New to arduino with ambitious project for work

Hi,

Im hoping to obtain some help as i have been tasked with a project for work including an arduino system and have to purchase the arduino units before i begin the project to allow for lead times.

My project is the remote control of an ISDN unit over satellite modems(analogue audio) with DTMF tones and arduino.

I have identified the MT8880 IC as the DTMF transiever on both sides and need only to specify and buy the arduino types i will require but as i have not yet touched an arduino this seems quite tricky.

If i supply a list of pre requisites and a brief system diagram can i grab some recommendations for arduino boards?

Requirements:

Arduino unit 1- data connectivity to MT8880 IC for recieving/transmitting DTMF code information.

  • connection to an LCD display for feed back from the other side of the satellite link
  • connection for a 4 x 4 keypad (numbers 1-9, * and #, A B C D
    Arduino unit 2- data connectivity to MT8880 IC for RX/TX DTMF code info
  • RS232 serial connection for control of the C500xr ISDN unit.
  • connection to the PD25 unit for an automatic (timed) disconnect command should the link be lost.

Any recommendations for the units i should purchase would be very much appreciated.

Thanks

Ian

but as i have not yet touched an arduino this seems quite tricky.

Maybe you should get your hands on an Arduino ASAP. Are you talking about a one off or a small series? A little bit more context would be helpful.

thanks for your reply,

Basically its a case of "see if this is possible and if so we will roll them out across the fleet" with a fairly short timescale, hence why i need a recommendation of arduino model capable of what i want to do. so i can learn and create the project without costing the company a second set of arduinos.

Having looked through tutorials and literature i feel it wouldnt take me long to get my head around the project itself i just need somebody with a little more experience to make sure i dont buy the wrong thing befre i know any better and cost my company money.

Ideally i need to place my arduino order tommorow so any recommendations would be appreciated.

Ian

How about posting links to some of the parts you want to interface to?

You could play it safe and just get a Mega for both sides, will give you plenty of IO that way and also 4 serial ports each if needed.

So for instance, you could use 8 pins for a 4x4 keypad, serial/SPI/I2C interface to the display, and however many pins are needed for the MT8880 (this part? http://www.digchip.com/datasheets/parts/datasheet/537/MT8880C-pdf.php) And similarly on the other side however many pins are needed for the MT8880 and the C500xr ISDN unit, and whatever the connection is to the PD25 for the disconnect command (not shown in your diagram).

If only 1 serial is needed for the display on one side, and the C500xr on the other, then having 1 port free for PC debug print statements is also helpful. If you envision more RAM being needed for data buffering, and the amount of IO is smaller (32 pins), than a '1284 based board with 16K of SRAM and dual serial ports would be a good option as well.

thankyou for that last one, very help full.

iv decided to drop the LCD display as all communication will be through the MT8880 (DTMF) for status messages anyway.

So broken down,

minimum 20 I/O for each side (ideally more so i have spare to play with so around 24) 9 pin serial for control of the c500xr (so as per your recommend 2 serials for that and PC comms) Ram figures im not to sure about to be honest but i cant imagine needing anything hefty

so do we think the 1284 based board is best?

thanks again

Ian

Yes, for that amount of IO. 32 IO and dual serial meets your needs, 50+ & 4 serial is overkill. Need a shield for the RS232 levels, Atmega itself only does 0/5V TTL. RS232 uses +/-5V to +/- 10V swing. Little adapter card like one of these can be connected http://store.nkcelectronics.com/rs232-to-ttl-3v--55v-convert232356.html http://store.nkcelectronics.com/rs232-to-ttl-converter-board-33v232335.html http://www.sparkfun.com/products/449

How are you with electronics? I would strongly recommend considering a demo with stock Arduinos, then a batch of PCBs and your own scaled down components for your field devices… If that’s reasonable with your skill level, or resources available to you.

I would lean toward the 1284 out of principle… serial I/O and pin count, mainly. The LCD is of negligible cost and complexity, so if it would be at all useful, you should include it.

Also – again, this assumes you know your way around a soldering iron, and can read a datasheet, consider brewing your own TTL-to-RS-232 adapter with a MAX232 or equivalent.

(Maybe I’m showing my DIY bias here, but I consider shields to be a prototyping tool only. That’s just me.)

My project is the remote control of an ISDN unit over satellite modems(analogue audio) with DTMF tones and arduino.

Basically its a case of "see if this is possible and if so we will roll them out across the fleet" with a fairly short timescale, hence why i need a recommendation of arduino model capable of what i want to do. so i can learn and create the project without costing the company a second set of arduinos.

You want to build a communication gadget to be rolled out "to a fleet" and are so short on budget that you can not or do not want to pay prototype parts? You are also on a tight schedule and your time is less valuable than said prototype parts? Also you have never done something like this before. So my conclusion is that you are not at all prepared for this kind of budget. This does not imply that you can not do it. But you have to decide what you value more: time or money? If time is more critical get some experienced engineer / developer to help you. If money is tight add sufficient time for development. But also consider the cost of your own time.

You should also setup a risk mitigation / fall back plan.

Just to give you an idea what it means to design a simple circuit have a look at my web page and my shield. This is as simple as it can get. Even this took 3 prototypes before it was good enough for mass production. There is a lot of stuff that can fail. Your stuff is going to run in cars. From an electrical point of view cars are some of the worst environments to get electronics to function reliably. So you have to expect more than 2 prototypes and quite some piloting.

Things that are bad for you in a car: the power supply is "very dirty". All connections must be vibration resistant (including solder joints). Everything must be able to deal with an extended temperature range most Arduinos are not prepared for.

I'm not sure in this case "a fleet" means "...of automobiles". O.P., does it? If so, that definitely has its own set of requirements.

Also, I think the point of asking for recommendations was so as not to buy (arbitrary value) Unos only to find out they aren't the right fit, so now we need (arbitrary value) Megas instead. :-)

Not Megas, that’s overkill.
Just some '1284 boards with an RS232 adapter.
I have 19 bare boards on hand, could have assembled boards with bootloader shipped in about 2 weeks (need to buy some parts).

Then can modify it by pulling all parts back into one board for final use.

Hello again,

I'm regards to the fleet I'm referring to, i am an SNG engineer and the fleet is around 14 satellite trucks with substantial ac power systems so that isn't a concern. Vibration is always a concern but is taken as an acceptable risk that comes with the territory

My electronics knowledge is far from basic but not extensive so I'm quite confident with the prototyping. Cost is not a huge issue I just want to avoid unnecessary company spend. The rush aspect is just to get the arduino boards ordered so I can get to work on this, hence I think custom boards. With a 2 week lead time is too long sorry, I just need a recommend for stock boards, shields that meet my needs without being massive overkill, but if that is necessary to get them quickly then so be it.

Thanks again for your help

Ian

Ok, than Mega's with an RS232 adapter.

Guess I need to build up some '1284 boards so I have them as stock items.

I can understand the need to get started quickly, but I don't see the sense in buying enough parts to equip your whole fleet before you've proved the concept. You could easily find that you can't make it work, or it is not viable for some other reason, or your choice of parts was wrong. The more sensible route imo would be to put together a working prototype and then once you have that working make the go / nogo decision and decide what hardware you want for the 'production' units. For example, you might decide that you want ruggedised Arduinos given the environment you're in. But it would be a waste of time and money to buy large quantities of these before you know that they will do what you want.

Agreed. That said, some Radio Shacks are purportedly carrying Arduinos.

Not mine. I live in an igloo. I'm lucky I can find 9v batteries without having to order from Amazon.