Netduino home hardware projects downloads community

Jump to content


The Netduino forums have been replaced by new forums at community.wildernesslabs.co. This site has been preserved for archival purposes only and the ability to make new accounts or posts has been turned off.
Photo

SPI clock pin needs amplified


Best Answer SteveS123, 20 January 2014 - 03:46 PM

So after many hours of testing I figured the problem out.  I wanted to post an answer so that this thread wasn't left unanswered.

First off I am running four 5 volt 60 amp supplies and each one is supplying power to 24 strands.  All the grounds are tied together but this is where the problem is.  I am using one of the supplies to supply the 5 volt logic on my controller board and it also supplies the 5 volts to the 24 strands.  If I move the supply wire to my controller to any of the four supplies I get different affects on the LEDs.  The affects are flickers and erratic behavior.  The supplies are switching supplies and are causing me to have this issue.  I am able to control all the LEDs perfectly if I use just one supply but if you do the math it can't support the current load.  I was able to test two segments off of one supply for a short time to verify that it worked.

Second, I did put the driver chip in on my CLOCK line to help stabilize it and that did make a difference ([color=rgb(40,40,40);font-family:helvetica, arial, sans-serif;]OPA633KP).[/color]

[color=rgb(40,40,40);font-family:helvetica, arial, sans-serif;]To sum it up I have to get a single high output supply and the driver chip.[/color]

Go to the full post


  • Please log in to reply
3 replies to this topic

#1 SteveS123

SteveS123

    Member

  • Members
  • PipPip
  • 21 posts
  • LocationMissouri

Posted 18 January 2014 - 07:40 PM

I am trying to drive many strands of ws2801 RGB LEDs. 

I have 96 rows of 45 LED strands.

My MOSI data is fine because I am using AND gates to control the data flow between each strand so the 74LS08 chips are working great.

The problem I'm having is that the clock pin is needed on all the strands at the same time so it is tied to each row, obviously the ND can't supply enough voltage or current for this so I'm trying to figure out how to create a driver for this.  I tried a 74LS244 which is a clock driver chip but the LEDs just flicker in a random order (not the programs intention).  I have isolated the clock pin to one strand of LEDs and was able to verify my MOSI is working through all the gates but as soon as I connect the clock pin to all the strands it fails.

So to sum it up, how do I create a clock driver to supply enough current to feed all the strands.

OP Amp?

Transistor?

 

Thank you for your help in advance,

Steve



#2 Mario Vernari

Mario Vernari

    Advanced Member

  • Members
  • PipPipPip
  • 1768 posts
  • LocationVenezia, Italia

Posted 19 January 2014 - 05:05 AM

I believe that the problem is not the current, because a single input requires an irrelevant amount of current respect the available from an output.

 

The actual problems are two, despite it worth a deeper analysis.

First off, the stray capacitance of a so long chain. The overall capacitance is the sum of each stray's, and that slows down both the rising and the falling edges.

This problem affects the whole strand set: if you see random flickers everywhere and cutting an half or more the problem disappear, that is.

 

Secondly, the delay between the clock edge and the actual data (i.e. MOSI) on the farthest strands.

This problem is more evident on the farthest strands: if you see the random just on the farthest, that is.

 

Just tell us how the chain behaves, then we'll see...


Biggest fault of Netduino? It runs by electricity.

#3 SteveS123

SteveS123

    Member

  • Members
  • PipPip
  • 21 posts
  • LocationMissouri

Posted 19 January 2014 - 03:59 PM

The LEDs do flicker random colors.  If I turn on just the red LED in each strand it is a lot more stable; I can also turn on just the blue or green with the same affect.  If I mix the color(use more of the RGB LED) I load the circuit down and get the flicker.  I did put large 10 AWG wire on the ground for each set of 24 LEDs which helped.  I also built a circuit that uses an OPA633KP buffer amplifier on the clock pin which helped the clocking issue.  I do agree with you that it is a capacitance issue, how do I go about correcting that?  

 

To tell you exactly what I have I will explain.

The layout is a 16 foot diameter floor that is lit up by the LEDs.

I have 48 supports under the floor that have three strands of LEDs on each support.

My controller is on one side of the floor.

I have four 5 volt 60 amp power supplies going to the LEDs and one supply handles 1/4 of the floor.

Each strand has 45 LEDs on it so the total LEDs are 6480.

I have all the grounds tied together.

The only difference between each strand is that the supply cable for each.  Since it is in a circle I sent half the cables around one side and half around the other.  They range in length from 4' to 16'.  All the LED strands are 54".

 

Hope this helps.

Steve



#4 SteveS123

SteveS123

    Member

  • Members
  • PipPip
  • 21 posts
  • LocationMissouri

Posted 20 January 2014 - 03:46 PM   Best Answer

So after many hours of testing I figured the problem out.  I wanted to post an answer so that this thread wasn't left unanswered.

First off I am running four 5 volt 60 amp supplies and each one is supplying power to 24 strands.  All the grounds are tied together but this is where the problem is.  I am using one of the supplies to supply the 5 volt logic on my controller board and it also supplies the 5 volts to the 24 strands.  If I move the supply wire to my controller to any of the four supplies I get different affects on the LEDs.  The affects are flickers and erratic behavior.  The supplies are switching supplies and are causing me to have this issue.  I am able to control all the LEDs perfectly if I use just one supply but if you do the math it can't support the current load.  I was able to test two segments off of one supply for a short time to verify that it worked.

Second, I did put the driver chip in on my CLOCK line to help stabilize it and that did make a difference ([color=rgb(40,40,40);font-family:helvetica, arial, sans-serif;]OPA633KP).[/color]

[color=rgb(40,40,40);font-family:helvetica, arial, sans-serif;]To sum it up I have to get a single high output supply and the driver chip.[/color]






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

home    hardware    projects    downloads    community    where to buy    contact Copyright © 2016 Wilderness Labs Inc.  |  Legal   |   CC BY-SA
This webpage is licensed under a Creative Commons Attribution-ShareAlike License.