Author Topic: Intro to the F32-B  (Read 1012 times)

Offline dkulp

  • Moderator
  • *****
  • Join Date: Sep 2013
  • Location: Framingham, MA
  • Posts: 1,500
  • Kudos: 87
    • KulpLights.com
Intro to the F32-B
« on: January 16, 2019, 11:36:09 AM »
The F32-B was designed to leverage the power of a BeagleBone to drive ws2811 pixels. 

F32-B Capabilities:
  • 32 fused and buffered outputs for driving strings of ws281x pixels.   Each string can have 750 pixels (for 40fps)(power injection required) or even more pixels at slower frame rates.   
  • 4 differential RJ45 ports that are compatible with the Falcon differential receivers.  The first 2 ports are dedicated to ws281x strings.  The other two ports can either be used as additional ws281x strings OR they can be configured to output DMX universes or PixelNet universes.   Each twisted pair in the port provides a full DMX/PixelNet universe.
  • v2.0 of the board will add an onboard Real Time Clock, a temperature sensor, and an I2C header for attaching an OLED display.
Requirements:
  • SD card with the latest FPP image.  The image can be flashed to the eMMC on the BBB and run from there.
Notes:
  • Initial (1.x versions) of the board only supported 5v and 12V input power.  With 2.0 and later, a 0ohm resistors were added near the input screw terminals  that can be removed to allow using 24V for that bank of 8 ports.  Removing the resistor disables the Fuse indicator LED's which would be fried if 24V is applied.
  • The BeagleBone mounted under the board.  Because of the size of the board, it is very difficult to get to the SD card slot and USB port.  v2.0 will alleviate that issue somewhat by having a large cutout.   When mounting, make sure to keep this in mind and allow for easy unmounting if access to the SD card is required.   Strong suggestion is to use an short USB extension cable for any usb attached peripherals.
  • Currently, the two "DMX/PixelNet" RJ45 ports are configured together.  If set to DMX/PixelNet, BOTH are set to DMX.  If set to ws281x strings, BOTH are set that way.  There is no way currently to configure one to DMX/PixelNet and the other to strings.  This is an FPP limitation, not hardware, so may change in the future.
Supported BeagleBones:
  • All versions support the BeagleBone Black, BeagleBone Black Wireless, BeagleBone Green
  • The gigabit port on the SanCloud BeagleBone Enhanced consume pin P8-10 so that port will not work.  On the 1.x board, that is port 2.  v2.0 will not use P8-10 and will work fine with the SanCloud BBE.
  • The BeagleBone Green Wireless WIFI adapter consumes a bunch of pins.  While the board will physically fit onto the BBGW, ports 3, 4, and 23 will not work along with DMX 3-6.   v2.0 uses a different pinout and will have even fewer ports working with BBGW.
Supported FPP Versions:
Official F32-B support was added to FPP in 2.0.  All versions of FPP since then support all capabilities of the F32-B.  However, v1.x uses the same pinout as the F16-B+expansion which was added in FPP 1.5.  Thus, on older versions, selecting F16-B should work.   The v2.x board will use a different pinout and will require an update to FPP.

Availability:
The F32-B is available from https://kulplights.com/product/f32-b/
« Last Edit: May 17, 2019, 01:25:26 PM by dkulp »

Offline najetset

  • Newbie
  • *
  • Join Date: Jan 2019
  • Location:
  • Posts: 9
  • Kudos: 0
Re: Intro to the F32-B
« Reply #1 on: January 17, 2019, 10:35:09 PM »
Well as you know I'm certainly interested.


Regarding the differences between your v1 and v2 boards: just to confirm the only differences between the two are:
1. v1 supports 5V and 12V whereas v2 supports 5V, 12V, and 24v
2. v2 supports I2C/RTC
3. v2 has larger cutout


anything else I'm missing?


Also you mention that the BBE's from SanCloud are supported.  Does this also extend to SanCloud's wifi versions of the BB Enhanced?
If so, is the same issue observed as with the BBGW regarding the loss of some ports (3,4,23,etc)?


And you mention the Raspberry Pi RTC's, what about the SanCloud BBE Wifi - 1G + RTC.


Anyways, with the above answers, I'd be interested in one of your v1's that you have left for my initial testing - if that testing works out as expected then I'd be interested in any more v1's you have or if they're all gone, then I'll certainly participate in the v2 group buy.


Offline dkulp

  • Moderator
  • *****
  • Join Date: Sep 2013
  • Location: Framingham, MA
  • Posts: 1,500
  • Kudos: 87
    • KulpLights.com
Re: Intro to the F32-B
« Reply #2 on: January 18, 2019, 06:47:28 AM »
Well as you know I'm certainly interested.

Regarding the differences between your v1 and v2 boards: just to confirm the only differences between the two are:
1. v1 supports 5V and 12V whereas v2 supports 5V, 12V, and 24v
2. v2 supports I2C/RTC
3. v2 has larger cutout

anything else I'm missing?
Surface mount vs through hole parts, but that's about it. 


Quote
Also you mention that the BBE's from SanCloud are supported.  Does this also extend to SanCloud's wifi versions of the BB Enhanced?
If so, is the same issue observed as with the BBGW regarding the loss of some ports (3,4,23,etc)?



The WIFI on the BBE is attached to the BBB via the USB bus.   Thus, it doesn't consume any pins.  However, the accelerometer and gyroscope on the BBE DOES consume some pins.   The good news, though, is that if we disable those (blacklist the modules), the pins are free'd up and things work OK.  That does mean that if you we planning on using the BBE + F32-B on a Drone to or something to have a "moving show", you'd lose some ports.  :)


That said, the antenna on the BBE wifi sucks.   Relatively spotty wifi connection compared the edimax adapters and such we usually use.   
Quote
And you mention the Raspberry Pi RTC's, what about the SanCloud BBE Wifi - 1G + RTC.

I don't have one with the RTC to verify (my BBE is wifi only), but I would assume the RTC on the BBE is wired to the same I2C bus.   Thus, it should work and the pins on the new F32-B would be irrelevant.    That said, it could also be wired to one of the two OTHER I2C bus's on the BBB in which case we would need to update the FPP bootup code to detect it.  Without one in hand, I wouldn't be able to check. 


I'm likely to order a BBE Industrial soon, but that doesn't have the RTC either.   Maybe I'll throw one with the RTC in as well just to have for testing.

Quote
Anyways, with the above answers, I'd be interested in one of your v1's that you have left for my initial testing - if that testing works out as expected then I'd be interested in any more v1's you have or if they're all gone, then I'll certainly participate in the v2 group buy.


Let me dig around to see what I have.  The one I thought I had turns out is missing one (out of 32) of the fuse indicator leds.  I may have one more, but I think the power buck on that is dead (but I may be able to replace).   I'll PM you.

Offline Kensington Graves

  • Full Member
  • ***
  • Join Date: May 2016
  • Location:
  • Posts: 122
  • Kudos: 1
Re: Intro to the F32-B
« Reply #3 on: January 18, 2019, 10:17:22 AM »
Question on power: I see 4 power ports.  Does this mean that each "bank" of outputs can be independently powered with different voltages?

Offline najetset

  • Newbie
  • *
  • Join Date: Jan 2019
  • Location:
  • Posts: 9
  • Kudos: 0
Re: Intro to the F32-B
« Reply #4 on: January 18, 2019, 10:58:34 AM »
That said, the antenna on the BBE wifi sucks.   Relatively spotty wifi connection compared the edimax adapters and such we usually use.   


I noticed that SanCloud offers an option during ordering of 'Onboard' or 'External' antenna.  I wonder if the external is just some sort of ufl connector(s).  Perhaps I'll order one up as well and see.


Interestingly they also purport to be experts in antenna design  :-X
https://www.sancloud.co.uk/?page_id=28

Offline najetset

  • Newbie
  • *
  • Join Date: Jan 2019
  • Location:
  • Posts: 9
  • Kudos: 0
Re: Intro to the F32-B
« Reply #5 on: January 18, 2019, 11:42:53 AM »

Offline klaasb01

  • Newbie
  • *
  • Join Date: Jan 2015
  • Location:
  • Posts: 7
  • Kudos: 0
Re: Intro to the F32-B
« Reply #6 on: February 08, 2019, 09:10:01 AM »
Anyone know the dimensions of the board?
Is there drawing available so I can make a mounting plate?

Offline dkulp

  • Moderator
  • *****
  • Join Date: Sep 2013
  • Location: Framingham, MA
  • Posts: 1,500
  • Kudos: 87
    • KulpLights.com
Re: Intro to the F32-B
« Reply #7 on: February 08, 2019, 11:16:53 AM »
Here's a screen shot with some dimensions on it.   Not sure if that helps.


Offline Twukken

  • Newbie
  • *
  • Join Date: Dec 2017
  • Location:
  • Posts: 6
  • Kudos: 0
Re: Intro to the F32-B
« Reply #8 on: May 17, 2019, 12:09:04 AM »
Got my F32-B board looks great.
Just a quick question on this board. Can I run 100 pixels per port without injection? Will the board be able to handle the current through the traces?
thanks in advance

Offline dkulp

  • Moderator
  • *****
  • Join Date: Sep 2013
  • Location: Framingham, MA
  • Posts: 1,500
  • Kudos: 87
    • KulpLights.com
Re: Intro to the F32-B
« Reply #9 on: May 17, 2019, 07:06:04 AM »
Got my F32-B board looks great.
Just a quick question on this board. Can I run 100 pixels per port without injection? Will the board be able to handle the current through the traces?


Most definitely yes....   


A note on my boards:  I design the pixel power traces to handle at least 6 amps of power using 1oz copper, despite having a 5 amp fuse.   However, when I manufacture the boards, I specify 2oz copper (other than the blank prototype boards I did send out, those are 1oz).   Thus, the power traces for the pixel connectors can handle a LOT more than the 5amp fuse would allow.  On the F32-B, the traces for the pixel power are sized for about 9 amps for a 10degree rise.    Thus, if you really need a bit more power on a particular port, it definitely is safe from a PCB trace perspective to replace the 5amp fuse with a 7 amp.  (note that the pixel connectors themselves are rated for 10amps)


HOWEVER, the screw terminal for the each bank of power is only rated for 30amps.  Thus, that is the thing you need to be careful of.  With 8 ports, if you you draw the full 5amp on each port, you will exceed the 30amp screw terminal specs already.  If you start increasing all the fuses to 7 amp, the chances of exceeding it increase quite a bit.


And to answer another un-asked question:  why pay for 2oz copper if the traces are OK with 1oz?   One reason is the ground.   If we max out 30amps of power to pixels, that means we need to get 30amps back though the ground.   30 amps on 1oz copper would require an equivalent of a 33mm trace from the pixel connectors to the ground pin.  That's a really big space that is needed and there is no way on any of the boards to be able to provide that.   The ground plane obviously acts as a giant heat sync and thus really could sync away most of the heat generated by the much smaller space that the board actually provides.  However, flipping to 2oz copper brings the requirement down to only 16mm which is way more reasonable.   I try my best to provide a quality and safe board so I do spend quite a bit of time optimizing not just the traces, but the ground as well. 


Anyway, lots of technical details to basically say that a 100pixel string at full 100% white draws about 3.3 amps and can easily be used without injection.  :)






Offline Emuney18

  • Hero Member
  • *****
  • Join Date: Nov 2015
  • Location: Apex, NC
  • Posts: 561
  • Kudos: 7
Re: Intro to the F32-B
« Reply #10 on: May 17, 2019, 05:08:27 PM »
That right there is a perfect example of the little details and the thought that goes into designing a quality board. 


Sent from my iPhone using Tapatalk

 

Back to top