I have just started work on a binding that uses the CUL transport to interact with eq-3 MAX! devices.
Work is being done here: https://github.com/cyclingengineer/openhab/tree/maxcul-binding and I have created and issue to track this here: https://github.com/openhab/openhab/issues/1085
Currently I have written binding configuration code and some bits that will be required for pairing mode. If you want more details then please ask.
I'll try and remember to update this thread as and when I have things completed and ready for testing so I can get feedback.
Thanks,
Paul
It looks like my binding is causing no problems build wise. 'Unstable' build result seems to be down to some other tests failing (not a regression on my part).
--
You received this message because you are subscribed to the Google Groups "openhab" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openhab+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
Visit this group at http://groups.google.com/group/openhab.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to a topic in the Google Groups "openhab" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openhab/T61K_PRC9o8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openhab+u...@googlegroups.com.
I suspected that is what happened. You will need to factory reset the thermostat (see the manual of how to do that. If I remember correctly you need to hold all three buttons and then insert the batteries) before pairing otherwise it just tries to pair again with the cube (i.e you will see a non-broadcast PAIR_PING message)
Hope that helps
Paul
Hi Ceepo,
Those strings we are receiving from the CUL are weird. They don't seem to make sense. Have you got any weird modes set from FHEM use?
Could you add the following to logback.xml and then send me the full log from a normal start up and then pairing attempt:
<appender name="MAXCULFILE" class="ch.qos.logback.core.rolling.RollingFileApp <file>logs/maxcul.log</file>
<rollingPolicy class="ch.qos.logback.core.rolling.TimeBasedRollingPolicy"> <!-- weekly rollover and archiving -->
<fileNamePattern>logs/maxcul-%d{yyyy-ww}.log.zip</fileNamePattern>
<!-- keep 30 days' worth of history -->
<maxHistory>30</maxHistory>
</rollingPolicy>
<encoder>
<pattern>%d{yyyy-MM-dd HH:mm:ss.SSS} %-5level %logger{30}[:%line]- %msg% </encoder>
</appender>
<logger name="org.openhab.binding.maxcul" level="DEBUG" additivity="false">
This will put all maxcul output (except exceptions) into maxcul.log and you won't need to start in debug mode.
Thanks
Paul
--
That serial message it gets is really very strange! You shouldn't see things like the plain texg serial number of the device in the string. It should just be a hex string preceded by a Z.
I use one of these-- http://busware.de/tiki-index.php?page=CUL I assume you use the same?
Its almost as if something is parsing the hex before it comes out of the serial port which is strange. But there is a lot of junk in there too...
Can you connect via minicom or something and paste the raw output of the serial port? It looks a bit more fundamental than the binding.
Also can you confirm what baud (bit rate) you expect your cul device to run at? Default is 9600.
Thanks.
Paul
V 1.61 CUL868
CUL_V3
Z170004000D1D14000000001001A04B455131303137313239
Z170004000D1D14000000001001A04B455131303137313239
Z170004000D1D14000000001001A04B455131303137313239
Z170004000D1D14000000001001A04B455131303137313239
Z170004000D1D14000000001001A04B455131303137313239
Z170004000D1D14000000001001A04B455131303137313239
Z0E75020205DF5D05FA2F0001080828z\0x17\0x00\0x04\0x00
\0x1d\0x14\0x00\0x00\0x00\0x00\0x10\0x01\0xa0KEQ1017129z\0x17\0x00\0x04\0x00
\0x1d\0x14\0x00\0x00\0x00\0x00\0x10\0x01\0xa0KEQ1017129z\0x17\0x00\0x04\0x00
\0x1d\0x14\0x00\0x00\0x00\0x00\0x10\0x01\0xa0KEQ1017129z\0x17\0x00\0x04\0x00
\0x1d\0x14\0x00\0x00\0x00\0x00\0x10\0x01\0xa0KEQ1017129z\0x17\0x00\0x04\0x00
\0x1d\0x14\0x00\0x00\0x00\0x00\0x10\0x01\0xa0KEQ1017129z\0x17\0x00\0x04\0x00
\0x1d\0x14\0x00\0x00\0x00\0x00\0x10\0x01\0xa0KEQ1017129Ah ha! You have a newer FW version of the stick.
Maybe there is a bug in the culfw that means that the X10 command was not properly implemented and it has since been fixed. I'll have to remove this command from the transport initialisation to stop it putting it into what appears to be raw binary mode or simply change the binding to actually accept raw binary data - which might be more sensible.
In any case it's unlikely I'll have time to make the above change until next week - so as a work around it might be worth downgrading your CUL to v1.58 and trying that to get started?
Thanks,
Paul
--
Hmm... given what Bernhard has said this is very strange. The culfw command reference states:
Z<func>[<hex>]
MORITZ (aka MAX!) mode. <func> is one of:
r - enable MORITZ reception. Note: only MORITZ messages will be received in this mode. Data is reported in hex. If bit 4 was set in previous X cmd (i.e. X10) reported data is binary.
This seems to suggest what ceepo is seeing is correct behaviour - data comes out in binary, but what we observe on our sticks is not correct according to the above statement i.e. we get hex reported.
At this point I don't know why there is the difference in behaviour...
I'll investigate some more when I get chance.
Berhard- glad you are having a good experience!
Thanks
Paul
Paul
Are you referring to this line?
2014-11-30 20:41:18 ERROR o.o.b.m.i.messages.AckMsg[:32]- Got ACK message with incorrect length!
This is in fact normal (I should probably fix this!).
Are your devices pairing with the CUL stick running with openhab? If not then please make sure your devices are factory reset before pairing otherwise they try and pair with the previous controller.
If you still have trouble can you send me a log of the pairing process with your stick.
Thanks
Paul
--
--
--
You received this message because you are subscribed to a topic in the Google Groups "openhab" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openhab/T61K_PRC9o8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openhab+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
Visit this group at http://groups.google.com/group/openhab.
To view this discussion on the web visit https://groups.google.com/d/msgid/openhab/9cec1e1d-8cbe-492a-9a91-aeca9efb4ec6%40googlegroups.com.
Hi Chris,
I have never heard of these devices so cant say yes or no.
We don't do any validation of the serial numbers with any third party so if that is the mechanism that they use then it won't affect the binding operation. If they have changed the actual protocol then it might be a bit trickier.
It would be interesting to here if it does work.
Thanks
Paul
--
You received this message because you are subscribed to a topic in the Google Groups "openhab" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openhab/T61K_PRC9o8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openhab+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
Visit this group at http://groups.google.com/group/openhab.
To view this discussion on the web visit https://groups.google.com/d/msgid/openhab/f0af0bd3-b85d-4a92-96ad-828737a29923%40googlegroups.com.