Replacing STD LED with RGB's

[BCBjorklund]

Icicles - Hmm... I wan to control each pixal and what color

Zones: I try to keep the lights controlled by zones  -just followed LOR suggestion, no other reason other then- for storage

always 12V - limit power injection if possible

 

 

[TheDucks]

28 minutes ago, BCBjorklund said:

Icicles - Hmm... I wan to control each pixal and what color

Zones: I try to keep the lights controlled by zones  -just followed LOR suggestion, no other reason other then- for storage

always 12V - limit power injection if possible

OK

All are Pixels (that just means a controller can do either type) , but that complicates things. Those Icicles go down (varied count? or KISS with all the same count) as well as across.  We need a pix count  per linear N foot group (BTW I have no clue what the standard, off the shelf has available. I have a feeling you are going to be doing a LOT of roll your own by reworking off the shelf strings )

 

 

[BCBjorklund]

TheDucks, 

I am using 10 icicle drops of 7 or 70 pixals per 6.5 ft which seems to be standard. I will have to cut and add as needed to fit the edges.

in my attachment I have a table showing the c9 & pixal count

 

Edited January 30, 2019 by BCBjorklund

[TheDucks]

70 seems to be a reoccurring number.   (A nice value for a single PORT. Leaves wiggle room for fudges a bit higher)

note: the drawing seems cropped on the right side: I only see 5 controller (locations)

I did a rough (chaining) grouping of your runs into 70's. I came up with 40 ports worth without considering AMPS per bank/board. (I am guessing 8-10 amps per port. that is seriously pushing any LOR BANK  total. You may need to power inject, just to keep that down)

I think you need a controller near the 24.6/4.0 location (would cover some of the ridge and icicles)

I am assuming you will put the controllers UNDER the eves (that makes cable runs ~3' longer inc drip loop)

In a couple of places, a short run of icicles were on the same string as ridge.

remember, nodes are logical, and can be worked upon as a group.

 

[BCBjorklund]

2nd page has the cropped side

I have C7 & 8 , are you suggesting a C9?  What if I put a controller on the top Ridge hanging on Exhaust Vent?

Yes all controllers are planned under the eve at locations indicated P1, C1 for example. a drip loop most definitely, the eves ate 24: overhang.

 

 

[TheDucks]

23 minutes ago, BCBjorklund said:

2nd page has the cropped side

I have C7 & 8 , are you suggesting a C9?  What if I put a controller on the top Ridge hanging on Exhaust Vent?

Yes all controllers are planned under the eve at locations indicated P1, C1 for example. a drip loop most definitely, the eves ate 24: overhang.

 

 

Ah P2, my printer skipped that when I blew the page up to fill the sheet. I gave up trying to match your zone items to the drawing.

What I am trying to accomplish: cables (inc drip) under 15"  for strings of 70 or more (worried about voltage. my nodes did not like 20' between groups of 25)

I think I now have 56 ports  The issue I keep hitting: A pixie4 can only handle 15A TOTAL,  a Pixie8: 30A (2 banks of 15A),  a Pixie16 can do 60A (2 banks of 30)

You need to use care to balance the heavy (high count) between banks and maybe NOT use a port on each bank just to stay in bounds.

OK I would move C6 to the back corner, does icicles back side and to chimney and hip ridge back there  (6 ports)

same with C5 Might move that to the right ~5'  , but that might just go and let C5 handle that area. Handles front hip ridge, and  icicles

Move C7  Left (closer to the ridge of 49 C9 intersection with those 2 12-4", would do  those 2@ 12-4" as well and the corner icicles  by the old C7 loc

Anyway, that is my shot.  Good luck.  Hopefully a TRUE expert will step up with advice

BTW It may be easier to assign a number to each PHYSICAL straight line piece ( 2 sets of icicles for 15' ) or partial . 1 for a 70-84  (you might borrow some nodes from a shorter adjacent run),

Then make a spread sheet. with columns for nodes, Amps with sub totals for ports (Short ridge sections may span a few numbers because of a direction change). Later you can add Controller-port and starting channel (if the PX is part of another physical string, the starting channel may be other than 1)

[k6ccc]

21 minutes ago, TheDucks said:

A pixie4 can only handle 15A TOTAL,  a Pixie8: 30A (2 banks of 15A),  a Pixie16 can do 60A (2 banks of 30)

You need to use care to balance the heavy (high count) between banks and maybe NOT use a port on each bank just to stay in bounds.

Keep in mind that you CAN power a string separately from the Pixie, or for long strings, power the first X pixels from the Pixie and the next Y pixels via power injection.  It's not all that hard with longer strings to exceed the per string amperage of most any controller.  With most pixel strings, at 100% full white, 100 pixels will exceed the per string capacity of most controllers.  I ran into that this year when I changed my pixel tree to 100 pixel long strings.  Had to set the SanDevices controllers to 50% to keep the current down to the per string limit.  Still managed to blow one fuse - so string 1 quit working between Christmas and New Years.  It was not worth getting out the tall ladder to replace the fuse that late in the season.  Likely no one but me even noticed.

 

[BCBjorklund]

Does this help?

[BCBjorklund]

or this

[TheDucks]

1 hour ago, BCBjorklund said:

or this

I like the chart improvement in this one and the Controller  locations and C9 assignments in the previous. (the drip lines need the same treatment, except they will have many more sections since there are 140 nodes per 15': 2 ports or power inject at node 70-71)

I was having trouble mapping your zone notation to the actual string run  (distance lines) to plan where feed points AND logical direction needed

eg C9 drip goes L>R but it mates with C8 R>L and continues C8 L>R to the corner  When you program, you work with the total nodes. When you connect and power, you work at the port level (n# per port=Amps and direction)

Minor nit (to avoid confusion)  They are both pixels. one is C9 the other Icicles. to me: RGB is dumb whatevers (and uses a different controller type)

[BCBjorklund]

I am still stuck planning contrillers vs power injection 

[TheDucks]

Like warfare, plans last until the first engagement.

Cover the basics, have plan B (inject the trouble spots), have SPARE parts, be prepared to do some place(area) differently.

Distance to first node (per port)

Number of nodes per port and power available

Having power supplies that meets or exceeds the bank draw (the excess could be used for fused Injection source)

Turn it ON (test pattern,single color at a time for a low draw) in place.does it all wok. Now kick it hard (white). Be reasonable, but harsh with your tests (don't do something you will never do in real, like the whole place steady white? That leads to way over-killed power)

Zip cord (SPT-1) has a Rib. That is always negative/neutral :). have a spool on hand for power inject (if needed). Have fuse holders/blocks and fuses for those leads. IMHO, power from the same supply as the boards bank to avoid power on sequence issues.