Custom controllers?
 
Notifications
Clear all

Custom controllers?

10 Posts
10 Users
0 Reactions
3,189 Views
Posts: 19
Topic starter
(@umodjm)
Eminent Member
Joined: 17 years ago
[#24388]

Just curious, how many of you out there are using a custom controller? Something you designed yourself?

When I first saw christmas lights synchronized to music I was immediately captivated. I googled for weeks and weeks on how it's done. I found Planet christmas first, and thought about building a Hill320. Then I found out about the Grinch, bought a truckload of shift register chips, then abandoned it for a custom board using MCP23S17 chips. I eventually got that working, but didn't like that it took so many "write" operations (in worst case) to change the state of every channel. I've now settled on a whole new design that scales infinitely with a linear "write op" to bank growth (not write-op to channel).

I'm just curious how many other people out there were inclined to roll-your-own vs how many out there had the "why re-invent the wheel?" mindset? For me, the driving force was to see if I could come up with something that was more efficient in bandwidth than the Hill320, Grinch, etc... that, and just wanting to do it for fun. To me it's more fun to design and build something than to build someone else's.


9 Replies
Posts: 3650
(@jhinkle)
Famed Member
Joined: 15 years ago

I got hooked when my Sister sent me an email showing me a Christmas Display - 12/20/2011.

Had my own 24 channel controller designed by March 2012 - it's the only controller I use.

Currently designing my own hi-channel count pixel controller - may use it this Christmas if I get any pixels.

Joe


Reply
dmcole
Posts: 2196
(@dmcole)
Noble Member
Joined: 19 years ago

I have had here on the bench a DMX512 controller for a unique application that I designed for a couple of years -- I may even get it to work in 2013! -- but otherwise have pretty standard stuff. I built a variety Lynx and RPM DMX controllers and have a bunch of the cheap three-channel controllers.

One of the beauties of DMX512 is that you can usually add controllers from a variety of makers (even yourself) without having to have a special output driver for Vixen/xLights etc.

HTH.

\dmc


Reply
DanielSkeen
Posts: 438
(@danielskeen)
Honorable Member
Joined: 14 years ago

I have been working with an Arduino unit to contral 16 RGB channels via ethernet. I have hacked some simple software on the PC side to provide command/control but when time permits I would like to cost reduce this design and off load more of the control on to the external processor(s) with the goal of controlling of 10^6 RGB channels plus special effects(ie this does more than RGB, a special command could make a group of lights chase or make a robot arm move). Part of this scheme would be distributed computing that would off load or eliminate a PC from the system.


Reply
ebrady
Posts: 670
(@ebrady)
Prominent Member
Joined: 17 years ago

Just finished my third Christmas season, all of my controllers and PCBs are custom made and my own design. Every year I do a new controller for fun, I totally dig this part of the hobby in the off season.

Year #1:
Custom 16 Channel Controllers - 64 Levels of Dimming - Custom Communications Protocol
4 Channel SSR Board

Year #2:
Custom 16 Channel Controller - 256 Level of Dimming - DMX Protocol
4 Channel DMX Splitter Board

Year #3:
16 Channel Low Current Controller using V02223 - 256 Levels of Dimming
8 Channel Low Current Controller - 256 Levels of Dimming - DMX
32 Channel SSR Controller version #1 - 128 Dimming Levels - DMX
8 Channel DMX Servo Controller (Hardware complete software still being worked on).
4 Channel Low Current SSR Board (V02223)
4 Channel Low Current SSR Inline Board (V02223)

Year #4: (Working on for 2013)
32 Channel SSR Controller version #2 - 256 Dimming Levels - DMX & Renard - Upgradeable Firmware
New version of 8 Channel Servo Controller - DMX & Renard
More to follow....


Reply
Posts: 2168
(@djulien)
Noble Member
Joined: 19 years ago
umodjm wrote:I'm just curious how many other people out there were inclined to roll-your-own vs how many out there had the "why re-invent the wheel?" mindset? For me, the driving force was to see if I could come up with something that was more efficient in bandwidth than the Hill320, Grinch, etc... that, and just wanting to do it for fun. To me it's more fun to design and build something than to build someone else's.

I started by first making slight adaptations to the standard wheel, then moved on to more extensive designs for custom purposes.

My first year with Vixen I built the standard Renard circuit based on Phil Short's design (Xmus-style with RS232/RS485 input) and AC SSRs on custom PCBs (stripboard) for use with the standard Renard firmware (non-DMX). My goal was to learn about the circuit and practice construction skills. I used the standard Renard circuit and firmware because I was not up to speed on PIC programming and not comfortable changing the circuit design yet, although I did stray a little by using a motherboard/daughterboard approach on the PCBs (112 channel controller). I also ran 7 signals per Cat5 instead of 4 to reduce cabling.

My second and subsequent years I designed custom PCBs for chipiplexing Renards and wrote new firmware for a custom Renard-like protocol, and PCBs for daisy-chained SSRs to reduce cabling. My goal for those was to become familiar with PCB design and fabrication and PIC programming, and to make more efficient use of the hardware (drive 56 channels per PIC instead of 8, with a lot fewer cables), and the data compression in the protocol gave me much larger bandwidth to work with.

Last year I reworked my earlier controllers and firmware that handled AC SSRs with my newer ones that handle RGB pixels, and to rewrite the firmware in C. The goal with that was so I could standardize on one type of controller for everything (AC SSRs, GECE strings, LED strings and strips, etc), and to move to a distributed controller architecture so I can put smaller, simpler controllers directly on the props instead of using a centralized controller.

don


Reply
ErnieHorning
Posts: 4434
(@erniehorning)
Famed Member
Joined: 19 years ago

I designed the first Renard-8 in existence (a prototype). It used Phil Short’s first Renard code.

picdimmerdy1

This was only a stepping stone for the first Renard-64 in existence. It used the same Renard code.
This has been my main controller for the previous 5 years and it used the Renard protocol.

renardtop2es9
renardbottom2tq8

I designed another Renard-8 and it used the Renard protocol last year and I reprogrammed it to DMX for this year.


Reply
dirknerkle
Posts: 10310
(@dirknerkle)
Illustrious Member
Joined: 18 years ago

My first controller was a home-brew, battery-powered, 16-channel parallel controller using two parallel ports, a couple ULN2803A's and a couple female Centronix connectors I had sitting in a box... probably leftover from dismantling a couple printers... Still have it -- you never know when you need something really, really simple.

Then I moved on to Olsen595's, experimented with a couple home-brew Grinch-16's (yah, they actually worked!), then a Ren-C opened up a new world for the Olsen595's... I added SS24's, SS16's, SS8's, Ren64XC's, Ren24LV's.... and on and on until Mrs. Dirk said one day, 'Hey, don't you have to put lights out too -- controllers are nice, but lights... that's what people look at, right?" That's when I realized I'd gone 'round the bend...

Soon I went to the dark side and wireless in 2009 when I made the first Ren-W... Now I have only wireless equipment -- a couple SS24's, a couple SS16s, and a whole bunch of the DIGWDF low-channel-count controllers -- MiniRen8/4XBs, MiniRen8/8XBs, in fact, we actually use all the equipment we sell in the store and I wouldn't have it any other way... If you don't use your own products, how can anyone else trust them to work?

parallel 16

Reply
jchuchla
Posts: 8698
(@jchuchla)
Illustrious Member
Joined: 15 years ago

I've started in this hobby by building others designs, for a few different reasons. First as you stated, why reinvent the wheel. Secondly, there's economies of scale by getting involved in group buys and ordering from established sellers here. I couldn't possibly roll my own one-off designs for the price of the kits. Then there's the parts availability. I don't know enough to design it all out on paper and get it right enough to make a single parts order. I'd end up needing one more resistor of a different value, etc, and then the costs get out of control quite quickly. One of the biggest factors for me is I don't really don't have the knowledge or experience yet to do my own from scratch.
That being said, I've always had interest in electronics as a hobby. I'm finding that this hobby is the perfect place to hone my skills. There's good working designs out there, and the creators are happy to talk about them. There's lots of discussion here around how things work. So I'm now starting to get the comfort level built up to do some of my own designs. I won't be designing another thing that does the same things the same way as something that's already out there. But I do have some "evolutionary" designs I want to try. I'd be using building blocks from existing proven projects and combining them into something that does things the way that works best for me.


Reply
Mactayl
Posts: 2491
Guest
(@mactayl)
Member
Joined: 17 years ago

I made this one a while back, I have used it for the past 3 Christmas's without any failures and it's a real work horse.

Renard Simple 1
Renard Simple 2

Reply
Share: