Making a Scoring Machine

[yba2cuo3]

I've been working on another fencing score box prototype when time permits.

Here are some of the supported features so far:

1. Firmware updates via USB port,
2. Multiple machines can be configured in a wireless network,
3. Incorporates a large (inexpensive) programmable color LED display,
4. Supports programmable keypad assignments on the remote,
5. <$300 in parts per node.

The design will likely be submited under the distribution terms of the GNU General Public License.

Stay tune for updates!

Marc

[griffindm]

Very cool display! I look forward to seeing what you come up with!

[yba2cuo3]

Reduced my original hand-wired GPL Scoring Machine prototype from two boards into one dedicated PCB. Have a look at the attachments.

Marc

[griffindm]

Great design. Do you plan to sell these units?

[yba2cuo3]

I haven't given it much thought outside of supporting our local city clubs. Our club http://www.kanatafencing.com started using a GPL prototype a few months ago & they have been very pleased with it's performance, size, ease of installation & use. Since then I've received a demand for 10 more units across the city. The initial protos were manually hand wired so I needed to come up with a PCB design to speed up the assembly process. I had 50 bare Printed Circuit boards made up to keep the price low. The final PCB implementation ends up being very sturdy with everything mounted on one board. Once assembled, the board simply slides into the extruded aluminium enclosure with the bezels installed at the ends.

The GPL fencing machine is definitely the way to go for small to medium clubs looking for a cheap and reliable solution that doesn't take too long to setup.

Marc

[Gambit36094]

A friend and I will be moving from our club to the local University next semester. Currently the club at the University is completely dry fencing, but we wish to change that. We debated trying to convince our coach to lend us a few of the old extra boxes they have lying around, but if this is a cheap alternative, we may have to give this a go. Or, if you decide to sell, buy a couple from you, considering I've got the engineering prowess of a 3 year old.

[22hansen22]

You could also consider the BlackBox Fencing Scoring Box

http://sites.google.com/site/blackboxfencing/

its pretty affordable, comes with 10 days trial period, if you dont like it, you can just return it!
It meant to be for small clubs with tight budgets!

I am for the idea of an open hardware/software fencing box, and Marcs option looks very good!
but if you just dont want to mess with electronics this could be a good option!
Regards!

[yba2cuo3]

You are probably better off getting the box pre-assembled if you are not familar with electronics. I can offer you a completely assembled & tested GPL scoring machine for $155 USD (without tripod and carrying case) + shipping. I would recommend using a miniature tripod and carrying case (add extra $25), or you can supply your own.

For those handy in electronics who would prefer putting their own together, experiment, etc, I can offer bare PCB's for $35 USD each + shipping.

I've attached the Operator's Manual for your perusal. The currently supported features are:

• Meets FIE 2006 timing requirements for all 3 weapon types,
• Updates available via re-programmable microcontroller & open-source software,
• Simple operation with auto weapon detection & selection,
• Piste support,
• Anti-fraud,
• Performs internal self-test during power-up,
• Ultra-bright solid-state visual indicators with wide viewing angle,
• Circuitry protected against Electrostatic Discharge (ESD) damage (15kV),
• Compact and sturdy aluminum case 4.7 x 4.1 x 1.2 inches (120 x 100 x 30 mm),
• Weights only 10 ounces (280 grams),
• Includes compact 240/110 AC adaptor,
• Includes sturdy tripod stand and padded carrying case (optional)
• Environmentally friendly construction: Lead (Pb) Free meeting Restriction of Hazardous Substances (ROHS) requirements (required for sale in California, Europe, China, etc).

The nice thing about the GPL design is that the code and hardware is completely open-source, which means any updates and/or improvements will always remain in the public domain.

Cheers,

Marc

[SJCFU#2]

Interesting.

The question I'm debating right now is should I buy a complete machine (they do look rather nice) or just buy the board and build something around it that will fit into the shell of an existing machine (perhaps a Leon Paul "bug-eyed monster" - or better yet, a Kessinger)? Buying a complete machine would certainly be easier (and once you factor in the time and labor, probably cheaper). But I already have a collection of scoring machines, most of which are rarely used. On the other hand, building a machine would require starting on yet another project, and I already have several others waiting to be finished.

Are you still using Luxeon Star/0 LED's? I seem to recall reading somewhere that they are being discontinued.

Also, are you selling the PIC16F57 microcontroller already programed?

Finally, how much current does it draw? (I'm wondering if it would be practical to build a 12V battery pack to power it).

[yba2cuo3]

Yes you are correct. The Luxeon Star/O LED's are being discontinued. They worked great in my original prototype (BTW lots of them still available from folks like Future Electronics. etc.), but in the end decided not the best choice due primarily to high cost.

The new design uses 4-pin SuperFlux LED's which are also very bright (flux 8-11lm ), use much less power, and come at a fraction of the cost of the Luxeon's (or any other 1-3W power LED for that matter). The Superflux LED's are also easier to mount with a much smaller physical footprint. Additionally, with the appropriate lens I can achieve full 180 degrees viewing angle, also better than the Star/O LED's.

Current requirements are as follows...

Steady state (micro running & no visual/audio indicators enabled): <50mA
Active (micro running & all visual/audio indicators enabled): <200mA

For a 2000mAh 12V battery, and assuming all indicators are on for only 1/4 of the time, one could expect approx 40hrs of continuous use before needing to replace/recharge the battery.

So battery operation is very feasible if one desires to go that way!

Marc

[yba2cuo3]

Originally Posted by SJCFU#2

Also, are you selling the PIC16F57 microcontroller already programed?

Yes. If you buy a bare GPL PCB I can include a pre-programmed PIC16F57 for an additional $10.

Cheers,

Marc

[griffindm]

Originally Posted by yba2cuo3

I've been working on another fencing score box prototype when time permits.

Here are some of the supported features so far:

1. Firmware updates via USB port,
2. Multiple machines can be configured in a wireless network,
3. Incorporates a large (inexpensive) programmable color LED display,
4. Supports programmable keypad assignments on the remote,
5. <$300 in parts per node.

The design will likely be submited under the distribution terms of the GNU General Public License.

Stay tune for updates!

Marc

Great stuff so far! Have you considered making a score/time add-on unit (w/remote), which could augment the many boxes out there that have neither? A nice extra feature would be a power pass-through that would break power to the box when time expires.

Given the price differential between boxes with and without this, it would probably be a winner in the $200-$300 range.

Of course, if your final design above is cheap enough, it might render it moot.

Keep up the great work!

Dave G.

[yba2cuo3]

Good suggestions. I never considered supporting the older hardware with these features! Makes perfect sense for those wanting to keep their existing units.

I'm estimating the cost delta between a score/timer add-on design (to support older units) vs. a new fully integrated solution would be less than $50 in electronics. Also with the older units, there's still the problem of how to handle FIE updates. The new units would be upgradeable via USB port, networked, programmable display format, etc. which would more than pay for itself after a couple of rule changes.

The programmable RGB LED matrix is the main cost in the new system; i.e. ~$125. However, it also comes with a loads of features; like being able to send any text messages or graphics to the display from a main control office. Examples: display fencer names & piste number before match, display advertisement or product placement in-between matches, etc. Would be very powerfull.

Thanks for your suggestions Dave. Keep the ideas coming!

Marc

[SJCFU#2]

Originally Posted by griffindm

Have you considered making a score/time add-on unit (w/remote), which could augment the many boxes out there that have neither? A nice extra feature would be a power pass-through that would break power to the box when time expires.

Personally I like this idea - in fact I once suggested something similar to the maker of another machine several years ago (unfortunately it never went any further).

The trick would be ensuring that the timer stops as soon as a touch registers, thereby allowing the touch that arrives milliseconds before time would have expired to stand while preventing touches which arrive milliseconds after time expires from registering. This would probably be fairly simple if the add-on box were combined with extension lights - simply use the signal which turns on the extension lights to stop the timer and have the add-on box either cut power to the scoring machine or otherwise block touches from registering (perhaps by breaking the A lines and shorting together the B and C lines on both sides prior to reaching the scoring machine) when time does expire (and maybe sound a buzzer so that everyone will realize that time has expired - similar to having the timekeeper say "TIME!"). Unfortunately extension light outputs vary from one scoring machine to another, and in some cases they may not even be available. In these cases the add-on box would have to receive input via some other means. What that means might be, I have no idea (perhaps a piezo cell mounted near the audio alarm on the scoring machine which activates whenever the buzzer on the scoring machine sounds?).

[brtech]

Photosensors taped to lights would be faster! Also gets you a universal extension light.

It's not a bad idea, although if it were also an extension light, the delta cost in electronics from a entirely new machine is roughly zero (two switches and a buzzer), since you already have the two sets of inputs, the lights and the S&T display). More code, and you lose whatever you like about the machine you have.

[Gav]

I'd just like to say that I love you guys. From a geek's perspective this is one of the more interesting topics I've read on this board recently.

Poor guy asks for help with his HS project (good luck!) and we get about 4 or 5 demonstrations of how it's done.

yba2cuo3: I really like the design of your little scoring boxes. Well done.

[SJCFU#2]

Originally Posted by brtech

Photosensors taped to lights would be faster! Also gets you a universal extension light.

I've seen this approach used with incandescent bulbs but I'm not sure how well it would work with LED's (which tend to emit light over a much narrower bandwidth). Not saying it won't work (for all I knot it may), only that I've never looked into it. But having four individual photosensors does seem like overkill when all you want to do is stop the timer (although I agree it would be an advantage for a universal extension light system).

It's not a bad idea, although if it were also an extension light, the delta cost in electronics from a entirely new machine is roughly zero (two switches and a buzzer), since you already have the two sets of inputs, the lights and the S&T display). More code, and you lose whatever you like about the machine you have.

If you're saying that the total cost of a basic machine plus an add-on unit that combines extension lights, score and time might approach that of a new machine with integral score and time then I probably agree (at the very least the difficult associated with handling multiple items would offset any cost savings). However I was thinking more along the lines of an add-on for existing machines, in which case a sizable chunk of that cost has already been paid and the add-on would merely serve to enhance an existing capability.

[jjefferies]

Originally Posted by yba2cuo3

The nice thing about the GPL design is that the code and hardware is completely open-source, which means any updates and/or improvements will always remain in the public domain.

Hello Marc,
Could you post a link to the GPL descriptions of code etc? I'm assuming you've put it up on SourceForge or some other site. I'd like to look at your software design and see how amendable it is to adding stuff on as a curiousity.

J.

[griffindm]

Marc,

One comment on the design of the simpler box above. From the manual, selecting sabre requires blade contact before turning the unit on. While this will work fine for portable installation, it would not scale as well to a design where a club may want to permanently install the unit above a strip, usually out of direct reach of the fencers. Often in such installations, all of the machines are master controlled by a single wall switch, and I doubt other fencers want you turning their machines off so you can do sabre on your strip.

Keep this in mind for your upscale machine with scoring and timing.

Keep it coming. This is great stuff.

Dave G.

[yba2cuo3]

Originally Posted by griffindm

Marc,

sabre requires blade contact before turning the unit on...

Dave G.

Yes, I can see that being a problem with any box which doesn't have some sort of remote for changing weapon type; or a means of individually turning the power off/on, as in this case. I did consider adding a selection switch on the GPL box at some point, but that wouldn't help resolve this issue.

The microcontroller used in the GPL design; i.e. 16F57, has no spare I/O's, so the design is pretty well at its limits when it comes to adding new features. However, one can always design a simple remote power switch add-on which would plug directly into the DC power connection at the GPL; i.e.

+12VDC supply --> Remote switch ---> GPL power connector

It could even be a wireless control; i.e. IR or RF, with receiver and control electronics being powered from the external +12VDC supply. The GPL box could then be turned off/on via handheld remote, as long as the power switch on the GPL was left in the "ON" position.

I have an alternative solution running on a 18F4550 USB PIC with a wireless remote. The 18F4550 has many more features when compared to the 16F57; i.e. A/D ports, serial ports + many spare I/O's, and uses a USB bootloader for upgrading the firmware.

Thanks for the feedback. I'm always amazed by the breadth of knowledge and experience in this forum!