Befuddled by Nodes

[Rainlover]

I have been building SSCs for the past couple days. I have attached them to 34 count node smart strings. Every thing has programmed successfully and worked in Vixen until..........
I connected a newly finished SSC to a new 128 node smart string and tried to program the SSC with the utility. Only the first 3 nodes lit up. I hooked up a working SSC and the same 3 nodes lit up. I hooked up my little controller that I bought from Ray and only the first 3 nodes lit up. I removed the 4th node and spliced the string back together. Only the first 3 nodes lit up. I removed the first 3 nodes and hooked up a SSC. Nothing lit up. I hooked up the little controller from Ray and everything lit up. I could light up each node in red, green and blue.
I have tried 3 different working SSCs. I rebooted the laptop. I turned off the power to the hub and turned it back on. Tried a different port on the hub, different cable and different dongle. I did a blinky dance and sprinkled fairy dust. Now I am going to take the dog for a walk and get his thoughts.
Anyone have any ideas what to check?

John

[tng5737]

I removed the first 3 nodes and hooked up a SSC. Nothing lit up. I hooked up the little controller from Ray and everything lit up. I could light up each node in red, green and blue.

Did you re-config the SSC AFTER you cutout the first 4 nodes so that the SSC had the correct node count?  I woul dtry that first.  If that does not work I would re-flash the PIC with the latest firmware. I have found that I have a situation like yours where ONLY the first n nodes work - many times it is NOT the first non-lit node which is bad - but rather the last lighted node not passing the signal on to the next. 

[tbone321]

The string count shold make no difference as long as the count is at or greater than the actual number of nodes.  The first thing that I would look at is the voltage on the string when you are powering it up.  I would do it both with them connected to the SSC and then on the controller from Ray.  If the voltage is ok when connected to the SSC then you may have a defective string.

[rrowan]

my guess would be the third node is the problem. Not sending any data and or power

Rick R.

[tbone321]

He removed the first 4 nodes and the problem remained. 

[Rainlover]

This SSC works with 3 other 34 node strings so I know it is configured right.
I have 11.77 volts from Ray's controller and 11.90 from the SSC.

I cut the last 34 nodes off the string and used them They work fine. Somewhere on the first 90 nodes is a problem child, I will have to figure out how to find it.

John.

[tbone321]

Here is a possible way to do it.  Solder a pin to a lenght of wire.  We will call this a signal probe.  Connect the power and ground of the defective string to the SSC as normal but not the signal wire.  Connect this signal probe that you just made to the data output of the SSC.  With this signal probe connected to the data output you can inject the control signal anywhere in the string that you want by simply pushing the pin thru the insulation in the signal wire of the string.  I would start at the far end and see if the node lights.  If it does then the probe method will work and you can begin probing further on down the line until you find the bad node.  If not then the bad node may be pulling the signal line high or low and the output of the SSC might not be high enough to compensate for it.  Ray's controller might be putting out more voltage on the signal line allowing it to control the string even with the bad node. 

If it doesn't work, then you can always use the binary search method to find the bad node.  That is a bit more cutting but it will find the bad node in short order.  To do that just the string in half and check each side.  Every time you find the bad side, just cut that one in half and in short order you will have the defective node.  Just remember that there may be more than one.

[Rainlover]

Here is a possible way to do it.  Solder a pin to a lenght of wire.  We will call this a signal probe.  Connect the power and ground of the defective string to the SSC as normal but not the signal wire.  Connect this signal probe that you just made to the data output of the SSC.  With this signal probe connected to the data output you can inject the control signal anywhere in the string that you want by simply pushing the pin thru the insulation in the signal wire of the string.  I would start at the far end and see if the node lights.  If it does then the probe method will work and you can begin probing further on down the line until you find the bad node. 
If it doesn't work, then you can always use the binary search method to find the bad node.  That is a bit more cutting but it will find the bad node in short order.  To do that just the string in half and check each side.  Every time you find the bad side, just cut that one in half and in short order you will have the defective node.  Just remember that there may be more than one.

tbone321,
I tried the "signal probe" method and couldn't get any lights to work. I did manage to impale most of my fingers with the needle.
I cut another 34 lights off the end of the string and connected them to a working SSC and they didn't work. I figured that the bad bulb must have been in that group of 34 so I connected the SSC to the remaining lights and they didn't work either. Both strings work with Ray's controller. I did notice that when I connect the lights to Ray's controller and turn the power off, 2 of the nodes stay on real dim and don't go completely out until I remove power.
This is really frustrating. I can't wait until I get home from work so I can work on them some more.

John

[tng5737]

[ I did notice that when I connect the lights to Ray's controller and turn the power off, 2 of the nodes stay on real dim and don't go completely out until I remove power.

When you shut the power off on on Ray's box you are just shutting off the box - power still remains on the string.  The nodes staying lit is normal and I wouldn't worry about that.   
Did you remeber to remove the jumper from the SSC Pgm jumper?  It is only used during config AND TEST if you have teh TEST firmware loaded.
Just to be clear are you saying that this particular string  does NOT work with any SSC?  Have you tried using the ssc and your string with a small Vixen sequence? 
Just some random thoughts!

[taybrynn]

What kind of sequence are you running it with?  Just make sure thats all good and no duplicate channel assignments, etc.

I recall in LOR software that if you put in a channel twice ... with one occurance saying "on" and the other duplicate channel saying "off" ... it would still be "off" ... and then you could sometimes go on a wild goose chase wondering whats wrong.

I doubt its this, but trying to think outside the box for you.

Also, your only use SS Nodes, not mixing with modules or any other type off SS items, right?  The other 34 node strings are nodes, not modules ,right?

[Rainlover]

Did you remeber to remove the jumper from the SSC Pgm jumper?  It is only used during config AND TEST if you have teh TEST firmware loaded.
Just to be clear are you saying that this particular string  does NOT work with any SSC?  Have you tried using the ssc and your string with a small Vixen sequence? 
Just some random thoughts!

I have 7 SSC that are configured and are working perfectly. I can connect any of these to any of the other working strings and they work. I can light them up with a Vixen sequence and I can light up each individual RGB. I have 1 string of 34 and one string of 56 that I cannot get to work with any SSC so I know I have at least 1 bad node in each string

Also, your only use SS Nodes, not mixing with modules or any other type off SS items, right?  The other 34 node strings are nodes, not modules ,right?

All I am using are SS nodes. No strips, no modules.

I may have to start the "binary search method" to locate the naughty nodes and when I find them, I am going to punish them severely

John

[rrowan]

Just verifying you got the wires correctly connected between the SSC and nodes

The input side of the string nodes are the side with the Big chip on it
The wiring should be Red to +, Green to -, Blue to Data

Hope you find the problem

Rick R.