DiyLightAnimation
Hardware => Lynx Smart String => Topic started by: RJ on July 18, 2011,
-
I had a string sent me with the issue a few have had. We have found that it is a node that is not exactly right causing but we could not understand why the tester from ray work with them but not the SSC's.
So With one in hand to see and work with I have found the issue. On the string I have to test there are at least one node that is drawing current on the data line. Now all digital systems draw current on data lines but it is so small it has very little impact. Until you get one that does pull an abnormal amount. So why does the ray tester over come this? Simple, it has a much higher drive current on the data line so it can function even with the higher load on the data line.
Whats the fix? In my cause changing the 330 ohm resistor on the SSC to a 150 ohm increased the current high enough to make the string work. You can either remove the 330 and change it to a 150 ohm from Radio Shack or add another 330 on the bottoms of the board hooked to the 330 that is in there. (Two 330 resistors in parallel equal a 150 ohm one.)
This is not needed unless your string is doing this but I think I will move to the 150 ohm on all of them going forward. this way if you get a node like this it will work. I will not promise the node will not fail at some point but at least you can find the bad one once it does. Having the string not work and no way to find the bad one is a pain!
If you have one of these strings try this and let me know some feedback. The 150 ohm will work the pic a little harder but it is not an issue.
RJ
-
So if I still have some SSC boards to assemble, would it make sense to just build them with the 150 ohm resistor? That is, is there any downside to using it even if I end up with pixel strings that don't need it? In some respects I'd just like to have all my SSC's the same and not have to worry about which ones have the mod and which don't.
-Mike
-
Sure, It will not hurt anything. I am going to change the Part List (BOM) and put a note in the manual unless someone wants to help me out and add the info in the manual for me <wd..
The pcbs will still show 330 until we run out and have more made but that could be a while since I stocked up from the coop since the price was so good. Thats why a few people who pm me for price sent pm's back asking if i had messed up on the cost ;D
Very busy working on some really important projects <pop..
RJ
-
Thanks RJ,
I've updated the manual.
We don't want any distractions for you. <pop..
-Mike
-
Thanks!
I will get the Parts list tonight when I take a break (about 4am most likely!) <fp.
RJ
-
Thanks RJ, i've had this problem several times before. You are awesome and I just want you to know how much I enjoy working with you. So many smiles are all because of what you do for us that are given by others. Hope to see you next spring for your Expo.
Ronniek
-
My experience trying this fix. As I had 2 other strings with this symptom besides the partial string I sent to RJ, I dropped by Radio shack on my way home from work. They only had the 330 ohm resistors so I soldered them onto the underside of the board as instructed. I tried to program the 1st string and the solid white symptom was still there. No flashing string. However after about 20 seconds the 1st node turned pink and stayed that way. I cut that one out and it worked. I soldered on a replacement node and tried it and all was well. I went to tackle the 2nd string and initially it was flashing like it was supposed too. About 10 seconds later the back 1/3 of the string started freezing solid on. I left it running and the freezing kept working its way toward the 1st node a little at a time until the whole string stayed on. I powered down and tried again but every time the string was frozen white. I had noticed the 1st node had momentarily turned ping then back white so I crossed my fingers and cut it out and all is well. I don't know if the 150 ohm resister would work better. I do know I never saw those nodes change to ping with the single 330 ohm resistor. Maybe because my strings are normally 126 nodes long the 165 ohms is not quite enough to override the problem? Any way I am about to put the test firmware on these and run them all night!
-
Well that kind of was my hope, that it would help people find the issues. The nodes that are causing it are having issues but at least maybe this will allow people to spot the ones with issue. The other of your strings works fine but would not do anything until the current was increased as you know.
RJ
-
Yes the 1st string was easy to spot. The second was much harder as it only changed for a second so I wasn't sure that was the problem. Huge sigh of relief when It was. Did you spot the bad node in the string I send you and mark it or cut it out? Strange I plugged in the 2 strings after fixing and loaded the test software and would think they would stay in sync with each other since they got power at the same instant. But the longer they run the more out of sync they get. Strange.
-
The loosing sync is to be expected any time you have indepedent clocks. Even atomic clocks drift and require periodic adjustment...
RM
-
... I dropped by Radio shack on my way home from work. They only had the 330 ohm resistors so I soldered them onto the underside of the board as instructed.
According to the RS website most stores should carry 5 packs of 150 ohm 1/2 watt and 330 ohm 1/4 watt resistors. They don't stock 150 ohm 1/4 watt, but the 1/2 watt version is fine to use so long as it fits the PC board (RJ can confirm). The watts value is just the maximum wattage the resistor can handle before failing, so going higher is not a problem, other than potential physical size issues.
-Mike
-
You are correct.
Sent from my Charge by Tapatalk
RJ
-
Could I use a bunch of spare 180 ohm resistors I have?
Two 330 ohm = 165 correct, so 180 just a teeny bit higher...
Thanks,
Brian
-
Could I use a bunch of spare 180 ohm resistors I have?
Two 330 ohm = 165 correct, so 180 just a teeny bit higher...
Thanks,
Brian
Likely will work fine. We are just upping the drive current. There was not really a scientific reason for 150 I would not go lower than 150 as the pic will start having issues pulling the line low enough to get a clean "0" bit.
RJ
-
Likely will work fine. We are just upping the drive current. There was not really a scientific reason for 150 I used two 330 as you said which is 165 ohms so the closest with out going higher and that was common was 150 ohm.
I would not go lower than 150 as the pic will start having issues pulling the line low enough to get a clean "0" bit.
RJ
-
I understand the phrase "white lock up". I just loaded test firmware to a ss controller and connected 40 rectangle nodes to the hub and turned on the power. All node are white and do nothing else but white. Not knowing just what the "TEST" firmware is supposed to do, I would say I have white lock up. I want to know what should happen if you do not have white lock up. Do the lights change color? Do they blink? Should the change quickly? Sounds like I may have to change from 330 ohm to 150 ohm.
Ron
-
Ron,
With the test firmware, the lights should cycle Rad, Green, Blue, White, Red, Green Blue White etc etc.
John
-
Thank You! I do then have white lockup.
Ron
-
If I am not going to use the test firmware, should I still change the 330 to a 150 or will it not matter running the show firmware.
Ron
-
If the program jumper is on = solid white
If the program jumper is off = colors should cycle
I plan on changing all of the 330 ohm resistors in my SSCs (only two built units left, 14 still awaiting assembly).
RM
-
Shoot! I did not have white lockup. I had program jumper on. Where is the info about using the test firmware like the jumper on or off for white and colors? I seem to have missed it.
Ron
-
The white lockup issue a few strings have had tends to cycle red,green,blue then white but then hangs on white instead of moving on again.
So I also wonder about the jumper but RM already posted since you said it came on white.
RJ
-
The white lockup issue a few strings have had tends to cycle red,green,blue then white but then hangs on white instead of moving on again.
So I also wonder about the jumper but RM already posted since you said it came on white.
RJ
Yes I had the jumper on. I had 5 sets of 40 on the table and the would only do white. When RM posted about the jumper I removed the jumper on all 5 controllers and turned them back on. They scrolled through the colors very smoothly. Thank you guys.
Ron
-
We are a team and we are all human.
I've already toasted at least one fuse... I'm just glad RJ designed them in - without it something else might have gotten toasted like a trace or wire. 40A at 12 volts can do some damage...
RM
-
Just stumbled across this. This is one of those items that should be made a sticky as I am not sure I would have found it otherwise and it has some pretty important troubleshooting info.
Just my $0.02,
-Keith
-
I just received the rest of my nodes from Ray and have started the burn in process. I have 3 strings so far that will cycle through a couple of time then lock up in "white mode", well sort of. Not all of the nodes are pure white. I used the 150ohm resistor on all of my sscs and have tried different controllers on each string. My question is "How do you pin point which node is causing the issue when they lock up?" I am using 128 node strings and have not cut any of my last shipment up yet. Any suggestions?
-
Fastest way is to just cut in half and just test 1/2 at a time and repeat. Always looks for 1 node to look a little different as sometimes that pinpoints the problem quicker.