Scoring machine repeater lights again

[aserapig]

Hello all -

Lucky I happened upon this wonderful resource. Won't go into the long story yet - but suffice it to say I have been drafted into this exact situation.

I've only been fencing a year, but at 34 years old find myself wearing t-shirts that are older than 90% of the club I'm at. Our armourer has taken a liking to me as I actually know what he's talking about since I have a background in Physics and am now in the computer industry. That said - I haven't dealt with basic electronics for 14 or 15 years!

Anyway - between Christmas and New Years he pulls me aside and says he has a project for me. Seems he procured some rope lights and would like to turn them into expanded scoring displays. They are your basic, residential, plug into the wall rope light from the Holidays (Red and Green - go figure....)

Anyway - Started to look into it and quickly got a head ache and a reminder of why I drank so much through college.

Now into what I have found on my own before finding this discussion -

- I am using Favero Full Arm 01's
- They do have a phone jack output with 6 pins.
- From a PDF off their site for their lights repeater (see link) They have a wondrous little nugget saying " you can cut the cable and extend it with only two wires and connecting one wire to the two central wires and the
second wire to the two lateral wires (the cable has four wires)"
- I was informed by the "little birdie" that Favero does output with a digital TTL signal.
- There are wondrous little guys out there called a picaxe that are very easily programmable to interpret the signals and kick out another command to put in front of a relay of you want to... all for around $4 to $24 depending on size.

I was originally going down the path that has been nicely summarized already with the the SSR's and hoping that the output would be as simple as the proverbial on/off with a designated pin for each light to trigger the relay.

So - as someone else put it - is it easier to "hack" the board and tap off the line feeding the existing lights? Take that and create my own "output connector" and feed into my little black box of simple relays?

I have to admit - now that I'm getting into this project, a bit of the old education is coming back and I am very much eager to actually make this....

[Sboard1019]

The way i would do this, though it might not be the easiest/best, is to use a relay by connecting the output of the scoring machine to switch the mains to the rope lights. Though this means you will need to plug in the scoring bock and the rope lights into separate outlets.

[KD5MDK]

I really like the idea of christmas lights as repeaters. Tacky but really big/cheap. You could even run them down the sides of the strip so the referee can stare at the floor (epee)

[SJCFU#2]

My experience has been that the hard part with extension lights isn't the lights themselves - it's the controller that drives them.

Most machines use a simple arrangement where one pin is energized for each light that comes on. The controller uses this signal to switch extension lights on and off as required. An opticoupler to usually included somewhere in there to isolate the machine from whatever power source the extension lights are using. Unfortunately Favero machines do things differently, output a serial signal rather than simply energizing a single pin for each light. I don't know if anyone outside of Favero knows exactly what protocol they use however you may want to contact Brtech (I believe that he has been looking into this).

[brtech]

Well aserapig, I told Tim that it was bit serial and that was indeed the problem here, but I don't think it would be hard to reverse engineer the protocol. He thought you were just the guy to do that

I will take one shot at asking Favero if they will reveal it. I doubt the answer will be favorable, but I have Mr. Favero's email, and I'll ask.

I do recommend:
http://www.futurlec.com/Relay_4.shtml or
http://www.futurlec.com/Opto_Relay_4.shtml
to drive the extension lights: 4 channels, 10A/channel, TTL drive, $15

You could probably use:
http://www.futurlec.com/PIC16F628_Controller.shtml
if you are in to PICs.

I'm playing with AVRs at the moment:
http://www.futurlec.com/ET-AVR_Stamp.shtml

[KD5MDK]

I can't imagine it's that hard to hook a scope up to the pins and watch what happens when a light comes on. Assuming, of course, that you have a scope.

[jjefferies]

I can't imagine it's that hard to hook a scope up to the pins and watch what happens when a light comes on. Assuming, of course, that you have a scope.

Actually a storage scope. And a digital analyzer is more of what is called for. Of course they can be had but who has the money outside of a lab. I've got a little analog scope. No go for digital.

I hope you share whatever you find out. I've got a small project with a PIC processor trying to implement the small scoring box that was touted around. Besides trying to follow the schematic and the software the hardest problem so far is finding a 74HCT595 chip to handle conversion of a serial to parallel output.

[brtech]

If it really is TTL (kind of doubt it, because TTL doesn't go down long wires too well), or you get the physical interface right, I think your PIC would do well enough as a digital sampler. It's not going to be high frequency.

Either it's a UART-like thing which self clocks, or one wire is the clock and the other is the data. Either way, can't be hard to figure it out. It could be a two way protocol, but it's probably one way (machine to extension and that's all there is).

[jjefferies]

If it really is TTL (kind of doubt it, because TTL doesn't go down long wires too well), or you get the physical interface right, I think your PIC would do well enough as a digital sampler. It's not going to be high frequency.

Either it's a UART-like thing which self clocks, or one wire is the clock and the other is the data. Either way, can't be hard to figure it out. It could be a two way protocol, but it's probably one way (machine to extension and that's all there is).

Not sure if you are speaking to the original poster's issues or to my off the cuff comments about a low cost scoring machine. Here is the original URL that I was working from.
http://home.iprimus.com.au/dougburbi...ontrolBox.html
Note that the eclipse box uses a similar, one microcomputer, chip sort of solution. This solution used a PIC16F627-10/P or -20/P processor. Then I found that Futurelec sells a board with a PIC16F628 processor
http://www.futurlec.com/PIC16F628_Controller.shtml
(Same as the 16F627 but with more flash memory) for USD $15. I thought why not give it a try. Problem is that the original solution calls for use of a serial to parallel chip to provide the signal light output. Fair enough, the chip only costs $.28 cents or $1.00 in single units but so far I've not found anyone interested in selling a single chip for prototyping. Which is where I'm currently stymied in trying this. The Futurelec board has a small area set up for extra chips and buffers the I/O pins from the outside with a 74LS04 and 74LS07 chips and also has a serial chip - which I have no use for. The previous poster had a comment about the speed of the chip. With clock speed of 10MHz and a sampling rate of around 2 MHz I would hope that is fast enough.

Hope this interests someone else. The original URL has a small program which runs the box as an epee or foil box. It can easily be combined with other boards from Futurlec to driver repeater lights etc. The Futurlec site is worth investigating.

[brtech]

We could use the PIC as a digital analyzer to reverse engineer the Favero protocol. Any fast computer where we could do a real time sampling of a couple of digital bits at a couple hundred kilohertz will do.

Futurlec doesn't sell the 74HCT595 but they do sell the 74HC595. That part can only drive 6ma, which really isn't enough for two LEDs (it's enough for one).

There are a couple of ways around that. The easiest is probably to use the '595 outputs to a 74HCT244. This is an octal line driver. Futurlec will sell you one of each. Connect the Q outputs of the '595 to the inputs of the '244 and connect the outputs of the '244 to the LEDs. Don't you need the LEDs and resistors, and other small parts to make the box out of the PIC board?

[brtech]

aserapig, I've been thinking about how to reverse engineer the protocol, and decided that the first thing we need to do it to figure out what the voltage levels on the wire are. Unless you have a storage 'scope, I think the easiest way to do that is to just figure out what is driving the lines.

I suggest you open up the FULL ARM and trace the wires back to whatever chip it's connected to (unless it's some transistor, which I doubt) and tell us what you find.

[brtech]

Wonder of wonders, Mr. Gino Favero replied to my email with news that they have a relay board available for building your own extension lamps (for 165 euros) and HE GAVE ME THE PROTOCOL.

I don't feel right just posting it, but I'll give you the part you need. It's indeed a serial line, 20 ma current loop. 2400 baud N/8/1. It sends 10 byte strings continuously. The first byte is always hex FF. The 6th byte has the lights:
Bit D0 = Left white lamp
Bit D1 = Right white lamp
Bit D2 = RED lamp (left)
Bit D3 = GREEN lamp (right)
Bit D4 = Right yellow lamp
Bit D5 = Left yellow lamp
The 10th byte is a checksum of the other bytes.

So, even a PC serial port could be used, with the right level converter. 20ma current loop used to be very common for this kind of application, because it can drive long wires with little difficulty. A simple resistor will work to receive it. For a 3.5V swing, .02A, R=E/I, 3.5/.02 = 175 ohms, and then you can put it into the PIC serial port (NOT using the RS232 level converter, right into the serial pin of the chip). It really probably would be best if an optocoupler were used. If I was building it, I would probably use an optcoupler. That's more parts though.

Current loops have two wires (it's a LOOP). It's undoubtedly the middle two wires in the RJ-11 jack, but they didn't specify.

So, sometimes it works just to ask nicely!!!