lights on without dongle plugged in

[lineman]

quick ? when I powered up my V5 LE all the channels came on mini lights were plug in . Then I plugged the dongle in and wireless and played a sequence and all worked fine when song ended all lights went off then when I unplugged the dongle and wireless I went outside to shut down LE and all channels were on again without any DMX going to it is this normal. Again LE works like its suppose to when playing a sequence lights only come on when I remove the dongle from the computer am I suppose to shut LE down first or does it matter.

[Dennis Cherry]

This might sound crazy, but swap the 2 RS485 chips and try it again.

I have found some 485 chips that the output will go low when no DMX signal is applied. The normal state should be high.

Just what you explained.

I sent RJ a few of these 485 chips for evaluation and he sent me replacements.

[egenoup]

The RS485 chips are highly static sensitive, so proper grounding when handling is essential.

[trekster]

Now you tell me.  After I have built several kits.  I guess I have been lucky.  I will use ESD precautions in the future.

[JonB256]

I've replaced 485 chips out in the yard using a pocketknife as a chip puller. I've found them to be pretty durable. Of course, now that I've said that, one of them will go Zap. 

[lineman]

FYI, I Powered up the same LE tonight and it did not come on with out dongle plugged in . A fellow member suggested that mabey it was the last command given when I turned LE off the last time ,somthing to the effect of 512 packet what ever that means over my head. Mabey someday I will understand all this but for now thanks for my fellow DLAers. SO all is well and mega tree looks great blinking to the beat, Ohh what a RUSH, OH BTW my little girl (Rachel) said to tell all yall THANKS SOOO MUCH she loves Blinky too

                             Thanks for all yall do

[RJ]

Jeff were you stand real close to the unit when you powered it up and all the lights came on?

If so then it is your magnetic personality turning the lights on! 

RJ

[jnealand]

I have a similar problem with my 16 singing pumpkins.  When I get up in the morning I find two pumpkins on that were off before I went to bed.  I have tried using test channels to turn them all on and then all off then went to bed, but they were back on in the morning again.  If I unpower the dongle the pumpkins go out, but I leave everything plugged in and powered up 24/7 until the end of the season.  Anyone have a clue as to what might be happening here?  This is an LE V4 that works just fine during a sequence.  I have another LE V4 with a different problem, but I will start a new thread for that one.

[Greg]

Jim, I suppose if you are keeping things powered 24/7 you could make a very short sequence with everything off and schedule it to run every so often during non-show times.  Doesn't address the root cause but would correct the errant channels.

Greg

[jnealand]

Further info.  On close inspection today of the two channels (12 and 14) that are staying on.  They also never go off during a sequence, they just dim.  It would appear that I have a power leakage somewhere in this LE.  To solve the problem I set another LE next to this one, gave it the same address and moved the 12 and 14 pumpkins over and no more dim bulbs and the show runs great.  All LEs are running on wireless and all have the terminate jumper on.  I also tried swapping out the transmitter (one more reason to have a spare) but the problem persisted.  So that LE goes back to the workbench on Monday after the Halloween show comes down.   This make two LEs with problems discovered during my Halloween setup.  Better now than Thanksgiving!!!  Makes me wonder about all the ones not used for Halloween.

[JonB256]

I'd try swapping the Optoisolator chips on those two channel to see if the problem moves. Happened to me before and it moved the problem, so I replaced the chip. Some of them just seem more sensitive than others.

[jnealand]

I'd try swapping the Optoisolator chips on those two channel to see if the problem moves. Happened to me before and it moved the problem, so I replaced the chip. Some of them just seem more sensitive than others.

Now if I can just figure out what an optoisolator chip is I will give it a try next week.

[JonB256]

There are sixteen, socketed Optoisolators on the LE. They are between the small green LED and the monster heat sink. Their purpose is to electrically isolate the 110vac or 220vac from the control chips. Inside is a very small LED and an equally small light sensor. So, the signal to turn on the TRIAC (they are the device screwed to the heat sink) goes through them.

I suspect that you have two (#12 and #14) that are too sensitive and triggering when they don't need to. If it is the Optoisolator's problem, it will move with them. If not, it won't move.

It is also VERY possible that moving them will make the problem go away completely. You may also find that one of the Optoisolator has a leg that is bent and not actually inserted into the socket, making partial contact instead of solid contact.

While you SHOULD use a chip puller, I usually just use a very small pocketknife blade and gently pry each end, alternately, until the chip pops out of the socket.

Do I need to tell you to unplug the board first? 

[tbone321]

It could be the isolators or it could be the triacs themselves.  Triacs can leak a little when off and some leak more than others.  I believe that you can simply remove the isolator chips from the channels that are not shutting down completely and then power up the board and see if the channels are now  completely shut down.  If so then the problem is with the isolators and if not then the problem must be the either the triacs or their connections on the board. 

[JonB256]

You didn't say and I didn't ask, but what is the actual load on channels 12 and 14? Are they LED lights or incandescents? If LED, then just a tiny amount of leakage current can keep them glowing.