Author Topic: CESM Searchlight Signals/Signal Controllers - Chapter 2  (Read 2013 times)

0 Members and 1 Guest are viewing this topic.

dcutting

  • Crew
  • *
  • Posts: 359
  • Gender: Male
  • Respect: -58
CESM Searchlight Signals/Signal Controllers - Chapter 2
« on: August 26, 2016, 03:57:19 PM »
0
Hi Everyone,

It's been almost a week since I released my searchlight products, and a lot has changed since then. Many have voiced concerns that the driver boards are too much to deal with, that the signals are too hard to construct, and that  the general lack of support around signals has deterred them from purchasing anything.

Just for reference, the first thread about the searchlights is located here: https://www.therailwire.net/forum/index.php?topic=39727.0

Seeing the underwhelming response from the modelling community, I have disabled the signals so that they are no longer available for purchase (http://www.cescalemodels.com/n-signals)

That being said, I still plan to offer the signals that I have spent months working on, but with a plethora of important changes to make them easier to purchase, assemble, install, and use. This thread will entail all of the ways that I am doing that. I'm not quite ready to show all of the changes that I am working on because I don't know if all or any of them will work, but I will show a few of them to start off.

Change 1: I am scrapping the driver board design. Too costly, too specialized, not very flexible, and overall just under-powered and redundant with the actual signal controller system. The durrent driver boards on the market are too much for what they do. I needed a better solution that would work with not only these signals but also detectors, servos, RFID recievers, etc.

In order to keep costs down I designed an arduino-based driver board that is powered by and gets commands from the regular DCC bus. The DCC commands are decoded by an on-board arduino pro mini ,and the arduino controls all of the I/O



The arduino doesn't have too much in the way of I/O though, and a lot of arduinos can get expensive fast, so I set up an I2C bus standard and designed several boards that you can use to get tons more I/O.

Expansion board one is a PWM output board. Just connect it through the four-pin micro JST interface to your I2C bus on both sides as part of the daisy chain, and then use the screw terminals on the sides to connect LED's. We can put resistors on that limit each individual LED or we can put zero ohm resistors on and let you add your own to either ground or VCC. Very versatile. You can chain up to 62 of these boards on a single arduino board... that's 992 LED's or servos



Expansion board two is a GPIO expansion board. Use these for non-PWM outputs and block detectors.. or come up with your own use. You can only chain 8 of these on one bus, but that's already 128 input/output lines.



The third board we're thinking about is an RFID reader board... but I wanted to see what you guys think about N scale RFID before I jump into this one. We'll see about that one.

The fourth board that we are seriously considering is an ADC (analog-to-digital converter) board that also runs on I2C so you can get signals from analog devices such as potentiometers, current sensors, or whatever you please.

Coding for these will probably work so that the arduino has one address per expansion board to keep everything clean-looking in the IDE. To program the arduino, I plan to make a GUI where you can select the board you are using, the settings for each channel, and it will auto-generate code for you. More on that later.

Change 2: I'm thinking about offering the signals as pre-made, ready-to-run models in addition to the kits. We'll see how the pricing and work load turns out.

Change 3: We're going to offer youtube videos and comprehensive tutorials entailing the setup of our signals with JMRI, your layout, and the arduino pro mini I2C interface.

There's more changes ahead, but these are just the start. If you think we should change something that you don't see worked on yet, let's hear it now so we can fix it.

David
David Cutting

pdx1955

  • Crew
  • *
  • Posts: 687
  • Gender: Male
  • Respect: +477
Re: CESM Searchlight Signals/Signal Controllers - Chapter 2
« Reply #1 on: August 26, 2016, 05:49:02 PM »
0
I would suggest not putting resistors on the driver board for the signal LED's. Many of the signal controller boards (Azatrak,Tam Valley, etc) already have dropping resistors installed and if SMD resistors were in both places then the LED's may be too dim. As it stands now , I can only use your boards for standalone "dummy" signals rather than pairing them up with current products.
Peter

"No one ever died because of a bad question, but bad assumptions can kill"

peteski

  • Crew
  • *
  • Posts: 35258
  • Gender: Male
  • Honorary Resident Curmudgeon
  • Respect: +6638
    • Coming (not so) soon...
Re: CESM Searchlight Signals/Signal Controllers - Chapter 2
« Reply #2 on: August 26, 2016, 06:45:26 PM »
0

Just for reference, the first thread about the searchlights is located here: https://www.therailwire.net/forum/index.php?topic=39727.0

Seeing the underwhelming response from the modelling community, I have disabled the signals so that they are no longer available for purchase.

That being said, I still plan to offer the signals, but with a plethora of important changes to make them easier to purchase, assemble, install, and use.

Change 1: I am scrapping the driver board design. Too costly, too specialized, not very flexible, and overall just under-powered and redundant with the actual signal controller system.

I saw that early on in the thread you mentioned above.  My post: https://www.therailwire.net/forum/index.php?topic=39727.msg486619#msg486619

Hopefully this revision will be better received. But it seems that you are creating a totally new signal control system. Modelers (and people on general) are slow to adapt to major changes.
. . . 42 . . .

dcutting

  • Crew
  • *
  • Posts: 359
  • Gender: Male
  • Respect: -58
Re: CESM Searchlight Signals/Signal Controllers - Chapter 2
« Reply #3 on: August 26, 2016, 10:55:01 PM »
0
I would suggest not putting resistors on the driver board for the signal LED's. Many of the signal controller boards (Azatrak,Tam Valley, etc) already have dropping resistors installed and if SMD resistors were in both places then the LED's may be too dim. As it stands now , I can only use your boards for standalone "dummy" signals rather than pairing them up with current products.

I said in my original post, I can either supply the boards with zero ohm resistors if your signals operate on 5V or I can supply the boards with resistors based on your LED specs. And just to be clear, these boards are intended as a stand-alone system that connects directly to the DCC bus and is controlled by the DCC controller, not by another signal board.
David Cutting

dcutting

  • Crew
  • *
  • Posts: 359
  • Gender: Male
  • Respect: -58
Re: CESM Searchlight Signals/Signal Controllers - Chapter 2
« Reply #4 on: September 07, 2016, 03:30:00 PM »
+1
Hi everyone,

The first PCB's arrived and I was able to put together a PWM expander board and a decoder board. I bought the wrong chips for the GPIO board so that will need to wait a bit. Here's the results:

Arduino on top of decoder. Some components were not populated because I don't plan to program the first version over DCC.


Arduino on Decoder with FTDI board for programming via computer.


PWM Expander Board. This board has 0 ohm jumpers on one side of the board and 330 ohm resistors on the other side so I can test some different options.


I assembled these boards using laser cut stencils, solder paste, an electric skillet, and some ESD safe tweezers. I am impressed at the quick results and because it didn't take much time at all I will probably build all of the boards this way. Special thanks to sparkfun on writing a tutorial on soldering like this.

Tests with code coming soon...
« Last Edit: September 07, 2016, 03:32:10 PM by dcutting »
David Cutting