touchpad right button seems stuck

Started by diegor, November 08, 2017, 11:23:35 PM

Previous topic - Next topic

diegor

Hi,

a couple of day ago, I've received the Teres, and I've already assembled. It worked well, and suddenly today, it started to behave strangly. At the beginning it get random "right click". Now it's like the right button is always pressed, but it's not.

  I've opened and double checked the ribbon cable, and removed the pcb touch button, and checked with a tester the right button, and it worked fine (short circuit, when pressed, open otherwise).

  Do you suggest any other check?

Thanks in advance,
  Diego.

 

JFP74

We assembled the Teres yesterday with 4 children (12,12,12 and 10). It was a great shared day.
Laptop works fine but we have also seen that right click append sometime when we use touchpad.
Usually, touchpad can be setup to make a right clic when you hit one time the surface whitout slide.
Maybe it comes from a touchpad failure ?

LubOlimex

Hey guys,

Thanks for the input.

This is a known issue that can be fixed by software means. You need to update the softawre of the AVR board that controls the keyboard and touch-pad (and its buttons). This is done on the laptop itself, using the command line interface. Step-by-step instructions are located here: https://github.com/d3v1c3nv11/TERES-KBD-RELEASE/tree/master/upgrade - refer to the README.md

Best regards,
Lub/OLIMEX
Technical support and documentation manager at Olimex

albertux

Same problem here, ad also the same problem after update touchpad firmware...what can I do? I re-open teres and check two flat cables from touch pad connector, and seems OK.... what can I do?  :(

Thanks in advance

albertux

diegor

Hi,

  I've updated the firmware today. It's more usable, but there's still problem. To analyze the problem, I've used evtest.

  $ evtest | tee test.log

Then I've tapped on a touch pad, pressed left button and then right button. The output was this:

Input driver version is 1.0.1
Input device ID: bus 0x3 vendor 0x15ba product 0x3c version 0x111
Input device name: "Olimex Ltd. TERES Keyboard+Touchpad"
Supported events:
  Event type 0 (EV_SYN)
  Event type 1 (EV_KEY)
    Event code 272 (BTN_LEFT)
    Event code 273 (BTN_RIGHT)
    Event code 274 (BTN_MIDDLE)
  Event type 2 (EV_REL)
    Event code 0 (REL_X)
    Event code 1 (REL_Y)
    Event code 8 (REL_WHEEL)
    Event code 9 (REL_MISC)
  Event type 4 (EV_MSC)
    Event code 4 (MSC_SCAN)
Properties:
Testing ... (interrupt to exit)


A tap on touchpad ->


Event: time 1510430472.660385, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90001
Event: time 1510430472.660385, type 1 (EV_KEY), code 272 (BTN_LEFT), value 1
Event: time 1510430472.660385, -------------- SYN_REPORT ------------
Event: time 1510430472.668411, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90001
Event: time 1510430472.668411, type 1 (EV_KEY), code 272 (BTN_LEFT), value 0
Event: time 1510430472.668411, -------------- SYN_REPORT ------------


Press left button ->

Event: time 1510430475.109700, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90001
Event: time 1510430475.109700, type 1 (EV_KEY), code 272 (BTN_LEFT), value 1
Event: time 1510430475.109700, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90002
Event: time 1510430475.109700, type 1 (EV_KEY), code 273 (BTN_RIGHT), value 0
Event: time 1510430475.109700, -------------- SYN_REPORT ------------


Release left button ->

Event: time 1510430475.277780, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90001
Event: time 1510430475.277780, type 1 (EV_KEY), code 272 (BTN_LEFT), value 0
Event: time 1510430475.277780, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90002
Event: time 1510430475.277780, type 1 (EV_KEY), code 273 (BTN_RIGHT), value 1


Nothing happen when I press the right button.

Thanks in advance,
  Diego.

ibu

Hi,

after assembling my TERES-I everything seemed to work (except for part of the desktop sounds) with the original ubuntu - congrats @olimex!

One day later, however, the touchpad stopped working (while the case was closed). (The same happened to a friend of mine.) I managed to fix this through readjusting the cable between the touchpad and PCB3-TOUCH. (Maybe it had changed position due to dilatation while the battery was loading and warming up?)

At the same time that the above error appeared I also found the right touchpad button to be stuck. For quite some time I tried to the following (with many shutdowns and some battery disconnects inbetween):


  • readjust the cable between PCB1-Main and PCB5-Keyboard
  • flip side of this cable
  • exchange the cable with that connecting PCB4-PWR to PCB1-Main
  • Use a screw to fix PCB5-Keyboard to the bottom from inside (because movements while closing the case might displace the cable)

Directly after exchanging cables, everthing worked for a few minutes. However the problem reappeared. The current situation is this:


  • Moving the cursor works, both with the touchpad and with an external mouse.
  • The left touchpad button somehow works; however, the right one seems to have been virtually pressed directly afterwards (context menu appears). As diegor (cf. previous post) I see this sequence of xevents: BTN_LEFT pressed; BTN_RIGHT released; BTN_LEFT released; BTN_RIGHT pressed
  • The right touchpad button also somehow works (context menu appears); however pressing and relasing it generates the same sequence of xevents as the right one.
  • For an external USB mouse the middle mouse button works; the left mouse button can close an already open context menu, but otherwise does have no effect; the right mouse button does have entirely no effect, except for this: When I keep the right mouse button of the external mouse pressed and press the right touchpad button, no context menu appears.
  • With evtest I see xevents from the touchpad only when I press a touchpad button or touch the touchpad.

It seems that either there is a short-curcuit between the left and right touchpad button, or the AVR keyboard software cannot discern both.

Of course, before doing all that, I've updated the AVR keyboard software as suggested previously in this thread.

Could it still be a software issue?

Best,
ibu

diegor

Hello,

  during the weekend I've made some more tests. And I've found out that:

1) vcc on the avr processor is 5v (ok)
2) avcc on the avr processor is around 3.4v (weird)
3) abut0 with no button pressed is 3.4v (wrong but consistent with previous point)
4) abut0 with left button pressed is 2.65v
5) abut0 with right button pressed is 3v

  I tried to disconnect the cable to pcb touchpad, and abut0 is 5v as expected.

  I removed the pcb touchpad, and while disconnected from everything, I misured the resistance at C1 and C2, and it gived me 8K and 15K, even if the button were not pressed. When pressed the resistance were 0 as expected. I don't know if it is a problem with the capacitor or the switches, but this parassite resistance would explain the low voltage of abut0, and low level of abut0 would explain the apparent right button stuck.

  I tried to solve the problem changing the level value in the firmware souce code, but it's not easy, because when I've measured abut0 another time, voltage level are slightly different:

  no button -> 3.10v
  L button pressed -> 2.64v
  R button pressed -> 2.9v

  Last test was to detach the touchpad cable, but there was not difference. So I think that should be a problem with the PCB3-TOUCH-BTN.

Thanks,
  Diego.

zinoviev

If this is not too difficult, would it be possible for you to open the laptop and make photos of TERES-PCB3-TOUCH and TERES-PCB5-KEYBOARD? I'd like to check something.

LubOlimex

As you guessed correctly the values here look different. Here is a median between a couple of boards:

AVCC=4.9V
L=2.74V
R=3.5V
M (middle button, both buttons pressed) = 2.24V

Technical support and documentation manager at Olimex


LubOlimex

Some of the connections seem loose. For example, the smallest ribbon cable between TERES-PCB5-KEYBOARD and the main board seems to be partially out and the lock seems to be disengaged. The locks are a movable part at the top of the connectors for the ribbon cable. You need to disengage the plastic lock (pull them out of the connector), then insert the cable and then close the lock (push the movable part towards the base of the connector). Also the smaller ribbon cable between TERES-PB5-KEYBOARD and TERES-PCB3-TOUCH-BTN seems out.

Best regards,
Lub/OLIMEX
Technical support and documentation manager at Olimex

diegor

Hi,

  the problem is not the cable, I've already check the cables, and even detaching and retaching the cable don't fix the problem.

  Also, the resistance measured over the capacitor of the button pcb, is low (8K and 15K ohm), even if the pcb is completly disconnected. In the next few day I'll make some more test (with new photo if you need).

Thanks,
  Diego.


nix

I am having a similar problem:

the touchpad works fine, but the buttons don't respond at all anymore (tested with evtest). Earlier, the left button was acting as if it was the right button, and the right button didn't work.

nix

Swapping the PCB3-TOUCH between two laptops also transfers the problem, so it really appears to be the PCB3-TOUCH that is faulty.

Also, on the faulty PCB3-TOUCH, the voltages on SWL1 and SWR1 are strange:
SWL1: 0.04V
SWR1: 3.52V

On the working PCB3-TOUCH, the voltages are 4.9V.