Diagnose OutletLinc communication

[ds7771]

Hi,

I am trying to diagnose a problem communication on a specific outletlinc relay device.

 

I am wondering what OL and RR codes mean in the ISY event viewer. All of the devices in my network will return "some" value for OL and RR. However this particular outletlinc device which is having problems, returns only zeroes for those values, as you can see below. It also consistently returns Hops Left = 1 (never 0), which I realize is not the best, but at least it is always consistent.

 

It has this behavior even with an access point as well as several nearby dual band devices confirmed to be on opposite phases using 4 tap test.

 

What do all those zeroes mean? Is the outletlinc device defective?

 

Thanks!

 

Tue 01/14/2014 02:31:02 PM : [iNST-TX-I1 ] 02 62 25 1B 12 0F 19 00

Tue 01/14/2014 02:31:02 PM : [iNST-ACK ] 02 62 25.1B.12 0F 19 00 06 LTSREQ (LIGHT)

Tue 01/14/2014 02:31:02 PM : [iNST-SRX ] 02 50 25.1B.12 25.28.3C 27 0B 00 (00)

Tue 01/14/2014 02:31:02 PM : [std-Direct Ack] 25.1B.12-->ISY/PLM Group=0, Max Hops=3, Hops Left=1

Tue 01/14/2014 02:31:02 PM : [ 25 1B 12 1] ST 255

Tue 01/14/2014 02:31:02 PM : [ 25 1B 12 1] OL 0

Tue 01/14/2014 02:31:02 PM : [ 25 1B 12 1] RR 0

[Xathros]

OL = On Level

RR = Ramp Rate

 

Both only apply to dimmer type devices.

 

-Xathros

[ds7771]

Thank you, that makes sense. (Does UDI publish a glossary of codes somewhere?)

 

Based on your explanation, it seems that RR and OL would not apply to an outletlinc relay (2473) device, which explains why this device returns RR=0 and OL=0.

Howver, I have 2 other 2473 devices, as well as a switchlinc relay (2477S) which all return OL=255 and RR=28.

 

Why do the other non-dimming devices return values for RR and OL? Still thinking there is something wrong with this outletlinc device.

Thanks again.

[Xathros]

These codes are Insteon. UDI May have a listing but I bet Smarthome would be the better source for that info. A relay will have an Onlevel of either 0 (off) or 255 (on). Not sure what the range of values is for Ramp Rate but that should make no difference to a relay type device.

 

You may be able to fine some useful info here: http://www.insteon.com/pdf/insteondetails.pdf

 

-Xathros

[shannong]

All my of 2473 devices also return OL=255, RR=28.

 

The RR=28 is interesting since the normal values are between 0-31 and you wouldn't expect a non-dimming device to be somewhere in the middle. Mine work fine though.

 

Have you tried a factory reset on the 2473 using the airgap switch on bottom?

[LeeG]

ds7771

 

The OL and RR values have no meaning.

 

If you want to change them issue these REST commands using your ISY IP address

 

http://192.168.2.3/rest/nodes/25 1B 12 1/set/OL/255

 

http://192.168.2.3/rest/nodes/25 1B 12 1/set/RR/28

 

I don't think it will change the problem you are trying to solve.

[ds7771]

Thanks so much... I will ignore the RR and OL values as a symptom of my communications problems.

 

Focusing back on hops left count=1, it seems that this outletlinc device has to make 2 retransmissions if I am understanding hop count correctly.

 

Why does this happen?

 

Now, to fill in the picture, I have several 2 band devices within 10-15 foot proximity of this poorly communicating outletlinc.

 

But working against me, all of my gang boxes are old school metal boxes, and to add insult, all of my walls are made of old school plaster and metal lathe. So I have plenty of opportunity for RF interference.

 

Do you think it is a correct assumption that this outletlinc device is on the opposite phase, and also not within good RF range of other nearby devices due to interference, and therefore needs to make so many RF retransmissions to another nearby device in order to reach the PLM on the other phase?

 

If that is the case, do you think that a signalinc phase coupler at the panel would help with phase bridging so that retransmissions are reduced? I have already tried locating an Access Point nearby to no effect.

 

To my knowledge, I do not think the outletlinc relay devices are capable of indicating phase bridging using the 4 tap test, so I have no way to tell what is going on with this device.

Thanks.

[shannong]

Did you try doing a physical reset of the device using the gap switch?

 

The ultimate test for the question "is it the device or communications issue with its location in my network" is to swap it with a known good working unit.

[shannong]

Thanks so much... I will ignore the RR and OL values as a symptom of my communications problems.

 

Focusing back on hops left count=1, it seems that this outletlinc device has to make 2 retransmissions if I am understanding hop count correctly.

 

Why does this happen?

 

Now, to fill in the picture, I have several 2 band devices within 10-15 foot proximity of this poorly communicating outletlinc.

 

But working against me, all of my gang boxes are old school metal boxes, and to add insult, all of my walls are made of old school plaster and metal lathe. So I have plenty of opportunity for RF interference.

 

Do you think it is a correct assumption that this outletlinc device is on the opposite phase, and also not within good RF range of other nearby devices due to interference, and therefore needs to make so many RF retransmissions to another nearby device in order to reach the PLM on the other phase?

 

If that is the case, do you think that a signalinc phase coupler at the panel would help with phase bridging so that retransmissions are reduced? I have already tried locating an Access Point nearby to no effect.

 

To my knowledge, I do not think the outletlinc relay devices are capable of indicating phase bridging using the 4 tap test, so I have no way to tell what is going on with this device.

Thanks.

 

Hops Left does not indicate a problem by itself. Messages can have a Max hops value between 0 and 3 depending on it's distance from the sender the PLM in this case. When a message is sent without an ACK the original sender increments the Max hop count one each time up until 3. Each time a device repeats the signal in the path the Hops Left count is decremented by one until zero where it is not forwarded by the next receiving device.

 

No reason to guess if the dual-band devices are on different phases as desired. It's easy to look and see. Each row of breakers is on the same phase. The next row down/up is on a different phase. All odd breakers are on the left column. All even breakers are on the right column. There are two phases on the panel. So breaker 1&2 are on the same Phase1. 3&4 are on the same Phase2. 1&3 are on both Phase1 and Phase2. 2&4 are on both Phase1 and Phase2. 1&5 are on the same Phase1. 1&7 are on both Phase1 and Phase2. It's easier to understand if you look at a breaker box as you read the description.

[Brian H]

Best to look at the chart found on most breaker boxes covers.

Some that can take half sized breakers can have two on the same phase adjacent in the row.

 

Of course. Most Dual Band Modules have a built in communications test. That can also be used to verify phase and communications, Between modules.

[LeeG]

The OutletLinc Relay is NOT Dual Band. The OutletLinc Relay is a Responder Only device. Note in the OutletLinc Relay User Guide that there are no instructions for linking the OutletLinc Relay as a Controller.

 

It would be helpful to know what is working and what is not. Are Direct On/Off commands always reliable and Scenes are not. Is it communication from the PLM or another Controller that is not working well.

 

With metal boxes and some metal in the walls I expect RF is not what one would like. The Hops Left count you are seeing in an Event Trace would be for messages sent to the OutletLinc from the PLM. Direct On/Off messages from the PLM are sent with Max Hops=3. How many devices have to repeat the message before the OutletLinc receives the message determines the Hops Left count. A consistent Hops Left=1 is okay. If the Hops Left=x count is jumping around indicates a problem. It is difficult to analyze the problem without knowing more about what is and is not working. Knowing if the circuit powering the OutletLinc is on the same phase or opposite would help as well.

[shannong]

Best to look at the chart found on most breaker boxes covers.

Some that can take half sized breakers can have two on the same phase adjacent in the row.

.

 

That's how I made room for my Signalinc's.

[ds7771]

Thanks so much for all the suggestions.

 

LeeG, I did not realize that outletlinc relays are not dual band. I assume that means powerline only communication, no RF. Is that correct? Does this mean that an Access Point will have no effect on an outletlinc relay device?

 

Here is a more descriptive view of what is going on:

I have 2 outletlinc relays (2473) in a room. Floor lamps are plugged into them.

One switchlinc relay (2477S) is installed at the entrance to the room. The switchlinc relay is wired to neutral, but carries no wired load. Instead, the switchlinc relay is part of an ISY scene consisting of the 2 outletlinc relays. The switchlinc is set as controller for the 2 outletlinc relays set as responders.

 

One of the outletlinc relays consistently reports hops left=2. The PLM is actually plugged directly into this device. The problem device is the other outletlinc relay in the room. The problem device consistently reports hops left=1.

 

Here is the communication problem:

Lets say I turn on and off the scene 7 times.

6 out of 7 times the devices all respond correctly.

However, 1 out of 7 times, the problem outletlinc (the one reporting hops left=1) will not respond to the "on" command or "off" command. It will remain in the current state, while the other devices will change their state.

 

But really strangely, the ISY event viewer does report the correct state for the command, either ST=255 for "on" state or ST=0 for "off" state. The ISY control panel will also show the device being in the correct state. However, that particular outletlinc device is actually not in the correct state. For example, when I send an "off" command to the scene, the outletlinc will appear to be in an off state according to the ISY, but it will still physically remain on.

 

Again, this only happens 1 out of 7 times. It gets it right the other 6 times.

Also, this behavior is exactly the same, regardless of whether the command is sent from the ISY or by tapping the switchlinc paddle.

 

Finally, it makes no difference if I place an Access Point directly next to the problem outletlinc device. Behavior does not change.

 

I am unable to check the phases on the panel right now, but I will do so in the morning. Its complicated by the fact that one of the circuits is on a sub-panel off the main one.

 

Really appreciate all the help.

Thanks!

[LeeG]

I would start with the idea that the OutletLlinc that is intermittent is on the opposite phase. Plug an Access Point into the OutletLinc and run the 4 tap Set button test from that Access Point. What color is the LED on PLM blinking?

 

I’m starting with the assumption the SwitchLinc and PLM are on the same phase, opposite to the OutletLinc that is intermittent. Since the OutletLinc fails to both On and Off commands from the SwitchLinc and the PLM is it likely going through a Dual Band device to get to the other phase. The color of the PLM LED will verify the assumption.

 

The ISY marking the OutletLinc On/Off is due to the PLM receiving the SwitchLinc turning On/Off so the ISY assumes the responders reacted as linked. Nothing goes back to the PLM regarding what the OutletLinc actually did in response to the SwitchLinc.

 

Yes, the OutletLinc Relay communicates over the powerline.

[ds7771]

I would start with the idea that the OutletLlinc that is intermittent is on the opposite phase. Plug an Access Point into the OutletLinc and run the 4 tap Set button test from that Access Point. What color is the LED on PLM blinking?

 

I plugged an Access Point into the problem outletlinc device and ran the 4 tap test from all three devices.

Here are the results:

 

4-tap on the Access Point:

AP=blinking green

PLM=blinking green

Switchlinc=blinking red

 

4-tap on the PLM:

PLM=blinking green

AP=blinking green

Switchlinc=blinking green

 

4-tap on the Switchlinc

Switchlinc=blinking green

PLM=blinking green

AP=blinking red

 

Im not sure what to make of those results, but I suppose they mean that the PLM and outletlinc are on opposite phases, however, the Switchlinc and outletlinc are on the same phase. So the PLM has to hop through the switchlinc to get to the outletlinc. Perhaps that is causing the problem. But if that is true, then why does the 4-tap test on the PLM show all devices blinking green? Seems like the PLM should be able to get directly to the outletlinc.

 

How do I fix it? Should I locate the AP somewhere that is opposite phase of the switchlinc? Would I get better results by installing a signalinc at the breaker panel?

Thanks for keeping with me on this.

[LeeG]

The color of the LEDs on the receiving devices indicates same(Red)/opposite(Green) phase. The fact that the receiving devices are blinking indicates receiving the test RF message.

 

The OutletLinc and SwitchLinc are on the same phase, the PLM is on the opposite phase.

 

The fact that the SwitchLinc will not control the OutletLinc at the time of failure indicates a powerline problem since they are on the same phase. The PLM not controlling the OutletLinc could be either but since the SwitchLinc will not control being on the same phase I think this is a powerline issue.

 

Is the SwitchLinc and OutletLinc a long distance apart as far as the wiring is concerned. It could be distance but I control a detached garage some 300 wire feet away over the powerline. What else is being powered on the circuit that supplies the OutletLinc? Perhaps something is generating noise or attenuating the powerline signal.

 

The wired SignaLinc is not a favorite of mine. It is passive so a low level powerline signal is coupled to the other phase at that lower level. I prefer a pair of Access Points for coupling.

[ds7771]

Think you are right about power line noise.

 

I relocated both the PLM and the AP to entirely different outlets in the house, and made sure they are phase bridging each other.

 

Now I have hops left=2 count on all the outletlinc devices, including the problem one, and I am getting consistent behavior with controlling their state.

 

Thanks for helping me think through this problem, and learn more about insteon communication.

 

In a few months I will embark on some major home renovations, which will include a new breaker panel for 200 amp service. I intend to ask the electrician to use this opportunity to isolate as many circuits as possible on separate breakers, to minimize the chance of overload and noise on any particular one. I assume he has to do this by code anyway, but I want to go above and beyond. I also plan on adding more insteon devices as part of the renovation.

 

Are there any tips I should consider, or best practices to follow when rewiring the panel, to maximize good Insteon communication? Now is the chance to do it!

 

Thanks.