DiyLightAnimation
Hardware => Lynx Express => Topic started by: winwin on December 02, 2013,
-
I have a new LE version 5.0 recently built. In the process of setting up my display, I ran into a little issue with channel 54.
_Using Xlights for testing, with LED lit on channel 54, the light string sometimes works (lights on), sometimes it does not.
_I then ran a sequence with vixen and channel 54 worked fine.
_I ran a different sequence with vixen and this time channel 54 did not work (LED is lit, but no lights on the string)
_All trial I did using the on-board test were successful with channel 54
Any ideas what might be happening and how to fix this? Thanks!
-
I would check and reflow those solder points on that Triac. This looks like a bad connection.
-
My first build I had the same problem. Didn't have enough solder on one of the optocouplers.
-
Thanks for the suggestions and I will reflow those solders. It does look like a bad connection, doesn't it?
I am not sure my first post was clear enough and I am sorry for that. the LE controller is reacting fine according to the signal it receives. If channel 54 is supposed to be on, LED 54 is on. If it is supposed to be off, LED 54 is off. It is just the lights attached to that channel that are erratic. I have ruled out that the light set might be the issue (connected it to a different channel).
I do not know the function for all components on the LE but if reflowing the solders on the triac and optocoupler does not do the trick, should I assume that those components needs to be replaced?
I will only have a few hours over the weekend to find the fix and complete my display so I need to make sure I have all my ducks in a row! If i need spare sparts, I need to order them too. Can you think of anything else I might need ? Many thanks in advance for your help and suggestions
-
I doubt that it is the opto, I wuld put the effort into the triac first. Make sure that you crank up the iron before you attempt to reflow the joint and use a touch of new solder to help with the heat transfer.
-
Good points and thanks for the soldering tips Tbone. I'll attempt the fix this weekend.
In the meantime, do you think it is unlikely that the pic could have such effects and it might need to be reflashed?
-
If the LED for that channel is lighting, then the PIC is working as expected.
-
makes sense, thanks a lot.
-
Also check your solder connections on the terminal strip itself, and ensure the opto is firmly in its socket.
-
Will do this this weekend Don, thank you for the suggestions.
Now that I think about it, I am not so sure that the problem has anything to do with a bad solder. if it was the case, why channel 54 would work with one sequence and not with another?
-
The sequence should have nothing to do with it. The LE is unaware of any sequence, it simply responds to commands. If the channel LED is responding to the sequence but the lights are not, then there is either a failed component or a bad solder joint. If the component had failed, then it would never work and you claimed that sometime it does. This almost always indicates a bad connection. It could be a bad opto but an easy test would be to move it to a new position and see what happens. If the problem follows the opto, then it is the problem and if not, then the problem is something else.
-
Now that I think about it, I am not so sure that the problem has anything to do with a bad solder. if it was the case, why channel 54 would work with one sequence and not with another?
If, as you say it works in one sequence and not in another that would be a sequence issue. You should fire up xlights and use the tester to turn whatever is on that channel off and on. You can use the twinkle function to cause it to keep going on and off while you monitor it or even put a multimeter on the terminal strip to check the voltage. Also there is no channel 54 on an LE only 1-16. It is a better reference to use 1-16 when you are posting a troubleshooting question.
-
I have tested voltage output when I built the LE and all was good. However I ran out of time this past weekend to do this again when I noticed the problem so I certainly will do this again at the next opportunity. Moving the opto to another position and see what happens seems to be a good troubleshooting step too. The odd thing is that the channel in question worked all the time when running sequence A but would not work when running sequence B. I am frustrated because I can't really give you more info until I work on the equipment next weekend.
Jim, channel 54 is on my LE #4 and is channel #6 and while it is obvious to me, it may not for others not familiar with my equipment. From now on, it will be channel 6; good suggestion!
I appreciate the help folks, thanks a lot.
-
It sounds like one of the things you need to do is truly confirm you actually have an issue. The previous couple posts give you some good troubleshooting suggestions; isolate the controller and run it thru its paces. If you find that on its own it works fine, and everything else in your display is working fine besides that channel, then you likely have something going on with the channel configuration for one or more songs.
-
Let's not run down too many rabbit holes here. In his OP, he said that on the second sequence, the channel LED was flashing but the light string was not. The LE was receiving the commands to light the channel but was unable to do so. This clearly indicates a hardware issue beyond the PIC. The possibilities are the opto, the triac, the output plug, or the light string. Since the channel would work sometimes, I really doubt that it is a failed component since when most solid state components fail, they fail and that is it for them.
He claimed that he used a known good string and it still didn't work so it's not the string. I would first check the output plug and make sure that the connection to the terminal block is a good one. With the power off and unplugged, I would also run a continuity test between the terminal block and plug outputs. If vampire plugs are being used, they sometimes have issues making a good connection. The next step would be to reflow the triac connection's. An iron hot enough to solder in a resistor or a cap will not always be hot enough to properly solder in a triac. Triac's have a lot of heat sinking capability and while the solder may flow to the board, there may not be enough heat to flow properly to the pin of the triac. This type of cold join is hard to see and can make or break connection for the smallest of movements and a few degrees of temp change can also make or break it.
The opto can also be an issue but after checking the sockets solder joints, the easy thing to do is to swap it with one in a different socket. If there is a bent pin on the chip, that will be seen when the opto is removed and unless the issue follows the chip, the opto is not the problem.
-
Good point..re-reading winwin's comments, it seems he does have proper channel setup in all cases..I missed that on first read.
Out of 12 LEs built this year, I've run into two problems...one, a bad output cord/plug...you make a couple hundred and you're bound to have a bad one in the mix, and another was an intermittent channel which turned out to be a poorly seated opto..it made it thru build and test, decided to be a problem once deployed in the display. Removed and reseated..and viola..working.
This should be an easy one to track down and find..just a matter of working back from plug to opto..
-
Not another rabbit hole, but if it worked in test mode and with sequence A, would we not also confirm that sequence B is not using a profile or has something to do with the amount of output as in a light curve on the board?
Steve
-
Cold solder joints will react this way as well. We don't know what other conditions changed between these tests. The LED's on the LE respond proportionally to the output of the channel it is connected to. If the output was so low that you couldn't see the light from the string, you wouldn't see the light from the LED either. The first step should always be to confirm that the hardware is fully functional before you head down the much more difficult road of configuration quirks and setup issues.
-
I still think a good test with xlights will help.
-
Good news at it seems that the LE is working fine and as Don pointed out, it may have been the case since day one.Unfortunately I cannot tell (yet, need more time) what the root cause may have been but I would be tempted to say that the main problem was xlights related with the tests functions (either not using properly or a bug?). They may have been a few other factors that contributed to muddy the water and distracted me from the main issue...mea culpa! While I do not recall the exact event, I was running xlights to test lights and noticed that when switching from one test to another the lights would not respond accordingly. It made no sense why a channel would work with one test and not with another...so i did the typical stuff in that case, shut down the program, rebooted and tried again and found the non-responding test to work. I ran again out of time this past weekend so I want to be cautious until the show is running (this weekend!) to confirm that all is good.
I know this is still odd but if it is working fine now i am not going to touch it! Seriously, you have given me good info to work with and should the problem re-occur, I am in position I can troubleshoot the system right away.
Thanks for your supports, you are really helpful folks!