Bizarre IR behavior Dish Network HD Satellite Rcvr
Hedberg
Posts: 671
Had more than a barrel of fun today with a new Dish Network HD satellite receiver. Captured some IR using the IRIS and when I tried to control the box with the captured codes it seemed as if the box would repeat the codes sent all by itself. For example, if I went into the "guide" and hit the down arrow, the box would continue to respond as if the signal were being sent over and over -- even if the emitter was subsequently removed from the front of the box. It does this for a while and then just shuts itself off and won't turn back on until you turn your back on it -- then it turns itself back on. I guess when it's good and ready.
A pulse of .1 seconds won't get the thing to respond at all but a pulse of .2 seconds sends it into orbit.
Is this common with these stupid things? I'm sure we've done Dish Network boxes before without difficulty, but this appears to be a pretty recent receiver. Is it most likely a matter of too strong of an IR emission? Maybe we need to try some black tape and a 1k resister.
A pulse of .1 seconds won't get the thing to respond at all but a pulse of .2 seconds sends it into orbit.
Is this common with these stupid things? I'm sure we've done Dish Network boxes before without difficulty, but this appears to be a pretty recent receiver. Is it most likely a matter of too strong of an IR emission? Maybe we need to try some black tape and a 1k resister.
0
Comments
Did you try .15?
Paul
Your best bet, if you can find them, would be to cut and paste hex. That way you can be sure you have it proper. I tend to use my IRIS only as a last resort.
I've attached an IR file I use with these boxes all the time, maybe that will help (ignore the model number, it was correct when I first made the file, but I've updated it for newer boxes as they have come out).
It doesn't appear to me to be a poorly captured file. Thing is that if I go to the 'guide' and get to the channel list, a .2 second (that's an integer 2 -- I'd try .15 if I could find an integer value when multiplied by .1 gave me .15) down arrow pulse makes the receiver respond as if it's getting hundreds of down arrow pulses. The master has no indication (red light on the port) that multiple pulses are being sent and the the behavior continues even when the emitter is removed from the box. That is, once the box thinks it's getting pulsed it continues even though it can't possibly be. The box is a new model that doesn't appear on the DishNetwork website (best I can tell). It doesn't have mechanical buttons on it -- the buttons are touch sensitive rather like a touch panel. When you manipulate it with the IR remote, the power indicator flashes once for every IR command sent. When you send it a down arrow, it flashes about 4 times a second for about a minute (even though there is no IR being sent beyond .2 seconds) for about a minute until the unit turns itself off. I couldn't find a model number or manufacturer name on the box -- it doesn't look like the EchoStars that I've seen, but it seems that EchoStar is what Dish Network uses, so perhaps it is.
The IR detector on the box is pretty sensitive. Using the IR remote, the box will respond to a signal bounced off a white wall about 6 feet away from the box. I tried positioning the emitter off the box but aimed at it but then it didn't want to respond at all. That sometimes works. Probably will try some more and maybe try the 1k resistor or obscuring part of the window. Problem is that once the thing starts taking off it's about five minutes before it gets back to stability and allowing itself to be turned on so experimentation is not easy. Man, I hate IR control.
Ignore the brand as well? The question is about DISH and your file is titled for DIRECTV.
Oh snap ... apparently, I've been having one of those cognitive disconnects where someone is saying one thing, and I'm hearing something else entirely. Well, forget that IR file
I'll second your hate for IR control, esp sat receivers
This stupid box has an ethernet connection but the documentation gives no indication that it has either a web server or telnet-like server that allows for control of the box. I don't understand the mentality of creating a functional IP port without function control as well.
This tends to be indicative of a dirty IR stream being received, caused by interference from surrounding light sources or a bad IR stream being sent out.
It is unlikely (but not impossible) that the amplifier stage in the receiver circuitry is being driven too hard; I have personally never seen this issue on any box. Attenuating the IR output sent to the box can smooth over these other issues in certain cases.
First check is always that a blockout cover is placed over the IR receiver on the box to obscure stray light sources. In this case, in a closet, this is unlikely to be the issue.
Second check is that the IR file is good. IR learners (even the IRIS which generally isn't too bad) don't always capture codes "correctly" (timing wise) or "sensibly" (repeat pattern wise).
That said, if the box is going off into the weeds like you suggest it sounds as though it has some firmware issues to add to the mix.
Harold, if you can post your IR file, I can have a look and see if anything appears odd about it. Without a remote for the box I can't be certain about timing, but it should be obvious if the repeat pattern hasn't been learnt properly.
Ok, that's way too kind of an offer for me to refuse. I think I have the file attached -- it seems that it has to be zipped before it will allow an upload.
I captured this file with the Iris and didn't pay a lot of attention to the channel name to function protocol, so the channels won't match the standard.
edit:
Just so you know, it's not just an offer too kind to refuse, it's a very generous offer which is appreciated very much. Thank you.
I would try learning it again to see if it captures differently. There's a good chance that this code is what's sending the box into orbit if you're pulsing digit-digit-[...]-enter or similar.
It looks like the second code could be the same code with a toggle bit inverted; when you capture you may need to learn the code once, aim the remote under the desk and press the button again and then do the verify in order to get it to learn correctly.
The waveform should probably look like this:
If you capture it again and it still looks like the first waveform, or you still have problems with it, let me know and alter the function manually and send back a few versions for you to try out.
My pleasure, hope I can help out.
Yes, I see in IR edit that this waveform is different from all the others. But, it seems that it's the channel up and down signals that drive the box crazy, though I can't be certain that the enter/select signal wasn't sent in there somewhere as well. When we go out there again, I'll recapture the enter/select function and see if it comes up different. I've got plenty of ideas to play with. Thanks again.
Harold
Thanks again.
Harold