data coming in from an X10 receiver, as against a house code, unit code, or a command. With the TW523, and for simple switching of loads, do not bother with this define as the defaults work fine.

3) Logic inversion is not required with a TW523 and PBP, but pull-up resistors are. Are you using them?

Regards,

Anand Dhuru

[RFsolution]

I have been looking with a scoop to the timings regarding to the zerocrossing as reference

In a 60hz system timings are as followed:
Bursts are at 120 Khz, 1 millisecond long
1 half periode = 1/120 hz (2x 60hz) = 0,0083333333333333333333333333333333 sec There are 3 burst per zerocrossing Timing between each burst regarding to zerocrossing from 60hz is 2.778 ms,5.556 ms and 8.333 ms

In a 50hz system timings should be as followed:
Bursts are at 120 Khz, 1 millisecond long
1 half periode = 1/100 hz (2x 50hz) = 0,01 sec
There are 3 burst per zerocrossing
Timing between each burst regarding to zerocrossing from 50hz is 3.333 ms,6.666 ms and 10.000 ms

I had the same problem with another US X-10 microcrontroller project and we arranged it this week

So a conclusion is: X-10 Xin/Xout can not work on a 50Hz system
I contacted Melabs to inform them, and I got some immediate support
I hope that they come with a solution soon

[ardhuru]

Hi,
You are right, there are 3 transmissions per zero crossing. The first is almost at the zero crossing, the second and third 2.778 ms and 5.556 ms later (in a 60 Hz environment) respectively. These last two are transmitted solely to cater to receivers which might be on other phases, and which would therefore see their zero crossings 2.7 ms and 5.56 ms later (again as per 60 Hz) .

Therefore, as long as the transmitter and receiver are on the same phase, they experience the zero crossing simultaneously, and the frequency itself is immaterial.

Regards,

Anand Dhuru

[RFsolution]

Well the first takes place 2.778 ms after zerocrossing for 60hz
To get this working properly it needs to be 3.3333 ms for 50hz

So even if you use one phase and the same phase this timing needs to be correct

[ardhuru]

So even if you use one phase and the same phase this timing needs to be correct

No, you're mistaken here; if you look at the graph, you'll notice the first transmission is *at* the zero crossing. The 2nd after 2.778 ms, and the 3rd after 5.556 ms. The next *zero crossing* happens after 8.33 ms (or 10 ms in the case of 50 Hz), but this is irrelevant as no transmission happens at this time. Anyway, trust me, the TW523 does work happily on 50 HZ.

Regards,

Anand.

[RFsolution]

Hi

With some support of Melabs i changed timings in the
PBP lib file and correct them for 50hz

Now X-10 is working

So timing is important, even if you work on a single fase

I now will work on Xin with Melabs

Will post if its work