Motorola home control

Thanks, Dave. I've interfaced AVRs to LINX serial radio modules, but never ASK/OOK. If there is constant radio noise, and therefore constant data coming in, how do you pick out the real data in the stream? jj, cinci

I have no idea what protocol they use nor whether they use FSK or ASK but most similar devices use ASK/OOK because it's simplest and lowest cost.

You can buy a 418MHz ASK receiver from Mouser - PN 509-RCR-418-RP FOR $4.26.

The data output will be in the audio range so you can record it as a .WAV file using a soundcard line-in (It's IMPORTANT that you use line-in.) and view it in almost any Wave editor.

I have illustrated this at...

The RF section is about half-way down the page.

For any f>Motorola is expanding there home automation products:

Ok, that looks doable without too much trouble. I'm using BASCOM...sooo easy

The only reason I looked at Motorola is that I really like the design of their wireless motion detector. What's the most reliable and powerful (hopefully they're not mutually exclusive) wireless X10 motion detector these days. I have a DM10A but what a huge beast it is. I only want to use them for "I'M HOME" determination.

jj, Do we have some nice weather this week or what!

You have to look for the first non-noise pulse. Some protocols are easier than others, using a lead-in pulse that also functions to set the AGC in the receiver. The NEC protocol used by X-10 is one of the easiest as it has a long lead-in but lots of others also have easily distinguishable lead-ins.

Noise pulses tend to be shorter than most data pulses so even protocols without a lead-in are fairly easy to handle. And, as you can see from my screenshots, the AGC quickly separates the wheat from the chaff, giving you a fairly clear signal.

Also, the specific RF receiver I mentioned has a linear output which can be used with an ADC input as an RSSI indicator. The RSSI of random noise will be low in comparison to a signal (as long as the signal is stronger than the random noise. You can use it as a rough Carrier Detect indicator.

The screenshots are a bit misleading in that the RF receiver has automatic threshold control which tries to make mountains out of molehills. See...

It's actually pretty easy to separate signal and noise as long as the signal is stronger than the noise.

But, how easy also depends on the specific protocol. I have seen some with no lead-in and short data pulse/spaces.

What language do you use with the AVRs? You can probably find example code somewhere as this is a frequent application.

BTW, I'm just across the river from c>Thanks, Dave.

Good question, I'll see if I can find it

Agreed. I think there are some examples in the app notes on the BASCOM site that will be helpful even though they are for IR - techniques are basically the same. It all depends on the protocol Motorola uses.

Although I'm now using PICs, I just measure pulsewidths until I get a start pulse and then capture the pulsetrain and decode it.

I like the X-10 MS10A security motion detector but it is also quite large. X-10 uses 310MHz so you'd need a different RF receiver and still have to build your own detector since it only talks to their security bases. The smaller X-10 *Eyes haven't been very reliable.

It's certainly welcome after the earlier heatwave.

There are a couple of Application Notes on the BASCOM site dealing with capturing and decoding Sony IR codes. The techniques will be similar. There are also application notes on Microchip's web site that show how to do it using PIC ASM. Even though the PIC ASM code won't be of any use, the techniques used have universal application.

The signal which I used at the very bottom of my article on RF Noise vs. Signal is about as tough as it can get. That signal is sent three times and has no obvious lead-in but you can find the beginning of a single code by looking for the space/pulse sequence composed of the final space of one copy and first pulse of the next copy. There will always be some such feature that you can use. Remember - the guy who designed the protocol also has to capture/decode it.

In this specific case, the code shown is the preamble sent by RF capable Pronto remotes. It precedes all RF codes sent to their RF extenders and identifies which RF extender should handle the code, the channel, and IR carrier frequency. But there's a special secret ingredient that simplifies things for the receivers in the RF extenders. The signal is double modulated by the both the data envelope and by ~38kHz. The superregenerative receiver I used (and recommended to you) cannot turn on/off fast enough to reproduce the 38kHz bursts so you get the data envelope directly. The Pronto RF extender's receiver can deal with the 38kHz and it will ignore any codes that lack it. IOW, its output will be as clean and noise free as the typical IR receiver.

Dave, After looking at your sit, I have a couple of questions for you.

1) Is the IR and RF generally about the same data rate? Could you send a Sony or RC5 message via RF as well.

2) I read about people referring to the X10 RF advanced security code. How is this different to the standard X10 RF codes. (I use AVR's to transmit X10 RF codes)

3) Have you had any experience with use the standard IR receivers I.e. Sharp UI60? (something like that) outside in the full sunlight.

Thanks for your time, Neil.

Yes. In fact, X-10 RF uses the NEC IR protocol. All of the RF extenders like X-10's Powermids are sending IR codes as RF.

I'm not sure what "advanced" codes are but the security codes just use a slight variation on the protocol (but one that still meets the NEC definition).

I describe both standard and security codes at...

No.

What do you use for a transmitter?

The easiest way to understand the security codes is to capture them as I described for the OP.

Standard codes use AB 'A'B CD 'C'D. Security codes use AB A'B CD 'C'D. For example:

9C 93 00 FF