EDIT: Admin, could you please remove this post.

Very nice, I particularily like the idea of creating ones own slave helper circuits to overcome certain shortcomings of the Netduino (or other boards). Also,taking on a challange and thus make oneself learning something new (like C programming in your case) is both impressive and clever. Congratulations and thanks for sharing! I recently bought a PSoC 1 for creating my own little ASICs in a way similar to how you did and I too will have to learn a new "language" and a new technology. We'll see how that goes though...

With the right controller, on, off, forward and backward (no speed control) would require 2 pins per motor. So depending on which Netduino you got, you could potentially control 8 to 11 motors and with an IO expander IC you could control many more.

I got a Netduino Plus that can be flashed to behave the same as a regular Netduino (i.e. flash image w/o Networking) that I'm not really using if that works for you. I'm in Sweden which might render some additional shipping costs but I don't know how much.

The STM32F405xx datasheet says "Up to 140 I/O ports with interrupt capability" depending on package:

http://www.st.com/st.../DM00037051.pdf

 

They don't explicitly say that you can enable interrupts on all ports simultaneously and I strongly doubt you can.

 

The reference manual says the interrupt controller has 82 interrupt vectors:

http://www.st.com/we.../DM00031020.pdf

 

Since the interrupt vector table only has 82 entries of which many are used by peripherals, there's no way all GPIOs can be configured for external interrupts. This is confirmed in chapter 10.2.5 that says there are 16 external interrupts.

 

The NP2 has has a maximum of 22 GPIOs and it seems only any 16 of these can be used with interrupts.

Yes, 8 bit mode would be faster - in 4 bit mode, you have to transfer the low and high nibble of each byte separately while as in 8 bit mode, you transfer a whole byte in each transaction. Will you be using a separate device (shift reg or similar) in between the Netduino the LCD and if so, may I ask which?

Hi there and welcome to the Netduino forums. Debug messages should show but I suspect you debug output is direct to the immediate Window. Either switch to immediate window while debugging or change Visual Studio settings to direct debug message to the output window.

Hi michel!

 

What is your setup?

 

Do you have a Netduino Plus attached to your network where the SD card you wish to modifiy is inserted into the CD card slot of the Netduino?

 

What kind of device needs to modify the SD card, is a PC, another Netduino or some other kind of network attached device?

I usually use newline 'n' only and not vertical tab 'r' and as far as I remember, it should work.

 

Of course, when all else fails, you can always issue multiple calls to WriteLine().

Please refrain from duplicate posts, you already have this one going:

http://forums.netdui...067/#entry49614

Hi!

 

I suppose that would depend on the rate at which data must be processed and how lengthy transmissions are.

 

I think it's possible to emulate the decoder using a interrupt pin connected to the data out of the RF receiver since interrupts are correctly time stamped and buffered until your app is ready to examine them. However, if the reception is continuous or comes in too lengthy bursts, the interrupt buffer would overflow and cause communication to fail miserably. 

 

It's probably harder to emulate the encoder depending on the modulation and data rate but maybe it could be done using SPI.

 

Could you perhaps refer to a description (or picture) on how data is modulated outside the RF link?

This is not directly related but looks an interesting way to build smartphone apps for interfacing with your Netduino, for example via sockets over a network.

 

As I understand, XMarine would enable you to use Visual Studio for writing apps for both Android and iOS devices.

 

http://xamarin.com/

 

To me it sounds very cool. Has anyone of you guys tried it?

I agree, a Netduino is not a good choise for those things and I don't even think the PI could cope. Any of those would probably be a really big detour from achieving what you want. As all of the bullets are pretty high level stuff, I'd recommend using a PC. Later, you could move to a micro or nano ATX if size is important. Of course, a Netduino could perhaps handle some of it but what's the point if you got s PC on the side. To be quite honest, I think the stuff you've mentioned taken together could take years to develop no matter what hardware you got.

Hi, That "first chance" is from an exception being caught in a try...catch and the second is an exception (the same?) not being caught. I suppose the error is what it is said to be, e.g. an object that unexpectedly has a value of null. Since the code has run for quite a while and the exception looks not to originate from your own code, my guess would be that it is due to unsufficient memory available or something else that builds up over time. I would think its hard for anyone to say much more about it having not seen your code.

Faith, what controller is it, got a datasheet you can point to?

You should use an interlocked exchange to change the value of the _serverReady varible. The Interlocked.Exchange method will replace the value and return the previous value in a single atomic (unbreakable) operation.

int _serverReady = 1;if(Interlocked.Exchange(ref _serverReady, 0) == 1){            // process request       .      .      .      // done, open unlock the door      _serverReady = 1;}else{      // server busy}

As indicated above, I think you have to declare _serverReady as an integer for this work. You might want to use the volatile modifier on your variable declaration but I don't know if that feally matters in .NETMF. There might also be a way to configure the listener as synchronous so that the server will never attempt to process requests in parallel. Could be this feature is not available in .NETMF, I'm not sure.

That makes it compile but it's not quite the same thing, I'd say it translates into this:

if ((this_led_color & mask) != 0)       SDI.Write(true);else       SDI.Write(false);

Checking for equality to mask is no longer the same when mask has other than a single bit set.

Hi! While looking for adjacent info, forum member emg mentioned this module: http://www.ebay.co.u...e-/270762577188 That and other alternatives was discussed in this thread: http://forums.netdui...for-rtc-backup/

Hi all!

 

The MCP41100 is a programmable digital potentiometer with SPI interface and comes in a breadboard friendly DIP-8 package. You can set it to any value between 0 (zero) and 100 kOhm in 256 steps. You can of course tweak it into doing other ranges by wiring in series or in parallel with other (fixed) resistors.

 

For another project, I needed to produce a variable voltage between 0 (zero) and 5V so I made a simple driver class for it and thought I'll share it with you. Attached you'll find the driver and demo app. It is written for my beloved mini but works with others too requiring only small modifications.

 miniMCP41100.zip   4.51KB   7 downloads

 

You simply use it like this:

var r = new MCP41100(new SPI.Configuration(Pins.GPIO_PIN_17, false, 0, 0, false, true, 10000, SPI.SPI_module.SPI1));// set half resistance (i.e. ~50k)r.Level = 128;// set zero resistancer.Level = 0;// ...or you can use the Resistance property to set ~50k like sor.Resistance = 50000;// set ~100kr.Resistance = 100000;

The demo app assumes the IC has been wired as a voltage divider like in this schematic (Pw = output):

When connected that way, you can use it as an a very slow "arbitrary signal generator" meaning you can pretty much draw any signal over time like I did in this post or a nice sine wave if you prefer that:

 

I guess you could think of the MCP41100 like a "poor man's DAC"  but since the mini does not have a DAC, it can actually very useful at times as it's also much simpler than to use PWM.

 

Enjoy!

 

EDIT: Note that the MCP41100 is not suited for signal generation in general.

Yes, I even mentioned PWM in the end of my post. Naturally, the MCP41100 is not at all intended for signal generation (for many reasons), it just seemed like a "fun thing" as opposed to being something of practical value. 

Speaking of PWM, below is a sine(-ish) wave form that I generated using a UART and an LP-filter. It was done using a regular USB to TTL UART converter cable with code running on the PC in an attempt to produce audio - see how beautifully smooth it is?   

 

Also, I was actually thinking of doing PWM audio on an MSP430 but someone beat me to it:

 

Very nice, at first I didn't see the Lumia but only an old 5110 but then I noticed the animated "fly by" arrows and reallzed the 5110 was never capable of that. Thanks for sharing! Edit: also, congrats on becoming a Nokia Developer Champion!

If you mean what the pin is used for, I believe it is for supply voltage 7.5 - 9V.

The shield is probably based around the Maxim MAX7456 chip. Sparkfun has a similar board based on the MAX7456: https://www.sparkfun.com/products/9168 That one has been discussed in these threads: http://forums.netdui...58-netduino-tv/ http://forums.netdui...-video-overlay/ I'm pretty sure I've seen it in other posts as well but I can't find that info right now. Hope that helps a little anyway.

Instead of Unit-testing, you might consider "black boxing" by writing a corresponding set of interfaces through which data Is introduced. You would then make your fake implementations always returning fixed or random dummy data as needed. Later of course, you would make real (non-fake) implementations.

I would have thought it be unusual for an SPI slave device to ignore chip select but could be it only looks at the raising edge of the clock as long as !cs is asserted, i.e. performing and AND operation on the two. Maybe that is defacto standard procedure. The parameters you set do matter and controls how the Netduino SPI module manages the cs and clock pins respectively but I guess you knew that already.