Buffer/line driver - null pixel without pixel

Question about power injection. I have a long run and will power inject at the prop with a second power supply. I know I have 3 leads (+DC, D, -DC) coming into the line driver from the controller. I know, since I'm using a second power supply, I don't want to use the +12VDC from the controller to supply power to the prop (where I'll inject with 2nd power supply instead). So here's a couple of questions.

1. Do I run the +12VDC from the output side of the line driver all the way to the prop and NOT connect it to the prop (where I know I use the +12VDC from the 2nd power supply instead)? Or do I just not connect a wire to the output side of the line driver thus not running the +12VDC from the line driver output downline to the controller? In other words...is there a reason to run or not run +12VDC down that long run and not connect it at the other end....Just curious about having a live power wire running that far and parallel to data and if that's a potential problem.

2. I know I need a common ground between the 2 power supplies....So the question is are we talking the grounds on the AC side of the 2 power supplies or are we talking about connecting the (-) DC between the 2 power supplies?
 
Santacarl said:
1. Do I run the +12VDC from the output side of the line driver all the way to the prop and NOT connect it to the prop (where I know I use the +12VDC from the 2nd power supply instead)? Or do I just not connect a wire to the output side of the line driver thus not running the +12VDC from the line driver output downline to the controller? In other words...is there a reason to run or not run +12VDC down that long run and not connect it at the other end....Just curious about having a live power wire running that far and parallel to data and if that's a potential problem.
Click to expand...

That is your choice and really depends on how long the run from data source to data destination is.
  • For runs under 20' I have just V- and Data going from the controller to the prop. Then I inject at pixel 1 with the local PSU.
  • For runs over 20' I run V+, V- and Data from the controller to a buffer/driver (or multiples for really long runs). From the last buffer/driver to the first pixel I only run V- and Data to the first pixel.
  • The longer you make your DC runs, the greater the voltage loss due to resistance in the wire. Keep them as short as possible. With that said, A long run of the DC 12V line that is unterminated acts like an antenna and introduces noise into your system.

Santacarl said:
2. I know I need a common ground between the 2 power supplies....So the question is are we talking the grounds on the AC side of the 2 power supplies or are we talking about connecting the (-) DC between the 2 power supplies?
Click to expand...
NO YOU DO NOT need a common ground between the PSUs. You need V- and Data from the data source all the way through to the data destination. In your case you only need V- and Data from the controller and they must be (travel, follow the same path, be tied to each other, be twisted around each other is even better) together for the entire distance to the data destination (your first pixel). The same rule applies for all injection points: V- and Data must travel together from a data source (a pixel) and a data destination (the next pixel) no matter how long the distance between the data source and data destination. Many people have only tied V- together at the PSUs and for some people it works out. For the rest of us, Murphy has a field day.
 
MartinMueller2003 said:
That is your choice and really depends on how long the run from data source to data destination is.
  • For runs under 20' I have just V- and Data going from the controller to the prop. Then I inject at pixel 1 with the local PSU.
  • For runs over 20' I run V+, V- and Data from the controller to a buffer/driver (or multiples for really long runs). From the last buffer/driver to the first pixel I only run V- and Data to the first pixel.
  • The longer you make your DC runs, the greater the voltage loss due to resistance in the wire. Keep them as short as possible. With that said, A long run of the DC 12V line that is unterminated acts like an antenna and introduces noise into your system.
Click to expand...

Thanks Martin....I appreciate the response.

Ahhhh...the third bullet makes a lot of sense....that's the way I was leaning. I'm running Cat 5 so the twists are there for Data and V- but I was concerned about the radio aspect of that live V+ wire..... I'm going to experiment with distance. I have 1 run that is 50' and I know you use multiple line drivers for that distance but I've been reading that Cat5 seems to be pretty forgiving....thought I might get away with just 1 line driver...we'll see....I can always add more line drivers until I get a signal down line....

MartinMueller2003 said:
NO YOU DO NOT need a common ground between the PSUs. You need V- and Data from the data source all the way through to the data destination. In your case you only need V- and Data from the controller and they must be (travel, follow the same path, be tied to each other, be twisted around each other is even better) together for the entire distance to the data destination (your first pixel). The same rule applies for all injection points: V- and Data must travel together from a data source (a pixel) and a data destination (the next pixel) no matter how long the distance between the data source and data destination. Many people have only tied V- together at the PSUs and for some people it works out. For the rest of us, Murphy has a field day.
Click to expand...

Really? I've read in sooooo many many threads that you need to have a common between two power supplies where 1 PS powers your controller and the other PS is used to inject at the destination....That's in addition to the data and V- from the previous pixel. You didn't mention if you use this method of no common between PS' or not??
 
17 psus in my show. zero dc wires between them. 52 pixel ports. 7 of them have v+ coming out because the lights start within 1 foot of the controller.
 
MartinMueller2003 said:
17 psus in my show. zero dc wires between them. 52 pixel ports. 7 of them have v+ coming out because the lights start within 1 foot of the controller.
Click to expand...

Wow....I guess it DOES work! :biggrin2:

I guess, after thinking about it, in a way the V- is connected to both...since you run V- from the controller through the buffer/driver to the pixel that also has a V- from the 2nd PS that's injecting....DUH on me....
 
V- is GND, so common ground means connect all V-‘s
Low voltage side GND is NOT the same thing as AC ground and they are not connected.
 
RobG said:
V- is GND, so common ground means connect all V-‘s
Low voltage side GND is NOT the same thing as AC ground and they are not connected.
Click to expand...

Thanks Rob....I had heard that on some Power supplies the AC GND and the V- are the same/connected internally...while other power supplies that's not the case... That was driving some of my confusion.

In a post above Martin mentioned that a V+ that is unterminated can introduce a 'radio' problem. Since I'm using Cat5 and don't connect V+ from the line driver output to the outgoing Cat5 what about the wires that now go unused in the Cat5? I have 3 wires in my Cat5 that are unused/unconnected running the length along with my D and V-.....
 
Santacarl said:
Thanks Rob....I had heard that on some Power supplies the AC GND and the V- are the same/connected internally...while other power supplies that's not the case... That was driving some of my confusion.

In a post above Martin mentioned that a V+ that is unterminated can introduce a 'radio' problem. Since I'm using Cat5 and don't connect V+ from the line driver output to the outgoing Cat5 what about the wires that now go unused in the Cat5? I have 3 wires in my Cat5 that are unused/unconnected running the length along with my D and V-.....
Click to expand...
As mentioned in another thread. Terminate the wires. All unused wires go to V- at both ends of each length of the CAT5 cable. If you have a buffer in the middle of the CAT5 cable then the buffer gets V+ from one side via CAT5 and that wire is tied to ground at both ends on the other side CAT5. With a buffer in there you essentially have two lengths of CAT5 that are independent of each other.
 
Santacarl said:
Thanks Rob....I had heard that on some Power supplies the AC GND and the V- are the same/connected internally...while other power supplies that's not the case... That was driving some of my confusion.

In a post above Martin mentioned that a V+ that is unterminated can introduce a 'radio' problem. Since I'm using Cat5 and don't connect V+ from the line driver output to the outgoing Cat5 what about the wires that now go unused in the Cat5? I have 3 wires in my Cat5 that are unused/unconnected running the length along with my D and V-.....
Click to expand...

When using Cat5, use all of the solid color wires for V- on your P/S, the orange/white wire for Data, and the rest of the solid/stripe wires for V+ on the P/S.
 
RobG said:
The most common way to combat effects of cable capacitance (and other issues) and improve signal quality is to use null pixel.
That means you have to make sure you add all those null pixels to your sequence.
The better solution I think would be to create a small buffer/line driver board which could be used instead of null pixel.
The good thing about it is that it will work with any pixel protocol, will be transparent to the sequence, and will work with any voltage.

View attachment 36834

Here is the wiki page.
And how to waterproof.

Note:

I will not be selling them on Tindie, you can get them directly from me using sign up sheet #5

In stock.
Click to expand...

Please forgive my ignorance but can’t figure out for life of me how to add myself to the list to place an order


Sent from my iPhone using Tapatalk
 
wilmar said:
Please forgive my ignorance but can’t figure out for life of me how to add myself to the list to place an order


Sent from my iPhone using Tapatalk
Click to expand...
These are a gem. I have a prop that has a "weak" last pixel. Put in a driver right at the pixel and the problems magically goes away.
 
You must log in or register to reply here.
Back
Top