Corrupt Insteon Messages in Specific Location

[auto_mate]

I have 2 dual band keypadlincs and when I install one in a particular wallbox I get corrupt communication problems.  The first byte of the from address is corrupt when the device sends a message.  There is another single band device in that box that does not show corrupt bytes.  When I move the device to a different location, the corruption goes away.  This happens with two different dual band keypadlincs.  I have tried unplugging everything else on that circuit I could find, and it did not solve the problem.  I have approx 50 insteon devices, most are single band, and this location is the only one that has shown this problem and has consistent communication issues.

 

For example:

 

[iNST-TX-I1  ] 02 62 3D 87 2A 0F 19 00
[iNST-ACK    ] 02 62 3D.87.2A 0F 19 00 06          LTSREQ (LIGHT)
[iNST-SRX    ] 02 50 4C.87.2A 0C.A6.AD 27 00 00           (00)
[std-Direct Ack] 4C.87.2A-->ISY/PLM Group=0, Max Hops=3, Hops Left=1

 

 

Has anyone seen this before?  How does a corrupt message get through the PLM and into ISY without having a correct checksum byte?  How could the checksum be correct if the first address byte is corrupted?

Thanks

[Techman]

I have 2 dual band keypadlincs and when I install one in a particular wallbox I get corrupt communication problems.  The first byte of the from address is corrupt when the device sends a message.  There is another single band device in that box that does not show corrupt bytes.  When I move the device to a different location, the corruption goes away.  This happens with two different dual band keypadlincs.  I have tried unplugging everything else on that circuit I could find, and it did not solve the problem.  I have approx 50 insteon devices, most are single band, and this location is the only one that has shown this problem and has consistent communication issues.

 

For example:

 

[iNST-TX-I1  ] 02 62 3D 87 2A 0F 19 00

[iNST-ACK    ] 02 62 3D.87.2A 0F 19 00 06          LTSREQ (LIGHT)

[iNST-SRX    ] 02 50 4C.87.2A 0C.A6.AD 27 00 00           (00)

[std-Direct Ack] 4C.87.2A-->ISY/PLM Group=0, Max Hops=3, Hops Left=1

 

 

Has anyone seen this before?  How does a corrupt message get through the PLM and into ISY without having a correct checksum byte?  How could the checksum be correct if the first address byte is corrupted?

 

Thanks

 

You might want to try doing a factory reset on the KPL , then a restore device from the ISY.

 

What type of load is connected to the KPL?

[auto_mate]

You might want to try doing a factory reset on the KPL , then a restore device from the ISY.

 

What type of load is connected to the KPL?

 

Tried it.  4 incandescent bulbs 60W.  The errors happen even if the load is off.

 

The device is within 15 feet of the PLM plus access point and other wireless devices work reliably.

[paulbates]

Welcome to the forums

 

Fitting 2 keypads in a dual gang box can be a tight squeeze. Is there a lot of pressure on them to get them installed and also when they are in?

 

Paul

[jerlands]

[iNST-TX-I1  ] 02 62 3D 87 2A 0F 19 00
[iNST-ACK    ] 02 62 3D.87.2A 0F 19 00 06          LTSREQ (LIGHT)

 

I'm not sure about this but I believe 62 indicates the record is in the PLM

 

[iNST-SRX    ] 02 50 4C.87.2A 0C.A6.AD 27 00 00           (00)
[std-Direct Ack] 4C.87.2A-->ISY/PLM Group=0, Max Hops=3, Hops Left=1

 

and these two records are coming from the device.

 

What does ISY show as the address for the device?

 

 

 

Jon...

[auto_mate]

Welcome to the forums

 

Fitting 2 keypads in a dual gang box can be a tight squeeze. Is there a lot of pressure on them to get them installed and also when they are in?

 

Paul

 

No they are not installed at the same time.  I am saying the problems occur if either one is installed so it isn't due to a faulty unit.

 

[iNST-TX-I1  ] 02 62 3D 87 2A 0F 19 00

[iNST-ACK    ] 02 62 3D.87.2A 0F 19 00 06          LTSREQ (LIGHT)

 

I'm not sure about this but I believe 62 indicates the record is in the PLM

 

[iNST-SRX    ] 02 50 4C.87.2A 0C.A6.AD 27 00 00           (00)

[std-Direct Ack] 4C.87.2A-->ISY/PLM Group=0, Max Hops=3, Hops Left=1

 

and these two records are coming from the device.

 

What does ISY show as the address for the device?

 

 

 

Jon...

 

 

Yes as you can see ISY sends the query to 3D.87.2A (its real address) and the device responds as 4C.87.2A.  Sometimes the messages from the device will be valid and show the 3D.87.2A address.  But many times the first byte will be wrong and it is not always 4C but most of the time it is.  3D and 4C are different by more than just a bit or two.  If the message is corrupt the PLM should discard it since the checksum would not be valid.

[jerlands]

The trace shows poor comms (Hops Left=1.)  Does the single band device also show bad comms or is it on the same circuit?

 

 

Jon...

[auto_mate]

Yes there are 2 other single band devices on the same circuit and they usually show 1 hops left as well. But even if the communication is unreliable it does not explain how the corruption is deterministic (almost always 4C) and how it passes the PLM checksum test.

[jerlands]

Is the box metal?  Odd that it's limited to this one particular box and apparently affects either KPL.  Seems to be either wiring or one of the other devices.

 

 

Jon...

[auto_mate]

No it is plastic. 

 

Maybe if a repeater is corrupting the message and fixing the checksum when it adjusts hop count, but still hard to believe.

[Brian H]

[iNST-TX-I1  ] 02 62 3D 87 2A 0F 19 00

[iNST-ACK    ] 02 62 3D.87.2A 0F 19 00 06          LTSREQ (LIGHT)

 

I'm not sure about this but I believe 62 indicates the record is in the PLM

 

 

The 02 62 command is the command sent to the PLM from the ISY controller. To sending a Standard or Extended Insteon command to an Insteon module and the PLM sending an ACK back to the ISY controller to indicate it was a valid command and processed. If the reply had a 15 {NAK} as the last digit. The command was invalid or not processed correctly.

[auto_mate]

Any other insights/suggestions?

[Techman]

Any other insights/suggestions?

 

What are the details of your PLM?

 

Do you have any UPS's, or anything else with a power supply on the same circuit as the KPL or the PLM?

[auto_mate]

PLM is single band but has access point attached.  Sure there are some power supplies but most are behind filterlincs, no UPS.  And like I said there are 50 devices spread throughout the house and only these new KPLs in this specific box are showing the problem.  They should be more reliable since they are dual band but they are less reliable than the single band device in the same box.

[stusviews]

You haven't indicated the actual difficulty, only unexpected reports. What about functionality?

[auto_mate]

When the device initiates the message, usually it isn't received by the PLM.  This is a problem for running programs and keeping status.  But other devices do respond to group messages most of the time.  When the PLM initiates a message, the reply is often corrupted but received.

[Brian H]

Are your 2443 Access Points the hardware revision one or two series?

I have had a few hardware two versions get flaky and do strange things. They where built on the same hardware revision one base 2413 main board that have the power supply issues.

[Techman]

PLM is single band but has access point attached.  Sure there are some power supplies but most are behind filterlincs, no UPS.  And like I said there are 50 devices spread throughout the house and only these new KPLs in this specific box are showing the problem.  They should be more reliable since they are dual band but they are less reliable than the single band device in the same box.

 

 

Have you tried running a diagnostics / compare links table on the KPL you're having an issue with?

 

Being that your PLM and most of your devices are single band I would focus on a communications issue. 

[auto_mate]

Are your 2443 Access Points the hardware revision one or two series?

I have had a few hardware two versions get flaky and do strange things. They where built on the same hardware revision one base 2413 main board that have the power supply issues.

 

They are rev 1.  I removed the access point at the PLM and the corrupt messages went away.  The reliability still isn't great.  So I will have to consider getting a new repeater to plug into the PLM or is there a way to repair the old access points?  Thanks for pointing me toward that.

[Brian H]

The version 1 2443 Access Points are built on the older power transformer designed unit. Used in a few of the earlier controllers.

 

I have not seen any specific repair information on them.

Maybe filter capacitors could be changed. They are completely different  capacitors from what we change in the 2413S and version 2 2443 Access Points.

I had a V1 fail with strange happenings.