Known issues with SSCv3 and TM1804 square pixels?

[Steve Gase]

Is this the module?
You are not allowed to view links. Register or Login

The page states 1.44 Watts

If this is it, it is 120mA per pixel, and 50 of them will be 6 amps. you may have to cut back to 34. If not do you have a link to it?

The pics here show 3-LED:
Product ID: 340442205 (but I requested it as 374338613)[/font][/color]12V/3.7A/45W waterproof power supply;AC11V/220V input;CE approvedthis is 20 nodes...  the pics for both show 3-LED  but they arrived as 4-LED  by extrapolation, 50 nodes is 9.25A/112.5W as these are the RJ-approved 120-node capable, then this would be 22.2A/270W
 

[keitha43]

Interesting the 3 led version link still works. Back in Sept. people were reporting Ray was ceasing production of the 3 led and only going to make 4 led versions. You are not allowed to view links. Register or Login

[RJ]

If you have leds that take too much current you can feed the power seperate from the SSC. 4 amp limit is for running the power from the SSC on the cat5.  You just add work and extra wires.

RJ

[rm357]

FYI - I think the 4 led pixels are wired as two parallel sets of two series connected LEDs with current limiting resistors - all hooked to the 1804 chip.

[Steve Gase]

If you have leds that take too much current you can feed the power seperate from the SSC. 4 amp limit is for running the power from the SSC on the cat5.  You just add work and extra wires.

RJ

yep...


I built my 4th coroflake tonight, this one also using the 4-LED square nodes.  It was working ok, and then it stopped after 15 minutes into the SSC test mode.  I overloaded the smart hub fuse and lost power to the SSC.


So... I guess I will alter the 4-LED flakes to split the nodes between 2 SSCs.

[Steve Gase]

FYI - I think the 4 led pixels are wired as two parallel sets of two series connected LEDs with current limiting resistors - all hooked to the 1804 chip.

That seems to match my observations... with 2 of the flakes with 4-LED nodes built so far, each has initially had 3 problem nodes (out of the 50).  With each failed node, I see 2 of the 4 LEDs doing one thing -- and the other 2 LEDs doing another thing...   kind of like a "split brain".

[zwiller]

Let me confirm what Keith posted.  These modules perform fine with V3.1 and V4 and when I placed an order for the 3 led modules Ray told me he is only selling 4 led modules from now on... 

Curve ball: I had flickering with short wiring to only 20 nodes with V3.2 as well as to a single module.  I never really subscribed to the lack of current idea since the TH2010 controller worked with them so well... 

Anyway, not sure if I have 50 but I will try and hook up as many as I can to the SSCV4 and see if there is an issue.  I really like these modules and would like to get them "approved".  I am willing to bet that SSCv4 will power 80 just fine...

Steve, those stars are gonna be bright!  I think these modules are perfect for illuminating objects.  Zwilluminary 2.1 beta    

[zwiller]

Sorry guys, I only have 30 modules...  Steve, would you be willing to test and post your results with a SSCv4 since it appears they should be arriving soon?  Real interested to see if it does the job...    

[Steve Gase]

Sorry guys, I only have 30 modules...  Steve, would you be willing to test and post your results with a SSCv4 since it appears they should be arriving soon?  Real interested to see if it does the job...   

No problem!
But, it needs to happen before I separate out the 50 into multiple controllers and seal up the coro enclosure.

[rdebolt]

Sorry guys, I only have 30 modules...  Steve, would you be willing to test and post your results with a SSCv4 since it appears they should be arriving soon?  Real interested to see if it does the job...    

Tell you what: Sam you send me your 30 and Steve you send me 20 and I will test them for you..... Results will be posted Shortly...And lights will be reurned no later than January 15 2014.   

[keitha43]

And then for accuracy sake I could confirm your results by independantly testing them in 2014 and have them approved for the 2015 decorating season.

[zwiller]

 

There's no way I will get these modules back, they're too cool! 

Steve, what about trying a second power line to the SSC?  IE - 2 pigtails or I think you use the 4 core connectors. 

[therealbigjim]

Instead of glueing the coro-flake back together I used a series of zip ties. My flake from last year used dumb dmx. I knew I wanted to change up in the future so I zip tied the crap out of it.

[Steve Gase]

There's no way I will get these modules back, they're too cool! 

Steve, what about trying a second power line to the SSC?  IE - 2 pigtails or I think you use the 4 core connectors.

I was considering having a second, power-only connection that pulls power from another jack on the active hub and connects to the backside of the 50 pixels... so, twice the carrying capacity, power injection, and twice the fuse...  that way I wouldn't need another SSC.
 
BTW: I just measured amps on a 4-LED, 50-node coroflake running SSCv3 test firmware.  During the white, it maxed out at 2.62A. 

Update: I just used another meter -- this one integrated into my test bench power supply.  It showed a max of 2.75A.  I then substituted a Ray Wu decoder and set it to full white -- and got the identical 2.75A.

Is it possible that the SSCv3 test mode does not do a full 100% white (100% R, 100% G, 100% B)?
 
My head is getting fuzzy around this.  under 3A should be fine -- right?  why did I have a blown fuse?  do I need to inject power and spread the load across multiple fuses?  what bearing does all of the other numbers we've discussed have on what I am seeing? 

[tbone321]

As I said before, these numbers don't mean much of anything.  It's not a voltage or current issue or removing a few feet of wire would have had no effect.  I would check your connections to make sure that you don't have anything loose that could be shorting and blowing your fuses.  2.75A is well within the spec of the SCC and the hub.