Print

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

[keyboardgnome]

Hi all. I'm now over the edge with X10's hardware. Their sales reps claimed the CM15A would work in Linux; it doesnt. I ordered switches only to have them change the behavior. 3 weeks later they were replaced with what was advertised. Now, I just had a switch explode and ruin damage some electrical wiring. Out of the blue. It had been running fine for months, and tonight when I went to turn on the lights there was an electrical explosion and the breakers threw the power. I just finished removing all the other X10 switches that I have and I fear they'll do the same.

I'm going to ebay all of the crap I've just purchased, because I'm sure X10 will not take it back now. I'll never go back to them.

I need alternatives. I want to do home automation and I especially want to do it with Linux. Anyone have any advise?

Thanks.

Here's the switch: http://handorf.org/include/images/messageboards/switch.JPG

[HA Dave]

keyboardgnome THANK-YOU for sharing the picture........ would very much like to see a picture of the inside of the switch.

[-Bill- (of wgjohns.com)]

Very detailed pic.

From close inspection... it appears the tip (i.e. point) of the mounting screw is actually melted... as well as the obvious char and such of the other end of the screw which appears to have arced to the switch's mounting plate.

Is that true?  Is the tip of the mounting screw melted?   

[KDR]

After looking closely at everything here is one possible cause...

In the picture below is the screw blown up a bit. You can see the tip of the screw has gotten hot and has also melted. You can see the forming of a drip. The second picture shows a closeup of the burn mark on the switch mounting ear where the screw head was located. You can also see a burn mark on the screw near the underside of the head. Since the arc took place here it would indicate that the screw was not tight and this connection was a bit loose. If the switches mounting screw end was very close to the open side of a wire nut say and the switch moved it may have come in contact with a live wire causing the short. The arc would have been strongest at the closest point to an uninterrupted ground. What that means is the screw would come in contact with the live wire and taken on that potential and the arc would happen at the next point where a poor connection exists. (If there would of been no other poor connections the arc would have taken place at the screw tip) Under this condition the screw tip gets hot but doesn't arc and burn like where it did at the screw head (loose connection) from which that point on you had a good connection to a ground. You do need to further examine the switch for damage because my possible explanation is only that. one possible explanation.





If the short was outside the switch the switch would still be functional. A test... You could try hooking up the switch on a test bench to see if it even works now. You would need to be very careful doing that and be sure it is powered up on a fused circuit.

^{----------------KDR}

[keyboardgnome]

http://handorf.org/include/images/messageboards/switch2.JPG
http://handorf.org/include/images/messageboards/switch3.JPG
http://handorf.org/include/images/messageboards/switch4.JPG
http://handorf.org/include/images/messageboards/switch5.JPG
http://handorf.org/include/images/messageboards/switch6.JPG

All of my switch and outlet recepticle boxes are metal. I think something inside the switch shorted out and grounded through the screw. I live in Philadelphia, PA.

I added a photo comparing the two screws next to each other. I think the tip is rounded like that from the cutting of the material to make the screw itself, and not from the shorting. The one that shorted is on the left.

I should note that when I opened it, my wife commented as to how much it made the room stink. It has that good old transformer oil smell, or popped capacitor (though, I dont see any).

I'll try and test it tomorrow.

[KDR]

keyboardgnome I see what you mean about the screws. From the looks of the other pics you posted it looks like it failed inside. My guess is there is no use testing it. I see a burn mark on the side of the switch also. Wonder what caused that.

^{----------------KDR}

[Charles Sullivan]

You've stated some electrical wiring was ruined, yet the wires from the switch look undamaged.  That plus the burn mark on the _side_ of the switch are a clear indication that a bare hot wire in the box came into contact with it.

[keyboardgnome]

Charles,

The box is grounded. I also figured out what happened with the wiring that caused me to think it was bad. When I pulled out the blown switch, on the black wire the copper contact under the plastic coating crumbled away (which I've never had this happen before, except once when I did blow all the wiring in the circuit). I pulled back more copper until this stopped (2 to 3 inches). It was not like this when I had installed it. I also checked the breaker box side which was fine. The light now works just fine, as well as the few others that are on this circuit.

Also, both light bulbs were dead after this. I had just turned them off a moment before, so I knew they were working.

As for the burn mark on the side of the switch, there's a screw there that goes directly to the inside of the switch itself and makes contact with what looks like a voltage regulator (looks like this: http://www.opamp-electronics.com/~opamp/catalog/images/LM7918CT_1A_18V_TO-220_Negative_Voltage_Regulator_012950.jpg).

[Charles Sullivan]

Charles,

The box is grounded. I also figured out what happened with the wiring that caused me to think it was bad. When I pulled out the blown switch, on the black wire the copper contact under the plastic coating crumbled away (which I've never had this happen before, except once when I did blow all the wiring in the circuit). I pulled back more copper until this stopped (2 to 3 inches). It was not like this when I had installed it. I also checked the breaker box side which was fine. The light now works just fine, as well as the few others that are on this circuit.

But the wire from the switch (a much smaller wire than the house wires) is intact, so it didn't draw excessive current.  I suspect a piece of the insulation on the house wire cracked off and the bare wire hit the plate on the side of the switch.

Also, both light bulbs were dead after this. I had just turned them off a moment before, so I knew they were working.

I don't know what to make of that.

As for the burn mark on the side of the switch, there's a screw there that goes directly to the inside of the switch itself and makes contact with what looks like a voltage regulator (looks like this: http://www.opamp-electronics.com/~opamp/catalog/images/LM7918CT_1A_18V_TO-220_Negative_Voltage_Regulator_012950.jpg).

No, that switch component is a triac, which is what actually controls the AC current through the switch.   The aluminum plate on the switch is used to dissipate the heat generated by the triac when it's on, so the screw keeps them in tight contact.

[keyboardgnome]

Makes sense that it's a heat sync point.

But, anyways. Anyone have any suggestions for linux friendly home automation bit's that aren't X10? I'm just a little fed up with them.

[Charles Sullivan]

Makes sense that it's a heat sync point.

But, anyways. Anyone have any suggestions for linux friendly home automation bit's that aren't X10? I'm just a little fed up with them.

If you weren't so soured on X10 (unjustly so in my opinion), I would have recommended you get a CM11A interface and use Heyu under Linux  (http://www.heyu.org).

Support for anything other than X10 under Linux is rudimentary.   You might take a look at Misterhouse (http://misterhouse.sourceforge.net).  Some of the guys on the Misterhouse mailing list have been making noises about Insteon and Z-Wave support.

[keyboardgnome]

I may try the CM11a, but X10 wont sell them anymore (according to the rep I just spoke with). My frustrations are with the company itself, not the technology. The sales folks that I have spoken with there just flat out lie; and when I brought one of my technical related issues to their staff one of the members accused me of "making things up" only to later find out that his organization had changed the devices behavior. My recent experience with X10 had all started with this thread regarding these switches:

http://www.x10community.com/forums/index.php?topic=13031.0

Since then, they replaced the "dimming switches" I had with the on/off ones I have now, and X10 doesn't plan to keep making them in this form (the dim on/off drove me and my wife nuts). So, if I blow anymore switches, I've exhausted my supply and only stock of the on/off forms (From this stockpile that I have, I already had one switch malfunction, this makes 2. I have only one functioning spare left, and X10 has no more of these on/off switches. They're dim on/dim off).

Then there was the frustrations with the CM15a:

http://www.x10community.com/forums/index.php?topic=13689.msg76011

When I have seen support go this route, with the vendor not providing any information, the device very rarely gets supported (all community driven, usually by a few people).

From my perspective, I have failing hardware that isn't supported with a computing interface that isn't supported for my environment despite being told otherwise from the X10 organization. HA is great, and I love the idea. However, I lack the faith that X10, as a company, is of any good anymore. What I'm looking for is an alternative to that company, rather than the technology.

[dave w]

Keyboardgnome
Not that this changes anything, but the isolated triac could have become "un-isolated". A nasty surge (caused by or resulting in the light bulb failure?) , or physical force or stress on the triac or the heat sink tab, might cause the die to come in direct contact with the metal substrate that the die is mounted on (?). If you are curious and have a multimeter, check the resistance between the triac's metal tab and each of the three leads. "0 ohms" or very low ohms indicate an internally shorted triac.
Sorry about your luck, but for many of us, X10 has been reliable and fun.

[Charles Sullivan]

Keyboardgnome
Not that this changes anything, but the isolated triac could have become "un-isolated". A nasty surge (caused by or resulting in the light bulb failure?) , or physical force or stress on the triac or the heat sink tab, might cause the die to come in direct contact with the metal substrate that the die is mounted on (?). If you are curious and have a multimeter, check the resistance between the triac's metal tab and each of the three leads. "0 ohms" or very low ohms indicate an internally shorted triac.
Sorry about your luck, but for many of us, X10 has been reliable and fun.

An internally-shorted triac would not account for the burn mark on the side plate.  There's no evidence of damage to the triad tab itself in the pictures.

[Charles Sullivan]

I may try the CM11a, but X10 wont sell them anymore (according to the rep I just spoke with). My frustrations are with the company itself, not the technology. The sales folks that I have spoken with there just flat out lie; and when I brought one of my technical related issues to their staff one of the members accused me of "making things up" only to later find out that his organization had changed the devices behavior. My recent experience with X10 had all started with this thread regarding these switches:

http://www.x10community.com/forums/index.php?topic=13031.0

Since then, they replaced the "dimming switches" I had with the on/off ones I have now, and X10 doesn't plan to keep making them in this form (the dim on/off drove me and my wife nuts). So, if I blow anymore switches, I've exhausted my supply and only stock of the on/off forms (From this stockpile that I have, I already had one switch malfunction, this makes 2. I have only one functioning spare left, and X10 has no more of these on/off switches. They're dim on/dim off).

Then there was the frustrations with the CM15a:

http://www.x10community.com/forums/index.php?topic=13689.msg76011

When I have seen support go this route, with the vendor not providing any information, the device very rarely gets supported (all community driven, usually by a few people).

From my perspective, I have failing hardware that isn't supported with a computing interface that isn't supported for my environment despite being told otherwise from the X10 organization. HA is great, and I love the idea. However, I lack the faith that X10, as a company, is of any good anymore. What I'm looking for is an alternative to that company, rather than the technology.

I think they still have these in stock:
  http://www.x10.com/promotions/hc60rx_vt_kit.html
If not, many X10 reseller stores on eBay have them, although at a higher price.

Read this about the "new & improved" WS467 Wall Switches.  (Heyu now includes models for them.)
  http://www.heyu.org/docs/New_WS467.txt

If you don't like the X-10 branded switches, compatible switches are available from Leviton and Smarthome.