Does anyone have a nice tutorial for wireless transmissions in between calculators like with a XBee transmitter. One with pictures or schematics would be great.

Unfortunately, no one has actually successfully put such a setup together yet. As we discussed in your other topic (this one), you have a lot of options, but you'll have to set out figuring most of them on your own. Of course, we'll be happy to help.

Well quite frankly, that sucks. I'm bored with high school and I need something to do. My new goal is to make a successful link that can be easily and cheaply repeated. I'm talking custom PCB's and and open-source code. Maybe even a website dedicated just to it. Considering how many people want this I think this would be the best thing to do.

Then you should! I've been wanting someone to make wireless hardware that plays nicely with CALCnet for ages, and if you take the lead on it, I'd be happy to contribute the knowledge of my Bachelors and Masters in EE as well as two decades of experiential knowledge, plus we have many other experts who can chip in ideas.

Thanks for the support! I'll try to at least get an Arduino version working by the end of this year.

I'd be happy to share my CALCnet Arduino code, then; your job is to figure out how to slap the Zigbee or other wireless hardware onto it. Actually, I tend to think a Zigbee might have sufficient onboard computing power to not even need an Arduino.

I Actaully got a ti rf kit that should play nicely with calcnet. I just need to finish learning c++ and hardware stuff to get it to work. The plan is to try and set it up to work with calnet which shouldnet be to hard (and it will run off the calc batteries just fine wheeeeee)

Thanks Kerm that would be great! And GeekBoy1011: Would you mind sharing some information on your setup?

Sure haven't really started yet but it was a thought Kerm gave me the other day.

I came into ownership of a ti msp430 ez430-rf2500. this is a low power micro controller that has a 2.4ghz RF transmitter/receiver on it. it runs off 2.2-3.5Vs and draws like 50ma max and has a bunch of gpio pins i can play with.


That's about as far as i got but it seems promising once i learn some more about how hardware and stuff works. I'm still learning my self ^^

Edit: does any one know a place where i can go to start learning about hardware and stuff like gpio pins I'm kinda stupid in that regard xD

If you can find documentation for this particular microcontroller, check that out. In my (limited) experience with microcontrollers, you can find a list of GPIO ports and code to initialize and access them in the documentation for that microcontroller.

In the link you provided, it looks like this document will have most of the information you need!!
http://www.ti.com/lit/ug/slau227e/slau227e.pdf

note: they refer to the gpio pins as "development pins," not gpio pins.

Yeah I already have this I just have no idea how to go about initializing them. More searching will ensue xD

(I've been pondering this for a while. I'm an experienced EE, and I've been considering this as a real thing, not just a breadboarded up hack)

XBees have a serial mode that would let them replace a serial data connection, and my thought was that using an xbee in that mode would be really simple, with just a small breadboard with a serial calculator interface and power supply. However, the last time we talked about this, someone pointed out that the chat program in Doors (I Think) ack's each bit, which would make doing a really simple serial replacement pretty inefficient and laggy.

Now, I also think that it would be possible to do a WiFi node (Arduino with USB connection to the 84+, WiFi from the Arduino, then porting gcnclient to the Arduino) But wifi support on the Arduino seems to be a bit bleeding edge, and I'd rather debug my own stuff. I'm also not up for the SW work, as I'm more of a hardware guy.

Of course, the biggest problem is that Xbee's are $25 each, and spinning a powersupply board to mount it and power it, will run another $10 for parts, so it's not a cheap prospect.

The Arduino/Xbee approach looks fairly simple from a hardware point of view, and a Case would probably be pretty simple, as there are a bunch of them out there that would probably fit.

The TI MSP430 is an interesting possibility. You could probably do a calculator/calculator chat program with the RF2500, and I've got an eval kit I'd be happy to send off to someone who was doing a link program so they had one for each end. I'm not aware of what's available off the shelf for MSP430 boards to embed into a module so more than a couple of folks could have one, but something is probably available.

The MSP430 approach probably doesn't help with a WiFi approach, I suspect that's a chat to chat only thing

well the msp430 usb debuging thingy that some of them come with (like mine) mimics a com port on the computer. So it would be rather simple to just forward the ports output over rf to the com port which the computer gcn client thinks is the calculator.

And do you mind linking the kit you have?

That's the dev-kit specific documentation. You'll also want a copy of SLAU144 (MSP430x2xx Family User's Guide).

For a fresh design I'd still rather use a CC430 rather than the hardware setup in the TI dev kits, since those have discrete controllers and radio cores. I don't think it would be too difficult to quantize the standard TI link protocol by ACKing bits as they arrive, then transmitting a whole block over the radio once your radio buffer fills enough or there's a longer delay between bits. Latency would still be poor, but it's still a significant improvement over ACKing each bit over the radio link (particularly since TI's protocol is still byte-oriented at the software level).
I think I mentioned it before, but you could provide a transparent radio link by sending messages over the radio on line transitions, in which case it would look to the calculators like a very long link cable, but the latency would (again) be very high since that would require a packet round-trip for each state transition.

With a CC430, costs are fairly low- around $6 for the micro in small quantities, and all it needs beyond that is a small PCB and power supply.

For a WiFi approach, there are some easy-to-interface modules (SPI or SDIO) from Redpine Signals, although they're still not exactly cheap. Some sort of USB bridge to connect the CC11xx radios to a PC is probably the most cost-effective way to do it, and you don't need a whole IP stack on the target when there's a bridge involved.

While we're on the topic, RFDave also weighed in with some interesting thoughts in this thread. I'm personally fond of the idea of somehow tossing the actual link line states around the network as fast as possible, which would allow any protocol, not just CALCnet, to work over the wireless links. However, I understand if for bandwidth's sake it's necessary to convert CALCnet to serial on the wireless transceivers.

Actaully last night i found a bunch of sample code for working with the development/gpio ports slowly working my way through it, hopefully I will be able to come up with something eventually.

That's great to hear, keep us posted. By the way, would you mind sharing the exact product of TI's that you have, as well as how much it set you back?

I linked the page to it Right here ti msp430 ez430-rf2500

And it did not cost me anything I got it from a family friend but the cost of it is $50 sadly. Though I am sure the launch kit with some cheaply sourced parts can work for much cheaper but not be as small (back to like 10-15 usd maybe have to look)

For everything that can do, I'd say that $50 is not an unreasonable price, personally.

oh I have to agree its a pretty neat device. The best part is though that it comes with a usb debugging dongle. meaning I can debug on hardware ^^ makes testing a little easier imo


EDIT:

Does any one know the ma rating on the i/o port? I need to get a resistor to bring it down to 3v so I don't fry my msp430