DALI on LM5

[LKBA]

Hello,

I have just migrated whole project from LM4 to LM5, and my DALI feedbacks stopped working.
All the commands go through - all the lamps go on/off and dimms with no problem, but when queryingactual to read the status i get such error messages:

Event for DALI READ ALL (15/7/3) 22.12.2016 15:51:10 * string: rx error

script:
require('user.dali')
res, err = dalicmd('internal', 'queryactual', { addrtype = 'short', address = 0 })
if res then
   log(res:byte())
else
   log(err)
end

when querying any other address i get:


Event for DALI READ ALL (15/7/3) 22.12.2016 15:52:44 * string: reply timeout

When querying all the ballast it's the same:

Event for DALI READ ALL (15/7/3) 22.12.2016 15:54:08 * string: reply timeout

script:

require('user.dali')
res, err = dalicmd('internal', 'queryshortaddr', { addrtype = 'broadcast' })
if res then
   log(res:byte())
else
   log(err)
end


Then i tried scanning through devices and after scan 16 of connected ballasts has disappeared, i cannot detect them again. They were visible before the scan.
Tried rebooting LM, turning off power supply of DALI line.
What else should i try?

Holidays are coming and customer is getting angry. Please help
If You can call me on +370 600 31726
or e-mail directly to l.korsakas@bustoautomatika.lt

[LKBA]

Thanks to Edgars, https://dl.openrb.com/lm-16.12-imx6/pkg/...2_imx6.ipk this package solved it.

[Morpheus09]

I'm having the same issue here.. 

I get "tx error" as well as "rx error" alternatively, I have a LM5p DW1 New CPU with the 20180828 version FW. 

The fix file is not available and I presume it wouldn't work either with the version I have. 

Thanks..

[Daniel]

This bug should be fixed in this FW but all the packages are here https://dl.openrb.com/lm-18-imx6/pkg/

[Morpheus09]

The "rx error" appears when I do a "queryactual" to certain groups, in one LM the error shows on the groups 1 and 2, and in the other LM groups 2 and 3 fails.

In each LM I have 7 Dali groups assigned, the rest of the querys work fine.

tried to reinstall the packages but there is no change.. the error continues.

[Daniel]

Do you do it to a group or to a individual ballast?

[Morpheus09]

Do you do it to a group or to a individual ballast?

To groups, I made a tool to add and delete ballasts to groups, each ballast only has 1 group assigned, and the other groups respond fine to the query, the groups not assigned return a timeout, but I don't care about them..

I use groups because the customer didn't like the effect when sending a command to each ballast at a different time...

[Daniel]

I don't think this should work to a group. What value would you receive when lights can have different values in a group?

This works fine

Code:

require('user.dali')
res, err = dalicmd('internal', 'queryactual', { addrtype = 'short', address = 0 })
-- read ok
if res then
 value = res:byte()
end

log(value)

[Morpheus09]

You are right, if the values are different, it should give me an error but if the value is the same shouldn't be a problem, also the devices are the same model..

ignoring the query actual to a group, how could I get the status of a group then? making a query to each ballast and making an average of the ballasts of the group?

[Daniel]

You are right, if the values are different, it should give me an error but if the value is the same shouldn't be a problem, also the devices are the same model..

ignoring the query actual to a group, how could I get the status of a group then? making a query to each ballast and making an average of the ballasts of the group?

In KNX dali gateways they select one light in a group and they just query one ballast.  If you control them all together then this is enough.