Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[JeffVolp]

1. Not sure I understand why the Leviton Repeater will not "couple the strong XTB signal to the second phase". Isn't that what it is supposed to do?

Nope.  A repeater does exactly what the name says.  X10 signals are normally transmitted in two identical halves.  A repeater receives the X10 signal during the first half, and then re-transmits it using its own internal transmitter in sync with the second half.  The strength of the signal on the opposite phase is determined by the output capability of the repeater itself.

A good tuned circuit passive coupler provides a low-impedance path between the phases at the X10 carrier frequency.  So the strong XTB output is coupled directly across the phases.

Unless the XTB is combined with a good tuned-circuit passive repeater, only the phase that the XTB is plugged into will receive the stronger signal.

2. Can you help me understand the "macro" thing? I have built some in the past but they really do not work right and when you say "recognize a signal from the RR501", that really looses me!

Think of a macro as a "mini program" that is initiated by some event.  For example:

If receive A1 ON then delay 5 sec and then transmit A2 ON.

I haven't used the CM11A in years, but I recall they were built graphically.

I don't want the CM11A that far away from my PC right now because interfacing with it is a real pain.

That can be an issue for many people.  If you want the XTB to drive both phases, you will need a good passive coupler at the panel.

I just don't understand why the difference? I would think that I would see all the X10 commands or none, but not specific commands such as TIMED "ON" codes because the manually sent "ON" codes from the CM11A are responded to.

That's one I don't understand either.  One thought was that the timed events were not being transmitted properly.  However, I recall your stating that other modules DID respond to those ON codes.  So the only thing that is left is a signal or noise issue.  Checking the signal strength at those modules should give some insight into what is happening.

Jeff

[BSarte]

2. Can you help me understand the "macro" thing? I have built some in the past but they really do not work right and when you say "recognize a signal from the RR501", that really looses me!

Think of a macro as a "mini program" that is initiated by some event.  For example:

If receive A1 ON then delay 5 sec and then transmit A2 ON.

I haven't used the CM11A in years, but I recall they were built graphically.

I'll give it try and see what happens.

I just don't understand why the difference? I would think that I would see all the X10 commands or none, but not specific commands such as TIMED "ON" codes because the manually sent "ON" codes from the CM11A are responded to.

That's one I don't understand either.  One thought was that the timed events were not being transmitted properly.  However, I recall your stating that other modules DID respond to those ON codes.  So the only thing that is left is a signal or noise issue.  Checking the signal strength at those modules should give some insight into what is happening.

Jeff

Yes, the real test here is that there are 5 LM465's that are set to code 7. Four of them are on other legs/circuits of the electrical wiring from the circuit panel and work 100% of the time responding to both "ON' and "OFF" codes. It is just this one particular set of wires (2 separate circuits each having its own distinct circuit breaker on separate phases of the electric and each having a differently coded LM465 on it) going out of the house for about 85 feet into an underground box which do not respond to timed "ON" codes, but do respond 100% of the time to timed "OFF" codes.

I also have a detached garage that has #2 wire running from the house circuit panel back to another circuit panel in it. The distance is minimum of 375' from one panel to the other. I put the ELK-ESM1 in an outlet in the detached garage today and had my wife use the handheld remote to send commands through the RR501 up at the house. The signal is 2 bars on the scale which would translate to about .5 volts. A little low but it is enough to control a Leviton Wall Switch (same at the WS467), so it looks like there is plenty of signal going quit a distance around here! The X10 green light also light, so I guess we can say the command is formatted correctly as well.

When I tested the signal strength at various outlets around the house, the signal strength from the handheld remote through the RR510 averaged just over 1.0 volts and again, a green X10 light.

I'll have to dig up the out door box and test that tomorrow.

Lets see what I come up with.....

I'll also try the macro to send out continous timed commands every 5-10 seconds and see what I get. I guess I'll use a code that is not being used on any of the modules so that way it will simulate but not actually do anything other than test the circuit for noise and good X10 command format.

Bruce

[JeffVolp]

As I recall, those "problem" modules did work when commanded via your RR501.  It would be interesting to see what signal levels you read coming directly from the CM11A/XTB combo.  Since the CM11A is not as easy to trigger as the RR501 unless you build a macro, you might just move the RR501 to the socket that your CM11A is now plugged into.  Then test signal levels again - perhaps with and without the XTB boosting the RR501 output.  I wonder what the signal level is at those "problem" modules when the source is the CM11A.  If the signal level doesn't change with the XTB, then they are on the opposite phase.

Jeff

[Boiler]

I also have a detached garage that has #2 wire running from the house circuit panel back to another circuit panel in it. The distance is minimum of 375' from one panel to the other. I put the ELK-ESM1 in an outlet in the detached garage today and had my wife use the handheld remote to send commands through the RR501 up at the house. The signal is 2 bars on the scale which would translate to about .5 volts. A little low but it is enough to control a Leviton Wall Switch (same at the WS467), so it looks like there is plenty of signal going quit a distance around here! The X10 green light also light, so I guess we can say the command is formatted correctly as well.

Bruce,
Be careful making the assumption that because the Leviton switch functioned your LM465 should function as well.  Depending on the model, your Leviton switch may have AGC and would be more reliable when dealing with low signal levels.

Another thought on the manual on vs timed on events - I've noticed in the past that the CM15a produces "malformed" power line commands when using timed events and macro's.  When manually activating (graphic control) the commands are clean. 

The following is what my testerlinc reports for a simple macro executed On command:

Macro E6 On
E6
BCY (bad 3 cycle gap)
E6 On

The above doesn't appear to affect my modules (they all respond).  I'm wondering whether it may further complicate things in a noisy or low signal environment.

Oldtimers, any theories out there?

[BSarte]

Another thought on the manual on vs timed on events - I've noticed in the past that the CM15a produces "malformed" power line commands when using timed events and macro's.  When manually activating (graphic control) the commands are clean. 

The following is what my testerlinc reports for a simple macro executed On command:

Macro E6 On
E6
BCY (bad 3 cycle gap)
E6 On

The above doesn't appear to affect my modules (they all respond).  I'm wondering whether it may further complicate things in a noisy or low signal environment.

Oldtimers, any theories out there?

Thanks for the input!

I use a CM11A, not the CM15A. Does the same distortion of commands happen from the CM11A? If it does, then why are all the other LM465's with the same code (7) responding fine? Kinda a rhetorical question. There has to be something on those two circuits distorting the "ON" commands.

As soon as I dig up the underground box and test, I'll have better info but I have to build the macro to send out continous commands first....so maybe today, but maybe not with family coming over...

Bruce

[Boiler]

Bruce,

I think I remember seeing posts complaining about the same "malformed" transmissions with the CM11a.  My CM11a has been stored for about 4 years now.  Maybe someone else can verify this?

At this point this is just an observation.  I can't think of anything that a "bad 3 cycle gap" would do to affect the turn on of a receiver.  Let's table this one for the moment (probably shouldn't have brought it up - it's a distraction).

Here's a better scenario -

• With your "problem" LM465 off you have a long weekly terminated transmission line with a fair amount of X10 signal loss.  Under these conditions your LM465 can't reliably detect the ON command.
• When you do manage to activate the LM465 your line becomes terminated through the load resistance.  Your LM465 is now able to reliably detect the OFF commands
• Your RF commands work reliably because your RR501 has a higher output level than your CM11a
• Your manual CM11a commands work because?? This is where I was theorizing about the "malformed" timer commands

If you wanted to test the above, you could plug another load into your underground box.  Your LM465 should turn on reliably with the second load activated.  This obviously won't solve anything.  But it may allow you to sleep better knowing it's not all smoke and mirrors.

Bottom line is that I believe that you and Jeff are on the correct track.  The XTB should boost your CM11a output and allow you to reliably communicate with the problem LM465 (as your RR501 does now).

This is an intriguing problem - please keep posting your results.

[BSarte]

Bruce,

I think I remember seeing posts complaining about the same "malformed" transmissions with the CM11a.  My CM11a has been stored for about 4 years now.  Maybe someone else can verify this?

What other unit can I use if the CM11A is the culprit? CM15A?

At this point this is just an observation.  I can't think of anything that a "bad 3 cycle gap" would do to affect the turn on of a receiver.  Let's table this one for the moment (probably shouldn't have brought it up - it's a distraction).

Here's a better scenario -

• With your "problem" LM465 off you have a long weekly terminated transmission line with a fair amount of X10 signal loss.  Under these conditions your LM465 can't reliably detect the ON command.
• When you do manage to activate the LM465 your line becomes terminated through the load resistance.  Your LM465 is now able to reliably detect the OFF commands
• Your RF commands work reliably because your RR501 has a higher output level than your CM11a
• Your manual CM11a commands work because?? This is where I was theorizing about the "malformed" timer commands

I tested the transmission levels to both circuits in the underground box, I'll elaborate more farther down, but on the ELK-ESM1 meter, I see a good X10 command and levels at about 1.1 volts for the RR501 commands, the CM11A timed & macro commands, and the CM11A direct commands.

If you wanted to test the above, you could plug another load into your underground box.  Your LM465 should turn on reliably with the second load activated.  This obviously won't solve anything.  But it may allow you to sleep better knowing it's not all smoke and mirrors.

When I tested the signal format & levels, there was the transformer for the ELK-ESM1 installed which I would assume should be enough load?

This is an intriguing problem - please keep posting your results.

'intriguing' is not quite the word I would use right now!

Well, I opened up the underground box today. Let me explani what is in there coming from the circut panel. from the circuit panel I have a #12 3 wire with ground cable terminated in a quad box with 1 (black) wire feeding one duplex outlet and 1 wire (red) feeding another duplex in that quad receptacle with the white and ground wires common between the two duplex receptacles, thus giving me two separate 110 VAC outlets. In the circuit panel the black & red wires are terminated in separate circuit breakers stacked one on top of the other thus giving me power from both phases of the electric coming in.

Now, there are 4 LM465's in that box with codes 4, 7, 14, & 15. 4 is not a timed only used when we want to turn those lights on manually. 7, 14, & 15 are all timed events from the CM11A with 7 & 15 turning on just after dusk and going off at 11 p.m.. 14 is timed to go on at 6 a.m. and shut off at 6 p.m. through the CM11A timed event(s).

Now what I did was to plug the transformer into one of the outlets (removing module with code 4) so I could watch the commands and signal levels. Before that I created a macro (code 2) to turn on codes 7 & 15, then 1 minute later turn off codes 7 & 15, then one minute later turn on codes 7 & 15, then one minute later turn off 7 & 15.

When I executed the 2 command from the hand held remote, I saw signal levels at one red bar over the 1.1 volt scale which should put the level at about 1.3-1.4 volts and the green X10 light went on. Each succesive command from the macro was about the same but the LM465's did NOT go on. When I simply executed the on command from the hand held remote (which executes via the RR501), the levels were not any stronger and the green X10 light went on AND THE LM465's WENT ON!

One thing I noticed was that the simple command via the RR501 appeared to be different. The signal bar showed several ( 3-4) pegs about 1/2 second apart while the macro commands only showed two very quick pegs to the signal level.

Now lets go back to the way I have these circuits wired up. Since I am using the white wire as a common between the two circuits, could the X10 commands be "colliding" and canceling each other?

All of the above tests were done without the XTB in place. One other test I did was to put the XTB in place in front of the CM11A and test the signal levels. One side of the duplex outlet showed a very marginal increase in signal level, maybe up to 1.4-1.5 volts.

I put everything back to together and quit for the day. Got family coming over and the ribs need to be marinated, cold beer needs to be drunk, and forget this whole 'intriguing' issue for now.

Feedback please!

Bruce

[JeffVolp]

When I tested the signal format & levels, there was the transformer for the ELK-ESM1 installed which I would assume should be enough load?

The ESM1 transformer presents very little load to X10 signals due to its relatively high inductance.

Since I am using the white wire as a common between the two circuits, could the X10 commands be "colliding" and canceling each other?

No, that should not be a problem.  The white will be the return path for both phases.  You measured plenty of signal, so the problem must be associated with the timed commands themselves.

All of the above tests were done without the XTB in place. One other test I did was to put the XTB in place in front of the CM11A and test the signal levels. One side of the duplex outlet showed a very marginal increase in signal level, maybe up to 1.4-1.5 volts.

Since you don't have a good passive coupler, this is exactly what I would expect.  The signal on that phase is coming from your Leviton repeater, not the XTB.  I'm starting to wonder whether the repeater might be part of this mix too...

Regarding possible "malformed commands" coming from the CM11A, I don't understand why some modules respond to those and some do not.  I recall your saying you had several modules set to the same codes, and only those in that underground box did not work.  The others responded to the CM11A timed events properly.  Would you confirm this is true?

If that is the case, the only possible difference is the signal level and perhaps the LM465s themselves.

The X10 spec says there should be a 3 cycle gap between commands.  It also says there should be no gap between bright/dim commands sent in sequence.  The RR501 does place a gap between commands.  It was interesting to discover that the X10 maxi-controller DOES NOT place a gap between commands.  When a button is held down it sends a sequence of commands one right after the other without any gap between.  It doesn't matter which key is pressed.  This is true for the old boxy unit and the new low profile one with the smoked plastic cover.  I can't check out the  CM11A because my computers are all running XP now, and none of them will recognize the CM11A through the serial port.

Jeff

[BSarte]

Since you don't have a good passive coupler, this is exactly what I would expect.  The signal on that phase is coming from your Leviton repeater, not the XTB.  I'm starting to wonder whether the repeater might be part of this mix too...

OK, time to consider getting a new 'passive' coupler. What should I consider?

Regarding possible "malformed commands" coming from the CM11A, I don't understand why some modules respond to those and some do not.  I recall your saying you had several modules set to the same codes, and only those in that underground box did not work.  The others responded to the CM11A timed events properly.  Would you confirm this is true?

Yes, code 7 has 5 separate LM465's around the house and only one of them is in this underground box, so it is not the specific code but all the LM465's in this box that do not respond to on codes in this box on those two circuits. All the other LM465's with that code (7) respond to both on and off codes. Could it be as simple as no load in that box on those circuits? Simply, put a load on those circuits and they seem to work as with the off codes as they do when they are on and the off code is sent, they shut off! All the other circuits with the other LM465's that work fine probably do have loads on them. Should I consider putting something in that box that would creat a larger load and be on all the time? Hmm. maybe a simple test would be to run the pond pump longer (past when the night lights are scheduled to come on)?

I can't check out the  CM11A because my computers are all running XP now, and none of them will recognize the CM11A through the serial port.

That is exactly the way I am running mine, Windows XP SP2 through the serial port to the CM11A with ActiveHome.

Bruce

[Boiler]

Bruce,

Sorry, I seem to have jumped in without reading up on your "saga".  I missed several important details in your installation (specifically your Leviton repeater).  I also apologize for my poor choice of words (i.e. intriguing).

...One thing I noticed was that the simple command via the RR501 appeared to be different. The signal bar showed several ( 3-4) pegs about 1/2 second apart while the macro commands only showed two very quick pegs to the signal level.

The difference in the indications between the RR501 and CM11a transmissions indicate that these units may be on opposite phases.  The RR501 sounds like it is on the same phase as your Elk tester.  The communication sequence would be as follows (four pegs of your ESM1):

RR501
Houscode, Unit code (EX : E6 - transmitted twice)
Housecode ON (E ON transmitted twice)

In contrast to the above, if your CM11a is on the opposite phase, the Leviton repeater will repeat only one of the two transmissions (two pegs of your ESM1)-

CM11a
Houscode, Unit code (EX : E6 - one transmission)
Housecode ON (E ON one transmission)

As Jeff has indicated, your Leviton repeater will block any signal increase from the XTB and will transmit at it's own output level.

OK, time to consider getting a new 'passive' coupler. What should I consider?

In order to get the maximum benefit from the XTB, I would say yes.  Jeff is far better qualified to give recommendations on the type/model.

As a test, try plugging a trouble light or other resistive load into your box.  I'm starting to wonder if your electrical run has been compromised (increased resistance/capacitance).  Have you had any "activity" in the area of this run that might have damaged the wire?  How old is this wiring?

The Boiler

[JeffVolp]

OK, time to consider getting a new 'passive' coupler. What should I consider?

An inexpensive passive coupler that works well is the SignaLinc 4816H.  It is essentially a .1uF cap with a 18uH inductor in series to cancel the capacitive reactance at 120KHz.  The one misgiving I have about that unit is that the sample I examined had a 400VDC capacitor, which is certainly marginal at 240VAC.  A capacitor rated for 250VAC would have cost them a few cents more.  The X10 XPCP and similar Leviton coupler is a much more complex unit with twin tuned circuits and transformer isolation from input to output.  The original version I tested drives the two legs "out of phase" when wired per the diagram.  The newer one in the "Decora" style case drives the two legs "in phase" when wired per the diagram.  More info on what this means can be found here:  http://jvde.us/x10/x10_couplers.htm

Both the SignaLinc 4816 and the original XPCP can be found on eBay for about $10 plus shipping.  The newer "Decora" XPCP is about $30.  I have a few of them left at $25 each.

For best performace, I recommend a coupler that wires directly to your distribution panel.  The phase of either XPCP can be flipped to match the output of your repeater.

Jeff

[BSarte]

Test last night by leaving the pond pump on later (unit creating more load on those two circuits) did not have any positive change, still nothing responded to the CM11A timed commands, but did respond to the off commands. I did not expect anything since the pond pump is plugged into one of the circuits and all the other equipment is plugged into the other circuit.

One other thought was to eliminate one of the circuits and have everything plugged into the same circuit. Might have some effect. Can't do that change for a while since Iwill be traveling until next week.

I just bought and paid for an X10 XPCP passive coupler. So hopefully that will be here when I get back and then I can install that as well as to make the change in the underground box to put eveything on one circuit and eliminate the other.

Any further thoughts on what could be effecting the "on" commands to these two circuits? Nothing around those two circuits has been changed that I am aware of that could have some effect on those two circuits. I ran this #12 3 wire cable out of the house encased in 1" electrical PVC all the way over to the underground box, so the wire should be well protected, and it is a straight run abut 85 to 90 feet from the circuit panel. I have three other runs of outdoor electric around the house and have done the same thing since I am always afraid of hitting the wire when digging if I don't encase it in electrical PVC underground.

Bruce

[Boiler]

Bruce,
Sounds like a nice install on the underground run.

I'm sorry but I don't have anything constructive to add at the moment.  Let us know if adding a load to your problem circuit or adding the XPCP changes anything.

Boiler

[BSarte]

I have been away on a business trip and just before I left, I ordered the X10 XPCP. When I got home on Monday, it was here.

Today, I got everything together to install it. I installed it very carefully as always, understanding and following the instructions. I used a 2 foot long tail piece of #12 3 wire with ground to connect the XPCP up to two circuit breakers, using the red/black for connection to the L1/L2 poles on the XPCP to the two differing circuit breakers on each phase of the panel and the ground/white for the N's to the common in the circuit panel. Unfortunately, when I turned on the first breaker, the XPCP shorted out violently. 

I double checked the diagram and all the wiring, but it appears that the XPCP shorted out internally. I double checked all the wire colors and the labels on the XPCP. All is wired correctly.

I guess I'll have to order another and give it another try!

Are there any Leviton equivalents of the XPCP?

Also, I was planning on installing a load in my underground box so as to try that to resolve this issue. Any idea's on what would be good to provide a sufficient load but not run up too much on my electric bill? Needs to be something that has a good long life before failure since this will be in the under ground box.

Bruce

[BSarte]

Just bought the Leviton 6299!