New SSC

[rdebolt]

I was waiting to see if anyone else posted, but I have not been able to get the new firmware to do anything.

If I use the original version 4 firmware, everything works fine. If I use the test, test2 or test3 version, I do not get anything from the SSC. When I run the Smart String Utility, I do not get anything flashing or anything.

I was curious how the ones who have gotten it to work are programming the PIC and what version of the Microchip software they are using. Also, are they running the smart string utility with a USB dongle or Etherdongle?

I have tried both version 1 and version 3 of the standalone pickit3 programming utility with the same results. I am using an Etherdongle to program start channel and test. I have not tried with a USB dongle yet.

Rick P.

Can't really tell from what you are saying where you problem is, but I made a comment earlier that when using the SSC Utility you MUST remove the Blue jumper before removing the cat 5 cable or the program will not take.

[twooly]

So far its looking ok using the last v4 test hex file.  Ill be doing more testing this afternoon and Ill report back.  Sorry I share the pickit with my dad and he had it for other testing

[twooly]

Going backwards its always a channel off for me

Couple Examples

14 nodes, null 0, grouping 1
    Going backwards it doesn't start the red until channel 2
    Going backwards with hybrid, channels 1,2,3 work on the how string but then the first node doesn't do the red until channel 5

Another thing that I really can't explain why but

14 nodes, null 1, grouping 4
  Works fine forward, channel 1 lights up the 5th node.  But in backwards it won't light up the last node until channel 35 and it lights up only the last 2 nodes.  I would expect the same as forward meaning the 5th node in from the end would like at channel 1



It does this for both nodes and on the flex strip.  Everything forward seems ok so far.

[rdebolt]

Going backwards its always a channel off for me

Couple Examples

14 nodes, null 0, grouping 1
    Going backwards it doesn't start the red until channel 2
    Going backwards with hybrid, channels 1,2,3 work on the how string but then the first node doesn't do the red until channel 5

Another thing that I really can't explain why but

14 nodes, null 1, grouping 4
  Works fine forward, channel 1 lights up the 5th node.  But in backwards it won't light up the last node until channel 35 and it lights up only the last 2 nodes.  I would expect the same as forward meaning the 5th node in from the end would like at channel 1



It does this for both nodes and on the flex strip.  Everything forward seems ok so far.

Make sure that you are using the latest Hex labeled "V4 test 3.hex" as I had the same issue, but this fixed it for me.

In RJ's post #81 of this thread

[rmp2917]

I was able to get it working. It was still doing the same as before. I tried the other suggestions listed here, but it still did not seem to work. Finally, I removed the cat5 cable while the SSU was transmitting and plugged it back in and the lights started blinking.

With this version of the firmware I seem to have to remove the connector while SSU is transmitting or wait until it is transmitting before plugging it in. I have to do this each time I program it. With the original version, I simply had to move the jumper while it was transmitting and then unplug it once it was finished to reset it.

Everything seems to be working. I will try to do some more testing this weekend.

Rick P.

[twooly]

Going backwards its always a channel off for me

Couple Examples

14 nodes, null 0, grouping 1
    Going backwards it doesn't start the red until channel 2
    Going backwards with hybrid, channels 1,2,3 work on the how string but then the first node doesn't do the red until channel 5

Another thing that I really can't explain why but

14 nodes, null 1, grouping 4
  Works fine forward, channel 1 lights up the 5th node.  But in backwards it won't light up the last node until channel 35 and it lights up only the last 2 nodes.  I would expect the same as forward meaning the 5th node in from the end would like at channel 1



It does this for both nodes and on the flex strip.  Everything forward seems ok so far.

Make sure that you are using the latest Hex labeled "V4 test 3.hex" as I had the same issue, but this fixed it for me.

In RJ's post #81 of this thread

Well poo yep I was in too much of a hurry this morning since my dad was taking the pickit back.  Thought I downloaded the latest but you're right I only did the test2.  Ill have to go steal the pickit back 

[RJ]

I was able to get it working. It was still doing the same as before. I tried the other suggestions listed here, but it still did not seem to work. Finally, I removed the cat5 cable while the SSU was transmitting and plugged it back in and the lights started blinking.

With this version of the firmware I seem to have to remove the connector while SSU is transmitting or wait until it is transmitting before plugging it in. I have to do this each time I program it. With the original version, I simply had to move the jumper while it was transmitting and then unplug it once it was finished to reset it.

Everything seems to be working. I will try to do some more testing this weekend.

Rick P.

Interesting I only change two lines of code in the reverse section of code, should have no change on the programming at all. I should be able fix the need to remove the cable. I will post a newer version later tonight.

RJ

[jnealand]

I have tested with both flexstrips and nodes just using the latest xlights test function.  Both the flexstrips and nodes seem to work fine with one exception on the nodes which may not be due to the ssc.  I am using a 50 ct string of nodes.  When I program the start address all 50 nodes light up and flash red.  When I go to operate mode and fire up xlights tester I only get 37 nodes working even though all of them light up during programming. The flexstrips that I hooked up for reference using V2 SSCs still work as expected with all 128 nodes working.  All SSCs are programmed to address 1 for the test and I have all 512 channels activated in the first universe.  I have to leave home for awhile now, but when I get back I will dig out some additional strings of nodes and see if the problem is only with this string or is consistent on other strings.  I can also bring one of my window frames out of storage down to my shop and test with its 64 nodes.  I will also try the reverse functions then.

[RJ]

Thanks I hope to finish what I am working on late today and can clear the bench and do some testing on it myself.

RJ

[jnealand]

I have switched to using the test hex v3.  I am using the etherdongle for all my tests.  I retrieved a 128 node string for my latest tests.  It works fine and does everything I expected.  Went back to my 50 node string, used the ssc utility to set it for 128 nodes (to lazy to change it back to 50) and still get the last 13 nodes not working even though they all flash during programming.  The last nodes do not light during any testing.  Reset the SSC to string mode and 100% of all nodes work fine in all testing.  Reset the SSC to backwards in string mode and all nodes work fine.  Reset the SSC to individual nodes, 50 count, reverse order and the last nodes still do not work.  I do not understand how the last nodes will work in one mode, but not another.  I've also noticed that the two SSC v2 that I have also been using are much warmer than the SSC v4.  Is that to be expected?  My infrared sensor indicates around 125 F which I do not think is too hot, but I never noticed it before.

[deplanche]

Got the replacement board, and built it last night.  Limited testing last night on rigid strips went well, with just the problems that others have noted above.

A few comments on building it, which I think are worth noting in the manual once that is written...

 - There are two different capacitors used, both of which look identical except for the very, very tiny numbers written on them, so it can be very easy to confuse the two.  Mine came in separate bags from mouser for one board, and I had to pay special attention not to confuse them.  I can only imagine how hard it is going to be for someone who is in a coop to keep track.  My recommendation is to install all of one type, take a break and then do the others. 

- There are 3 other parts with 3 pins (I am not tech savvy enough to know what they all are and do), but they also all look very similar except for the tiny numbers on them.  I am sure if there aren't installed in the right places, they aren't going to work.  So just need to emphasize that they need to go in the right spot.  This really isn't that much different from the old boards, just smaller parts and smaller numbers to try to read on them.

I do like the compact size of the new board and parts!

[RJ]

I have switched to using the test hex v3.  I am using the etherdongle for all my tests.  I retrieved a 128 node string for my latest tests.  It works fine and does everything I expected.  Went back to my 50 node string, used the ssc utility to set it for 128 nodes (to lazy to change it back to 50) and still get the last 13 nodes not working even though they all flash during programming.  The last nodes do not light during any testing.  Reset the SSC to string mode and 100% of all nodes work fine in all testing.  Reset the SSC to backwards in string mode and all nodes work fine.  Reset the SSC to individual nodes, 50 count, reverse order and the last nodes still do not work.  I do not understand how the last nodes will work in one mode, but not another.  I've also noticed that the two SSC v2 that I have also been using are much warmer than the SSC v4.  Is that to be expected?  My infrared sensor indicates around 125 F which I do not think is too hot, but I never noticed it before.

I will be working on this today and hope to post new firmware for you guys to try. Yes the new unit will run cooler.

RJ

[RJ]

Got the replacement board, and built it last night.  Limited testing last night on rigid strips went well, with just the problems that others have noted above.

A few comments on building it, which I think are worth noting in the manual once that is written...

 - There are two different capacitors used, both of which look identical except for the very, very tiny numbers written on them, so it can be very easy to confuse the two.  Mine came in separate bags from mouser for one board, and I had to pay special attention not to confuse them.  I can only imagine how hard it is going to be for someone who is in a coop to keep track.  My recommendation is to install all of one type, take a break and then do the others. 

- There are 3 other parts with 3 pins (I am not tech savvy enough to know what they all are and do), but they also all look very similar except for the tiny numbers on them.  I am sure if there aren't installed in the right places, they aren't going to work.  So just need to emphasize that they need to go in the right spot.  This really isn't that much different from the old boards, just smaller parts and smaller numbers to try to read on them.

I do like the compact size of the new board and parts!

Yes we will have to have the member pay attetion do to the smaller parts being harder to read.

RJ

[twooly]

I have switched to using the test hex v3.  I am using the etherdongle for all my tests.  I retrieved a 128 node string for my latest tests.  It works fine and does everything I expected.  Went back to my 50 node string, used the ssc utility to set it for 128 nodes (to lazy to change it back to 50) and still get the last 13 nodes not working even though they all flash during programming.  The last nodes do not light during any testing.  Reset the SSC to string mode and 100% of all nodes work fine in all testing.  Reset the SSC to backwards in string mode and all nodes work fine.  Reset the SSC to individual nodes, 50 count, reverse order and the last nodes still do not work.  I do not understand how the last nodes will work in one mode, but not another.  I've also noticed that the two SSC v2 that I have also been using are much warmer than the SSC v4.  Is that to be expected?  My infrared sensor indicates around 125 F which I do not think is too hot, but I never noticed it before.

I saw the same kind of thing on a flex strip but I also saw it on a v2 ssc so I figured it was the strip (bad at 36 area) and put a new strip in to do more testing.  On both the v2 and v4 the entire strip flashes on programing but when testing it wouldn't go past node 36 (a strip of 42 nodes)

[zwiller]

Built mine yesterday and it went well.  A few things: With V4test3 hex I got a checksum of E697.  Also new pigtail (gray) is larger than the cream one and required same 15/64" hole as disconnect (I actually like that they are the same size now and the new pigtail seems beefier)

After spending some time with driver issues for pixelnet dongle since I am testing conductor I got blinky, well, sort of.  With Xlights and vixen 2.1 tested new square nodes, rectangles, square modules (4 led), and flex and they all light and no flickering.  I am having issues with the square modules but I think it is a bad few modules unrelated to new ssc and will fool with these later... 

For the purposes of this discussion I am using 128ct of the new square rear entry nodes.  With the utility programming for individual nodes of 128 forward, I get blue flashes.  HOWEVER, the resulting output is hybrid mode only where first 3 channels drive the string and that's all.  Tried a few times.  Thoughts? 

Once I get around this I will do some side by side testing and video.