Describe the "rawdata" format?

[papadeltasierra]

I might try decoding the protocol from a IR controlled fan I have but it's not on your supported list. It sounds like the raw data will tell me much about the signal but is there a descrition of what the raw data contains? Timing for on/off? 1/0 based on a configred timing window?
Apologies if this is in the docs already and I've just missed it.

[NiKiZe]

The basis of IR signals, better covered elsewhere.
LLM summary:

The raw data of an Infrared (IR) signal is the exact, unprocessed timing sequence of the IR light being turned on and off. It is recorded as an array of alternating numerical durations (usually in microseconds) representing bursts of IR light (marks) and the pauses between them (spaces).

There is a modulation frequency involved as well.

[papadeltasierra]

Just so I am clear, I think I'm right in saying that a data "1" is typically a series of IR bursts at the modulating frequency for a period defined by the protocol whereas a data "0" is IR silence. However some protocols can "encode" 1 and 0 so that there are enough IR bursts being sent even if the data is all zeros to ensure a receiver locks on. The raw data is the individual IR bursts without any prior knowledge of the modulation frequency and/or encoding. Is that correct? Thanks.

[NiKiZe]

Rawdata is timing for mark and spaces, additionally you need to know the frequency.

The actual ones and series in a message is encoded in different ways, there is a fewwell used encodings.

[crankyoldgit]

Just so I am clear, I think I'm right in saying that a data "1" is typically a series of IR bursts at the modulating frequency for a period defined by the protocol whereas a data "0" is IR silence. However some protocols can "encode" 1 and 0 so that there are enough IR bursts being sent even if the data is all zeros to ensure a receiver locks on. The raw data is the individual IR bursts without any prior knowledge of the modulation frequency and/or encoding. Is that correct? Thanks.

As @NiKiZe said, rawdata is series of Marks & Spaces pairs (the last Space at the very end is implicit).
A Mark is the length of an IR modulated series of pulses at a given carrier frequency. Eg. A pwm signal at typical 38khz at the given IR wave length.

Spaces are a length of time of complete absence of any signal (zero transmission at all).

Combined, a Mark & a Space often encode a 0 or a 1, or even 2, 3, 4, ... etc (10, 11, 100, ... etc)

Most protocols alter either just the Space or the Mark length to define a logical 0 or a 1. Some alter both to keep the length of time of a bit to be the same. It depends on the protocol.
However, there are exceptions to this where clocking occurs in the encoding of bits such that a bit doesn't neatly line up or appear as a distinct pair of a Mark & Space.

Thus rawdata is the format for just what the signal looks like to the receiver module (except demodulated). It has no concept of bits , thus is always safe to replay a signal, but useless to construct a new (different) message.
See the FAQ/wiki for other descriptions.

So, in short, your assumption is not correct, but kinda close. Hope they helps.

[papadeltasierra]

Thanks, very useful.