Am I barking up the wrong tree?

[Frenchy232]

Hello All,

First post...

So my background is many years ago I started life as a computer engineer, I am far from being an expert but am not phased by technology and can get my head around stuff generally.

In more recent times I have started a business with friends and we organise endurance running / multisport events which are growing in popularity.

To help with the logistics overhead and improve the events overall I have started looking at various timing systems on the market.....you can see where this is going

To date I have been unable to find a system which meets our requirements exactly although some could be adapted but would be very expensive as a result.

The requirements are;

To be able to time an competitor over a course from start to finish and varying number (max. 10) checkpoints on route.

The competitors would ideally carry a physical device (RFID chip maybe).

The checkpoint reader must be robust and have enough battery power or the ability to hot swap batteries to last say 24 hours.

The checkpoint reader should be able to store read data locally to be downloaded later to a laptop or similar and ideally (nice to have) if it could broadcast in real time via GPRS or similar then great!

Costs need to be sensible but do have a few K to invest but would want to be confident the end product is achievable before we spend to much , so a cheap POC needed.

Interested in peoples thoughts on whether I am being far to ambitious or is this something you guys knock up on a Sunday afternoon or a regular basis

All advice / questions welcome.

Thanks,

Frenchy.

 

[hzuiel]

I'm guessing a few K isn't going to cut it. GPRS is a cellular service. If you're talking about having cellular devices you're also talking about 10 subscriptions to the phone company for cellular data, every month. That's quite a large recurring bill. What kind of distances from your central location would these checkpoints be?

 

[Frenchy232]

Hello,

I am less concerned about the GPRS element and view this as a nice to have, whether there is a pay as you go option which the cost could be offset against the events is where my head was at.

I can find no end of RFID readers but it is the battery life element and ability to write to memory which is challenging me, not sure how challenging finding and integrating these components would be?

Cheers

 

[Frenchy232]

Sorry to answer your question they will vary up to 50 ish miles in mountainous terrain.

 

[Solidus]

It is ambitious, but I'm sympathetic to ambition - I've had a history of biting off more than I can chew with designs - I also intend to startup a semiconductor fabrication company in the future, so "no" has never been my method of operating haha.

That being said, there are several hurdles you have to overcome here.

Does the device have to identify a runner? That is, does it have to be such that if Runner 1 and 2 pass the checkpoint at the same time, they are distinguished and time-logged individually?

If so, implement an RFID with a unique key - 8-10 bits should be enough - for each runner, and have the reader log that to memory. If not, you could implement it with a laser on one end, detector on the other - simple "tripwire" concept.

If you let me know more about how you intend to carry this out, I can help work up schematics for you and construction details.

With the advent of high-capacity batteries (think batteries for high-powered RC cars), battery life shouldn't be too much of an issue.

Writing to memory on a microcontroller is a simple operation - probably the cheapest method is to store it to memory until the device is recovered later - as hzuiel posed, GPRS and cellular communication is expensive, and mountainous terrain is liable to not have good reception. (Many carriers offer "business" rates, if you decide it's the best, you might look into that)

Or, at each checkpoint, you could have a wireless transceiver - a radio - and have someone ride the course on a bike with the "master" device that could poll each checkpoint and pull the data.

If you come up with a more detailed outline I can probably help you further.