Controller odd-predictable flashes

[eurbani]

Hello,

I have one controller that is behaving oddly. At first I thought this was a random behavior but the same thing happens at the same place in the sequences, Sometimes this behavior shows itself as a quick flash of a channel on (less than .25 seconds) other times, the channel does what another channel is doing and sometimes it does what another channel is doing, but to a lesser intensity (i.e. channel 4 is on a dim down 100 to 25 and channel 8 appears to be dimming down from 75 to 10).

Here is what I have done so far:

1) replaced the comm cable (no change -- I didn't think this was the issue since the two controllers down-line from the problem controller worked fine)

2) re-arranged the network so that this controller was closer to the computer (no change)

3) power cycled the controller

4) connected the computer directly to this controller (no change)

5) removed all other controllers from the network (no change)

6) deleted everything from the sequence except the suspect controller (no change)

7) deleted everything from the sequence except the suspect controller half (no change)

8) removed most of the load from the channels (no change)

9) reset the controller (no change)

Some more information:

This is only happening on channels 1-8.

There is a heavy load on channels 7 and 8 (5.18 amps each), but the total for channels 1-8 is only 13.06 amps.

The odd behavior seems to be equal when the high current channels (7 and 8) are on, changing or off.

This 1/2 controller is plugged into a dedicated 15 amp GFCI circuit.

I will open a ticket with LOR, but wanted to see if anyone had a suggestion as to what may be going on.

[OurMikeyRocks]

Eurbani

I have have a current post about "Phantom channels." I'm not sure it's the same as your problem, but I will be interested in seeing what you come up with

Mikey

[-klb-]

Have you run the LOR verifier against your sequence, or your complete scheduled show?

[eurbani]

I had run the verifier against the complete show. I just did it again. All that I have is a warning error (28) for the channels that are not on during some sequences.

[-klb-]

Any CFL? Any heavily LED channels without phantom load? Are the channels with issues LED? Is there possibly some tails of fades not turned off?

[eurbani]

Any CFL? Any heavily LED channels without phantom load? Are the channels with issues LED? Is there possibly some tails of fades not turned off?

Thank you klb.

No CFL. Channels 9-16 do have a couple of LED channels without phantom load. But those channels work fine. Also, I ran a version of the sequence with only the first 8 channels and the issue still existed. I checked the sequence. No tail off fades.

[eurbani]

Here is the sequence for the half controller with the issues:

http://www.thechinsells.com/files/937026/Silent%20Night%20one%20half%20controller%20only.lms

Here is a video starting at 35.8. http://youtu.be/5MvADFWdaKM

The red and green blink at 1.3 in the video corresponds to 37.1 in the sequence. It makes no sense to me.

[plasmadrive]

It could be that you have a small remnent of something in your sequence. I had several of those that I had to find.. needed to expand the sections out and up to see them... They were left over from ramps that got deleted.. I missed either the very begining or the very end cell..

[Dennis Cherry]

What comes to mind in other posts with problems like this, you cannot use two different power sources on one controller unless both are on the same phase of the AC power.

Your power is 220VAC 2 phase coming into your house, your main breaker box has one bank of breakers on one phase the the other bank of breakers is on the other phase.

I looked at your LMS file in another program and did not see anything wrong, just the 2 green channels ramp down from 100% to 0% and the two red channels ramp down from 50% to 0% ending at time 37.2 seconds.

Try reducing the load load on channels 7 & 8 and see if that makes a difference.

[plasmadrive]

What comes to mind in other posts with problems like this, you cannot use two different power sources on one controller unless both are on the same phase of the AC power.

Your power is 220VAC 2 phase coming into your house, your main breaker box has one bank of breakers on one phase the the other bank of breakers is on the other phase.

Actually that is not accurate... Sorry

You can indeed use power from both sides of the 220vac power line... (it is considered single phase BTW, not 2 phase). You have L1 for Line 1 and L2 for Line 2. They share a common neutral and ground but are 180 degrees out of phase with each other. L1 and L2 each carry the full connected load for that phase. The neutral however only carries back the imbalance of the load between L1 and L2. (ie... 12 amps on L1, 10 amps on L2 = 2 amps on N). That is why you can use a single neutral conductor for two 120VAC circuits providing you use L1 for one circuit and L2 for the other. It is still 120VAC from N to L1 and N to L2 and 240VAC from L1 to L2. So, I doubt that is your issue...

Edited December 29, 2012 by plasmadrive

[Dennis Cherry]

Read this: http://en.wikipedia.org/wiki/Two-phase_electric_power

And also this statement. Three-wire, 120/240 volt single phase power used in the United States and Canada is sometimes incorrectly called "two-phase". The proper term is split phase or 3-wire single-phase. The two live outputs of a 3-wire single phase transformer secondary winding are properly called "legs".

The reason is the dimmer circuit on the controllers us the 60Hz to make the triacs dim, if you have the dimmer circuit on one phase the other circuit 180 degrees out it will not dim correctly.

[plasmadrive]

Read this: http://en.wikipedia...._electric_power

And also this statement. Three-wire, 120/240 volt single phase power used in the United States and Canada is sometimes incorrectly called "two-phase". The proper term is split phase or 3-wire single-phase. The two live outputs of a 3-wire single phase transformer secondary winding are properly called "legs".

The reason is the dimmer circuit on the controllers us the 60Hz to make the triacs dim, if you have the dimmer circuit on one phase the other circuit 180 degrees out it will not dim correctly.

The dimmers will work fine from a three wire system. I have one of my LOR controllers working that way now because the load is too much for one circuit. I used to design dimmers back in the day.....

Ok.. I just typed a whole explanation as to how it all works then realized it was a bit overboard so I deleted it.. sorry.. suffice to say, the LOR dimmers work very well off the 240/120vac 3 wire +ground system.

[eurbani]

They are plugged in to circuits 37 and 38 (different phases). In the controller, I have noticed that the status LED lights when the 9+ side is powered. I guess that means that the transformer that runs the control board is powered on that side which is having no issues. So, this makes some sense.

I'll check it out and let post the results here.

FYI: post 7 above is just one small example of many places where I have issues with circuits 1-8 on this controller.

[plasmadrive]

They are plugged in to circuits 37 and 38 (different phases). In the controller, I have noticed that the status LED lights when the 9+ side is powered. I guess that means that the transformer that runs the control board is powered on that side which is having no issues. So, this makes some sense.

I'll check it out and let post the results here.

FYI: post 7 above is just one small example of many places where I have issues with circuits 1-8 on this controller.

The xfmr is indeed powered from the 9-16 side. The 1-8 side being on a different phase out of your house panel would have nothing to do with it.. However.. if you have something on that other phase that is causing glitches in the power.. that is another story... You see this happening regularly.. unless you have something that is causing your power to do funny things... should not be an issue... Triacs can be triggered by a couple of other things such as exceeding the dv/dt rating of the device. I doubt that is happening at your place.. but truthfully, I have nothing by experience to base that on... so.. I could be wrong in your case..

You could also be causing the issue by voltage drop into your 9-16 circuit. You said you have heavy usage on that side..You said in your OP that you removed most of the load.. was that from the high or low numbers? Just for grins.. try unplugging some of the load from the high numbers.. see if that doesn't help, unless that is what you did.

[Max-Paul]

And for anyone reading the above #9 post. The two hot wires coming into a circuit breaker panel is not split between the banks of breakers. The two hot wires alternate down both banks. Thats why your hot water tank, A.C. & Dryer for a few examples have breakers side by side, not on opposite sides of the panel.

[eurbani]

I reduced to the loads to a safe margin and then put them on the same circuit. Same results.

If no one has any other suggestions, I'm going to open a ticket with LOR.

[eurbani]

You could also be causing the issue by voltage drop into your 9-16 circuit. You said you have heavy usage on that side..You said in your OP that you removed most of the load.. was that from the high or low numbers? Just for grins.. try unplugging some of the load from the high numbers.. see if that doesn't help, unless that is what you did.

I actually deleted every channel from the sequence except 1-8 and still had the same issue. I could unplug the loads from 9-16, but since the above sequence using only circuits 1-8 and still has issues, I don't think there would be any change. Right?

[plasmadrive]

Unplugging should make no difference, Right... From the info given it doesn't sound like a controller issue. It sounds like the controller is doing what it is being told to do. Since it doesn't do the flashy thingy all by itself when there is no data coming it's way.. I would rule out power hardware.

I have an idea if you are game.

1. create a completely new sequence with only this controller in the channel configuration.
   
2. do some sort sequence making sure you leave that part where it usually messes up blank. See what happens then. Use the same song.. It should flash there also if it is not sequence related..... I would think.
   
3. If that stops the flash.. load up your other channel config and try it.. Something tells me you could have a config issue that you are looking at over and over and it is hiding from you..
   

[eurbani]

From the info given it doesn't sound like a controller issue.

It isn't a controller issue. I swapped controllers and had the exact same issue. This was an odd combination of several issues.

1) Having controllers plugged into two different phases DOES (at least in this case) cause an odd flash at the beginning/end of any dimming. Sorry Dennis. I didn't mean to contradict you earlier.

2) On some of my sequences there were a lot of data going out over a short period of time. On those sequences, I was able to minimize (but not eliminate) the unresponsiveness of that controller by making it controller 1 (instead of 9). I'm not sure how the data is sent to the controllers, but I assume that 9 (00001001 binary) is taking more band-width than 1 (00000001 binary). Really this is just a guess on my part. I get the ones and zeros, but I don't really understand how they get from one point to another.

3) There were a couple of mistakes in my sequences. (two timings close together with lights off in between)

4) The last issue was something about the loads. I am 100% certain that I didn't overload any channel or side. (5.2A max on two channels, 13.2A on low side controller and 12.8 on high side) I'm still not clear when or why this was happening exactly. After 4 hours sitting with my computer on the concrete driveway, I gave up.

Thank you everyone for your input. I was ready to send it back to LOR. Swapping controllers is really a good test.

Edited January 9, 2013 by eurbani

[plasmadrive]

It isn't a controller issue. I swapped controllers and had the exact same issue. This was an odd combination of several issues.

1) Having controllers plugged into two different phases DOES (at least in this case) cause an odd flash at the beginning/end of any dimming. Sorry Dennis. I didn't mean to contradict you earlier.

2) On some of my sequences there were a lot of data going out over a short period of time. On those sequences, I was able to minimize (but not eliminate) the unresponsiveness of that controller by making it controller 1 (instead of 9). I'm not sure how the data is sent to the controllers, but I assume that 9 (00001001 binary) is taking more band-width than 1 (00000001 binary). Really this is just a guess on my part. I get the ones and zeros, but I don't really understand how they get from one point to another.

3) There were a couple of mistakes in my sequences. (two timings close together with lights off in between)

4) The last issue was something about the loads. I am 100% certain that I didn't overload any channel or side. (5.2A max on two channels, 13.2A on low side controller and 12.8 on high side) I'm still not clear when or why this was happening exactly. After 4 hours sitting with my computer on the concrete driveway, I gave up.

Thank you everyone for your input. I was ready to send it back to LOR. Swapping controllers is really a good test.

If you have some sort of issue with a flash at the beginning of a fade up there could be issues with your power to that controller because having them on opposing phases should have absolutely no effect for that reason alone. If your neutral isn't electrically clean, that can cause odd problems as well. Dimmers (at least most all triac type) use zero crossing detection to start the dimming cycle delay. If the neutral is not clean, the cycle zero crossing can be off.

Ya know............... if there is a lot of capacitance on the load side, that could be a problem. Dimmers start at some percentage of the cycle to turn on.. the less light you want, the longer they wait in the cycle before triggering.... Once triggered on the triac conducts until it commutates off (load current goes below the hold on threshold) at the zero crossing. If you have a bunch of capacitive loads and the current does not go to zero at the appropriate time it could stay triggered into the beginning of the next half cycle.. which would mean an odd flash because it would remain on for the entire next half cycle..

OK.. that is way more then I am sure you needed to read from me.. but I was thinking "out loud" (on the keyboard) in case anyone else wanted to chime in along my thought process and comment. Anyone? This is an interesting problem.. at least I think so! Perhaps not so interesting to the OP..... but troubleshooting is always fun if it is not your own problem.. right??