brichi Posted April 6, 2017 Share Posted April 6, 2017 after I fix this I wondered if I should've left the ground connected and only cut the + I have a tree with 100 nodes tied to it, The data line runs the full length of output 1 on a pixie 8 but at the 51st pixel I cut the + and - and with my second 12V p/s I injecyrd power at node 100 to feed 51-100. Is this ok or should I have left the ground connected too and only cut the + after node 50? here is a sadly drawn pic of the string of lights, pixie 8 and 2 power supplies https://www.dropbox.com/s/knrvt5bvo143h3n/INJECT.jpg?dl=0 Link to comment Share on other sites More sharing options...
TheDucks Posted April 6, 2017 Share Posted April 6, 2017 Splice the - back (but keep the PSU - connected). It is also part of the Data circuit Link to comment Share on other sites More sharing options...
brichi Posted April 6, 2017 Author Share Posted April 6, 2017 so on each strip pixel connect the ground back? i did connect the grounds between the power supplies too as per another diagram Link to comment Share on other sites More sharing options...
Ebuechner Posted April 6, 2017 Share Posted April 6, 2017 You can leave the ground wire connected at the power supply but it also should be continuous on the strips. Link to comment Share on other sites More sharing options...
brichi Posted April 6, 2017 Author Share Posted April 6, 2017 thanks, ill reconnect them on the trees, i hope i left enough sick in the wire to strip and add Link to comment Share on other sites More sharing options...
TheDucks Posted April 6, 2017 Share Posted April 6, 2017 2 hours ago, brichi said: thanks, ill reconnect them on the trees, i hope i left enough sick in the wire to strip and add It is all about keeping the Signal path Timings intact.. (if these were dumb RGB, it would not be an issue) Link to comment Share on other sites More sharing options...
k6ccc Posted April 6, 2017 Share Posted April 6, 2017 As already stated, leave the ground on the strip intact. To explain the issue, here is a little information (at least some of which you already know, but may not have put it all together). In most (or all) pixel strips or strings, each pixel regenerates the data to be passed to the next pixel. That data signal is electrically referenced to the ground signal at that pixel. If the ground from one pixel to another is substantially different, the next pixel (the 51st in this case) may have problems decoding the data properly. Now how would the ground be different between pixels 50 to 51? Let's say that pixels 1 - 50 are all full white, and pixels 51 - 100 are all off. Because of the current being drawn by pixels 1 - 50, the voltage of the ground wire at pixel 50 will not be the same as the voltage of the ground wire at pixel 1 of that string. I edited your drawing to show voltage measurements of the ground wire at various locations All measurements are referenced to the ground terminal of the pixel controller board. For this example we will assume that the pixel output of the controller shown is being fed from power supply #2, and the power injection is coming from power supply #1 (as shown in the drawing brichi linked in the first post). Also assuming a 12 volt system with the power supply outputting exactly 12.000 volts. You will see that there is over a half volt of difference in the ground wires between pixels 50 and 51. Do you think that this voltage difference might cause pixel 51 to have an issue decoding the date? The answer is yes. To make it worse, think about the variations on the ground wire if the pixels are not on all the time, but instead flickering very fast (such a pulse width modulation - the way we get varying levels). Yes, this can make it worse. What can make this really hard is that it might work most of the time, but get just the right combination of voltage drops (and fluctuations thereof) based on differing light levels and all of a sudden it stops working right. And as any technician will tell you, there is almost nothing worse than intermittent problems! Link to comment Share on other sites More sharing options...
brichi Posted April 6, 2017 Author Share Posted April 6, 2017 wow, thank you so much for that info, makes perfect sense AND would explain why last night i was getting random flashes of green when others where whit and with fast stroking it seemed like there were delays on some pixels also i know the voltage is most likely a little different between p/s1 and 2, I always measure the voltage first when i build a new box before i plug in the Pixie board to make sure its not too much to damage it and it always reads between 12.2-12.5 volts between different power supplies Link to comment Share on other sites More sharing options...
k6ccc Posted April 6, 2017 Share Posted April 6, 2017 19 minutes ago, brichi said: also i know the voltage is most likely a little different between p/s1 and 2, I always measure the voltage first when i build a new box before i plug in the Pixie board to make sure its not too much to damage it and it always reads between 12.2-12.5 volts between different power supplies And that's why you only tie the grounds together. You can get power supplies that are intended to parallel, but there is little to no need to spend the money for that higher capability for our purposes. For example, 15 feet from where I'm sitting are 13 power supplies that are outputting about 53.5 volts at up to 2000 watt each, but those cost hundreds of dollars each. Link to comment Share on other sites More sharing options...
TheDucks Posted April 6, 2017 Share Posted April 6, 2017 1 hour ago, k6ccc said: And that's why you only tie the grounds together. You can get power supplies that are intended to parallel, but there is little to no need to spend the money for that higher capability for our purposes. For example, 15 feet from where I'm sitting are 13 power supplies that are outputting about 53.5 volts at up to 2000 watt each, but those cost hundreds of dollars each. That sounds like a Telco Office 'Battery" supply. I remember when I was a frameman, laughing at the 3000A 'spare" fuse hanging on the wall in the MG room (Reseda had 1 'Rectifier", the rest were MG's) Link to comment Share on other sites More sharing options...
k6ccc Posted April 6, 2017 Share Posted April 6, 2017 3 minutes ago, TheDucks said: That sounds like a Telco Office 'Battery" supply. I remember when I was a frameman, laughing at the 3000A 'spare" fuse hanging on the wall in the MG room (Reseda had 1 'Rectifier", the rest were MG's) You are close. Those are floating 3600 AH of 48V battery that is feeding the master site for a regional trunked radio system along with one trunked site with 14 channels and several microwave paths. We have about a day of battery here - in addition to the installed generator plus connection points that would allow us to add a trailored or portable generator. Link to comment Share on other sites More sharing options...
Steven Posted April 12, 2017 Share Posted April 12, 2017 I don't cut the + lead on my 5v pixel tree where I inject power. The Meanwell power supplies I use don't seem to care. Perhaps the 60' of ribbon wire isolating them makes it work. Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now