KNX Init allignment problem

[Domoticatorino]

Hi all,
I wrote a script like this below in order to allign the knx group to the graphics:

-- wait for 10 seconds
os.sleep(10)
for variable = 1, 50 do
grp.read('0/1/' .. variable)
  os.sleep(0.5)
end

-- for scansione grupp 4 da 1 a 50
for variable = 1, 50 do
grp.read('0/4/' .. variable)
  os.sleep(0.5)
end

I have seen that at the init machine send read request and devices reply. But it seem that icon some icons are not allign and updated hence icon is light on hence send off but actuator channel is off hence there is not updated. 

Have you any idea please?

Thanks.

[admin]

Which firmware version are you running?

[Domoticatorino]

Which firmware version are you running?

Hi,
honestly machine is branded by Schneider but surfing on web I discover your web site and I saw that they are very very similar. So I tried to ask this information.

I hope to be not unpolite.

Thanks.

Claudio

[Domoticatorino]

Which firmware version are you running?

Hi,
honestly machine is branded by Schneider but surfing on web I discover your web site and I saw that they are very very similar. So I tried to ask this information.

I hope to be not unpolite. Could you help me?

Thanks.

Claudio

[Erwin van der Zwart]

Hi,

Where do you see the requests / response ? Ets monitor?

If yes do you connect over HL/SL or USB?

Does none work or just some?

Did you check all read flags inside ets project?

Do you use line couplers?

Is physical address of HL/SL set to the correct line?

What FW are you using? 

BR,

Erwin van der Zwart
Schneider Electric

[Domoticatorino]

Hi,

Where do you see the requests / response ? Ets monitor?

If yes do you connect over HL/SL or USB?

Does none work or just some?

Did you check all read flags inside ets project?

Do you use line couplers?

Is physical address of HL/SL set to the correct line?

What FW are you using? 

BR,

Erwin van der Zwart
Schneider Electric

Hi Erwin,
thank you for your response. This is great for me use this forum because I would like to increase proposal of HomeLynk to my customer. Nowadays I have 2 HL working but it is fantastic.

By the way, I see on ETS Monitor all read requests sending by HomeLynk at INIT and consequently I see all responses from devices. Unfortunately it seems that icons on HomeLynk are not update. I am linked to the bus throug HomeLynk but in any case customer told me that sometimes there is not allignment.  

At the moment it is installed firware 1.3.1.

Should I use grp.update? But this function writes to HomeLynk. So I should fix the value received by read request to the value on HomeLynk. Is it? But it seems too heavy.

Allignment it is not automatic due to the fact the on project I keep separate the command address from the reading address as it shown on the attachment?

Thank you for your help.

Regards.

Claudio

[Erwin van der Zwart]

Hi Claudio,

Good to hear you like the system (;

If you see all the trafic in the ETS monitor when connected through the HL it means the telegrams are received.

FW 1.3.1 has for as far as i know not any issue that could cause this. You can upgrade to 1.5.1, but i don't think this would change anything. 

If you look at the object value of your status object, is this matching your installation? Or is this also wrong?

If this is correct, please try to clear cache, might be caused by older images, however is see you have SVG files so i don't think those are not refreshed due caching.

I also don't understand why you would use script to update the object value, the KNX TP is sending this value as reply on a read request and are already received by HL
and object should follow the KNX telegram without additional code.

Another question is why do you read the sending (main) address when using status object? I think only syncing with status objects should be enough.

You can do the syncing also with TAGS instead of a loop..

You can put a TAG for example 'sync' to the objects (can be done fast by filter the range and add TAG with mass update) and put this script inside init script:

Code:

syncobjects = grp.tag('sync')

syncobjects:read()

If all of this doesn't help please send me a private message with IP and (temporary) credentials or send me the HL backup including .knxproj file.

Regards.

Erwin

[Domoticatorino]

Hi Claudio,

Good to hear you like the system (;

If you see all the trafic in the ETS monitor when connected through the HL it means the telegrams are received.

FW 1.3.1 has for as far as i know not any issue that could cause this. You can upgrade to 1.5.1, but i don't think this would change anything. 

If you look at the object value of your status object, is this matching your installation? Or is this also wrong?

If this is correct, please try to clear cache, might be caused by older images, however is see you have SVG files so i don't think those are not refreshed due caching.

I also don't understand why you would use script to update the object value, the KNX TP is sending this value as reply on a read request and are already received by HL
and object should follow the KNX telegram without additional code.

Another question is why do you read the sending (main) address when using status object? I think only syncing with status objects should be enough.

You can do the syncing also with TAGS instead of a loop..

You can put a TAG for example 'sync' to the objects (can be done fast by filter the range and add TAG with mass update) and put this script inside init script:

Code:

syncobjects = grp.tag('sync')

syncobjects:read()

If all of this doesn't help please send me a private message with IP and (temporary) credentials or send me the HL backup including .knxproj file.

Regards.

Erwin

Ciao Erwin,
honestly I do not check the status of HL object. I only know that icon was light on, hence object was sending off but considering that channel actuator was already in off icon did not receive the feedback.

This is the reason why I would like to send reading request at init. Because after a shut down of the power some channels actuator are in off state even if before the shut down they were in ON STATE.

For this reason at INIT I send a reading request to all feedback address (0/1/xx and 0/4/xx) and I think that group state of HL has flag "U" enable in order to be updated after a reading request. Is it?

Why do you ask me the reason I send reading request for the allignment? It should be automatic?

Thank you.

Claudio

---

(can be done fast by filter the range and add TAG with mass update)

How can add a TAG to some group objects in massive way? For example add "FB" to all group address 0/1/xx?

Thanks.

Claudio

[admin]

Set object address filter to 0/1/*, then click Mass edit in table footer. There you can select which fields to edit for all selected objects.
As for status issue, can you check if Objects tab values are correct after sending read request?
Also, sensing many read requests at once might cause some devices to reply with Busy status. Having a small delay between read requests is a correct approach. You can do it like this with tags:

Code:

syncobjects = grp.tag('sync')
for _, syncobject in ipairs(syncobjects) do
  syncobject:read()
  os.sleep(0.5)
end

[Domoticatorino]

Thank you Erwing!!