I measured the remotes from my Toshiba LCD, Fios box, Denon DVD and amp. All had a modulated pulse width of about 18us. OK, 56K. Not surprised. When I set the output on my Xantech to match, nothing worked at all. I reduced it to 36Khz, and everything works.
Anyone have a clue?

You say you measured the pulse width ... did you measure rising edge to rising edge or just the width of the pulse? What did you use to measure it?

All of the Denon's I've run across are 32Khz carriers.

Jonathan

If you really want to get into IR signal formats, etc you should come over to the forums at http://www.hifi-remote.com/forums where there are plenty of experts hanging out.

And just FYI, even though 36kHz works, I bet 38kHz is a better match.

I will try that link. Thanks Rob.

Yes, rising edge to rising edge. I used a blaster. Put a scope across it and got real close with the remotes. A bit noisy, but enough to use the LED as a photo diode. I just got the "new" scope off e-bay. Tek 2235. Maybe I should put it on my freq standard and be sure it is close.

36K was just by "about the center of where it worked", then measured the result. If I were to stick a 10K pot inline, I could do some testing by reducing the level until it just worked and make it more sensitive to find the center range. Good project while waiting on my digital crossover.

I am waiting on Xantech to explain why covering the back side of the blasters ( blinking kind) made a big difference. I wonder if the red LED also spews out copious amounts of broad band IR and was swamping the detectors.

I am working for now. I would love to pull out the Xantech receiver and use the one built into the Toshiba TV as it is active when the TV is off. I found it sends 36K or so, not what was sent in, so it is a detector and modulator, not just a passthrough. So far, I have not convinced it to give me a nice square wave 50 feet of cable away.

If you measured 36K coming out of what I presume is your TV's IR out, then I don't think it's your scope's timebase being out of calibration. I would more likely suspect that your test jig using the IR blaster as a photodiode was displaying parasitic oscillation when excited by the remotes IR blaster.

When I've viewed any of my 'blinking' no-name IR blasters with a camcorder, they appear to just have one emitter. I suspect that they just use a led with a peak wavelength somewhere near the bottom of the visible spectrum around 700nM. The combination of the lower spectral half-width of the led, and the upper half-width of the peak detection frequency of the receiver's photo diode probably make these 'blinky' emitters work. If you 'covered' your emitter with something like electrical tape, the shiny back of the tape may have actually increased the output to your sensor through reflection. All just speculation on my part.

Jonathan