Water heater eating X-10 signal

| Since the transformerless power supply cannot provide that much energy, I | don't think the output impedance is in the fractional ohm region.

I was considering the impedance of the driver separately from the supply characteristics. That wouldn't be a nice thing to do in an advertisement, but for our purposes it allows an easy measurement. There doesn't appear to be anything special about ACT's reactive power supply and it should be possible to see it slump in the presense of heavy 240V loads if that is indeed an issue.

| I thought we were discussing summing of in-phase X10 signals on the neutral | reducing the end-point signal strength. That would only be true if there | were multiple X10 loads (or signal suckers) on both legs of that circuit.

I should perhaps have chosen my words more carefully, e.g.: any load with impedance (at the X10 carrier frequency) too low to ignore. Multiple loads per leg are not required. In the case of a single load on the opposing leg with in-phase drive that load hurts you; in the case of out-of-phase drive it helps you. Repeat as needed for additional loads. To the extent that the loads approach balance with an out-of-phase drive the impedance of the neutral between the supply and the first load (possibly along with other segments) drops out of the picture. In the best case this is like getting a free 50% reduction in cable length.

This may or may not be a material effect in the real world; however, given that the analogous effect for power transmission is viewed as significant and given that the relative loss at X10 carrier frequencies is considerably greater I'm not willing to dismiss it out of hand.

Dan Lanciani ddl@danlan.*com

Reply to
Dan Lanciani
Loading thread data ...

I wanted to follow this up. The SmartHome SignaLinc 4816H apparently consists of nothing more than a 1.uF capacitor and 18 uH inductor in series. So, that drives both legs in phase.

Jeff

Reply to
Jeff Volp

| > | That was an easy test to do - just confirmed an old Leviton 6299 does | > flip | > | the phases. | >

| > That certainly confirms the behavior I saw. | | I wanted to follow this up. The SmartHome SignaLinc 4816H apparently | consists of nothing more than a 1.uF capacitor and 18 uH inductor in series.

Is that a .1uF capacitor? That would give about the right resonance frequency with standard components.

| So, that drives both legs in phase.

I'll bet it would have worked a lot better for me than the 6299. I also noticed that we went through approximately this same discussion

10+ years ago, e.g., see:

formatting link
which suggest that either there was a different 6299 or else Phil Kingery wasn't aware that it flipped the phase. Oddly, in the Hometoys article Dave referenced earlier in this thread, the simplified schematic of the passive coupler appears intended to flip the phase but Phil still talks about the problem of 240V devices as if it did not.

Dan Lanciani ddl@danlan.*com

Reply to
Dan Lanciani

Yes. Incidentally, it is exactly what I recommended here:

formatting link
The XTB and XTB-II use .22uF and 8.2uH.

Interesting. My Leviton 6299 is an old unit I picked up some time ago on eBay.

That was not a simplified schematic. The two blue blobs are MOVs. The series caps are .22uF 360VAC (yes, VAC). The small caps are both 1000pf. The inductors are little tunable transfomers, like those used in X10 modules. I re-checked the 6299. It is definately out of phase, and peaks at 134KHz. That probably means the tunable transformers weren't quite adjusted properly.

The Leviton 6299 and X10 XPCP used to be identical. X10 recently changed the XPCP to a "Decora" form factor. The older version is still available on eBay. The new one is riveted closed, so I don't know what is in it. However, it is IN-Phase, and peaks at 118KHz. It still has a 4-wire connection, and both output lines must be connected. Also, reversing the output flips the phase, so it must still have transformer coupling.

It looks like X10 decided that in-phase is the way to go.

A side note - street prices for the old XPCP and SignaLink are similar. The XPCP / 6299 would seem to be a better deal.

Jeff

Reply to
Jeff Volp

Sorry for the delay, finally got time to do a little data gathering on this issue.

I find my measurement results a bit confusing.

In my breaker panel I have a Leviton HCA02-10E coupler/repeater. It's wired via an existing 2-pole 30A breaker feeding a dryer, and was part of my working configuration before this problem started.

To do some measurements, I wrote a small HomeSeer script consisting of a loop which sends "B1" "ON", waits 3 seconds, sends "B1" "OFF", waits another 3 seconds, then repeats.

With this script running I measured signal levels in the breaker panel using an ESM-1 connected via alligator clip leads.

Here are my (somewhat puzzling) results. I've arbitrarily labeled the phases "A" and "B", and expressed the observed signal strengths in terms of number of bars lit on the ESM-1:

Condition: HCA02 breaker OFF, water heater breaker OFF Phase A: approx 1.5 bars Phase B: 0 bars

Condition HCA02 breaker OFF, water heater breaker ON Phase A: approx 1.5 bars Phase B: 0 bars

Condition HCA02 breaker ON, water heater breaker OFF Phase A: 5 bars Phase B: 4 bars

Condition HCA02 breaker ON, water heater breaker ON Phase A: 5 bars Phase B: 4 bars

Apparently the HCA02 is performing its repeater function quite well. But what's puzzling is that with the coupler in-circuit, closing the breaker for the water heater apparently has no visible effect on the signal coupling as shown by the ESM-1 at the breaker panel, even though I've repeatedly confirmed that it debilitates my control of my front yard lighting.

Shouldn't I expect to see a big drop in Phase B signal with the water heater circuit connected?

Thanks.

Reply to
graftonfot

| Sorry for the delay, finally got time to do a little data gathering on | this issue. | | I find my measurement results a bit confusing. | | In my breaker panel I have a Leviton HCA02-10E coupler/repeater. It's | wired via an existing 2-pole 30A breaker feeding a dryer, and was part | of my working configuration before this problem started. | | To do some measurements, I wrote a small HomeSeer script consisting of | a loop which sends "B1" "ON", waits 3 seconds, sends "B1" "OFF", waits | another 3 seconds, then repeats. | | With this script running I measured signal levels in the breaker panel | using an ESM-1 connected via alligator clip leads.

Where did you connect the clips in the panel?

| Here are my (somewhat puzzling) results. I've arbitrarily labeled the | phases "A" and "B", and expressed the observed signal strengths in | terms of number of bars lit on the ESM-1: | | Condition: HCA02 breaker OFF, water heater breaker OFF | Phase A: approx 1.5 bars | Phase B: 0 bars | | Condition HCA02 breaker OFF, water heater breaker ON | Phase A: approx 1.5 bars | Phase B: 0 bars | | Condition HCA02 breaker ON, water heater breaker OFF | Phase A: 5 bars | Phase B: 4 bars | | Condition HCA02 breaker ON, water heater breaker ON | Phase A: 5 bars | Phase B: 4 bars | | Apparently the HCA02 is performing its repeater function quite well. | But what's puzzling is that with the coupler in-circuit, closing the | breaker for the water heater apparently has no visible effect on the | signal coupling as shown by the ESM-1 at the breaker panel, even | though I've repeatedly confirmed that it debilitates my control of my | front yard lighting. | | Shouldn't I expect to see a big drop in Phase B signal with the water | heater circuit connected?

Are you sure that one or both elements in the water heater come(s) on when you enable its breaker? Alternately, do you know that the yard lighting control is debilitated even when no elements are on? Or maybe simplest of all: were you checking that the yard lighting control came and went as you closed and opened the breaker during your above tests?

Dan Lanciani ddl@danlan.*com

Reply to
Dan Lanciani

What about the thermostat? Unless the heater element switches on, turning on the circuit breaker would have little or no effect.

Jeff

Reply to
Jeff Volp

What does the "X-10 Good" LED on the ESM1 do when the HCA02 and water heater are active?

Also, as Jeff notes, for the tests you need to make sure the water heater element is actively conducting (i.e. heating the water) by cranking the thermostat up.

Reply to
Dave Houston

OK, this is weird - I replied to this thread last night, but I don't see my reply in this thread today.

So I'll try again:

I have an active coupler (Leviton HCA02) connected to a 2-pole breaker supplying a clothes dryer. I wrote a short HomeSeer looping script to repeatedly send X-10 control signals while I measured the signals with my ESM-1.

Here are my results:

Coupler breaker OFF, Water heater breaker OFF: Phase A: 1.5 bars on the ESM-1 Phase B: 0 bars on the ESM-1

Coupler breaker OFF, Water heater breaker ON: Phase A: 1.5 bars on the ESM-1 Phase B: 0 bars on the ESM-1

Coupler breaker ON, Water heater breaker OFF: Phase A: 5 bars on the ESM-1 Phase B: 4 bars on the ESM-1

Coupler breaker ON, Water heater breaker ON: Phase A: 5 bars on the ESM-1 Phase B: 4 bars on the ESM-1

I can see the HCA02 is doing its job, repeating the Phase A: signal and coupling it to Phase B.

But what puzzles me is that, in terms of signal strength displayed on the ESM-1, there's no difference when the water heater breaker is open vs. closed, in spite of the fact that closing that breaker definitely causes loss of control of some of my X-10 devices (particularly, my front yard lighting.)

Reply to
graftonfot

I might have missed it, but where did you meter from? I'd rerun the tests with the ESM1 located where the problems occur so that you can see what effect the above combinations have at the devices giving you trouble.

-- Bobby G.

Reply to
Robert Green

Ooops, one other thing I meant to add: all the measurements I made were performed simply by switching the new water heater's circuit breaker on and off. In particular, I made no effort to determine whether or not the water heater was active, i.e. one or both of heating elements were energized and heating water.

Reply to
graftonfot

As Jeff and I suggested earlier you need to make sure the water heater is active when you conduct the test. Current has to be flowing through the heating element(s).

Reply to
Dave Houston

Hi, and thanks for your reply.

I used alligator-clip-leads to connect the ESM1 meter to the phase/ neutral buses in my main panel.

My front yard lights are controlled by a WS467 wall switch. Were you suggesting connecting the ESM1 at the site of the wall switch?

Reply to
graftonfot

I connected one lead to the neutral bus bar, then alternated the other lead between the phase buses (the metal tabs that the breakers stake onto.)

Actually, I'm more sure that the elements *aren't* on for all these tests. I'll need to figure out a reasonable way to confirm this, though (I don't have a clamp-on ammeter.)

All of the phenomena/tests I've described have occurred during times of no water usage. I realize this doesn't absolutely guarantee that one of the heater's elements is not in-circuit during the test, but it makes it much more unlikely. As I said I'll try to confirm this.

I'm not familiar with water heater circuitry - are they as simply as an SPST switch controlled by a heat sensor? I'm wondering if they typically include some sort of spark-suppression RC stuff to preserve the thermostat contacts...perhaps that's doing something weird to the X-10 signals.

Thanks for the continued suggestions...

Reply to
graftonfot

I'm fairly certain that the element(s) are not in-circuit when I close the breaker and lose control of the lights. I know that sounds crazy. I'm going to try to confirm this somehow.

I'm considering trying your X-10 signal booster at this point. But I'd sure like to find out what is really going on.

Appreciate your suggestions, thanks.

Reply to
graftonfot

Drat, I didn't think to check the status of that light. I'll do that next.

Not sure why I would need to ensure the water heater element(s) are active. This failure condition started occurring immediately after we installed the new water heater, and has consistently happened with the breaker on, while the X-10 operation has consistently gone back to normal with the breaker off. I've tested it many evenings, and, as unbelievable as it sounds, I'm pretty confident that the water heater was not active during those times.

As I stated in other replies, though, I'll try to confirm this somehow.

Thank you for your help.

Reply to
graftonfot

Do you know if there are any outlets on that run? I would try to get the meter as close to the actual problem load (as opposed to the problem source) as possible. My guess is that somehow, the water heater wiring is shunting the X-10 signal to ground somewhere. It sounds like whenever that circuit is active, whether drawing large amounts of power or not, it reliably "steps" on your porch light signal.

What brand/model of water heater is this? Maybe the documentation can shed some light on what's happening.

-- Bobby G.

Reply to
Robert Green

In article , snipped-for-privacy@yahoo.com writes: | On Apr 25, 12:56 am, ddl@danlan.*com (Dan Lanciani) wrote: | > In article , snipped-for-privacy@yahoo.com writes: | >

| > | Sorry for the delay, finally got time to do a little data gathering on | > | this issue. | > | | > | I find my measurement results a bit confusing. | > | | > | In my breaker panel I have a Leviton HCA02-10E coupler/repeater. It's | > | wired via an existing 2-pole 30A breaker feeding a dryer, and was part | > | of my working configuration before this problem started. | > | | > | To do some measurements, I wrote a small HomeSeer script consisting of | > | a loop which sends "B1" "ON", waits 3 seconds, sends "B1" "OFF", waits | > | another 3 seconds, then repeats. | > | | > | With this script running I measured signal levels in the breaker panel | > | using an ESM-1 connected via alligator clip leads. | >

| > Where did you connect the clips in the panel? | | I connected one lead to the neutral bus bar, then alternated the other | lead between the phase buses (the metal tabs that the breakers stake | onto.)

You might want to try hooking on to the "output" of the breaker feeding the yard light circuit.

| >

| > | Here are my (somewhat puzzling) results. I've arbitrarily labeled the | > | phases "A" and "B", and expressed the observed signal strengths in | > | terms of number of bars lit on the ESM-1: | > | | > | Condition: HCA02 breaker OFF, water heater breaker OFF | > | Phase A: approx 1.5 bars | > | Phase B: 0 bars | > | | > | Condition HCA02 breaker OFF, water heater breaker ON | > | Phase A: approx 1.5 bars | > | Phase B: 0 bars | > | | > | Condition HCA02 breaker ON, water heater breaker OFF | > | Phase A: 5 bars | > | Phase B: 4 bars | > | | > | Condition HCA02 breaker ON, water heater breaker ON | > | Phase A: 5 bars | > | Phase B: 4 bars | > | | > | Apparently the HCA02 is performing its repeater function quite well. | > | But what's puzzling is that with the coupler in-circuit, closing the | > | breaker for the water heater apparently has no visible effect on the | > | signal coupling as shown by the ESM-1 at the breaker panel, even | > | though I've repeatedly confirmed that it debilitates my control of my | > | front yard lighting. | > | | > | Shouldn't I expect to see a big drop in Phase B signal with the water | > | heater circuit connected? | >

| > Are you sure that one or both elements in the water heater come(s) on when | > you enable its breaker? Alternately, do you know that the yard lighting | > control is debilitated even when no elements are on? Or maybe simplest of | > all: were you checking that the yard lighting control came and went as you | > closed and opened the breaker during your above tests? | >

| > Dan Lanciani | | Actually, I'm more sure that the elements *aren't* on for all these | tests. I'll need to figure out a reasonable way to confirm this, | though (I don't have a clamp-on ammeter.)

Turn the thermostat way down?

| All of the phenomena/tests I've described have occurred during times | of no water usage. I realize this doesn't absolutely guarantee that | one of the heater's elements is not in-circuit during the test, but it | makes it much more unlikely. As I said I'll try to confirm this.

As long as you are sure that both the measurements and the debilitation happen under the same circumstances it should be fine.

| I'm not familiar with water heater circuitry - are they as simply as | an SPST switch controlled by a heat sensor?

I think you will find two DPST switches (one for each element) plus an additional high-limit cutout feeding both of them.

| I'm wondering if they | typically include some sort of spark-suppression RC stuff to preserve | the thermostat contacts...perhaps that's doing something weird to the | X-10 signals.

It's certainly possible. Whatever the exact mechanism I'm interested in this type of effect where one circuit causes impairment of another without having any observable effect at the repeater. We shouldn't ignore the possibility of a noise source (rather than attenuation). What type of X10 modules are controlling the yard lights?

Dan Lanciani ddl@danlan.*com

Reply to
Dan Lanciani

If X-10 is affected only when the elements are active, it suggests the out-of-phase feed condition described by Phil Kingery and suspected (I think) by Jeff, Dan and myself.

If X-10 is affected whenever the breaker is on regardless of the state of the elements, it suggests something unusual in the water heater control unit or in the manner in which it's wired.

I would want to understand it before trying Jeff's device.

Reply to
Dave Houston

| My front yard lights are controlled by a WS467 wall switch.

When its operation is impaired, have you confirmed that you can neither turn it off nor turn it on remotely?

Dan Lanciani ddl@danlan.*com

Reply to
Dan Lanciani

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.