X10 Limitations
Home Up Next

The method X10 uses to send commands from controllers to receivers is a bit like the old 'party-line' telephone system.  In those systems anyone who wants to talk on the line picks up the phone and listens for the dial tone before speaking or dialing, so they don't disturb anyone who is already talking.

In a 'one-way' X-10 system, there is little chance for this happening, because there is never more than one 'talker'.

In a 'two-way' system, you have sensors or wireless remotes which can need to 'talk' at any time. Although they don't 'talk' for very long, and even though more recent units from X-10 do try to be 'polite', there is the possibility that two may try to talk at the exact same time.

This means that sometimes X-10 signals may get lost!

Obviously, the more 'talkers' you have, and the more frequently they do try to 'talk', the more likely it is that you will see lost commands.

So, just as you push a button to turn on a light, someone arrives at the front door and the motion sensor sends a signal. If they occur at precisely same time they will interfere with each other.  Not only will the light fail to turn on, but the motion at the front door will be missed.

This is the "problem" that is expectable with any system which is based on the 'party line' concept. This is why you will hear some people say that "X-10 isn't reliable"

The counter to this is that X-10 is a very adequate solution to the problem that it is supposed to address.

BUT EVERY SOLUTION HAS IT'S LIMITATIONS !

 

The limitations of the X-10 powerline concept are:

bulletSpeed
bulletCollisions
bulletSignal strength

Speed here refers to how long it takes to send a single command on the powerline. The basic rule is that it takes about 6 tenths of a second minimum to send a single command.

For example, if you want to turn on three separate lights which are on different addresses, you will have to send three commands.

It will take 3 times 6/10ths or 1.8 seconds minimum to send the three commands.

This may seem to be a small thing, but in more complex systems, it can amount to a significant amount of time to turn on a group such as "Evening Lights", which might require numerous "Dim"s and "ON" commands.

It may not be personally bothersome that it takes 30 seconds to do this, but it does mean that for 30 seconds, we are 'saturating the powerline' with commands.

 

Collisions are what happens when two talkers talk at the same time.   For example, if the schedule entry for the "Evening Lights" group is happening just as someone walks up to your door and triggers a motion sensor which sends a X-10 ON command, it is possible that the command will cause a collision with one of the commands of the "Evening Lights" sequence.

True collisions of commands on the powerline will most often result in the loss of both commands.

Most motion sensors today attempt to be 'polite' and check the party-line before talking, but even so, it is a real possibility that even with trying to be polite, the motion sensor might try to talk at the same time that we issue a command to our X-10 interface, to "turn on Kitchen Lights"

In this case, we would 'miss' the motion at the front door, and the Kitchen Lights would not turn on.

Of course you will notice the fact that the Kitchen Lights are not on, and can turn it on directly, but what about the person at the front door?

This coincidence of fate can really occur at any time, not only just when long sequences of commands are being issued. Considering the statistical probability of this happening, it is truly insignificant.

But, considering that a motion sensor can trigger at the same time that you issue a command to close the Garage Door (or maybe you are sitting on a wireless controller?), you could end up with the person at the front door robbing everything in the garage.

That is the down-side view.
The up-side is that you can do things with HCA which will improve the probability of recovering from these collisions, and you can create programs and schedules which are 'presumptive', and in the case of the Garage Door, a program might send the command again in one or two minutes, just to make sure.

 

Signal strength is not so much the ability of a 'talker' to send a good strong signal, but rather the 'volume' of the signal when it reaches a 'receiver' on the other side of the house.

There are many evils that can exist between the 'talker' and the 'receiver' !

There are basically five things that affect the ability of X-10 signals to reach all over your house:

bulletProper operation of the transmitter
bulletProper operation of the receiver
bulletElectrical noise from appliances
bulletPoor wiring connections
bulletSeparation of power lines

The proper operation of the transmitter and receiver is of course determined by the quality of production. Because of the tremendous quantity of devices produced by X-10, there is a certain probability that some X-10 devices will fail early in their lifetimes and should be returned for replacement (one year warranty).

Electrical Noise comes from appliances in your home which are improperly produced. There are laws and regulations regarding the amount of noise which can 'legally' be caused by a home appliance, but sometimes this goes undetected in final production.

This noise can sometimes interfere with X-10 signals so much that they cannot be reliably sent all over the house.

Although it may be very aggravating to locate one of these things, there is a simple device from X-10 which can filter the offensive signals and still let you use the 'illegal' device. You simply plug it in between the device and the powerline.   These filters are available from many vendors of home automation gear.  Google for online sales sites for automation gear.

As an aside, it would be a nicer world if more people did discover and complain to manufacturers of devices which are discovered to be 'noisy' in this sense.

Poor wiring connections are notorious for reducing the strength of X-10 signals. Twist-type, push-in, punch-down and screw type terminals are all subject to corrosion and vibration and poor installation.

These kinds of things can often cause 'intermittent' behavior, and are most aggravating!

There have been reports of problems which disappeared when a sofa was moved from one side of the room to another! The solution was that whenever someone sat on the sofa, the stress was enough to physically disturb the wiring connections at the socket behind the sofa that it really stopped signals reaching the 'aggravating lamp'! The problem was solved by simply turning off the circuit breaker, and tightening the screws on the wires going to the socket.

Separation of power lines is especially a problem with larger homes, and is an intrinsic problem with all electrical systems in the USA, and in countries which have 'multi-phase' power.

In 110 volt systems, your power is delivered as 220 volts, 2-phase, and this is 'split' into two 'legs' or 'phases' at the service entry point for the house.

This is somewhat difficult to explain without a picture, so let's refer back to an old familiar image. If you live in the USA, you have this kind of electric service:

Twophase.gif (7806 bytes)
Here we have the 'interface'
plugged into 'Leg A'


and two lamp modules
each connected to a
different 'LEG' of your power.


and we have a
"Electric Oven"
which is connected
to both of
the 'legs' of your power.

In the worst case, and even in some small homes, commands which are sent from HCA to the powerline interface, are 'seen' by all receivers on Leg 'A', but may not be seen by devices which are on Leg 'B' !

If you've every heard the 'X-10 wisdom' which says:  If you turn on your electric oven, X-10 works better !
then you can look at this drawing and see why. The electric oven is connected to both of the 'legs' of your power, and therefore acts as a 'bridge' between both 'legs' of your electric service.

 

This is the last straw !
Right again! If you do have a larger home or if you are having problems getting simple commands sent across the room, then this is the last thing to consider!

There are several different solutions, each with its improvement in signal condition, and of course an attendant increase in cost.

There are three kinds of X-10 signal bridges: those that simply help the signals 'pass' from one 'leg' to the other, and those that actually 'reproduce' these signals.

The different devices are called :

bullet'couplers'
bullet'repeaters'
bullet'repeater-amplifiers'.

Couplers simply 'pass' X-10 signals from one 'leg' to another.

Repeaters actually receive X-10 commands from one 'leg' and re-transmit them to the other 'leg' automatically. Thus, the signals are at least 'boosted' to their original strength.

Repeater-Amplifiers not only repeat the signals, but they also increase the 'strength' of them. This is probably the best solution that you can buy.

There is also a device which repeats, amplifies, and "detects multiple repeaters". This is only necessary when you have a VERY large home, and you have multiple power distribution panels, and multiple "repeaters".

If you think you may need one of these type of devices, contact one of the X10 vendors and they will help you determine what you may need.

 

So, what have we learned ?

bulletSome things 'take time' to happen
bulletSome things can happen at the same time
bulletX-10 signals can be disturbed by common appliances.
bulletLimitations of the X-10 system are known and 'correctable'

There is no known system in the world which is perfect without the inclusion of a knowledgeable and caring person to occasionally check and verify that everything is performing as expected. If you ignore it, you will be 'disappointed'.

Just the same as if you ignore the little drip underneath your kitchen sink !

 

Much of the information in this article was written by Sand Hill Engineering, the maker of a Home Automation application for the Macintosh, and used with their permission.