Troubleshooting Pixie 4 to communicate on network

[IndependenceLights]

Short version: I've got a blinking red light on the controller and I'm trying to narrow it down. 

1. I can confirm the cat5 cable is good.
2. Other controllers are communicating, showing up in the hardware utility, and working as expected. I think that rules out anything on the USB side or comm port etc.
3. The controller was used for two seasons prior to this one. As far as I know (and can tell) nothing has changed on the board. No jumpers, switches, etc. should have been touched.

Here's a more detailed run-through of the setup and what I'm working with. I thought this might be easier. Let me know what other details would help. Just looking for things to try at this point.

 

[IndependenceLights]

I've tried resetting. Also tried setting device ID manually or setting it to have the hardware utility manage it. Same result. I definitely feel like it's not a software thing at this point. What else can I test on the board?

[stevehoyt]

I did not have the same issue as you. I had corrupted firmware. I had a fast flashing light and could not discover my controller either. 

Using the procedure below I was able to repair mine.

You might give it a shot to see if it discovers yours. 

 

 

1. Make sure all other Light-O-Rama programs are closed before opening the above program.
2. Once the HU is open, select the COM port from the drop down on the left.
3. Click the Firmware button in the bottom right.
4. Select the Only one device connected option and open the correct Pixie firmware
5. Click download.

 

[TheDucks]

Down to basic Q's: Do the connected (to that Pixie) lights run the test pattern when you short press the white button?

 

BTW I really suggest any FW upload be done with only that target board connected. (been there.🙄 Works most of the time until it does not)

[Jimehc]

Bad COM Chip.................... Contact LOR Help Desk to have them send you one....

Edited Friday at 05:50 PM by Jimehc

[IndependenceLights]

9 hours ago, TheDucks said:

Down to basic Q's: Do the connected (to that Pixie) lights run the test pattern when you short press the white button?

 

BTW I really suggest any FW upload be done with only that target board connected. (been there.🙄 Works most of the time until it does not)

I have not been able to run the test this way. I only get the first pixel on (sometimes but not always). 

[IndependenceLights]

8 hours ago, Jimehc said:

Bad COM Chip.................... Contact LOR Help Desk to have them send you one....

I'm really leaning towards this based on what I can tell. Wish I could find a good way to confirm on my own but this definitely seems like the issue. 

[Jimehc]

45 minutes ago, IndependenceLights said:

I have not been able to run the test this way. I only get the first pixel on (sometimes but not always). 

Makes me think it's more then a COM chip issue anymore, or if a COM chip issue at all....

[TheDucks]

3 hours ago, Jimehc said:

Makes me think it's more then a COM chip issue anymore, or if a COM chip issue at all....

I agree. BAD (noisy) DC power will cause erratic behavior.   1) The button fails. No Com chip involved.  2) I just recently had this happen setting up a new Pixie 8 . it had ONE 10 node string attached so I used a 1.2A 12V wall wort I had. The board was iffy on HU discovery and the test color really erratic.  (1.2A is more than enough for 10 nodes). I switched supplies and it was all good and solid.

one more test: Only 1 port connected (smallest  string). Does the white button now behave?  (Smart nodes draw a small bit of power even when the lights are off)  If it does, your PSU can't hack it. change it out (Be sure you have one with enough watts: rule of thumb is it should have ~20% MORE WATTS than you will use when all white )  (Some folk program the board to limit the full % instead of a bigger PSU)  

 

[IndependenceLights]

It's a 10A power supply and I don't run much off of it. I plugged in a 50 pixel strand. No test sequence. The only thing I can do with the white button is the reset function when I hold it down. When I do that I get the quick flashing red light and then it goes back to slow flashing. I also have a couple of some 10w floods that are for smart RGB (3 wire) and I couldn't get the test sequence to run on those either. I'm not sure which one uses less juice but either way I got the same result. 

 

I may have a smaller 60w power supply I could swap out and just see if something different happens. 

[Orville]

22 minutes ago, IndependenceLights said:

It's a 10A power supply and I don't run much off of it. I plugged in a 50 pixel strand. No test sequence. The only thing I can do with the white button is the reset function when I hold it down. When I do that I get the quick flashing red light and then it goes back to slow flashing. I also have a couple of some 10w floods that are for smart RGB (3 wire) and I couldn't get the test sequence to run on those either. I'm not sure which one uses less juice but either way I got the same result. 

 

I may have a smaller 60w power supply I could swap out and just see if something different happens. 

Did you press the white button, then immediately release it?   If you hold it too long, you could possibly cause damage to the controller.

To get it to run a test pattern, you press the white button and release it almost instantly, basically almost like a tap on the button to start the lighting test.

Otherwise the light test won't run/work.

[TheDucks]

120W should be plenty for a single 10W flood or 50 node strand test.  

Do you have a DVM or even a cheapo Analog meter. Is the 12V stable (12.7 is my target: +5% to compensate for wire losses while staying under 10%, typical 12V tolerances)

Things to remember.  Light shows are HARD on supplies. We go from 0 to full lights quickly and frequently.