Re: CM11A lockup (broadcast storm)

["Jeff Volp" <JeffVolp@xxxxxxx>]

"Robert Green" <ROBERT_GREEN1963@xxxxxxxxx> wrote in message
news:U7ednSnEVaZ1D1LbnZ2dnUVZ_j-dnZ2d@xxxxxxxxxx
> "Jeff Volp" <JeffVolp@xxxxxxx> wrote in message news:A1Ezi.51864
>
>> > <(more) stuff snipped>
>>
>> A house should only need one XTB-II or XTB-IIR.
>
> However, if you've purchased an XTB-II you could find yourself needing
> both
> if you really needed good repeater capability.  Can the XTB-II be wired to
> a
> single leg of the panel and act as a smart interface to something like the
> HAC 8x8 video matrix switcher that I currently have plugged into a TW-523
> and an XTB-I?

Sure.  The XTB-II works just fine as a high power TW523 when connected to a
single leg.  Actually, it does a bit better because it recognizes sequential
and extended commands, which the TW523 cannot.

>> Yeah, not looking forward to rolling these features over into that other
>> PIC.
>
> Well, *I* am!  But I'm not the one doing the coding! (-:

Lets just say I'm going to take a little vacation from coding after I
finally finish the XTB-IIR.

(deleted earlier discussion on how the XTB-IIR will shut down its
transmitter in response to a command storm)

> I think it's a nice feature for the XTB-IIR to have, but I'd also like to
> have a unit dedicated to nothing but scanning for storms.

You know how hot the CM11A gets when it is stuck in a continuous
transmission mode.  The XTB-II transmits with at least 25 times as much
energy.  So disabling the transmitter during a storm will prevent stressing
and possibly damaging the output driver stage.  That's why I'll eventually
roll that feature over into the XTB-II PIC.

>> Actually, it doesn't matter what the controller does as its output will
>> be
>> inhibited (if it interfaces through the XTB-IIR).  I remember an early
>> post
>> saying you couldn't find the cause of the problem because the XTB
>> transmitted with so much power.  The XTB-IIR will just sit there flashing
>> its LED, and returning line traffic back the digital port.  It will not
>> transmit anything until the line has cleared for 10 seconds.  It should
>> be
>> easy then to go around with a signal monitor to find out where the source
>> of the problem is.
>
> That was a concern of mine originally.  If the XTB stops repeating the
> storm
> it might be possible for the user to believe the problem was squared away
> when in fact the XTB had only stopped repeating the storm.  Having the LED
> flash when it's in this mode is a great idea as it gives immediate visual
> feedback that something's wrong.  If there's ever an XTB-III I would
> suggest
> adding more LED indicators like the ACT repeaters to indicate various
> error
> states like broadcast storms.  That's a personal preference, though, and
> reflects a visceral dislike of one button (or LED) serving multiple
> functions.  I realize that every hole that has to be drilled in the case
> adds quite significantly to the cost of the unit, so perhaps it's cost
> prohibitive on home manufactured items.  Even so,  I'm a big fan of red
> LEDs
> indicate a problem and green ones indicating normal operation.  Now that I
> think about it, you wouldn't have to drill more holes if you used a
> multi-colored LED that can flash either red or green depending upon
> polarity.

I agree with you that more info would be useful.   It would also be nice to
provide DIP switches for the mode programming rather than the 9-8-2-X
sequence.  I had considered a bi-color LED, but sharing the PCB limited what
I could do.  The PIC being used in the XTB-IIR is already fully utilized.  A
larger PIC would have forced tradeoffs I didn't want to make.

FYI, I had thought about a XTB-III (3-phase version for Europe), but I think
this is the end of the line.  I would consider adding some intelligence to
the plug-in version with surface mount components on both sides.  But that
will only happen if someone steps up to the plate to offer it as a
production unit.

(deleted earlier discussion on modifying a lamp module as a storm monitor)

>> Not so simple.  The modules only monitor a single house/unit code.
>> The IC would have to be replaced with one that includes something
>> like the XTB-IIR line monitor, and its storm detector. There may have
>> to be some power supply components too.
>
> I'm confused.  Doesn't the chime module suffer from the same limitations
> (ability to monitor only one house code at a time)?  Perhaps the best
> device
> to modify would be the ESM1, which already lights two diodes in response
> to
> legit X-10 traffic as well as strong noise.  If the LED's stay lit longer
> than 30 seconds, ring the bell.

Yes, the chime IC would have to be replaced with some sort of
microcontroller.  That can be done with a tiny daughter board that would
have pins to solder into the same footprint.  Snip out the IC, unsolder the
pins, and install the daughter board.

However, I think the market for something like that would be too tiny to
pursue.

> I'll bet it's even possible to design a snap-in storage cradle for the
> ESM1
> that has two photodiodes that line up with the built-in LEDs.  No
> electrical
> interconnections would be required at all as the connections are already
> optically isolated.  Even better, it would give me a place to store the
> ESM1
> which I only need rarely now.  I could even set up the cradle with a cheap
> CCTV board cam so that I can monitor and even record the ESM1's readings
> remotely.  That sort of set up could be used as a collision detector as
> well!  If the noise LED lights up too often it would indicate that
> something's colliding or otherwise garbling commands.
>
> I could even use the "round robin" technique I described a while back with
> multiple transformers located around the house feeding into the single
> ESM1
> through a stepper relay and probably locate the storm's cause from any
> video
> monitor in the house.  Best of all, this is all within my limited skill
> set.
> I've designed a number of successful timer circuits using 555 IC's so this
> should be a piece of cake.
>
> Off to the Batcave and the drawing board!

Sounds like you have the solution in hand.

Jeff