## Re: IR Transmitter circuit.

 Re: IR Transmitter circuit. From: Jim Paris - 2001-08-18 01:34:29 ```> Well, the serial port will put out 11V, so as long as it doesn't drop below > 7V I should be fine. How long of a cable can I expect to have before I lose > 4V? Solid 22 gauge wire has a resistance of about 0.016 ohms per foot. Assuming (conservatively) that your transmitting LED takes an average of 50 mA to drive and you leak another 50 mA through the regulator and transistor, your total voltage drop along the wire is V=IR, or 4V = 0.1 A * 0.016 ohms / foot, so you could go 2500 feet. Half a mile. By that point, the impedance of the wire and any noise would probably start to become more of a problem. Use stranded 20-or-larger gauge wire if you want to be really safe, but you really don't have to worry about much of a voltage drop across the wire (assuming I didn't do something stupid in my calculations). I'd be more concerned that the serial port voltage is unregulated and as soon as you start to draw your relatively high (compared to RS232) current from it the voltage would drop below 7V. -jim ```

 Re: IR Transmitter circuit. From: Sam Varshavchik - 2001-08-18 01:02:35 ```Jerome Kaidor writes: > Hi Sam, > > As somebody already pointed out, your main problem will probably be > head room for the regulator. If that does turn out inadequate, you > might try a zener diode instead. These don't require any head room at > all, although you still need a bit to deal with varying current demand > from the load. ( And if the current demand from the load did not change > at all, you could get by with no VR at all :)). Well, the serial port will put out 11V, so as long as it doesn't drop below 7V I should be fine. How long of a cable can I expect to have before I lose 4V? > I had a similar situation - I wanted LIRC to work through a fifty-foot > serial cable. I punted and ran the transmitter off a wall wart. I considered using a battery. In that case, however, I would want a low battery indicator LED. I'm trying to find a schematic for a low voltage indicator that I can actually understand. I am not having much luck, and I don't want to wire stuff together until I can understand how it works. radioshack.com is really beginning to suck, lately. Most of the stuff that they serve up is mail order only, and the rest of the stuff rarely has listed specifications. I guess I'll have to try my luck and hope that they print the specs on the packaging. -- Sam ```
 Re: IR Transmitter circuit. From: Sam Varshavchik - 2001-08-19 00:29:26 ```Enrique Vidal writes: > You may want to try the Transmitter circuit I posted some time > ago to this list. You can find it linked at: > > http://www.lirc.org/transmitters.html > > (in fact, here: http://www.geocrawler.com/archives/3/3680/2001/1/0/5037984) I've seen it already. You're using DTR to both charge and signal, because apparently that's the only pin lirc pokes. For now, I do not need to use lirc for my application, so I think I can simplify the circuit quite a bit by charging from a separate pin, and using custom software. If necessary, I'll tweak lirc later to use my transmitters. It shouldn't bee too difficult. pull up dtr, wait a few seconds to charge the capacitors, then use rts to signal. If necessary, I'll just swap the pins, and hold rts up for a few seconds, then signal via dtr normally. lirc may already be holding rts high by default, when opening the port. But it seems more natural to use rts to signal. But anyway, the reason for the voltage regulator and the capacitors is so that I can have a known voltage even if the voltage on the pc serial port goes down because of the drain. This should allow me to safely push the maximum current through the led, even if the port voltage dips. Perhaps its overkill, but that's ok. It looks like radioshack is now stocking some very high output infrared leds. Rated at 100mA, with a maximum of 1.2amps! So, with a 10R and 5V I'll have a steady 500mA going through the LED, even if the port voltage drops down to 7V. -- Sam ```
 Re: IR Transmitter circuit. From: Sam Varshavchik - 2001-08-19 01:59:29 ```I wrote: > for a few seconds, then signal via dtr normally. lirc may already be > holding rts high by default, when opening the port. But it seems more > natural to use rts to signal. Hmmm... It looks like lirc does pull RTS up high: #define LIRC_OFF (UART_MCR_RTS|UART_MCR_OUT2) #define LIRC_ON (LIRC_OFF|UART_MCR_DTR) ... and uses DTR to signal. Is there any specific reason to keep RTS high and signal via DTR, instead of the other way around? -- Sam ```
 Re: IR Transmitter circuit. From: Enrique Vidal - 2001-09-12 08:35:27 ```On Tue, 11 Sep 2001, Scott Baily wrote: > At 01:45 AM 8/19/01 +0200, you wrote: > > You may want to try the Transmitter circuit I posted > > some time ago to this list. You can find it linked at: > > http://www.lirc.org/transmitters.html > > (in fact, here: > > whttp://www.geocrawler.com/archives/3/3680/2001/1/0/5037984) > > Can you please list what NPN and PNP transistors, and also > what diodes you're using. Also are the capacitance and > resistance values critical? Are you tuning the circuit to > some frequency or should any capacitors in a simliar range > work. Any type of low/medium-power diodes and transistors are suitable. The only critical componnets (apart from the IR diodes themselves) are d1-d2. These diodes are marked as "silicon", meening that they must be "normal" modern (hence silicon) diodes, which have a significant offset voltage (+0.7 volt, approx.). Other kinds of diodes (e.g., germanium diodes) may have much lower offset and are inadequate for the intended purpose. Tolerances are not critical for any component of this circuit (+-10% or even +-20% should be O.K). > I tried building your circuit with some parts i had lying > around, but it didn't improve my range (~2 meters). I'm > using a 2N 3904 transistor an some resistors with 2 IR LEDs. OF course, the range depends on how much power is used to drive the IR LEDS and of the LEDS themselves. I have several kinds of LEDS (intended for IR remotes) with no much difference. As to the power, it depends on a) The value of the (22 Ohm) resistor from GND to the emmiter of the PNP transistor and b) on the large (4000 microF) capacitor. The range can always be increased by lowering the value of the resistor, but than the value of capacitor needs to be incresed correspondingly. Hope it can help. Enrique. ```
 Re: IR Transmitter circuit. From: Jim Paris - 2001-08-18 01:34:29 ```> Well, the serial port will put out 11V, so as long as it doesn't drop below > 7V I should be fine. How long of a cable can I expect to have before I lose > 4V? Solid 22 gauge wire has a resistance of about 0.016 ohms per foot. Assuming (conservatively) that your transmitting LED takes an average of 50 mA to drive and you leak another 50 mA through the regulator and transistor, your total voltage drop along the wire is V=IR, or 4V = 0.1 A * 0.016 ohms / foot, so you could go 2500 feet. Half a mile. By that point, the impedance of the wire and any noise would probably start to become more of a problem. Use stranded 20-or-larger gauge wire if you want to be really safe, but you really don't have to worry about much of a voltage drop across the wire (assuming I didn't do something stupid in my calculations). I'd be more concerned that the serial port voltage is unregulated and as soon as you start to draw your relatively high (compared to RS232) current from it the voltage would drop below 7V. -jim ```
 Re: IR Transmitter circuit. From: Enrique Vidal - 2001-08-18 23:46:07 ```You may want to try the Transmitter circuit I posted some time ago to this list. You can find it linked at: http://www.lirc.org/transmitters.html (in fact, here: whttp://www.geocrawler.com/archives/3/3680/2001/1/0/5037984) It does have a considerable range and does not suffer from the voltage drops you are discussing about. It can also be attached using rather long wiring (four meters in my case, without troubles). Very long wiring is not possible because of signal degradations (nothing related with voltage loose!). Best, Enrique. ```