Show Posts
Pages: 1 ... 357 358 [359] 360 361 ... 462
5371  Forum 2005-2010 (read only) / Bugs & Suggestions / Re: USB connector wired backward in online CAD fil on: November 05, 2007, 06:46:50 pm
Just as an update, this was wrong, but it's been fixed in the USB V2.1 cad files
5372  Forum 2005-2010 (read only) / Bugs & Suggestions / USB connector wired backward in online CAD files? on: May 12, 2007, 01:49:37 am
I was laying out a serial arduino that would use a USB connector for power, and the
layout I came up with from "first principles" wasn't matching the layout of the USB
connector in the downloadable CAD files of the USB arduino  (arduino_usb_8.)
Further looking around shows that the online schematic for NG has the pins wired
exactly backwards from usb_8 (the ng schematics match my pinout understanding.)
Did you wind up with some odd USB connectors or what?
5373  Forum 2005-2010 (read only) / Bugs & Suggestions / Re: I like adruno but in needs 2 things to sell it on: May 04, 2007, 06:16:25 pm
I wonder if you could implement a useful debugger by having the arduino "pre-processor" insert (explicitly in the C code) a call to a debug function after each user source-code statement.  Yeah, the code for a "debug mode" program would be much larger than a non-debug version, but how big IS the average arduino program?  Is the typical basic stamp program less than a a couple hundred statements? (BS2 only has room for about 500 statements.)  Or how about an arudio simulator?  (although gdb is not very beginner-friendly, it does have a standardized remote communications protocol.)
5374  Forum 2005-2010 (read only) / Interfacing / Re: Using the arduino for an 8X8X8 LED CUBE on: January 06, 2009, 11:40:45 am
You might want to take a look at this thread:
which has some discussion of a 384-led display using the Holtek HT1632 driver chip.  In particular, there's a demonstration program that updates all 384 LEDs (serially, one at a time) and is still "fast enough for animation."
A couple of these chips would be a good start on an 8x8x8 cube, I think; they do all the period refresh, so your arduino would actually have time to do other things...  (However, I'm not sure where you'd buy them...)

(an 8x8x8 cube would be quite a substantial MECHANICAL undertaking as well...)
5375  Forum 2005-2010 (read only) / Interfacing / Re: Sure Electronics 6432 5mm board on: January 07, 2009, 11:25:42 am
I wish Sure had used the ht1632 drivers in their larger displays as well at the 24x16 version; it's a pretty nice chip that does the multiplexing for you, and there are a couple of ways that it could have been used in larger configurations...

I estimated that the whole "shiftout" sort of thing would get close to 10x faster using direct IO instead of DigitalWrite(); you can try that if you get desperate, I guess.

A quick speedup is to make most of your "int" variables be "byte" instead.

I didn't see anything massively wrong (speed wise) with your code.  It does seem like the arduino OUGHT to be fast enough to do a less flickery display...

5376  Forum 2005-2010 (read only) / Interfacing / Re: SureElectronics 3216 Led Matrix on: December 17, 2010, 06:07:56 pm
these use the new HT1632C not the older HT1632 Led Driver chips. The RC Master command ($18) is used differently now
How so?  The wording in the datasheet has been changed slightly, but it looks like the actual functionality is identical...
5377  Forum 2005-2010 (read only) / Interfacing / Re: Long-term reliability of Wiznet-based Ethernet? on: December 14, 2009, 06:03:27 pm
On boxes with similar reliability requirements, we would make "educated guesses" as to whether network interfaces might need a reset.   Issued a "send packet" command and not seen a "transmission complete" back within 5 seconds ?  Reset the interface!  Not seen ANY received packets in the last 10 minutes, when you expect to see one every 10 seconds?  Reset the interface!  Got some weird error status?  Reset the interface!

You can consider doing this for your entire project, of course.  The "penalty" for resetting an embedded cpu like an arduino is a lot smaller than for a mainframe or desktop system, and rebooting every 24 hours might hide a lot of weaknesses in your software that might show up in 25 days when millis() turns negative (for example.)
5378  Forum 2005-2010 (read only) / Interfacing / Re: Electret Microphone with Arduino on: January 08, 2011, 08:17:19 pm
Exactly how do you have it connected, and exactly what code are you using to look at it?  The analogRead() function only runs at about 10000 samples/second, which makes it "just barely" able to process audio (somewhat worse than a telephone), and it would be easy to have code that did enough "not quite right" to get essentially random results when trying to sample audio...
5379  Forum 2005-2010 (read only) / Interfacing / Re: Non-linear resistance in a digital potentiometer on: January 09, 2011, 11:03:16 pm
I need to source at least 50 mA, which is more than the Arduino PWM pins put out.
It's also more than the digital POT is supposed to be able to source...
5380  Forum 2005-2010 (read only) / Interfacing / Re: Reading 800 times a pin on: December 27, 2010, 12:03:42 am
Are we still talking about looping 800 times, or have we devolved into talking about how to get these particular radio transceivers working?

In the latter case, I see the following issues from looking at your photographs and the data sheet link you provided (and in particular, looking at the photo of the board in the data sheet.)

1) You seem to have only connected the transmitter GND power pin to GND on both transceivers.  In the data sheet, it does not appear that the transmitter and receiver grounds are connected on the transceiver itself.

2) Since there IS some sort of connection between the receiver and transmitter, I think that powering only one of them is a bad idea.  Power both and disable the transmitter of the receiver by connecting the TX input to GND.

3) Likewise, there are multiple GND pins for the receiver that do no appear to be connected together.  If there are multiple GND pins on a chip or module, it is generally a good idea to actually connect ALL of them to ground.

4) No Antennas at all?

You initial program with its 800 analogRead() calls within an otherwise empty loop would have completed in about 1/10th of a second.  If you were planning on using it as a sort of real-time monitor, it's doubtful that you would have captured the right instant in time.  Try something like:
int lastval = -1;
void loop()
  int curval;
  curval = analogRead(pin);
  if (curval != lastval) {
     lastval = curval;

5381  Forum 2005-2010 (read only) / Interfacing / Re: Reading 800 times a pin on: December 22, 2010, 08:32:37 pm
with all the code in setup()
it is only executed once
so there is no chance to read different values!!!!!!
You can still put a loop in setup() that does as much as you want.  In fact, your code never needs to return from setup() at all, if you put everything there (and it makes it look more like a traditional C program with setup() == main())

The code in the first post looks fine to me.

Except that (as Groove mentioned already), you're probably using more RAM memory than the chip actually has.  Does it work if you only do 500 samples?
5382  Forum 2005-2010 (read only) / Interfacing / Re: Power over Ethernet? on: May 27, 2009, 01:06:57 am
I found these that are not too expensive:
No personal experience...
5383  Forum 2005-2010 (read only) / Interfacing / Re: On Bug Zappers and the like. on: December 28, 2010, 02:24:26 am
2300V, eh?  I always figured these things used surplus/excess/recycled guts from disposable camera, which would put them at 300-700V.  I suppose there are assorted tricks to get higher than that, but it hardly seems worthwhile for bugs...
(note that most "wire" is rated for up to 600V.)
5384  Forum 2005-2010 (read only) / Interfacing / Re: How reliable is the USB serial port connection on: December 28, 2010, 02:21:55 am
If USB transparently handles error detection and retry, there should be no need for me to implement this myself in arduino software right?
The connection between the PC and the USB chip should be error-corrected.  The connection between the USB chip and the Arduino is NOT, and is only "barely" capable of speeds faster than 115200.  (Try v0022; the serial code has been optimized a bit...)  (and the closest actual bitrate supported by a 16MHz AVR is 250kbps; 2.3% away...)
5385  Forum 2005-2010 (read only) / Interfacing / Re: millis (crystal) accuracy & thoughts on: December 20, 2010, 03:22:08 pm
The secret is not to count millis between it doing something but to work just with the big number since it was started.
Yes, but he's been doing that all along.  From the first message:
void loop(void) {
  //main loop
  while(1) {
    //update display
    current_time = (millis() - millis_offset)/100;

Note that the divide by 100 is throwing away relatively important precision, and probably explains some of the non-linearity in the times reported in your first post.
Pages: 1 ... 357 358 [359] 360 361 ... 462