X10 Community Forum
🖥️ActiveHome Pro => ActiveHome Pro General => Topic started by: burntcrispy on April 05, 2005, 09:12:54 AM
-
I have mine in upside down, and I don't notice any difference than before (doesn't mean there isn't for sure, but if there is it is not big), I liked the unpolarized because some of my outlets are upside down. Just my two cents worth. Just for kicks I might try it the other way again later.
OP:
Regular unit had a polarized plug, Diagnostic unit is NOT polarized, which brings up 2 points...
- /1/ There was a change in some components. Though I am not sure if it was intentional or not, if one phyiscal component was changed maybe others were too.
- /2/ I believe for best performance X10 signals should be polarized. Now due to the size / shape of unit, when plugged into a wall outlet, it would almost certainly be plugged in "correctly" - however my unit is in an extension cord, so I must now pay attention to how I plug it in...
[/size]
Re: Hardware Differences between Reg and Diag. Units [2 of 3] (http://www.x10community.com/forums/index.php?topic=12352.msg30820#msg30820)
Re: Hardware Differences between Reg and Diag. Units [3 of 3] (http://www.x10community.com/forums/index.php?topic=12353.msg30943#msg30943)
[TTA Edit #2: Experiment to determine whether or not DELETING the OP (*AFTER* copying it into Reply #1) in 'Highly-Viewed' / 'Highly-Replied' but currently INACTIVE threads from LONG, LONG AGO (I'm starting with ~2 years) will allow current ACTIVE threads to appear in the Top 10 Topics (by Views) section of the Statistics Center (aka More Stats) (http://www.x10community.com/forums/index.php?action=stats) Page.]
[TTA Edit: Experiment to determine whether or not SPLITTING 'Highly-Viewed' / 'Highly-Replied' but currently INACTIVE threads from LONG, LONG AGO (I'm starting with ~2 years) into parts will allow current ACTIVE threads to appear in the Top 10 Topics (by Replies) section of the Statistics Center (aka More Stats) (http://www.x10community.com/forums/index.php?action=stats) Page.
NOTE: I *WILL* address the <BR> problem...]
-
Arf: I checked the schematic for the CM15a
and the X10 signals travel through an
isolation transformer, so it shouldn’t
matter which way it is plugged in. The
only circuitry that is connected directly
to the hot and neutral lines is the clock
detection circuit (it uses an opto-coupler
for isolation), and even that shouldn’t
care which way it is plugged in.
-
I have 2 diagnostic units. The first one
has the unpolarized plug, the second one has
a polarized plug and some unspecified
hardware change. The second one is my "in
service" controller, the other diagnostic
unit is my activity monitor.
The unpolarized plug is certainly more
convenient when connecting to extension
power strips.
Andy
-
AndyD -
How do oyu know your second diag. unit has
some "unspecified hardware change"?
-
ARF.. Because I was told so by the nice X10
person that sent it to me. (Obviously X10
knew what the hardware changes were but I
was not told).
Andy
-
X10 Pro -
So are there "unspecified hardware changes",
or simply code changes to store and transmit
diagnostic info?
-
I didn't know about the change of the
polarized plug on the factory until it was
reported here, but that's the only true
hardware change I'm aware of. We did have a
couple different versions of the diagnostic
firmware which we sent to people (most of
them forum users) before going with the final
one.
-
X10 Pro: Is it the plan to not use a
polarized plug on all CM15As? I can’t see
any reason for it to have a polarized plug.
-
X10pro:
I see that you mentioned that there were a
couple of versions of the firmware in the
diagnostic units you sent out. How can we
tell if we have the final version and if you
would, can you enlighten the group as to what
those variants did?
Joe (a diagnostic unit user)
-
Roger H: I don't think there's any
significance to the change in the plug. That
kind of thing happens from time to time at
the factory.
Joe: The only versions that were different
were ones that had hand-written labels on the
back. We were adding more data to the
diagnostic output, and it took a couple
versions to get everything we wanted.
-
Fundamental design flaws still exist, no
amount of lemon aid will make the cm15a
reliable.
-
Carmine -
What specific fundamental design flaws are
there?
-
Based on X10 pro input on this thread it
seems that I may have been mislead that my
second diagnostic unit had a hardware change.
I was told "Thanks for your help. I do have
a test version of changed hardware that I
can send that may solve this problem. Should
I send you one?"
I think I should now assume that "hardware"
should have said "firmware" and that the
second unit had the enhanced diagnostic
support.
In any event, that second unit has always
worked ok whereas the first diagnostic unit
didn't.
Andy
-
It needs an independent Hardware Real time
Clock with a Lithium battery back up(the
clock will keep working with power off for
the shelf life of the battery, at least 10
years).
Else the chip used has only 255 byte of
memory with 24 used by the USB buffers, do
the math: how many macros and timer you can
possible have active at the same time
before having a memory overflow?
I feel sorry for the firmware programmer
and tech support, this product will never
be reliable. I have been designing
microprocessor based industrial controller
for over 20 years, I wish I had seen the
cm15 schematic before I purchased, to
late!, $49 down the drain.
-
Carmine -
Not too late. If it doesn't work, and they
can't fix it, they must give you your money
back regardless of the 30 day guarantee
(assuming you are still in the 1 year
warantee period).
However, while I am sure there is some truth
to your analysis, how can you explain that
some units appear to be working fine with
multiple and complex macros, and while
disconnected from PC?
-
Indigo on my date code 05C10 as version 2. I
don't think AHP reports the firmware verion
that I can find anyway.
-
ARF1410
It may work for some, it depends on the
application, I can create macros that will
cause the clock to be overwritten and fail
every time a certain sequence of events
occurs. I have been designing hardware and
firmware for industrial application for
over 20 year. It is impossible to write
code that will never fail, we employ
several techniques to recover and the
instrument keep on working (failure it’s
not acceptable), I would not even attempt
to use a software clock, if the processor
reset (and it will) the clock it’s lost.
Reliable Clock it’s essential for this
application. Based on some test I done if
the cm15 is in the middle of sending
multiple DIM commands, if several motion
detector are triggered it will lock up and
not longer run any macros that depend on
time. If you connect the cm15 to the
computer the clock it’s reset before you
have a chance to se what it was, and in my
case it will work again.
X10 is run by a very smart marketing team,
because the ways the cm15 and smart macros
are bundled it’s not economic to return
them, so most people byte the bullet and
move on; in my case I will use the cm15 as
all house code transceiver, it’s not a
total loss.
-
Carmine -
didn't the CM11A also have a software
clock? It runs essentially flawlessly (up
to the potential of the X10 protocol, which
of course is limited).
-
ARF1410
You are absolutely right, but the cm11a had
only simple macros, and timers(very limited
capabilities).
If I compare the Chip used for the cm11a(
Microchip) versus the cm15( cypress ) the
cm15 is grossly under power for the
application, I believe the only reason they
selected this part was for the build in USB
capability( let’s call it USB for dummies)
and simple RF interface, by the time these
2 function are implemented there is very
little memory left for anything else.
PS they are several way to lock up the
cm11a , in my book , any macros or timer
the customer program should never ever
crash the cm15; If I my instruments where
this unreliable I would be unemployed or in
jail, considering instrument failure could
cost someone their life.
-
ARF: You're right -- the CM15A clock design
is based off of CM11A, and they're pretty
much the same.
-
None
they are both lemons
-
Carmine,
With your years of experience designing
devices like this, what would you say the
price point would be for a controller that
would do what the software can pass to it?
I think that all over the forums, everyone
has been coming to the same conclusion: for
the 50 bucks, X10 can never get the
functionality of a real, fully functional,
fault tolerant controller. When I see
discussions about the other controllers
available, they all seem to have drawbacks.
I'd appreciate your insight.
-
GeorgeP
I design instrument for oil and gas and
recently for other industrial applications.
In our industry price can be an issue, but
we have very stringent regulation that we
must meet so quality and safety comes
before price.
Having said that, if X10 would have opted
for a design similar the one I suggested in
one of the forums” Problem with CM15A as
whole house transceiver” it would
probably add $5 to 10$ to the cost, I
would gladly pay $100 to $200 for a product
that works.
Consider all the effort in attempt to fix
the CM15a how much is costing X10.
In 2002 we faced a similar problem, we had
our new flagship product being certified,
when we tried to add more features
suggested by a large customer we discovered
some serious short coming with Microchip
new flash microcontroller. I suggested to
redesign using a different microcontroller,
by boss reluctantly agreed to allow a
processor change( he thought it would take
many months); a month later I had new
hardware and firmware was converted over.
With new hardware we were able to add
customer features, our sales went trough
the roof and by the end of 2003 our company
had grown 70%; our gross sales probably are
at par with X10, but the company is
privately owned and we do not owe anybody
money.
Listening to customer pays off, after all
they are our real boss.
Our web site http://www.net-safety.com
-
Carmine:
I would also pay more for something that
works.
-
X10 Pro:
Listen to Carmine and listen to those of us
who want you to succeed. We have helped you
fix bugs, we have beta-tested software, and
we want this to work. Carmine - Thanks for
the response, sometimes we have to remember
who the real boss is.
X-10: Maybe you should consider a "power
user" controller? I would consider it if it
did everything we needed.