I'm running 2.0-b11 and Hijack 253 and everything had been working great, until I added a RM-X4S. I purchased it new from Crutchfield and plugged it into the tuner module. Unfortunately it rarely works. Occasionally on power up all buttons on it function, but normally only the "source" and "off" buttons if I'm lucky. I've checked and re-checked all the connections without any success, and the button test option in Hijack doesn't show any results on these "dead buttons". Is the problem a bad RM-X4S (exchange with Crutchfield), a bad tuner module (still under warranty), software problem, or something else. I should mention that I have also tried my "spare" RioCar running v1.03 without Hijack and get similar results. Is there a way to test the RM-X4S? I could sure use some advice on troubleshooting this, since it is driving me mad. Regards, Mark

- Is the tuner module itself working properly (i.e., you don't have the "bad tuner module extension cable" problem covered in the FAQ)?

- Are you running 2.0 beta 11? Have you tried it without Hijack?

- Can you check the output of the plug of the RM-X4S to make sure it match the resistance values expected (found in the Developer Info section of RioCar.org)?

Thanks Tony! I hadn't realized that the RM-X4S worked by changing resistance values. I will dig out my multimeter and check this out. I did inspect the extension cable which does not appear to be wired incorrectly. All other features of the tuner (FM, AM & RDS reception) work fine. I'll let you know what I find out.

Update: The resistance values of the RM-X4S all checked out within the proper ranges (thanks for your suggestion Tony). I've tried clean installs of v1.03 and 2.00beta11 (without Hijack). I've tried removing the tuner extension cable, and I have checked for loose or disconnected wires. None of these things have made a difference -- only the OFF button (61 ohm) results in any result. Capturing the stalk commands to the serial port verified that the only buttons that consistently send a value are OFF and pressing the stalk down and releasing it. Last week other buttons would intermittently work, but even they would not result in the correct commands. Now these buttons result in nothing.

I have written Tech Support, but have more or less reached the conclusion that something is wrong within the tuner module. Hopefully they can suggest a fix which won't require that I mail my tuner module off to the UK. ...Mark

I s'pose you have already discovered this debugging aid for config.ini:

[hijack]

trace_tuner=1

Then watch for messages on the serial port while pressing buttons.

I didn't try "trace_tuner=1". Would that be more informative than the "stalk_debug=1" command which I did use?

I had also tried connecting with hyperterm with the following commands:

stty -echo raw 19200 < /dev/ttyS0
od -txC -v -w4 /dev/ttyS0

however the results were similar to "stalk_debug=1".

David of SONICblue's Technical Support believes that it is, "a dry (solder) joint at best or a faulty opamp at worst". Hopefully a trip to England will do the trick for my tuner/RM-X4S problem. I'll let you know how it turns out. ...Mark

As promised, I am reporting back with the results. I returned the tuner to SONICblue (UK) and Hugo got it working. He reported,

"...we think it may have been a stray bit of solder causing a join where it shouldn't."

I re-installed it the same day I received it back, and it has worked flawlessly for a week now. It was well worth getting working, since it does make it much easier to operate all the features of the empeg while still keeping my eyes on the road. My thanks go out to David and Hugo for all there help!

Thank you for the excellent thread, espeacially the complete follow up.
I too have recently acquried an RM-X4S and had a follow up question.
It seems to work, however the response to button presses is intermittant. Half of my button presses result in no response. I am wondering what repsonse people are getting? I understand that there is an intentional response "release delay" behavior. I have yet to resistance test the stalk and/or view the Hijack button codes. I really do hope to hear that something is awry with my stalk/tuner, because its almost un-usable this way.

Also, I see how to get into the menu but once in it seems Skip should allow you to scroll the menu. Without that all I can do is down-down-down...pretty limiting. What am I missing besides doing my own ir-mapping...or could it be a faulty stalk/tuner?

Regards,
Michael

In reply to:

---

I too have recently acquried an RM-X4S and had a follow up question.
It seems to work, however the response to button presses is intermittant. Half of my button presses result in no response. I am wondering what repsonse people are getting? I understand that there is an intentional response "release delay" behavior. I have yet to resistance test the stalk and/or view the Hijack button codes. I really do hope to hear that something is awry with my stalk/tuner, because its almost un-usable this way.

Also, I see how to get into the menu but once in it seems Skip should allow you to scroll the menu. Without that all I can do is down-down-down...pretty limiting. What am I missing besides doing my own ir-mapping...or could it be a faulty stalk/tuner?

---

There are 2 known bugs in the 2.0 Beta 11 software (which is what Hijack also uses as its base), that affect Stalk operation.

1. There are delays on processing Stalk commands - this is a known bug and is mentioned in the bug list forum for 2.0 B11.

2. Menu navigation via stalk is 'broken' in 2.0Beta 11 software (also in the 2.0 Beta 11 bug list).

There is a thread somewhere around here about how to map the Stalk skip forward/back into front panel left/right buttons as a work around for this bug.

The first problem you just have to live with I think.

Other than that the Stalk buttons generally work fine - the buttons I use the most are the Vol up/down and skip forward/back plus next/previous visual (shift Vol+/shift vol-), Mute, source (button on end of knob) & cancel.

If you think your stalk is busted/not working properly consult the FAQ, there is some information on how to see if your stalk is working properly.

properly consult the FAQ, there is some information on how to see if your stalk is working properly

I think that's all in the Developer Info section of riocar.org rather than the FAQ.

you're right Tony,
however if one does a search on sony stalk on the riocar.org site FAQ area you get this link
Hijack Config.ini options - [amoungst a few others].

Which does indicate how to enable Stalk debug mode, which dumps out all Stalk packets to the serial port.

I would however suggest that this part of the Hijack config.ini section in the FAQ should possibly have a link to the Developer section on riocar.org which explains what the debug info means/how to interpret it.

Yeah, good idea. HEY LOREN!

One other thing - Loren can you add this to the FAQ somewhere when you update the Stalk FAQ entry.

If you have Hijack installed, then enabling 'Button Codes display' when your Empeg is in your car will enable you to see the Stalk 'button codes' as you tweak/manipulate the Stalk buttons.

This feature of Hijack provides a simple check' that the Empeg is in fact getting Stalk commands when you fiddle with the buttons.

Not sure where the button codes (hex numbers) Hijack displays when Buttons Codes mode is enabled come from, they may be the symbolic codes created by Mark when trapping the stalk codes for mapping/translation - if so there may be a table we can show of Hijack Stalk button codes as this might aid someone to workout what 'button' hijack thinks there Stalk buttons is.

I have Hijack v278 installed, but this feature would have been in the kernel for a while, I never needed it before now, so never tried it on earlier version of Hijack.