Hardwired Lighting System Using Cat5

[beelzerob]

Well, I don't have any dimmers left not connected to the controller...and if my wires at the controller weren't held together with glue and twist ties, I'd disconnect them for a test....

I never tried the double OFF thing before having it all tied together, so I don't know if that worked then.

I can say with certainty that the OFF behavior works as advertised, with it connected to HV and LV. Single OFF, jump to 0. Double OFF, slow ramp down.

As for any delay on the double ON....it all depends on how quickly I do the double tap. If I do it reallly quickly, then I don't notice any delay at all. The slower I do the double tap ON, the more noticeable the pause at the preset becomes....and of course that makes sense...it's just treating it more and more like 2 single taps instead of an actual double tap. It's possible that the switches don't actually register double taps ON, just react to two single taps, and the faster they are, the more likely the 2nd tap is registered before the lights have risen to the preset level.

I'd almost guess the switches don't register double taps at all, just react to single taps, but there is CLEARLY a double tap action on the OFF, not 2 single actions.

I could disconnect the serial port add-on and try some things, and maybe you could disconnect your Elk add-on and try some things, since that's one of the most major differences in our setups.

 

[felixrosbergen]

Will do tomorow...for sure my dimmer WILL NOT do a slow ramp to OFF on a double tap on the OFF button...

 

[TonyStewart]

ok...maybe somebody can test to confirm this for me.

I am quite consistently (but not always) seeing a 2 stage ramp up when doubletapping the on. It will go from 0 to say roughly 50% then hesitate for a split second and continue to ramp up. This absolutely does NOT happen when the dimmers are only connected to the HV and the LV is not connected.

On the dimmers that are only connected to the HV the double tap on the OFF consistenly brings the nice slow ramp down that beelzerob is such fan of. However once connected to the polling loop (and the ELK<>ALC interface) this behaviour disapears and only the quick not so nice off works.

This is very strange and i want to make sure I'm not crazy.

I will put together a mixture of ALC stuff, add an M1 and interface and try to emulate these things on the test bench. Saying I will do this get's the committment out of the way. Next I need to squeeze out the time from somewhere. Stay tuned.

TS

 

[TonyStewart]

Well, I don't have any dimmers left not connected to the controller...and if my wires at the controller weren't held together with glue and twist ties, I'd disconnect them for a test....

I never tried the double OFF thing before having it all tied together, so I don't know if that worked then.

I can say with certainty that the OFF behavior works as advertised, with it connected to HV and LV. Single OFF, jump to 0. Double OFF, slow ramp down.

As for any delay on the double ON....it all depends on how quickly I do the double tap. If I do it reallly quickly, then I don't notice any delay at all. The slower I do the double tap ON, the more noticeable the pause at the preset becomes....and of course that makes sense...it's just treating it more and more like 2 single taps instead of an actual double tap. It's possible that the switches don't actually register double taps ON, just react to two single taps, and the faster they are, the more likely the 2nd tap is registered before the lights have risen to the preset level.

I'd almost guess the switches don't register double taps at all, just react to single taps, but there is CLEARLY a double tap action on the OFF, not 2 single actions.

I could disconnect the serial port add-on and try some things, and maybe you could disconnect your Elk add-on and try some things, since that's one of the most major differences in our setups.

I bet you are right! The switch functions as two taps. This makes the most sense. With that as the gospel word, all other things fall into place and make sense as compared to design specs.

TS

 

[beelzerob]

But see, that's where the double OFF function really doesn't make sense, though....

If the switch only functioned based on single taps, such that if you hit ON once, it goes to preset, and ON again, it goes to full....then if you hit ON twice fast enough, it is still "going to preset" first, but then it continues on to Full as it processes the second tap.

But for going to OFF...the two single taps would ONLY make sense if the default for one tap was a slow ramp down. Because then you could hit OFF twice real fast, and it would go straight to level 0. Then, you could do it again, but have more of a gap between each OFF press, and it would still go to level 0 on the 2nd press. In other words, the first OFF would start it slowly ramping down, and before it got very far, the 2nd OFF would send it to zero.

However, in its current configuration, you can't have really any delay at all between OFF taps, because within half a second of the first OFF press, it has already gone to level 0. So there really is a "double tap" detected for OFF, but doesn't seem to be one for ON, or at least doesn't NEED to be one for ON because of what the first tap does.

So that's why the current functionality of the OFF pressing really doesn't make sense to me....

 

[TonyStewart]

But see, that's where the double OFF function really doesn't make sense, though....

If the switch only functioned based on single taps, such that if you hit ON once, it goes to preset, and ON again, it goes to full....then if you hit ON twice fast enough, it is still "going to preset" first, but then it continues on to Full as it processes the second tap.

But for going to OFF...the two single taps would ONLY make sense if the default for one tap was a slow ramp down. Because then you could hit OFF twice real fast, and it would go straight to level 0. Then, you could do it again, but have more of a gap between each OFF press, and it would still go to level 0 on the 2nd press. In other words, the first OFF would start it slowly ramping down, and before it got very far, the 2nd OFF would send it to zero.

However, in its current configuration, you can't have really any delay at all between OFF taps, because within half a second of the first OFF press, it has already gone to level 0. So there really is a "double tap" detected for OFF, but doesn't seem to be one for ON, or at least doesn't NEED to be one for ON because of what the first tap does.

So that's why the current functionality of the OFF pressing really doesn't make sense to me....

For no controller attached.......
There is no "off" preset. You can set the on level by adjusting either the off or the on. Once you get to the brightness point you are happy with, this become the "on" preset.

Off is always fast off. But...But...But....
In our classroom we have a portrait framing projector that slowly fades the bulb off. THERE IS NO ELECTRONICS in the fixture to do this. It is a part of the bulb itself. This is to avoid a premature death of the bulb. It always ramps up and always fades off. It has nothing to do with the switch attached. I have seen this type incandescant bulb in houses before. It is not as rare as you may think.

When connected to a scene learning interface which is then connected to the HAI controller, the operation I described is exactly the same with an instant off (I just tested that).

When connected to an Elk using the OnQ ALC to M1 scene learning interface, the operation should be exactly the same (not tested by me yet). Because..... the same circuitry exists on all scene learning interface. The only difference is a section not related to dimming features that holds the protocol for the controller of choice.

As for the operation when connected to a serial interface for use with CQC or any other serial method. I can not testify to how the switch works by overwriting the switches factory operation through direct programming. Here, many things are possible. And... the switch remembers your previous configuration changes and will hold that characteristic until you send it a new serial command via developers kits etc.

In summary, the brighten rate is inherent in the switch electronics not in the scene learning interface unless you override it with serial commands. But neither the HAI or ELK controllers will override the switches factory settings.

There is one exception to adjusting the on time (no exceptions to adjusting off time). When using an HAI controller you can program the on rate (called ramp over time) to take as long as several hours to come to a preset gradually. But I am not aware of this capability in the Elk and I am not aware of any way to fade the off.

I will beat on this some more, but as it stands now, I can't make any ALC Dimmer fade off!

TS

 

[felixrosbergen]

wow...what a mystery...

Seems liek both beelzerob and have onserved the fade off function on the dimmers. I can consistently (on at least 2 dimmers) repeat this (tapping off twice and watching it fade to off over about 10 seconds time frame) when these dimmers are not connected to the LV. I though the dimmers conencted to the LV didn't do this, but i couldnt resist by try on my way up the stairs last night before bed and what do you know...one of the dimmers did it. I tried it several times in a row (very effective at reducing the WAF ) and it did it a few times and then stopped. Maybe it depends on the preset value whether it uses this function or not.

All my bulbs all plain old vanilla bulbs.

Tonight I'm going to unplug the controller and see how the lights react then.

Too bad the RAMP OVER TIME ON/OFF functions don't work when controlling under ELK. They have the FADE function under ELK...but as confirmed it doesnt do anythign for ALC...so close though...just a few letters difference. Hopefully Spanky from ELK is reading this.

Beelzerob: with your CQC ALC driver can you control all these ramp rates and such? If this is the case I'm contemplating getting rid of the ALC<>ELK interface, getting the serial interface and put a serial router in such that i can control the lights both from CQC (Server<>Router<>ALC Serial Expansion<>ALC Controller) and from the ELK (ELK<>M1XSP<>Router<>ALC Serial Expansion<>ALC Controller). wow...writing it up it looks like i need a ALC Controller, ALC Serial Expansion, Serial Router and a ELK M1XSP....this would be an expensive excersise....

 

[beelzerob]

Ya, I didn't mean to imply there was or should be an OFF preset. I just meant that a slow fade off was something that would work for a single OFF press in order to have the next OFF press be instant off. It just doesn't work the other way (where one OFF press is instant off and the next OFF press is slow ramp....doesn't work! Two single presses don't work with OFF like they do with ON).

So are you saying you can't get a dimmer to fade OFF no matter what? Not even with double pressing OFF? Or you just can't reprogram the switch so that it fades OFF on a single press?

It's funny...I didn't even know that the double OFF would do the slow fade, and then once I was told about it, I'm disappointed I can't reprogram the switch to do it with a single tap.

Oh well. For now, I'm using CQC to set the levels to 0 instead of sending an OFF command whenever I change the switch via an interface screen. Setting the level to 0 does the slow fade down, whereas sending the OFF does the instant. Not as convenient as pressing the wall switch, but at least I get my effect.

 

[beelzerob]

Beelzerob: with your CQC ALC driver can you control all these ramp rates and such? If this is the case I'm contemplating getting rid of the ALC<>ELK interface, getting the serial interface and put a serial router in such that i can control the lights both from CQC (Server<>Router<>ALC Serial Expansion<>ALC Controller) and from the ELK (ELK<>M1XSP<>Router<>ALC Serial Expansion<>ALC Controller). wow...writing it up it looks like i need a ALC Controller, ALC Serial Expansion, Serial Router and a ELK M1XSP....this would be an expensive excersise....

If you go to the beta docs page for CQC, Supported Devices, you'll find the OnQ/ALC docs, and what the driver supports. Some of it is out of date though (the switch press events, mainly) and I'm going to get Dean an update.

I'm not aware of anything in the RS232 protocol for controlling ramp rates, except extended ramp rate, which can be set from 2.33 seconds to something like 10 hours.

HOWEVER...I am not able successfully set that ramp rate with the RS232 protocol currently (it rejects my attempt), and I'm in communication with ALC tech support to try and figure out why. Under the assumption that I am doing something wrong, then when I fix my code, you'll be able to ramp up or down at any rate you set.

HOWEVER2....this ONLY applies to sending commands via the RS232 (as in, from CQC). It does not modify the switch behavior in any way. That's why I can click a button on my CQC interface and have it ramp the lights down nicely to 0...but when I press the physical OFF switch once, it still snaps to 0.

 

[felixrosbergen]

With ELK the transition is smooth when using the SET LEVEL TO XXX rather than TURN ON TO LEVEL XX. The nasty thing is that ELK will not let me set the level to 0% only to 1% which i assume is probably not a good idea for the bulbs as they go buzzing below 15% or so.

So when using RS232 you can SET the level to 0. I wonder why ELK cannot do this. Still the issue of the local control, but at least turning scenes on and off would look 'smooth' and professional. At the moment it doesnt have that 'high end' feel when the transitions from OFF to ON and vice versa and between levels are so abrupt.

 

[beelzerob]

If you look at the driver description I linked, you can see you can either set the state (on / off) or the level (0 to 127) of dimmers. You can also read/write the preset level.

If I were rabid about having the soft off feature, then I GUESS I could always replace all my ALC switches with aux switches, capture that button press (via digital input) and then command ALC to set the level or however else I wanted it to act. But of course, you'd have to hide the actual load switch somewhere, AND you'd no longer have working lights when the CQC server was down for any reason...so I'm obviously not going to do it.

But I do agree that having lights that softly dim off simply gives it a higher-class feel. Too bad you have to do a special action to enjoy it.

 

[Jimmyjam]

What are the ALC hardware pieces that you would need to have a basic system set up to interface with either CQC or Elk?

 

[beelzerob]

I'm running mine right now (CQC) with the Controller and Serial Expansion (sorry, don't have the parts numbers). I think that'll give you 32 lights and 500' of wire without having to need anything else.

 

[beelzerob]

Ok, I unplugged the controller (power only, wires were all still intact), and checked the switch. Still works just the same as before, no difference.

 

[felixrosbergen]

I came home and unplugger the distribution module (where all the wires are landed) from the ELK interface. With it unplugged all dimmers consistently do a nice soft off (maybe 15 seconds) upon double tap on off AND when ramping up they do not have the hesitation that they had before.

Plugged the distribution module back in and the soft off on double off tap is gone and the hesitation is back. TS I'm not crazy...i swear.

PS I replaced the 2 flaky switches i decribes earlier in the thread with new ones, same location, same wiring method and the new ones are working just fine. These 2 switches will be in the mail back to you for investigation.