Jump to content

Can't control device with scene, but direct control works


chris.com

Recommended Posts

I moved to a new house about six months ago and went with an ISY 994i/IR Pro. (Currently running v.4.5.4) This is my first real foray into home automation since messing around with X10 20 years ago. I replaced every switch and dimmer in the house with Insteon switches and dimmers. (2477D's and 2477S's, all v.45) I added several 8-button keypads. (2334-222)  I successfully used the ISY interface and got the keypad buttons into scenes with various lights.

I soon learned that if I wanted to use a program on the ISY to control a device, I really needed to control the scene. If I controlled the device directly, the button on the keypad wouldn't reflect the change in the device's status. So I decided to put all my devices into scenes whether I had a keypad button associated with it or not. All my programs control scenes rather than devices. That way, if I decide to make a device part of a scene later on, I don't have to go back and modify my programs; they'll already be talking to the scene.

I have four devices, however, that cannot be controlled with a scene. (Three 2477D v.45's and one 2477S v.45.)  I have created scenes for these devices. They are the only devices in those scenes and they are controllers.

When I send an on or off command to the scene, nothing happens. However, when i send an on or off command to the device itself, it works flawlessly. I have many other devices set up exactly the same way and I am able to control them with their scene.  While doing this, I watch the event log and the command that goes out appears to be identical whether I'm talking to the scene or to the devices. Strangely one works; the other doesn't.

I have tried removing them from the scene, creating brand new scenes, and re-adding them. No luck. For the life of me, I cannot see any difference in the way these switches are set up versus other devices and scenes that are working just fine.

The ones that fail, do indeed fail the scene test under diagnostics.

 

Here's a scene that's failing.

Fri 02/03/2017 10:30:59 : [GRP-RX      ] 02 61 44 13 00 06 
Fri 02/03/2017 10:31:00 : [CLEAN-UP-RPT] 02 58 06 
Fri 02/03/2017 10:31:00 : [INST-SRX    ] 02 50 41.98.C2 40.51.8F E5 13 FF    LTOFFRR(FF)
<html><font color="red">----- Kate's Bathroom Light Test Results -----</font></html>
<html><font color="red">[Failed]</font> Kate's Bathroom Light (41 98 C2 1)</html>
<html><font color="red">----- Kate's Bathroom Light Test Results -----</font></html>

Here's a scene that's working fine for comparison.

Fri 02/03/2017 09:09:01 : [GRP-RX      ] 02 61 46 13 00 06
Fri 02/03/2017 09:09:01 : [CLEAN-UP-RPT] 02 58 06
Fri 02/03/2017 09:09:01 : [INST-SRX    ] 02 50 41.B1.0A 40.51.8F 65 13 46    LTOFFRR(46)
Fri 02/03/2017 09:09:01 : [Std-Cleanup Ack] 41.B1.0A-->ISY/PLM Group=0, Max Hops=1, Hops Left=1
<html><font color="red">----- Loft Light Test Results -----</font></html>
<html><font color="red">[Succeeded]</font> Loft Light (41 B1 A 1)</html>
<html><font color="red">----- Loft Light Test Results -----</font></html>
Fri 02/03/2017 09:09:09 : [INST-ACK    ] 02 62 00.00.46 CF 13 00 06          LTOFFRR(00)

I have read many forum threads about device interference, and the need to install a phase coupler or range extender. I don't think that's the issue here. If I send a command directly to the device, it works 100% of the time. If I send a command to the scene, it works 0% of the time.

I have performed a "Restore PLM." That seemed to go okay. No indications of any errors in the event log during that process. But it did not yield any change in the situation.

I have a feeling I'm missing something obvious and once I find out what it is, I 'm going to feel very silly about myself. Any ideas come to mind?

Edited by chris.com
Link to comment

Just a thought but maybe try adding one in the scene at a time and see if you can possibly identify the location or device giving you a issue? I had a dual band 3 way dimmer that would not work once 2 of the devices were added. I only came up with there was something in the circuit that seemed like it was extending too far or some kind of a loop. Strange enough single band devices don't do it.. Almost exactly opposite of what I would think. 

Link to comment

Just a thought but maybe try adding one in the scene at a time and see if you can possibly identify the location or device giving you a issue? I had a dual band 3 way dimmer that would not work once 2 of the devices were added. I only came up with there was something in the circuit that seemed like it was extending too far or some kind of a loop. Strange enough single band devices don't do it.. Almost exactly opposite of what I would think. 

 

I probably wasn't clear enough in my wording on the original post. In the case of these four devices, there are four difference scenes... one for each devices. Each device is the only device in its scene.

Link to comment

Hi Chris and welcome to the UDI Forums

 

Try a "restore device" to one of the keypads and repeat your test on its keys... right click on the kepad in the main device tree and select restore device.

 

Paul

 

Thanks Paul! That did the trick. Thanks so much for the tip. Now I need to go read up on "restore device" and learn more about exactly what that does.

 

Many many thanks,

-Chris

Link to comment

You're welcome, glad it worked. Restore device copies the isy onboard version of links back to the device. These links tell the device the plm's and other devices links so that they talk to each other. It's something to try when you have problems like you describe initially.

 

 

Sent from my iPhone using Tapatalk

Link to comment

I just used "restore device" to fix a weird problem, too.  There was clearly a bad link somewhere between a motion sensor and two SwitchLincs based on some erratic behavior, but dumping the device links and comparing them to the ISY showed no meaningful differences (all either Identical or Ignore).  I did notice that there were several "ignore" links in the middle that I suspect were inactive entries, so I did a restore to rewrite the link table and it fix the problem.  Luckily, it's usually quick and painless.

 

Rob

Link to comment

As per Paul's advice I like to do load links from the device, itself, and then click on compare to find the offending link, so I know who to spank, first. :)

 

I find your idea quite interesting. 

 

I poked around, looking for options similar to what you're describing. I'm assuming you mean  device->diagnostics->show device links table

 

When I do that on a device, I get a list of hexadecimal codes. (No friendly names unfortunately,) I see the "compare" button. When I do this to any of my devices, they almost always have at least one "[Record mismatch]." But I don't know if this is cause for alarm and, if it was, I wouldn't know what to do about it.

 

If you have a few minutes and wouldn't mind elaborating, I'd be very interested in knowing what sort of information is to be gleaned from this and what to do with said information.

 

If it's too long and complicated to explain, don't feel obligated. :)

 

Thanks so much,

-Chris

Link to comment

FWIW, I would suggest that if you're just getting started with your ISY, dive into some of the other features of the ISY that get you automating.

 

How Insteon & scenes work "under the hood" is a worthy area of learning, but for later. Investing time in it right now will not give you dividends as much as learning about the ISY's programming environment, modules, etc

 

I would remember this technique for future situations that are similar, eg devices that respond partially but not fully. Restore device or factory resetting and restoring devices, in that order, are both relatively easy to do in most cases and solve a lot of problems.

 

Paul

Link to comment

I find your idea quite interesting. 

 

I poked around, looking for options similar to what you're describing. I'm assuming you mean  device->diagnostics->show device links table

 

When I do that on a device, I get a list of hexadecimal codes. (No friendly names unfortunately,) I see the "compare" button. When I do this to any of my devices, they almost always have at least one "[Record mismatch]." But I don't know if this is cause for alarm and, if it was, I wouldn't know what to do about it.

 

If you have a few minutes and wouldn't mind elaborating, I'd be very interested in knowing what sort of information is to be gleaned from this and what to do with said information.

 

If it's too long and complicated to explain, don't feel obligated. :)

 

Thanks so much,

-Chris

Each line contains an address (link) of another Insteon device.

Minimum you have to have a link to your PLM or ISY would be blind.

You will also have a line for each scene the device contains.

At the end you will have a terminator line with all 00s.

 

If you have a multiple node device like a stat you will have more links due to setpoint, temp, humidity, fan etc,,,,

 

If the compare disagrees this indicates a disagreement between ISY's records and the links inside the device. The Restore is the fix for that.

 

I can't tell you much more without getting into the bit twiddling nitty-gritty and I can't say I  have spent enough time on it myself to understand it all. As per Paul above, it's not something to pursue, at this point or hopefully, even forever. :)

Link to comment

I use 'compare' after 'show device links table' when I find something behaving not as expected in my Insteon network. When the result displays one or more 'records mismatch' lines, performing a 'restore device' always solves my problem. I believe power glitches can cause corrupt device link tables.

Link to comment

I have read the section in the manual about device links... I'm a very conceptual learner so sometimes I have trouble really understanding something with the cursory and technical explanations often found in manuals. Many thanks to everyone in here who put their own spin on it. I totally get this now. The combination of everyone's explanation got it through my thick skull! You have my gratitude!! :-)

 

-Chris

Link to comment
Guest
This topic is now closed to further replies.

×
×
  • Create New...