Buttons work individually, but not in macro

[forestial]

I created an upgrade in RM for my Sony VPL-HS10 projector. This u/g just defines 4 discrete codes for OFF, ON, S-VIDEO1 and COMPONENT: these are mapped to buttons 0, 1, 3 and 6 respectively.

Loaded the upgrade into IR as a device. I use the TV button for this device and when pressing the buttons manually they operate the projector just fine. The remote is an 8910.

I also loaded in the protocol for delay (the standard one mentioned elsewhere in this forum). This projector needs something over 10 seconds from power on before it will accept another command. So I create a keymove to implement a delay with $D0 as the parameter, which seems to be plenty long enough. I have this keymoved to VCR/1104, the Pause key. (Don't have a VCR so this is just a spare key.)

Now I created a macro to power on my system. For the projector, the important things in the macro are to turn it on (so I have TV;1), wait a long delay (so I have VCR/1104;Pause) and then select the component input (so I have TV;6). Interspersed in the macro, but (I think) irrelevant here, are other commands to switch on my receiver and DVD player.

The macro successfully switches the projector (and receiver and dvd player) on, and the long delay works as expected, but the projector does not switch to component.

I have another macro to turn everything off: this includes TV;0. This also doesn't work. There is no delay in this macro: shouldn't need it.

But again, all of these buttons successfully operate the projector when used outside of the macros.

To make it really strange: I had this working previously. I've retraced my steps, reverted to saved IR files, and scratched my head for about 3 days, but I cannot figure out what went wrong.

Anyone else have experience with buttons that work by themselves, but not in a macro?

[Nils_Ekberg]

Make sure you are in the device when you execute the buttons. If that is not the problem upload the IR file to the jp1 files/diagnostics section so we can get a look at it.

[forestial]

Make sure you are in the device when you execute the buttons.

I think I am. The macro has 'TV;6' and TV is the button I've assigned to this.

If that is not the problem upload the IR file to the jp1 files/diagnostics section so we can get a look at it.

I'll do this later this evening

Thanks.

[forestial]

Checked the device button was correct (I use the TV button for my projector upgrade, under Type: VIDACC Setup code: 1957).

M1 is the key with the macro. The shutdown macro on the Sleep button also fails to turn the projector off ("TV;0" should do it but doesn;t)

I tried creating a couple of short macros: one with just "TV;6" (select component on projector) and one with "VCR;Pause;TV;6" (delay then select component on projector). They don't work any better.

One other observation from today: the macros work very sporadically. I had a couple of successes today after two days of failures. But a success rate of 1 in 30 is not going to get the job done.


I have uploaded the IR file to http://groups.yahoo.com/group/jp1/files ... os_bad.txt

Any help gratefully received.

[forestial]

Anyone have an idea what I may be doing wrong here? Any obvious errors in the file I posted?

I've tried many different ways of ordering things within that macro but it almost always fails to switch the projector to the component input. It has worked a few times but failed many, many more.

I tried entering the macro directly onto the remote rather than from IR, but it made no difference.

The same command, when sent by manually pressing a button, works reliably. So I don't think the problem is with the projector.

Also, the projector always turns on (I send a discrete ON command at the start of the macro - 'TV:1'). Why would one command work from a macro, while others do not?

[gjarboni]

Anyone have an idea what I may be doing wrong here? Any obvious errors in the file I posted?

Some protocols send commands differently in a macro then when pressed on the keypad. That would explain your problem.

Two things to try. Put that command that isn't working in the macro twice or three times.

Second, (and you'll have to switch to KM to do this) create your upgrade in the device combiner so you can specify duration. This is how long the protocol sends the signal. I would imagine increasing duration would increase your chances that your projector will recognize the signal.

[forestial]

Put that command that isn't working in the macro twice or three times.

I've tried putting it twice (didn't help): will try three times.

Second, (and you'll have to switch to KM to do this) create your upgrade in the device combiner so you can specify duration. This is how long the protocol sends the signal. I would imagine increasing duration would increase your chances that your projector will recognize the signal.

I've been using RM so far but I can give KM as shot. (Is this possible with RM?)

Thanks for the pointers

[gjarboni]

I've been using RM so far but I can give KM as shot. (Is this possible with RM?)

Nope you can't use RM with the device combiner (yet). That's why I mentioned that you'd have to switch to KM.

[jon_armstrong]

forestial,

I have a Sony VW-10HT projector and your symptoms remind me of when my wired IR distribution system became disconnected from the Projector.

Are you pointing your remote at the projector or do you have some sort of wired distribution system? I also know from controlling it with a home automation system that does IR, that for Sony commands l need at least three repeats (that you should be getting in a macro) to be reliable.

I don't see your file in the diagnosis area anymore but I completely agree with gjarboni's suggestions. I would also try varying the order that you send the commands.

[forestial]

A couple of new findings, prompted by the suggestions from gjarboni and jon.

1. I do have a wired IR system, and it is part of the problem. When I point the remote directly at the projector everything works. I had not considered this possibility before, because the manual commands work fine through the wired system.

2. I have experimented with repeating the command in the macro; 2 times seems to make no difference, 3 mostly works though I've seen it fail once, and 4 has been perfect so far.

The wired system is Xantech gear: an 'eye' at the front of the room wired to a connecting block in my equipment closet at the back. From the connecting block there are serveral outputs to IR emitters for the components, one of these being a 25 ft run to the projector which is ceiling mounted. I have had no problems with this system on any of the other components, only with the Sony projector in macros.

I believe the wired system is correctly connected, and the signal is certainly getting to the projector in some form because I can see the emitter flash when the macro runs. So it sounds like the wired system is degrading the signal sufficiently to cause the problem: but only the macro form of the command is susceptible to this degradation.

I plan to test the '4 repeats' a bit more and use that for now. But there's something unsatisfying about that solution so I also plan to look into gjarboni's suggestion on extending the duration of the commands. Now I need to learn KM.....

(Actually this stuff is fun to debug: software, hardware, comms protocols.... wish I had more time to spend on it.)

[mr_d_p_gumby]

I believe the wired system is correctly connected, and the signal is certainly getting to the projector in some form because I can see the emitter flash when the macro runs. So it sounds like the wired system is degrading the signal sufficiently to cause the problem: but only the macro form of the command is susceptible to this degradation.

By nature, any general purpose IR repeater system is going to "degrade" the signal because it's receiver has to have a wide bandwidth to be able to receive IR commands using many different carrier frequencies (or no carrier in some cases). That tends to reduce the signal to noise ratio, and the system may do other things to try and compensate for that. For example, it may have a longer time constant on it's AGC circuit. In non-technical terms, that may mean the very first one or two transmissions may be distorted to the point where the equipment won't recognize the signals.

When you execute your commands in a macro, the remote sends the absolute minimum transmissions of the command, whereas when you press the button manually, it probably sends it more times. If you can modifiy your setup to make the remote send more "minimum repeats" of the command (using Device Combiner), I think it will help your problem. As Jon said, most Sony equipment needs to receive the command 3 times before it will respond. If you are losing one or two to distortion then you may have to go as high as 5 repeats.

[jon_armstrong]

In my case the wired IR emitter had become disconnected. So the commands were working from reflections off the walls. Since Sony commands are relatively short, physically pressing a button will send many commands, but the three in a macro werent sufficient.

I have also had experience with noise (due to a ground loop) in my IR distiribution system. If you have the blinker type emitter, they will flash randomly or at a low level continuously. If you have the non-blinker emitter then if you have a web cam or camcorder they can detect IRand you can look at the output when no signals are being sent.

Also IIRC the Xantech IR targets (the "eye") requires 12VDC REGULATED. Most 12 VDC wall warts are unregulated and apparently that can cause reliability problems. The moral of the story is to buy the wall warts from Xantech and they cost a few $$ more. It is also possible that if you have the DC power coming down CAT5 and there is a long cable run then the resistance of the wire can become significant. If you have a multi-tester then you might measure the DC voltage at the "eye".

[forestial]

Jon,

I have the blinking emitters: I don't see any spurious flashing.

I do have the Xantech wall wart. I am using CAT5 for the cabling. It is about a 25ft run from the eye to the connecting block and then another 25ft from the block to the projector. When wiring it I doubled up by using two conductors from the CAT5 cable in parallel for each connection. I'll have to check on the voltage.

One more thing I am wondering about - is using the Device Combiner and adjusting the duration really equivalent to putting multiple commands in the macro? The term 'duration' seems to imply 'one command, but keep sending it longer' versus repeating it several times.

[mr_d_p_gumby]

One more thing I am wondering about - is using the Device Combiner and adjusting the duration really equivalent to putting multiple commands in the macro? The term 'duration' seems to imply 'one command, but keep sending it longer' versus repeating it several times.

The duration number specifies the number of times the command will be transmitted. When you do it in a macro, there will be a larger gap between commands (unless you are using an extender). So, no, they are not equivalent. The macro approach may take longer than is really necessary because of the delay between commands.

[forestial]

I implemented the suggestion of using the Device Combiner and specifying durations for the commands that were not working. This seems to have done the trick. It seems to need a duration of 5 for the Component command, as was suggested above. Now the macros are working reliably.

As a matter of curiosity, I wonder why some commands (discrete ON) seem to not need the extra duration, while others do.

Much thanks to everyone who contributed their knowledge in the replies above.

(Jon, since you said you have a Sony projector, does it need a long delay from power on before it will accept other commands? With the HS10 it seems to need around a 10 second delay.)