Re: XTB-II Enhanced Repeater

["Jeff Volp" <JeffVolp@xxxxxxx>]

Thank you for the detailed and thoughtful reply.  Just a few things to add:

Updating the XTB-II firmware only requires turning off power, wiggling out
the 8-pin PIC that is already in there, and plugging in a new device, making
sure it is oriented the same way.  I believe all early units were updated to
1.10.  Two additional mods were made - one to prevent repeater ping-pong,
and one to correct a bug in 3-phase transmission.  The firmware has been
stable at 1.12 since early this year.

I agree with you on the "polite" mode.  It will probably be useful in
installations with motion detectors and a TM751 transceiver.  That unit is
not polite, and will transmit whenever it receives a signal from a motion
detector.  That would cause a collision if the XTB-IIR is repeating at that
time.  If so enabled, the XTB-IIR would immediately terminate the
transmission, and re-transmit the repeat after the line has cleared.  A
receiving module would not respond to the XTB-IIR command if it also saw the
bits from the other transmitter.  So, the delayed re-transmit gives a chance
to get that module to respond properly.  Since it is more likely that an
installation will not need the polite mode, I have changed its default to
OFF.  Both the abort on collision and auto re-transmit can be enabled
through mode options.

While it is easy to count error conditions, providing that information in a
useful format is not.  A whole housecode could be dedicated to diagnostics,
and a two digit code issued in response to a request.  But, would anyone
would really use that kind of information?  It gets back to the question is
it worth the development time.

Another issue you had brought up was relative signal strength.  I had
actually looked at that myself.  It might be possible to make signal
strength measurements.  However, the issue is again how to make this
information available, and whether it would really be useful.

Well, back to the soldering iron...

Jeff

"Robert Green" <ROBERT_GREEN1963@xxxxxxxxx> wrote in message
news:o4OdnaSkZ5flQ37bnZ2dnUVZ_s-pnZ2d@xxxxxxxxxx
> "Jeff Volp" <JeffVolp@xxxxxxx> wrote in message news:iRzEi.77842
>
>> >> I've been trying to make this the best X10 repeater available.
>> >
>> > Don't take this the wrong way but as an owner of the XTB-II that's not
>> > going to be upgradeable to the XTB-IIR the subject of the world's best
>> > repeater is primarily an academic one for me now.  On the other hand,
>> > my interest in house-wide, real-time X-10 signal strength readings and
>> > broadcast storm detection are quite high on my list, hence my
>> > discussion
>> > of the details of  that project in the "broadcast storm" thread.
>>
>> While there are certainly some features that cannot be ported over to the
>> 8-pin PIC, things like enhanced error detection, abort transmit on
>> collision, and command storm shutdown can be.  It may also be possible to
>> add the smart bright/dim repeat for sequential commands.  The real
> question
>> is whether it is worth the effort if only a few people are interested in
>> a
>> firmware upgrade.
>
> It depends on a lot of things.  The most important, to me at least, is
> whether the upgrade is simple to perform and bulletproof.  Even if you
> volunteered to do it for a fee, the XTB-II is a critical component to be
> missing from a house for very long.  You've got a wide range of customers
> from those with no electronics experience to whose who could reprogram the
> PIC without blinking.  That's a factor that will determine at least some
> of
> the upgrade decisions.  Based on my experience with PC firmware, upgrading
> is often an iffy proposition.
>
> Then there's the issue of whether they'll need those features.  Polite
> mode
> (abort on collision) is nice, but it could be a liability as well as an
> asset.  If it fixes the rare cases where two transmitters send an X-10
> command simultaneously, it's an asset.  If it however fixes a problem that
> shouldn't be happening in the first place, it may be a liability in terms
> of
> discovering and fixing the colliders, at least for me.  I'd want to know
> if
> the repeater was being polite more than a few times a day.
>
> (You've alerted me to the need to add such detection capabilities to the
> BSD
> cradle.  I need to monitor the "X-10 Good" LED and the two or more
> segments
> of the ESM1's bargraph LED for collisions and send a message, light an LED
> or take some other action to alert me that more than a handful of
> collisions
> a day are occurring.)
>
> If you're still looking for features to add to the XTB-IIR, a way to read
> the incoming signal strength and to alert the user to an unusual number of
> collisions would be nice to have for most people, although perhaps not for
> me.  I am afraid that the BSD would probably conflict with the XTB-IIR
> because the IIR might detect and shut down the broadcast storm before the
> BSD ever had a chance to detect the storm and sound an alarm.  It would
> depend on whose time limit was set shorter.  I believe you've allowed for
> different broadcast storm lengths in your device, so if my timing and
> yours
> are both variable, that should be a problem that's easy to work around.
>
> I also realized that detecting a broadcast storm has some interesting
> issues
> associated with it, namely making sure that a legitimately long string of
> X-10 commands doesn't get misinterpreted as a broadcast storm.  In my
> case,
> I'll have to make assumptions based on the way the ESM1 parses the X-10
> command.  It's only after the bargraph LEDs go out that the "X-10 Good"
> detector lights.  I'm working on setting up my Sanyo CCTV recorder to film
> the ESM1 display as it reacts to X-10 traffic.  Since it allows for very
> high speed capture, with frame by frame viewing and millisecond time
> stamping, I should be able to make very good estimates of how long each
> LED
> stays lit and the exact on and off times.  It's probably overkill, but
> I've
> found that examining reams and reams of raw data can often point out the
> nasty exception cases that often prove quite troublesome to resolve after
> the design is complete.
>
> Sorry.  I'm hijacking your thread again.  (-:  It's hard to avoid because
> discussing the XTB's functions naturally links up in my brain to the
> sketching out the full spectrum of the BSDs capabilities that I've been
> contemplating. I want to try to get it right in "wetware" before I go
> soldering any wires.  (While writing this, I had the "ah ha" experience
> about how I need to set up the TV remotes which I'll describe in another
> thread.)  Even this tiny project has given me a greater respect for all
> the
> time and effort that's gone into building the XTBs.  There must have been
> an
> incredible number of decisions you had to make regarding component size,
> cost, operation, interactions, etc.  The BSD universe is quite tiny by
> comparison and yet it is still perplexing.  Every time I think I'll do X,
> either some design constraint or unexpected behavior means I have to do Y.
> Which leads us right back to your quandary:  To upgrade or not upgrade the
> XTB-II?
>
> Since most of the transmitters I use now for extemporaneous commands
> (ControlLinc Maxi's) have polite modes, it's probably not terribly
> important
> to have it in the XTB-II.  But it would be nice to have for the few
> devices
> that still collide.  If I need the IIR, I'll just remove the II and use it
> as a standalone device.  It's not as if the newer IIR obsoletes the
> original
> unit into worthlessness.  So far, the HAC-8X8 switcher works fine with the
> TW-523 plugged into the XTB-I outlet so perhaps when I finally get the
> CPU-XA doing something useful again I will use the II as its powerline
> interface.
>
> Getting back to original question of who wants the upgrade, you're the
> only
> one that's got the XTB customer list, so if you're losing sleep over the
> upgrade you know who to ask!  I realize that it's the kind of question
> that
> you might not want to ask because the (legitimate) fear is that everyone
> will answer "yes I want it" and then never follow through.  So I'll quash
> those expectations here and now by just saying "no."  (-:
>
> If I need the better repeater, I'll buy it from you because I know you're
> not buying a fleet of huge flatscreen TV's based on the profits you've
> made
> on this venture.  I shudder to think how small a TV fleet you could
> actually
> buy with XTB profits.  When I look at the cost of switching to Insteon or
> Z-wave, which I was seriously considering before the XTB arrived, it's
> still
> quite a deal even if I have to buy another repeater.  I just snagged a
> whole
> bunch of X-10 gear from Ebay NIB at 1/4 what I used to pay from
> Worthington
> so I'm not unhappy about passing that savings on to you by buying the IIR.
> If I didn't have the XTB's, I would be paying a lot more per controlled
> load
> than $5 and I would have perhaps 1/4 the overall capability and choice of
> components.
>
> The XTB-II's had their purpose.  It's reasonable to make people buy the
> IIR
> if they want a super-duper, "handles all cases" repeater.  There will be
> at
> least a few grumblers out there that won't agree, but if I were you I
> would
> consider the cost of porting the IIR code to that of a quiet trade-up
> program for any extremely noisy and unhappy XTB-II owners.  Not sure what
> you can do with the returned XTB-II's, but I am sure that over the years
> at
> least a few will be fried by direct lightning strikes, home fires, etc.
> I'll bet you could do the trade and sell the used ones cheap on Ebay and
> still sort of break even.  It all depends on the value you place on your
> own
> time.  You couldn't get me to do much of anything for $1 an hour.
>
> So the bottom line is, no, I don't think it would be worth the effort to
> port the code.  Even though I would like to have it, there are good
> business
> reasons not to, mainly because your time is probably better spent
> perfecting
> the XTB-IIR.  If there were no other uses for the retired XTB-II, I would
> feel differently, but there are.  Also, since you inspired me (and led me
> to
> a great vendor, Fruit Ridge Tools*, for the parts) to use a twist-lock
> 220VAC outlet to mount the XTB-II, there's very little switchover cost
> involved between the two devices.  FWIW:
>
> * "Fruit Ridge Tools specializes in electrical surplus tools and supplies.
> Offering a wide variety of switches, dimmers, plugs, wallplates, etc"
> Also
> has a store on Ebay.
>
> www.fruitridgetools.com/
>
> (No commercial interest or kickbacks involved, just a fairly priced and
> very
> helpful merchant.)
>
> --
> Bobby G.