Help with 38khz ir transmitter using multivibrator

[aruna1]

hello
i was wandering can i use simple multivibrator circuit to generate 38khz signal to use as a IR transmitter for TSOP1738?

i did some readings and calculations and came up with resistor and transistor values that gives 38khz.
but since i don't have a oscilloscope i don't know whether this circuit will work.

and most importantly all 38khz IR transmitter circuits on internet are based on NE555.i wonder why nobody use a multivibrator?

calculations are based on article
http://en.wikipedia.org/wiki/Astable_multivibrator"

any help would be great
thanks

here is my circuit diagram
http://img529.imageshack.us/img529/5172/multivibrator38khz.jpg

 

[Pumblechook]

Possibly not too stable.

Crystal oscillator would be best. LC second choice.

Not sure about availablilty of 38 KHz crytals. 32.768 KHz are made by the million. You could use a high freq and divide down and then low pass filter if nec. Would be extremely stable. Multivib will tend to be square wave anyway,

 

[vk6kro]

You could use an discrete component astable multivibrator. However the diagram you give has some problems.

One is that there is no way of adjusting the circuit to get it exactly on frequency. The calculations will get you close, but individual component tolerance errors will possibly cause frequency errors.

The maximum current available is about 3 mA (3v / 1000 ohms) and this probably wouldn't be enough to supply a IR diode. Actually it would be less than this because the diode would need about 2 volts, so maybe only 1 mA would be available. So, an additional amplifier would be needed.

There should be a capacitor across the battery as an aging battery will cause frequency variations. It will anyway.

You need some way of applying data to the output too.

The circuit already uses 10 components and needs more amplification to drive a IR diode. A 555 can do the same thing with 3 components.

 

[Pumblechook]

Multivib will be sensitive to supply volts.

Best stick with the 555.

 

[aruna1]

You could use an discrete component astable multivibrator. However the diagram you give has some problems.

One is that there is no way of adjusting the circuit to get it exactly on frequency. The calculations will get you close, but individual component tolerance errors will possibly cause frequency errors.

The maximum current available is about 3 mA (3v / 1000 ohms) and this probably wouldn't be enough to supply a IR diode. Actually it would be less than this because the diode would need about 2 volts, so maybe only 1 mA would be available. So, an additional amplifier would be needed.

There should be a capacitor across the battery as an aging battery will cause frequency variations. It will anyway.

You need some way of applying data to the output too.

The circuit already uses 10 components and needs more amplification to drive a IR diode. A 555 can do the same thing with 3 components.

Multivib will be sensitive to supply volts.

Best stick with the 555.

my problem s supply voltage.ne555 requres minimum 4.5v.but i need a circuit runs on 3v.
any ideas how to run ne555 with 3v?

 

[Pumblechook]

LMC555 CMOS version . Works as low as 1.5 V I think.

 

[skeptic2]

Try a unijunciton transistor 2N2646. You can't beat the circuit for simplicity and it should work fine on 3 V.

http://baec.tripod.com/DEC90/uni_tran.htm

Try the circuit of Fig. 7 and take the output from the junction of the resistor and capacitor but don't load it down.

 

[Bob S]

In your multivibrator circuit, the circuit is symmetric, and the feedback is ac coupled, so sometimes when you apply voltage, both transistors will saturate and the circuit will not toggle. Also, I don't see the value of the two 1N914's. If you leave them out, you will save two components and have more volts for your transistors. Maybe the 555 route is easier.

 

[aruna1]

ok guys i made it.it can generate 38khz.i tested it with HRM3800 infrared reciver.here is the diagram
http://img33.imageshack.us/img33/1179/multivibrator38khzworki.jpg


but now my problem is its working distance is very very low.only about 2cm

can someone help me to amplify its working distance?

 

[vk6kro]

ok guys i made it.it can generate 38khz.i tested it with HRM3800 infrared reciver.here is the diagram
http://img33.imageshack.us/img33/1179/multivibrator38khzworki.jpg


but now my problem is its working distance is very very low.only about 2cm

can someone help me to amplify its working distance?

How do you know it is on the right frequency? It would take a frequency counter or a well calibrated oscilloscope to get it on frequency.

You can get an idea of the output of the LED by looking at it on the screen of a digital camera or camcorder. You can't tell the frequency like that though.

To amplify the signal, you would put the LED in series with a 100 ohm resistor and a NPN transistor.
Connect the transistor in common emitter mode and drive the base via a 10 K resistor from the collector of Q1.
See this page and look at Fig 1 but add a LED in series with the collector resistor and a 10 K in series with the base.

http://en.wikipedia.org/wiki/Common_emitter_amplifier

 

[aruna1]

How do you know it is on the right frequency? It would take a frequency counter or a well calibrated oscilloscope to get it on frequency.

You can get an idea of the output of the LED by looking at it on the screen of a digital camera or camcorder. You can't tell the frequency like that though.

To amplify the signal, you would put the LED in series with a 100 ohm resistor and a NPN transistor.
Connect the transistor in common emitter mode and drive the base via a 10 K resistor from the collector of Q1.
See this page and look at Fig 1 but add a LED in series with the collector resistor and a 10 K in series with the base.

http://en.wikipedia.org/wiki/Common_emitter_amplifier

thanks i'll check

 

[skeptic2]

There are two potential problems. One is that due to the low supply voltage you may not be getting enough current through the LED. The second is that the circuit may not be operating at or near 38 kHz.

One way to determine if your IR LED is getting enough current is to replace it with a red LED. The red LED has a slightly higher forward voltage drop so if your red LED is bright, the IR LED should be too. If the red LED is bright but you're not getting any range then the problem is probably the frequency.

It is difficult to tune a multvibrator with a pot to find the frequency where your detector is most sensitive. Perhaps it would be better to use a different circuit that is more easily tuned. The unijunction circuit is one such tunable circuit and below is another possible circuit. The circuit below requires a low voltage opamp such as the LMC6036.

http://www.national.com/mpf/LM/LMC6036.html

Since I do not have of spice model of that opamp, I'm not claiming the resistor values are correct and you may have to experiment a little. By replacing R3 with a pot you can tune the frequency until you find the best sensitivity.