IR & Ocelot troubles

Sacarino

Member
Hi guys.

I need some assistance with my new setup here. I've moved to a new place and am slowly rebuilding my IR zone in my spare time.

My first problem is the wiring of my zone. I have a sensor next to the TV which pipes the IR into the Xantech block in the closet and then out to the emitters. I would like to insert the Ocelot in parallel with the Xantech block so my HS install knows what IR was sent and can react if needed. This works GREAT - as long as I only want the Ocelot to monitor the IR. If I attempt to send IR, it appears to lock up. I read a post from Guy in John's IR thread that said the Ocelot can't send and receive at the same time - or did I misunderstand?

My second problem is my Ocelot doesn't seem to want to work with my SA 3250HD box. I had a similar SA box before, so I know it can handle the 58Khz IR - I'm a little stumped. It apparently learns sucessfully - from what I can tell - but I get zilch when I transmit. It's transmitting something, just not something that works.

I would appreciate any insight you guys might have into my issues!

Thanks,
Shane
 
The Ocelot will work fine in the loop with you Xantech stuff, there will just be a slight delay between sending commands.

Normally, if you trigger a series of IR commands in the Ocelot it would send:
send command 1
send command 2
send command 3

If you have the Ocelot seeing it's own output there's a delay becuase it's now doing this:

send command 1
receive command 1
send command 2
receive command 2
send command 3
receive command 3

I think there will be even more of a delay if the commands are one that the are in the first x banks (user defined) for the ocelot to react to.

Using an SECU16IR, for some reason, speeds this up a tiny bit, but not much.

The best way to do this, and what I plan on doing soon, is to use a Xantech 2-zone block first.

- Connect your receiver to the 2-zone block's zone 1-in.
- Place an emitter from the 2-zone block's zone 1-out on the Ocelot's IR receiver
- Use a SmartHome IRLinc to connect the Ocelot's IR output to the 2-zone block's zone 2 input, Guy's also designed a cable with a diode in it that works as well...
- Run a cable from the 2-zone block's common emmiter port to your Amplified connecting block's IR input
- Run emitters from the connecting block to your components and additional connecting blocks as needed
- connect an emitter to the zone 2-out on the 2-zone block and tuck it away (unless you need something that only the Ocelot can control)

What happens with this setup is:
- The Ocelot and all your components see what your receiver sees
- Only your components see what the Ocelot sends
- The Ocelot does not see what it sends and therefore sends its commands fast

Here's a diagram:
ocelotin2zone.jpg
 
Sacarino said:
My second problem is my Ocelot doesn't seem to want to work with my SA 3250HD box. I had a similar SA box before, so I know it can handle the 58Khz IR - I'm a little stumped. It apparently learns sucessfully - from what I can tell - but I get zilch when I transmit. It's transmitting something, just not something that works.
OK, problem #2...

Are you learning the IR from HomeSeer or from CMAX?

I've found, if you need to learn something other than the Ocelot's defualt IR freq, you have to do it from CMAX, as HS does not let you specify what freq to learn at.

It takes a bit more planning to learn IR through CMAX, but it's not that hard.

Go ahead and setup all your IR labels in HS. If you're using the ADIOcelot plugin rather than the HSTech Ocelot plugin, you can have HS map all your IR labels to your CMAX project.

Now, while still in HS, attempt to learn the first command - We already know it won't work, but the purpose for doing this is to learn what location the command would be stored in. Let's say it's going to be stored in location 112.

So, now get out a piece of paper, and write down 112 and the lable for that command. Now that you know the first command is 112, you know the next is 113, the next 114, and so on. Just make a list of all the commands and their locations.

Now use the ADIOcelot plugin to start CMAX, or close HS and start CMAX. Connect to the controller and learn IR, starting at (as in the example) 112, and with the freq set to 58. learn yuor first command to 112, the second to 113, and so on. As long as you follow the list you made, you should be good to go!

Hopefully this will work!
 
A quick search on the ADI forum shows that several people have reported problems with learning Scientific Atlanta IR codes (just search for "atlanta"). In one case, all you had to do was specify a 58kHz carrier frequency. However it appears that you were already aware of this requirement. In other cases it seems that it still didn't work, and I remember a specific case (although I'm not sure it was a SA device) where the timing was very critical.
 
Could he go the pronto code to IRMax route too? Get the pronto codes from RemoteCentral and go from there?
 
Yes, you could try that too. Ideally, if you could find someone that has the same box in his setup then you ask for his LIR file and extract the needed codes from it using IR-Max.
 
A little off topic, but please let us know if you got the SA box to work. I currently have a Pace box, but Time Warner is using the new SA boxes now as well, so I am ready to call them and get it switched out, assuming there is a discrete on/off for the SA box.
 
John: Great diagram! I'll re-rig my wiring as soon as I get my hands on a 2 zone block. I never would have thought to use a multizone block, either. That's brilliant - thanks!

Side note: What's the verdict on a legit source to buy that from? I checked AO, Martin didn't have much listed that I saw... I'm not keen about Smarthome, but I'll go where I have to.

Guy: Yessir, I've spent many an hour browsing around trying to find the problem. I have been attempting to learn the IR directly from CMAX with the 58Khz freq spec'd (and I had a lower model SA box that worked great previously), but I'm not getting any love this time around. I'm wondering if this particular SA box is different. It's a different remote than I had last time, I know that much.

When you mention critical timing, what are you referring to? Interference occuring while learning? Or something during transmission?

I looked into the IRMax/Pronto route once before but backed away because I hadn't the foggiest idea what the hell I was doing. That was back when I was just getting started with all this though, so perhaps I know enough now to be dangerous. I'll give that a go, but I'd sure like to troubleshoot the cause of the other issue.

Thanks for the brainpower, guys. I'm in your debt.
 
Worthington is the only place to get Xantech gear from. They're inexpensive and ship fast.

Go to Xantech's site first: http://www.xantech.com/ to figure out exactly what you need. The Worthington site is not big on descriptions, so you need to go there already knowing EXACTLY what you need.

I've talked to Martin about carrying Xantech, but there's not enough demand for it, I think.
 
By critical timing, I mean the actual pulse durations themselves. Like I said, I don't remember if this was a SA box or something else but just fudging the timing a bit made a difference between working or not. Normally, IR codes are quite tolerant on timing and only measure pulse to pulse time instead of absolute time from the lead-in pulse (like RS-232 serial does). This self-clocking scheme allows for things like weaker batteries in a remote, etc affecting the timing.
 
Shane, have you tried all the learning "tricks"

- push learn, push button on remote
- push button on remote, push learn
- put a card between the remote and the ocelot, press button, press learn, remove card
- learn the code to something else, then use it to teach the ocelot useing any of the above techniques
- try all of the above with a "tunnel" so no stray light gets in
- try all of the above varying the distance from point blank to a foot or two in small increments
 
I have my Ocelot stored waiting for some free time to install it, so I dont have any practice with it yet. However I wonder, how feasible it is to turn the IR receiver OFF temporarily while you send the IR commands either by directly handling the IR receiver setup or by using a relay to open the input of signals?

Assuming that nothing else is possible, sending a command would have the following rules:

1. Open relay connecting the IR receiver
2. Send several commands
2. Close the relay connecting the IR receiver

Is this possible? Would this speed up the sending of several IR commands at the same time? Like when setting up an scenario in my HT?
 
I've never seen anyone do that but it could be tried. I suspect that it might be a bit slow to turn the relay on and off however, limiting the usefulness of doing this.
 
Guy Lavoie said:
I suspect that it might be a bit slow to turn the relay on and off however, limiting the usefulness of doing this.
Come on Guy! Your are my Electronics & Ocelot oracle. I was expecting you to reply with a paintbrush-drawn design inspired by the gods! :lol:

Now, serious. What solid state relay would you recommend? Would a transistor work? I dont know, probably a MOSFET.
 
What I'm saying is that controlling a relay in order to prevent the Ocelot from receiving IR would be a slow process if you're counting on a module like a SECU16. Because of the variable bus latency, you would need to actually wait a second or so before sending any IR commands to be sure the IR input is disconnected.

Another approach might be to have a 555 timing circuit directly in parallel with the IR output that would disable the IR input through some kind of gate (like a AND gate); a type of "missing pulse" detector circuit that would wait for any IR output to have stopped for a few extra milliseconds before re-enabling the IR input. But this would be kind of kludgy at best.
 
Back
Top