HS Watch Address

[Warichet]

By checking all posts again , I've re-read Gaston's post

You may then use the Status object in zthe HS as central address for the actual switching object to keep it in sync with central commands.

Soooo, same player shoot again:

Suppose we have an actuator with no status feedback, say: 1.1.101
In ETS, following GA are defined
0/0/101 Spot L5A. On/Off (linked to port #H of the actuator 1.1.101, flags: K L E T)
0/0/102 Spot L5A. Status (linked to nothing, will be used in the HS)
Those GA are imported/created in the HS.
So, in the HS, under 0/0/102, I've created a CENTRAL address pointing to 0/0/101, which is the only one "knowing" the status of the actuator for sure.
In the HS, Visu, a dynamic icon points to 0/0/102, supposed to reflect the current status of the actuator.

It works.
Oh well .....

[Warichet]

Hi Makki,

Thank you for sharing the results of your tests.

after sending a vlaue to a KO used as second listening address on an actuator without feedback, the values get read and are also visible in the HS-Monitor.

I fear I understood you . To make things clear, let's take a simple example.

Suppose we have a @#tua&r with no status feedback, say: 1.1.101
In ETS, following GA are defined
0/0/101 Spot L5A. On/Off (linked to port #H of the actuator 1.1.101, flags: K L E T)
0/0/102 Spot L5A. Status (linked to nothing, will be used in the HS)
Those GA are imported/created in the HS.
So, in the HS, under 0/0/102, I've created a watch address pointing to 0/0/101, which is the only one "knowing" the status of the actuator for sure.
In the HS, Visu, a dynamic icon points to 0/0/102, supposed to reflect the current status of the actuator, it doesn't work.

One alternate way of doing is to bypass this watch address stuff all together and link the status of the Visu to 0/0/101, basta !

Thank you for pointing me to my mistake(s)

[makki]

If I got you right, after flashing the HS 20 times today anyway (Troubleshoot APC..) I tested this:
It works fine, after sending a vlaue to a KO used as second listening address on an actuator without feedback, the values get read and are also visible in the HS-Monitor.

It just confused me a little bit that you have to add the Watch-address under the KO to Query, as I understood from the docu it's the other way around..

Makki

[makki]

Will try tomorrow, promised, as this thread brought me the knowledge (today) of existance and what Watch-Adresses are good for; there're some ideas saving me several current logic-constructions to read adresses after changes to other ones..

Makki

[Warichet]

Sooo, my tentative conclusion:
the HS Monitor doesn't display the Watch Address correctly.

Any chalenger ?

Digging an old topic.
So far, no reply for the challenge.

After having read & re-read what has been said here, I'm still at square one, no clue about this crummy watch address.

Would someone be so kind to test it for me ? please ?
No, no, don't tell me how it should work, I'm convinced. I'd like to know IF it works as advertized.

Thank you very much for the effort

[Warichet]

in order to check whether the actuator replies to a READ-Telegram correctly, just send a value-read telegram on the appropiate Groupaddress via the ETS or via the "Befehl" "Abfragen" via the HS.

Thank you for the simple, yet clever tip

If it works, you should receive the actual state of that channel

Yes, indeed, I see it in the ETS GA monitor, very consistent .
The display in the HS Monitor doesn't change

or you will see the answer in the HS-EIB-Monitor accessible via hslist if activated at "Systemlisten" in the Experte.

Nice, haden't used it, yet.

Sooo, my tentative conclusion:
the HS Monitor doesn't display the Watch Address correctly.

Any chalenger ?


PS: I know, the HS Monitor is not a supported tool.

[Michel]

Unfortunately I do not use the HS-Monitor, so testing is not possible.

But in order to check whether the actuator replies to a READ-Telegram correctly, just send a value-read telegram on the appropiate Groupaddress via the ETS or via the "Befehl" "Abfragen" via the HS.

If it works, you should receive the actual state of that channel or you will see the answer in the HS-EIB-Monitor accessible via hslist if activated at "Systemlisten" in the Experte.

[Warichet]

not all devices provide the possibility to use the READ-Flag.

Right, I overlooked this possibility. But I've only one actuator with this problem, so I cannot test with another device.

See the appropiate documentation whether you can use the "L" / "READ"

It's about a Merten 647829 8x220V. I've the 8_4_04_D_PRODUKT_KURZ_647829.pdf and it doesn't say much (it is really "kurz").

If not, no READ - telegrams send by the HS and other devices will be answered by such devices.

OK, fine. But I'm still puzzled about the HS Monitor behaviour.
In your opinion, should the HS Monitor display a Watch Address correctly ? If you have a HS and the HS Monitor a quick test will put a definite answer to the question.

[Michel]

Well, if it works for you and not for me, then I've something fundamentally wrong in my installation.

Not necessarily!
It depends on your devices!
Remember that not all devices provide the possibility to use the READ-Flag. See the appropiate documentation whether you can use the "L" / "READ" - Flag or not.

If not, no READ - telegrams send by the HS and other devices will be answered by such devices.

[Warichet]

the problem is that you don't have anything to trigger the watch address and make it read the actual state.

Hmmmm, well yes.
But the original question is explicitely about Watch Address and 1 bit object and the display in the HS Monitor. Indeed, in this case "no one" is reading, or forcing a read of the main GA

According to a specialist, the HS monitor doesn't make any difference between a Watch Address and any other. So, I've to take this for granted.

I persist to have some doubts, until some one has done the test and says "works for me".

[Warichet]

No, it works with every object!

OK, good news

Just remember: the HS performs an READ request.

OK, I got it

So it is necessary to have L-flag set in order to be able at all to read out the object status!

Do you mean the ETS read flag or the HS read flag ?
Anyway, I've put all flags on in ETS, and the HS read flag as well. No luck.

Well, if it works for you and not for me, then I've something fundamentally wrong in my installation. Maybe a total misunderstanding.

Do you agree that the set-up of my GA is correct ? if not, any suggestion about how to set it right ?

Thank you for your patience with a HS newbee

[jef2000]

From what I understood, the problem is that you don't have anything to trigger the watch address and make it read the actual state. Then you have to see when the actuator value is likely to change (when the actuator would transmit it if there was a status object) and trigger the watch address.
A periodic timer (e.g. every 60 sec) would do the job but at the price of unnecessary bus traffic. For which task is this actuator used?

Kr,

Jean-François

[MatthiasS]

No, it works with every object!

These objects don't even have to belong to one specific device.

You could, if it would make sense, let the HS read the current temperature from the weather station if someone rings the doorbell.

Just remember: the HS performs an READ request. So it is necessary to have L-flag set in order to be able at all to read out the object status!

[Warichet]

HS LISTENS to watch addresses, if watch address was sent, HS will READ main address.

Thank you very much Matthias
If I get you right, this behaviour is only valid for dimm objects ?
Hence it is normal that is doesn't work for a 1bit object ?
Well, well, it is not so obvious by reading the help. (Indeed the doc takes the dimm example)

[MatthiasS]

Watch address means:

If HS receives a telegram on this group address, it will read the overlying adress.

For example:

If a dimmer is "switched" on by value, sometimes the switch object itself will not show the correct status.

If I add the dimming or value address as a watch address to the switch address, very time someone dims the lights or sets the dimming value, HS will READ the switch objekt.

Conclusion:

HS LISTENS to watch addresses, if watch address was sent, HS will READ main address.