Cannot communicate with Switchlinc

[jwshome]

I have a fairly large Insteon network (maybe 25 Switchlincs, some of which are dual band; about a dozen lamplincs, also some dual band) and communications are highly reliable using HS and the iSY. I recently installed several additional dual-band switches, replacing some manual switches that I had not yet automated. I cannot communicate reliably with one pair of these switches (that make up a three-way circuit). They are not far from the main switch panel and PLM. I've done the following, none of which has worked:

 

1. Replaced the switches

2. Unplugged all other devices that were plugged in (most of which were filtered anyway) on the same circuit

3. Installed a dual band lamplinc on a different circuit about 6 feet away.

 

This is driving me crazy, as I would have thought at a minimum that the Lamplinc would have communicated wirelessly with the switch if nothing else had worked.

 

Is there something else I can do? Any ideas?

 

Thanks in advance for your help.

[oberkc]

Is there any other insteon device on this circuit. Possible loose wire connection.

[ELA]

Hello Jwshome,

 

Are your boxes plastic or metal?

Confirm the new device works at all by installing on plug w/pigtail -wire nutted on. Plug in same outlet as PLM.

If that works move the new device's location by plugging into outlets closer to the final intended location to see at what point it stops working.

[jwshome]

Yes, there are two switchlincs on the same circuit (it is a three-way switch) - and I am having trouble communicating with both of them. The circuit boxes are metal (but so are all of the other circuit boxes in the house, and I have no trouble with them).

 

I'm wondering if the Switchlincs are wired incorrectly for a three-way circuit - I'm going to check again.

[oberkc]

I'm wondering if the Switchlincs are wired incorrectly for a three-way circuit - I'm going to check again.

 

reminder...insteon devices are not wired the same way as normal three-way installations.

 

There are several possible clues to look for. Are the LED indication lights on? if so, you have power applied (at least for now) and you should be able to communicate with them. If the lights are off, there is possible wiring problem. Are the LED indications lit for now, but go off when you flip other switches? Possible wiring issue.

 

Bottom line...if they are wired incorrectly, there should be other clues which would indicate a problem.

[arw01]

What kind of bulbs are in the lights these switches control? Swap those bulbs to another fixture and see if the problem follows them.

 

I have a CF bulb that if put into a keypadlinc dimmer, causes the insteon devices on the circuit to just turn off and back on a second or two. Took that out, put it in a lamplinc controlled lamp, and don't have the issue anymore.

 

I would second the pigtail approach to see whats up. I am chasing some gremlins of my own. I have three older switchlincs that I cannot program if I don't have the PLM plugged into the outlet right below them. However, they always, always respond regardless of where the PLM is.

 

I put an access point right into the same spot as the plm, and they will not program with the access point as the bridge.

 

All the computers, except a couple of laptop cords, have filters on them, and I have an elk signal meter for the old X-10 signals that works for insteon signal strength. Let's me know right away if a filter did anything and it's amazing it (the filter) can take a 1.5v signal to 2.5v, and it still won't make a programming difference!

[ELA]

jwshome,

Metal boxes have the potential to block nearly all of your RF signal (thus making a DB unit basically only SB worthy). Especially possible if a metal cover plate were used.

Not saying that is happening, just that it is something to consider. Testing with a pigtail will help you determine that.

 

 

arw01,

I have heard comments from a few other members using an old ELK signal meter for signal strength measurements. Do you know if the value presented is an averaged value over a certain time period vs. a measurement of individual hops strength. I suspect it averaged based on what you describe and the fact that it was designed for X-10. It is therefore possible that the initial send (or hop 0) is very weak yet the other hops are fairly strong. I have encountered such a situation as that where the communications fails when hop 0 is very weak even though the hops 1-3 may be very strong.

[Brian H]

I have tried an ELK ESM-1 and don't think it averages anything.

Has a bargraph display that seems to be nonlinear. First bar is .1 volts tenth bar is 5 volts.

There is a schematic of the ESM-1 on this web page.

http://www.davehouston.net/ESM1-TEST.htm

 

Users Manual for ELK ESM-1

http://www.smarthome.com/manuals/4813_man.pdf

 

Looks more like maybe a few op-amp sections with maybe some tuning around the 120 KKz X10 frequency and a Zero Crossing detector. So it can give a X10 Good Message Decoded LED to light.

It uses an external wall wart type AC transformer so there are no high voltages on the simple analog amps or Zero Crossing detection circuits.

[arw01]

I have to concur, it delays the signal display out so a human can observe the signal onto and off the line. E.g. the lights dance up and down the meter display.

 

In another location I had a controllinc that would only show about 2.5v, put an access point down the line from it and now I get a FULL 5v on that circuit.

 

I also noticed today, that my microwave oven puts a steady 1volt on the display the whole time it runs. I may be moving a 10amp filterlinc to that microwave plug in and use an older X-10 on the computer that has that filterlinc on it now. But I have my PLM plugged into it's bypass line, which is awfully convenient.

[ELA]

Thanks Brian,

I had checked the manual before I posted earlier but had not seen the schematic. The manual stated 120Khz +/- 10%, just barely covering the Insteon band so signals may be somewhat attenuated at 131Khz. Since they do not state if the +/-10% represents the -3dB points?

 

Very interesting. If it uses a wall wart with a transformer I wonder how do the 120-131Khz signals pass through? Maybe a custom wall wart that capacitively couples the high freq. from the line onto the secondary?

 

Agreed from the reference that it probably does not average. What would you then estimate that you are actually seeing on the display? Consider that with Insteon you have signal bursts lasting ~ 50 msec.

Using 3 hops you have 4 of those bursts, assuming no ack and a standard message. Which of the hops are then displayed. Possibly all, but your eye would not be able to retain any except the last hop's, very last signal packet ( assuming it "holds" the last reading until the next begins).

[arw01]

I will endevour to make a video of it in action so you can see for yourself how it works. I have a nearby switchlinc that lets me put a signal on the line at will.

[jwshome]

Thanks for all of the comments. There are no CFL bulbs involved, so that's not the problem, and I've confirmed the wiring is correct (the LED lights work properly). Also, I forgot to mention: the two switches can communicate with each other perfectly fine (I've set them up in a typical 3-way functionality, where turning off one turns off the other, and vice versa). They just cannot communicate with the PLM reliably.

 

One other piece of information that may be relevant: the communication problem happens near the end of the setup sequence every time. In other words, the following happens:

 

1. I delete the device from the iSY

2. I put the iSY into linking mode

3. I hold the 'Set' button on the Switchlinc in fopr three seconds until linking begins.

4. Sometimes the iSY does not pick up on the linking. But it usually does

5. The iSY says it is writing links, which show in the samll font underneath the main status message on the Admin tool.

6. Then the admin tool keeps going, with the progress bar moving and the admin tool still saying "DOn't do anything to any devices until this is finished"

7. Then, near the end of this process, it says it canot communicate with the device.

 

Is it conceivable that I am running into a capacity problem with the iSY or the PLM? Could running out of space be the cause of the problem? Would the symptoms be the same?

 

Also, one more question: how do I wire up the pigtail?

[LeeG]

It is not a capacity problem. Disconnect a SwitchLinc from the existing wiring and physically remove it from the junction box. Connect the Black wire of the Appliance cord with the Black wire on the SwitchLinc. Connect the White wire of the Appliance cord with the White wire on the SwitchLinc. Cap the SwitchLinc Red wire with a wire nut. Plug the SwitchLinc into the PLM outlet and add the SwitchLinc to the ISY.

 

I think it likely the SwitchLinc will add successfully. If not run Tools | Diagnostics | Event Viewer at LEVEL 3. Repeat the device add and post the event trace.

 

If the SwitchLincs were linked together with the Set button use the "remove existing links" when adding both SwitchLincs. Once added successfully use an ISY Scene to cross-link the two SwitchLincs.

[oberkc]

Regarding the pigtail, I believe the suggestion is to do somthing like sacrificing a cheap extension cord. Cit off the female end and connect it to the problem device. Plug it into the same outlet as the PLM. If this solves the problem, it is near certain that you are having comm issues.

[jwshome]

Excellent, thanks for the advice. I will give it a try.

[arw01]

I will endevour to make a video of it in action so you can see for yourself how it works. I have a nearby switchlinc that lets me put a signal on the line at will.

 

I have uploaded some videos showing the display of the unit in action.

 

1st video shows in the same outlet as with a filter linc using my PC to send an off command to another room via the admin console.

 

http://www.webbyhome.com/images/VID_20130614_201748.mp4

 

Second video shows after I plug the filter linc in the same outlet that filters the washing machine.

 

http://www.webbyhome.com/images/VID_20130614_201824.mp4

 

Third video shows same outlet with the washing machine plugged straight without the filter.

http://www.webbyhome.com/images/VID_20130614_201952.mp4

Then I moved to the exact same outlet as my PLM works on, note the difference in signal strength.

 

http://www.webbyhome.com/images/VID_20130614_202308.mp4

 

I then moved it to another outlet and used a control linc, you can see we get even more power with the controlinc vs the plm.

 

http://www.webbyhome.com/images/VID_20130614_202443.mp4

 

Moved to another outlet across the room, used the control linc again and you can see the drop in signal.

http://www.webbyhome.com/images/VID_20130614_202555.mp4

 

 

Alan