Originally Posted by jumpjack

Well, isn't this exactly what I need?!? The same signal, but with opposite phase!
One LED plays original signal, the other LED plays the inverted one, and together they "create" a 38000 Hz signal...?

Or maybe I didn't understand anything of your project?

I gave you the answer (TLC272) and told you at the end "There is no way to produce a "real" stereo signal from a mono one."

This means you can work it out, but only if both channels are identical.

Originally Posted by XianLi

Do you know the principle of an LED ?

Anode --|>|-- Cathode

Anode = +
Cathode = -

If you invert the polarity of your signal,[...]

Yee, I know this.
What I don't understand is why we are putting each LED among LEFT and right rather than among LEFT and GND, and among RIGHT and GND!

I think I'm missing something about the audio file: is one channel 180° of the other or not? Maybe it's not 180?? 90°? 23.75°?

Originally Posted by XianLi

I gave you the answer (TLC272) and told you at the end "There is no way to produce a "real" stereo signal from a mono one."

This means you can work it out, but only if both channels are identical.

Yes, I can emit same signal from two channels; but I can also invert one channel using a transistor.



But if it's not a 180°-phased signal I need, then t would be quite more complex...

I did a long study trying to explain the principle which the IR-to-sound converter is based on.
These are my results:

Excel spreadsheet
Humongous image!

If they are correct, it should mean that connecting two IR leds in oppiste way to a single MONO audio signal should result in a double-frequency IR signal.

Please let me know if instead you find some "flaws" in my work.

I supposed that any light signal can only be positive, so a negative signal passing thrugh a LED is turned into positive light signal; so a sinusoidal signal should become a pulsed signal at double frequency... or not?

ANyway I still do not understand why symmetrical ouput results in double-frequency output with one IR LED!

It's confirmed: a single channel is enough to produce an IR signal which controls a device. At least, it works on my PC. For unknonwn reasons it does not on my mono phone, as only one of the two leds lights up. (???)

I created a project in Google Code:
http://code.google.com/p/ledrem/source/browse

Here you can find BASIC sources of programs which creates WAV files starting from sampled remote controls: feel free to port them to any language/platform you like!

BASIC programs are written in RapidQ basic, which can produce executables for both Windows and Linux. Does it mean they'll run on MacOSX too? No idea, sorry.
Here you can download RapidQ, if you want ti verify MacOSX compatibility:
http://g.yi.org/?f=2372


http://g.yi.org/?f=2372

any news?
any porting of LedRem to MacOSX?
Any new HW released?

Hi
I'm planning to build this but I'm having trouble finding the IR-sensor named QSE156
I will buy from elfa here in sweden (in English https://www1.elfa.se/elfa~eu_en/b2b/...browserminor=5)

The only IR-sensor I could find on elfa are these:

Code:

Quantity	Stock no. 	Name 	Currentconsumption 	Supply      voltage 	Price/each     1— 	25— 	250— 	
	
	Buy
	75-205-96 	IRM-3638N3 IR re 3,3V 	≤2 mA 	2,7–5,5 V 	1,46 	1,25 	0,63 	
	
	Buy
	75-202-73 	IRM8602S IR receiver 38kHz 	≤3 mA 	4,5–5,5 V 	1,75 	1,18 	0,60 	
	
	Buy
	75-202-81 	IRM8608S IR-receiver 38kHz 	≤3 mA 	4,5–5,5 V 	1,77 	1,21 	0,61 	
	
	Buy
	75-202-99 	IRM8752 IR receiver 	≤3 mA 	4,5–5,5 V 	1,76 	1,22 	0,61 	
	
	Buy
	75-303-89 	IRM8601S IR-receive 38kHz 	≤3 mA 	4,5–5,5 V 	2,00 	1,36 	0,69

Should anyone of those be suitable for the circuit instead of the QSE156?

When buying a IR-LED, what specs should I look for?

/Niclas

Ok, I just got my first cell phone this year, an iphone, so take this with a grain of salt... believe it or not, I thought of this idea a couple weeks ago (little late), then found this thread (of course it was too simple to be novel). But I still imagine there must be a few thousand people who would pay $20 to control any TV/STB - say if you were in a bar and wanted to put on espn - anyway, just read threw this thread.

apologees for replying without reading attached speadsheat and other thing:

If they are correct, it should mean that connecting two IR leds in oppiste way to a single MONO audio signal should result in a double-frequency IR signal.

This sounds right to me. As long as the mono signal is BIPOLAR, meaning it swings positive and negative, so that BOTH diodes will see the necessary forware voltage drop. If it only goes up or down, then only ONE of the diodes will light up.

Yee, I know this.
What I don't understand is why we are putting each LED among LEFT and right rather than among LEFT and GND, and among RIGHT and GND!

One reason I can think of off hand is to get double the available voltage drop across the diode; basicly more UMPHH. Say that with one driver (channel) you can get +- 1V (I have no idea what it really is, although I would like to know), then by having the second one out of phase (as you hypothesized) you could get +- 2V across the diode. Depending on the diode (its forward voltage and size, which is the current requirement traded off for optical power) and/or limiting resistor, this could mean the difference between light or no light. --- Note if you think about it, this means that current will be passed internally between the driver grounds, which would not be in the ideal design spec of any driver, as I imagine it. This is the only (legitimate) reason I can think of why Apple asked for the Griffin Total Remote to be pulled from the shelves.

What I actually don't understand is which is the final waveform produced by your/Griffin's device:

Well, isn't this exactly what I need?!? The same signal, but with opposite phase!
One LED plays original signal, the other LED plays the inverted one, and together they "create" a 38000 Hz signal...?
Or maybe I didn't understand anything of your project?

I think that one fine point that may not have been specifically mentioned yet here, is that the dioded is OFF for voltages below its drop. So lets say I can only UPDATE a Voltage every MILLISECOND (i.e. 1KHz) , with just ONE diode, even with bipolar output, I can make it +1, then -1, then +1 (assume diode turns on at .9V),,, etc. every ms. But the diode only turns on at +.9V, so I will get ONE pulse of light every TWO milliseconds, because it has to go ON and OFF - this is modulation at 1/2Khz.

If INSTEAD I have TWO diodes, tied in REVERSE, Then as I go from +1 to -1, there is a time when both diodes are OFF, because the voltage is below the threshold, and then when the voltage gets to -Vthreshold, the other diode will be forward biased; SO you get an ON - OFF - ON, for the price of an ON - OFF - this is modulation at 1KHz - twice the frequency as with ONE diode. Alternatively, you can also see it as getting a pulse for every rising or falling edge, instead of just for the rising edge. Like DDR.

Furthermore, from this logic it would seam the carrier emitted this way would not be a 50 % duty cycle.... sinces it depends on the slew rate of the driver and actual diode drop... but then again it shouldnt matter since the largest component by far would be that at the carrier frequency, since the pattern is strongly periodic at that freq, so it is not crazy to imagine it still being picked up by the reciever using a diode with a typical response time ( ~ 1us ) and typical slew rate (~ 1 us).

whats cool to think about is that the power in a remote signal is roughly the same (within an order of magnitudE) as that coming out of your headphone speakers, albeit for a much shorter time. I mean, headphones can be loud...

DUDE!!! what happened to you? are you out of the iphone scene? whats up with your rep? u were one of the most important mods in here, what happened???? wheres all your rep??????????????

Originally Posted by seseberg

DUDE!!! what happened to you? are you out of the iphone scene? whats up with your rep? u were one of the most important mods in here, what happened???? wheres all your rep??????????????

nice question...

About your previous post... I didn't understand most of it , but I'll tell you a thing which will "mess up" all your thinking: connecting ONE single led to a nokia 6680 with this pinout, it WORKS! I can control my TV, using a 19000 Hz tone!!
But it does not work to connect TWO leds to THIS port:
http://pinouts.ru/CellularPhones-Nok...t_pinout.shtml