Current Cost Support Forum

Leaving my EnviR or a Chanel with no IAM worked spike free for two days but at 1am this morning I got a 60kw spike.

So it may reduce the corruptions output by EnviR's it is not 100% fix

Automan.

The ENVIR datastream has no error checking or error correction capability. So an error in the watts value would simply pass through as a valid number. These errors could come from the 433MHz RF noise or interference or in the serial data transmission to the bridge / host pc. GPS units use a XOR byte 'CRC' checksum to validate the sent data matches what was received and if it doesn't match the information is discarded. Error correction bytes could be used to ensure the data stream is intact and correct errors.

Try a couple of things:

1) Keep the EnviR monitoring a single appliance, eg '1' and see if you only get spikes on channel 1 'EV', try this for a few days.
2) Keep the EnviR monitoring a non registered appliance channel, eg '4', so it displays '---' then see if you get any spikes on the other channels.

I expect in case (1) you will only see spikes on the channel you are monitoring, and in case (2) you will see a lot fewer spikes.

I have some measurement problems with recently buyed IAM modules. I buy a kit of 4 IAMs (EU Type) and I have linked to my Envi Monitor on channels 1 to 4.

Channel 1 (sensor 1) is for the washing machine.
Channel 2 (sensor 2) is for the dishwasher
Channel 3 (sensor 3) is for the living room air conditioner
Channel 4 (sensor 4) is for the bedroom air conditioner

For sensors 3 and 4, I connect to the output of the IAM a X10 controlled appliance on/off module. For this tests, this appliance modules was always turned off, so the energy consumption for this IAMs are 0 watts.

I use the serial output of the Envi to connect to a PC with Linux Debian distribution and a program made by myself in C language to upload data to pachube.

The problem is the reading from the IAM's modules sometimes reports spikes (about 32767 watts!!) through the envi monitor serial output. I set some debugging facilities to my program for investigate the cause for this spikes and with this information and with the previous posts, it seems to be a problem with the IAMs or with the Envi monitor processing.

I detect the fail when measurements are bigger than 1000 watts, in my case the appliances are turned off (Air Conditioner) and standby (washing machine and dishwasher).

Normally it takes some time to fail, next you can see the serial output and the time from I turned on my Envi monitor to get a fail on the serial output:

Time to fail = 1h 10 min.
Fail: Sensor 4 (ch1=0; ch3=32767)
Real condition: No consumption on this IAM
Message: <msg><src>CC128-v0.11</src><dsb>00905</dsb><time>19:11:00</time><tmpr>29.6</tmpr><sensor>4</sensor><id>00448</id><type>1</type><ch1><watts>00000</watts></ch1><ch3><watts>32767</watts></ch3></msg>

Time to fail = 1h 15 min
Fail: Sensor 3 (ch1=4095; ch2=32767; ch3=32766)
Real condition: No consumption on this IAM
Message: <msg><src>CC128-v0.11</src><dsb>00905</dsb><time>21:08:00</time><tmpr>29.6</tmpr><sensor>3</sensor><id>01279</id><type>1</type><ch1><watts>04095</watts></ch1><ch2><watts>32767</watts></ch2><ch3><watts>32766</watts></ch3></msg>

Time to fail = 1h 58 min
Fail: Sensor 2 (ch1=4095; ch2=32767; ch3=32767)
Real condition: The appliance connected to this IAM was in standby mode.
Message: <msg><src>CC128-v0.11</src><dsb>00905</dsb><time>23:09:50</time><tmpr>29.4</tmpr><sensor>2</sensor><id>00180</id><type>1</type><ch1><watts>04095</watts></ch1><ch2><watts>32767</watts></ch2><ch3><watts>32767</watts></ch3></msg>

Time to fail = 1h 17 min
Fail: Sensor 4 (ch1=4095; ch2=32767; ch3=32767)
Real condition: No consumption on this IAM
Message: <msg><src>CC128-v0.11</src><dsb>00906</dsb><time>13:01:40</time><tmpr>26.9</tmpr><sensor>4</sensor><id>00448</id><type>1</type><ch1><watts>00000</watts></ch1><ch3><watts>32766</watts></ch3></msg>

Time to fail = 1h 57 min
Fail: Sensor 3 (ch3=32767)
Real condition: No consumption on this IAM
Message: <msg><src>CC128-v0.11</src><dsb>00906</dsb><time>15:09:50</time><tmpr>27.9</tmpr><sensor>3</sensor><id>01279</id><type>1</type><ch3><watts>32767</watts></ch3></msg>

Time to fail = 1h 48 min
Fail: Sensor 1 (ch1=1; ch3=32767)
Real condition: No consumption on this IAM
Message: <msg><src>CC128-v0.11</src><dsb>00906</dsb><time>17:15:50</time><tmpr>27.8</tmpr><sensor>1</sensor><id>02468</id><type>1</type><ch1><watts>00001</watts></ch1><ch3><watts>32767</watts></ch3></msg>

somebody knows how to solve this problem?

Have a look at my post on this thread: viewtopic.php?f=11&t=1996&start=10

Especially the bit about "LOCATIONS".

In my data (I have 26 IAMs!) I find that IAMs occasionally "jump" to the wrong <sensor> channel. I identify each IAM using the radioID number and NOT the <sensor> number. I sometimes find that a single radioID will "jump" to a <sensor> number which it was not paired with. I believe this is one outcome of an RF collision.

The fix for this problem is fairly easy in my code: I just identify each transmitter using its radioID and I effectively ignore the <sensor> number.