How to identify IR signal?

[kookie22]

Hi,
I need to measure the frequency of infrared pulsing from an IR transmitter. A quick search tells me that i need a spectrometer to do this? they look mighty expensive to buy, is there any other way to do this?
The reason i need to measure this frequency, is so that i can make a receiver compatible with this transmitter, which would trigger it. I cannot make my own tx, i have to use the existing one. Thanks!

 

[Papabravo]

If you build an IR receiver and measure the output with a scope or a frequency counter you will have at least a datapoint. It goes without saying that your measurement will be subject to uncertainty and error.

 

[Nigel Goodwin]

Except don't 'build' an IR receiver, use one of the many three pin ones available - they just require 5V, Gnd, and output the data train from the third pin. Easy to use, and you can scope the ouput to see what it does, it's usually fairly simple to reverse engineer the signal - note that there are a great many different types in use, with the Philips RC5 being the most common.

 

[philba]

I assume you mean modulation frequency.

Who makes the IR transmitter? Most manufacturers pick a single frequency and stay with that. Could be a simple matter of a little googling.

If not, go the OScope route. clip onto the LED's output and watch the signal. a little tricky with an analog scope but there really are only a few possible frequencies.

 

[Nigel Goodwin]

I assume you mean modulation frequency.

It doesn't really matter very much, they are all pretty close, and a standard IR receiver will pick up all of them - it's the data that's important, and not the carrier.

 

[philba]

the prebuilt modules filter on the carrier. if you are only worrying about 38 and 40 khz then it will probably be ok. For example, the tsop11xx series of receivers shows some sensitivity at 10% on either side of it's center frequency. however there are other standard frequencies: 56K isn't uncommon and 30K, 33K, 36K, 36.7K are not unheard of either. I would guess that a 38K receiver could handle all but the 30K and 56K frequencies. Maybe 33K would be dicey. I doubt that 56 Khz could be received by a 38K receiver.

 

[Nigel Goodwin]

I doubt that 56 Khz could be received by a 38K receiver.

Possibly not, but those are extremely rare cases - generally all those you come across are around 38KHz, and work interchangeably quite happily.

 

[philba]

I dunno, maybe not that rare. Denon uses 32K and scientific atlanta uses 56K. It runs in my mind there are others that fall outside the 36-38-40 cluster. I think it's better to know than just take a stab in the dark.

 

[Nigel Goodwin]

I dunno, maybe not that rare. Denon uses 32K and scientific atlanta uses 56K. It runs in my mind there are others that fall outside the 36-38-40 cluster. I think it's better to know than just take a stab in the dark.

Well Scientific Atlanta are presumably USA specific?, I think you'll find Denon remotes work fine with normal 38KHz receivers?. But you're only talking a VERY small percentage (probably well under 1%?) that fall outside the normal range.

Incidently, years back B&O used to use 100KHz modulation, which is why 'One For All' didn't support B&O - this was pre-IR receiver IC's though!.

 

[philba]

SA made analog cable boxes for the US market - plenty left in use.

I think you'll find Denon remotes work fine with normal 38KHz receivers?

do you know denon works for sure or are you guessing?

My point isn't to quibble over what is the majority is but rather that knowing your target is better than not. I don't see how you can disagree with that. And like I said, all he has to do is google about for the brand he has and he will know.

 

[Nigel Goodwin]

SA made analog cable boxes for the US market - plenty left in use. do you know denon works for sure or are you guessing?

So SA are pretty obscure then?, a niche market just in the USA.

Partially 'guessing', because I can't remember definitely testing a Denon remote - I've built a number of IR remote testers (using 38KHz receivers), which are in daily use for testing remotes - and NEVER found one yet that they don't work with.

My point isn't to quibble over what is the majority is but rather that knowing your target is better than not. I don't see how you can disagree with that. And like I said, all he has to do is google about for the brand he has and he will know.

It's probably just as easy to connect up an IR receiver IC and check the output from it - google may provide useful information on a particular remote, and it may not.

 

[philba]

I dunno, google is faster than any other option.

I like how you define several, if not 10s, of millions of units as a niche.

 

[Nigel Goodwin]

I like how you define several, if not 10s, of millions of units as a niche.

It's a very specific product in a single specific area, that seems like a 'niche market' to me?.

 

[audioguru]

My Canadian cable TV company has Scientific Atlanta boxes made and packaged with their name on them.
Stores that sell home entertainment electronic goods also sell the re-named boxes.
I wouldn't call most of North America "a niche market".

 

[hjames]

I'm wondering if the poster was asking about the actual wavelength of the LED?
The usual standard wavelengths are 850nm, 880nm, and 950nm. There are also a couple companies which make other wavelengths - 780nm, for example are good for CCD illumination.

In order to tell the difference between frequencies, you might be able to do some comparisons using some IR filter material (look for differences in response of different filter materials). A long shot would be to jerryrig a spectrometer using a CCD camera and a diffraction grating. In all these cases you'll need to buy some LEDs to use as references, and I have no idea how reasonable/possible these guess are...

 

[Paul Obrien]

if you have a Pocket PC google for VitoOscilloscope_PPC as it will show IR pulses and decode most, you could also try a IR dongle for a PC and look for similar software.