Back
[09:54:19] <lambdanaut> Does anyone have any experience with
http://www.sparkfun.com/products/9825 ?
[09:55:03] <lambdanaut> I burnt out my first 4 avr chips with it, not understanding exactly what it means by the board has the "ability to power the target".
[09:55:30] <jacekowski> that's standard term
[09:55:57] <lambdanaut> Does it mean that I don't have to supply any extra 5v power source?
[09:56:37] <theos> heh
[09:56:42] <jacekowski> yes
[09:56:45] <lambdanaut> Oh, derp
[09:56:52] <jacekowski> unless you have a lot of things on board
[09:56:55] <lambdanaut> Well that explains why my chips broke
[09:57:03] <jacekowski> lambdanaut: it shouldn't kill it
[09:57:11] <jacekowski> it could have killed your programmer
[09:57:20] <jacekowski> but it shouldn't kill an avr
[09:57:24] <lambdanaut> uh
[09:57:38] <lambdanaut> well, my programmer still turns on
[09:57:44] <jacekowski> then problem is somewhere else
[09:58:03] <lambdanaut> And can be recognized by my computer, but I can't get it to recognize my AVRs
[09:58:20] <jacekowski> what are you trying to program?
[09:58:33] <lambdanaut> AVR328P
[09:58:47] <lambdanaut> ATmega328P
[09:59:02] <jacekowski> on what sort board?
[09:59:09] <lambdanaut> Just stuck in a breadboard
[09:59:18] <jacekowski> how are you connecting it?
[09:59:40] <lambdanaut> Wiring it directly from the 10 pin programmer's connector to the AVR
[09:59:55] <lambdanaut> And originally I was giving it an extra 5v from a wall wart
[10:00:01] <jacekowski> and all pins are connected correctly?
[10:00:05] <lambdanaut> yes!
[10:01:02] <jacekowski> hmmmmmmm
[10:01:15] <jacekowski> have you ever seen it work?
[10:01:20] <lambdanaut> yes, once
[10:01:25] <lambdanaut> and then, it stopped v_v
[10:01:35] <lambdanaut> I got avrdude to recognize my avr chip once
[10:01:45] <jacekowski> two things
[10:01:47] <jacekowski> fuses
[10:01:58] <jacekowski> breadboard capacitances/connections
[10:02:13] <lambdanaut> I've got two caps
[10:02:23] <lambdanaut> well, I did,
[10:02:27] <jacekowski> i'm talking about breadboard itself
[10:02:31] <lambdanaut> right
[10:02:31] <jacekowski> it has shitloads of capacitancies
[10:02:38] <lambdanaut> oooo
[10:02:48] <lambdanaut> I didn't know that
[10:02:57] <lambdanaut> is it because the underlying copper is so close?
[10:03:02] <jacekowski> yes
[10:03:05] <lambdanaut> woah.
[10:04:32] <karlp> and because of all those leaves of metal that let you push insert things in
[10:05:27] <lambdanaut> well, my Programmer itself is still recognized by my PC. Does that pretty much rule out it being blown?
[10:43:51] <abcminiuser> Heyo
[10:43:57] <abcminiuser> specing, no
[10:44:05] <abcminiuser> IIRC we don't make them any more
[10:45:51] <OndraSter> abcminiuser, that is regarding those FPGA+AVR cores?
[10:45:58] <abcminiuser> Ja
[10:46:00] <OndraSter> ah
[10:46:02] <OndraSter> too bad
[10:46:06] <OndraSter> I liked the idea as well :D
[15:47:38] <OndraSter> IT IS DEADDER THAN DEAD here
[15:47:40] <OndraSter> ..
[15:48:13] * Tom_itx opens up all the zoo cage doors
[15:48:22] <OndraSter> mmm tigers
[15:48:24] <OndraSter> cute cubs
[15:59:39] <Corwin> mmm tigers? you ate them ?
[16:43:52] <Kre10s> How does that edge detector on a Tn pin work? I am looking at the 16-bit Timer/Counter Block Diagram of an atmega644 and it shows tn->edge detect->clktn...
[16:45:00] <Kre10s> If i want to run the counter at a higher frequency than the mcu, i would have to use an asynchronous timer right?
[16:54:34] <jacekowski> Kre10s: you shouldn't do that
[16:55:03] <jacekowski> Kre10s: i mean, you shouldn't exceed maximum frequency
[16:55:48] <jacekowski> Kre10s: details are in the manual
[16:55:51] <jacekowski> how fast you can run it
[16:58:37] <Kre10s> the chip? i can run at 20MHz
[17:00:09] <OndraSter> the maximum speed is in the datasheet
[17:00:20] <OndraSter> you'd need external counter to pre-scale it :P
[17:02:38] <jacekowski> 15.9 in datasheet
[17:03:11] <jacekowski> The CPU main clock frequency must be more than four times the Oscillator frequency.
[17:03:15] <jacekowski> saying that
[17:03:27] <jacekowski> i've done it with clock running at the same frequency and it did work
[17:59:32] <Kre10s> what about asynchronous mode?
[18:38:59] <Casper> Kre10s: afaik, you can't go faster than F_CPU/2
[18:39:16] <Casper> but I might be wrong, check the datasheet
[18:39:29] <OndraSter> with async I suppose you, in theory, could
[18:40:43] <Kre10s> I've got an app note doing frequency measurment with a pic up to 50MHz... I'm trying the same with an avr, but my signal generator only does 10MHz max
[18:41:20] <OndraSter> Kre10s, underclock the MCU :)
[18:42:47] <Kre10s> eh? I dont get it/
[18:42:57] <Casper> divide your frequency by 5
[18:43:05] <Casper> and at 10MHz your code will show 50
[18:43:23] <Casper> but that won'T test the circuit
[18:43:24] <OndraSter> Kre10s, if you underclock the MCU to let's say 1MHz, you can check how much faster can the input clock me
[18:43:26] <OndraSter> be
[18:43:58] <Kre10s> yea. but if in asynchronous mode the couter counts irrespectively to the CLK_IO, then that won't tell me anything.
[18:44:11] <OndraSter> no?
[18:44:12] <OndraSter> it will
[18:44:39] <OndraSter> how does differ 10MHz signal + 2.5MHz device from 50MHz signal and 20MHz device?
[18:44:46] <OndraSter> it still goes 4 times faster
[18:44:49] <OndraSter> err, fix the math*
[18:44:55] <OndraSter> but you got the point I hope
[18:45:11] * GuShH removes the booze from OndraSter's desk
[18:45:15] <Kre10s> yea I understand what you mean.
[18:45:18] <OndraSter> okay
[18:45:29] <OndraSter> GuShH, :P
[18:45:37] * GuShH idly drinks it
[18:45:59] <drgreenthumb> hey pass that over. didn't your mother teach you to share?!
[18:46:13] <GuShH> Okay, you can have the empty bottle
[18:46:25] <GuShH> Recycle it.
[18:47:14] * drgreenthumb buys Night Train with the recycling money
[18:47:28] <Kre10s> Theres got to be an electrical limit though... not like "i can't count it because my code is slow", but "I can't count it because my IOs don't respond fast enough."
[18:47:44] <GuShH> what are you counting?
[18:47:50] <OndraSter> his fingers
[18:47:56] <OndraSter> (why did I say that loud?)
[18:47:59] <GuShH> Fine.
[18:48:04] <Kre10s> I have code that is working for 1 5 8 and 10Mhz. and should mathematically work for 50...
[18:48:30] <OndraSter> if you mean "bandwidth" of the IO - you will have to try
[18:48:48] <OndraSter> some megas (tinys?) could grab input clock of upto 60MHz or so and then divide it before feeding it to core I think
[18:48:59] <OndraSter> so I suppose it should work just fine
[18:49:06] <OndraSter> if you lay out the board properly
[18:49:14] <OndraSter> although 50MHz is still nothing
[18:50:37] <Kre10s> GuShH, i am measuring frequency... so counting the edges of a wave.
[18:51:22] <GuShH> you need -several- samples for a reliable count.
[18:53:53] <Kre10s> GuShH, yes. so i increase the 8bit counter resolution in software.