News:

LATEST RELEASE:  FPP 8.4 - Download from here - https://github.com/FalconChristmas/fpp/releases/tag/8.4

+-+-

+-User

Welcome, Guest.
Please login or register.
 
 
 
Forgot your password?

+-Site Stats

Members
Total Members: 16850
Latest: rjslightshow
New This Month: 28
New This Week: 1
New Today: 1
Stats
Total Posts: 135308
Total Topics: 16950
Most Online Today: 117
Most Online Ever: 7634
(January 21, 2020, 02:14:03 AM)
Users Online
Members: 2
Guests: 51
Total: 53

Adding additional PSU to R2R F16v5

Started by aohlson, December 11, 2024, 09:51:04 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

aohlson

Hello,

I have thoroughly searched forums and internet for info before posting.

I have started purchasing hardware for my first show (2025).

I bought a R2R F16v5 controller from pixelcontroller.com. I am looking to add a second PSU before I start to get things setup.

Does the PSU need to be wired in parallel? Or can I use one PSU to supply ports 1-8 and another to supply 9-16.

**Would be adding an additional 12V Meanwell 350W PSU

Poporacer

Quote from: aohlson on December 11, 2024, 09:51:04 AMDoes the PSU need to be wired in parallel?
No

Quote from: aohlson on December 11, 2024, 09:51:04 AMOr can I use one PSU to supply ports 1-8 and another to supply 9-16.
Yes, it speaks about it in the description page and although there isn't an F16V5 Manual, many things in the V4 manual apply.
If to err is human, I am more human than most people.

aohlson

#2
Is there any advantage to wiring them in parallel?

I may be answering my own question but would the advantage be to only need one power cord?

If I decide to run them in parallel would the attached diagram be correct?You cannot view this attachment.

Poporacer

Quote from: aohlson on December 12, 2024, 09:11:53 AMIs there any advantage to wiring them in parallel? 
Absolutely none, and most of the power supplies that we use do not have current sensing and wiring them in parallel could cause one of the power supplies to fail.
If to err is human, I am more human than most people.

aohlson

Excellent, sounds easier (and safer) to run them seperately. 

k6ccc

I think there may be some confusion between the people involved.  From the drawing, it would appear that aohlson is talking about having the AC inputs paralleled, and poporacer's answer relates to paralleling the DC output.

Using a single AC cord for the two power supplies is fine - in reality, even if you have two cables, but plug them both into the same outlet or even two completely separate outlets on the same circuit, they are functionally the same except for there being only one cord so the current capacity is less (not generally an issue).  Your drawing is correct for using one AC cord for both supplies.

HOWEVER, your drawing has a massive error on the DC side.  You tied the positive of one supply with the negative of the other supply - DONT DO THAT!!!!!  The two NEGATIVES need to be common, but that will happen in the controller board since at least every pixel controller I have ever seen that has dual DC power inputs has both negative inputs tied together.  You can tie the negative together between the two power supplies, but it is not generally needed - but won't hurt anything if you do (and has some potential advantages).
Using LOR (mostly SuperStar) for all sequencing - using FPP only to drive P5 and P10 panels.
My show website:  http://newburghlights.org

Jim

Poporacer

Quote from: k6ccc on December 12, 2024, 03:46:27 PMI think there may be some confusion between the people involved.
And the way the OP explained it.

Quote from: k6ccc on December 12, 2024, 03:46:27 PMFrom the drawing, it would appear that aohlson is talking about having the AC inputs paralleled,
The drawing was not included in the post when I responded.

Quote from: k6ccc on December 12, 2024, 03:46:27 PMand poporacer's answer relates to paralleling the DC output.
And that was based on the statement made:
Quote from: aohlson on December 11, 2024, 09:51:04 AMOr can I use one PSU to supply ports 1-8 and another to supply 9-16.
That didn't sound like he was paralleling the AC side.

Quote from: k6ccc on December 12, 2024, 03:46:27 PMYou tied the positive of one supply with the negative of the other supply - DONT DO THAT!!!!!
@k6ccc is totally correct there!

Quote from: k6ccc on December 12, 2024, 03:46:27 PMYou can tie the negative together between the two power supplies, but it is not generally needed - but won't hurt anything if you do (and has some potential advantages).
It not needed, but it can cause problems. I don't know of any advantages either.
If to err is human, I am more human than most people.

k6ccc

You are correct, most of the description was about paralleling the DC output, but some implied the AC side and the drawing clearly showed paralleling the AC input.  I was just trying to make sure everyone was on the same page - avoiding confusion.  The primary reason for making the post at all was the glaring DC error in the drawing.

One advantage of tying the power supply negatives together at the power supply end relates to users that have somewhat longer runs between the power supplies and the loads.  It can potentially reduce voltage drop assuming that only part of the prop is drawing more current at any given instant.  Not much of an issue if the power supplies are in the same box as the controller with only a few inches of wire.  However if the power supplies are a further distance from the controller, the ability to split the current between the negative wires (or using one larger wire) could reduce voltage drop.  The other is for users with for example a F16 controller and a expansion board that are each fed by different power supplies.  if you don't tie the negatives together (somewhere), there could be a quite different negative voltage between the two which could make the data buss between the two not very happy.  In the case of my pixel tree, both of those situations apply.  The power supplies are about 8 wire feet from the controllers, and the controller box has a F16v3 and the 16 port expansion board (likely two of those next year).
Using LOR (mostly SuperStar) for all sequencing - using FPP only to drive P5 and P10 panels.
My show website:  http://newburghlights.org

Jim

Poporacer

Quote from: k6ccc on December 12, 2024, 05:32:52 PMIt can potentially reduce voltage drop
I don't understand how that is possible, current is a continuous path and voltage drop is largely due to the path through the wires. It goes through the positive wire that is connected to the PSU, through the pixels and returns down the negative line back to the PSU. I don't see how in a normal path, it would reduce voltage drop. Can you elaborate?

But have you considered the case where there is an internal failure in one of the power supplies and the V- is disconnected?
 
What most people don't realize is that the ground wire carries the exact same current as the V+ wire. We put fuses on the V+ wire to protect the wires in the circuit. In most "normal" applications (one power supply per circuit), there is only one path for current, out the V+ and back through the ground. So, the current on the ground could never exceed what is going through the fuse on the V+ line- so all is good and no need to protect the V- wire from overcurrent. But when you add a second power supply AND tie the grounds together AT the PSUs, you are creating an alternate path to ground and the possibility of overloading the ground wire if one of the "internal" ground lines fail. I don't think anyone uses an oversized ground wire between the 2 PSUs (large enough to handle the load from 2, or more power supplies) or put a fuse on the V- wire? Overloaded wires can lead to fires. It isn't a very common form of failure.
If to err is human, I am more human than most people.

k6ccc

Quote from: Poporacer on December 12, 2024, 08:17:31 PMCan you elaborate?

Sure.  Because of the lighting at some particular instant, one power supply is supplying 50 amps and the other supply is supplying 20 amps.  Voltage drop across a resistor (or a wire in this case) is E=IR.  So the wire with 50 amps is going to have 2.5 times the voltage drop.  If the negative wires are tied together at both ends, the negative current is split between them fairly evenly so the voltage drop is about even, and the resulting negative voltage potential between to two controller cards is essentially zero.

If the negative lead in one of the power supplies opens, that power supply can't output any current because there no return path to the power supply internals.

BTW, to be picky, the negative may or may not be grounded to the green wire ground.  In most of the types of supplies we are using, the power supply negative, it floating from ground.  Although commonly referred to as "ground", a better term would be "return" or just negative.  Also remember that for our purposes, we use a negative return, there are applications that use positive return (Telecomm is almost universally run on -48V DC). 
Using LOR (mostly SuperStar) for all sequencing - using FPP only to drive P5 and P10 panels.
My show website:  http://newburghlights.org

Jim

aohlson

1) Thanks for the discussion.
2) Let's try this again.

I am looking at one AC unit. 

I swear I watched a YT video that was tying PSU together, and they had the V+ tied to the V- on the other unit, but it sounded wrong, and it was somewhat difficult to find a video on someone wiring PSU together. 

And if I understand correctly, tying the V- together isn't necessary but doesn't hurt to do so (and can have some potential advantages) 

k6ccc

Using LOR (mostly SuperStar) for all sequencing - using FPP only to drive P5 and P10 panels.
My show website:  http://newburghlights.org

Jim

Gilrock

I have a controller box that has a another box right beside it with only power supplies for power injection.  If I don't connect the 2 DC grounds together between boxes I get flickering in the pixels where power injection is done.  I'm pretty sure most of our power supplies do not DC couple the V- to earth ground so the ground reference inside the power supply is basically floating and tying two supplies together forces them to float to the same reference.  I've wondered whether I could solve this by putting a jumper between the V- and earth ground inputs on both supplies.  I know with audio equipment like guitar amps a lot times they actually give you a ground lift switch that breaks that connection because it can couple noise into the system so not sure if the same thing would happen with our pixels.

Support FPP

+- Recent Topics

FPP 8.4 released! by lrhorer
Today at 11:30:42 AM

Swapped PiCap V1 with V2 - Pixels will not light in test by twseeman
Today at 10:24:13 AM

Does the Falcon PiCap v2 support Renard Output? by dakingus
January 18, 2025, 05:45:25 PM

K2-Pi-Servo by breese
January 17, 2025, 01:47:06 PM

Big button url question by darylc
January 14, 2025, 05:54:32 PM

Display TEXT to LED Panels form REST API by CaptainMurdoch
January 14, 2025, 11:55:25 AM

Differential Board: PSU & Enclosure by jnealand
January 14, 2025, 07:42:55 AM

wine from matrix with octoscroller control by Poporacer
January 12, 2025, 10:40:05 AM

xLights.org forum registration waiting for approval for several months. by darylc
January 11, 2025, 12:42:54 AM

FPP sequence fade by CaptainMurdoch
January 10, 2025, 08:19:33 AM

Powered by EzPortal
Powered by SMFPacks Menu Editor Mod