Bongofish
December 10, 2019, 08:45:06 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News: You may have to login twice the first time,  but we don't know why - Erm I mean it's a security thing yeah that's it - security.
 
   Home   Help Search Login Register  
Pages: 1 [2]
  Print  
Author Topic: OQO tablet PC digitizer  (Read 15422 times)
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #15 on: May 15, 2011, 04:13:14 AM »

Will, I just went through reading your OQOasis thread.  (And also a few posts on DIYElectric!! hehe)

I was wondering: OQO is out of business. Did they release any internal data? Like Service manuals or even internal documents about schematics or connector pinouts?  It would be nice to have more data -- I am sure this could actually help in repairing OQOs (and finding the wacom signal pins on the connector).  Picasso would be the one to ask I suppose? Does he have any contact with any of these guys?
« Last Edit: May 15, 2011, 05:02:06 AM by bernard » Logged
Wilcorp70
Jr. Member
**
Posts: 63


View Profile
« Reply #16 on: May 15, 2011, 05:32:47 AM »

Yeah, apparently I am leaving quite the digital fingerprint; I never could get those guys on DIYElectric to focus and pick a wheel motor design.  Picasso would be the guy to ask; he hasn't been replying the last few days as I need some info about the extra OQO motherboards he sent me.  Once he starts posting again, I'll ask him about it.  Tomorrow or Monday,  I will start prodding around with the continuity tester on the digitizer and see what I get.
Logged
Wilcorp70
Jr. Member
**
Posts: 63


View Profile
« Reply #17 on: May 16, 2011, 06:16:10 AM »

Well, I started poking around a little bit with the continuity tester; there is a chunk of 7 pins that seem to be the area of the rectangular connector that deal with transferring digitizer information.  When a voltage is applied to where the digitizer connects to the daughterboard 5 of the pins in that area respond with a voltage reading (2 pins in the middle of the 7 don't read anything).  When I tested the the touch scrollers; I got readings from the pins on the opposite side of the connector.  I couldn't find a good spot to test the ambient light sensor.

Now I realize this is fairly arbitrary, but it does give me a starting point.  I am pretty sure that I've at least isolated the group of pins that are responsible for the digitizer control, now I just have to figure out the signals.  I'll post a diagram of where the pins are I've identified on Monday. 
Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #18 on: May 16, 2011, 06:27:54 AM »

Nice detective work!  Cool   

Talk about where you are injecting power also.  This is 3.3v? 

Try to find the pin numbers if you can (where is pin 1?) One thing to try to is to map the wacom signals in the same pin order. (but that is just a wild guess)

What type of tools do you have to do this?  What is your power source? 
« Last Edit: May 16, 2011, 06:30:50 AM by bernard » Logged
Wilcorp70
Jr. Member
**
Posts: 63


View Profile
« Reply #19 on: May 16, 2011, 10:48:58 PM »

Last night was just a quick test.  I don't know the voltage supplied by the continuity tester on the multimeter.  I simply placed the multimeter into continuity mode and placed one probe onto a random pin of the connector at A on the diagram of the board then I placed the second probe onto the connector at B.  I then checked the pins one by one on connector B and found that pins 1-7 were the general area that showed a voltage of 500 (it doesn't give units, but my guess is that means 5 or .5 volts).  Pins 1-3 and 6-7 all showed the same voltage while 4 and 5 showed no voltage; simply displaying 1 on the meter.  It doesn't seem to matter which pin on A I connect the meter to, there is no difference in which pins show a voltage or the voltage shown.  When I applied the probe to C pins 31-34 showed the same voltage (500), and all other pins showed 1.  

* daughterboard.png (46.33 KB. 1357x787 - viewed 680 times.)

Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #20 on: May 17, 2011, 02:08:24 AM »

Often multimeters in continuity mode also gives a "Ohm" reading (the resistance). But that might be different in your case of course.

Nice diagram!!  Very clear.

Do you think you would be able to connect wires to those signals? Is this why you were lurking a cable? How far apart are the pins?

First thing to find is the ground.  Next the power.  I would think that this type of boards runs at 3.3v -- All TabletPCs babyboard that I have seen ran at 3.3v (but all the standalone wacom tablets run at 5v internally).   

Do you have 3.3v source? 

Do you have more than one multi-meter?  I say this because I always keep one multi-meter "in series" in the incoming power set to "DC Amperage" -- I _always_ keep an eye on the amperage that is being pulled off. If too "high", it's time for an immediate disconnect!!! Something is wrong and might be "burning" internally. I also pass my fingers onto the ICs to see if any one is becoming super-hot.  BTW, if you happen to have "dust buster" cans (compressed air) you can simulate "freeze" by putting it upside down and let the little liquid "drip" by pushing lightly on the button -- it is like freezing liquid (not for your fingers).  Putting onto too-hot chips can help saving it.

Also the Amp reading gives you a hint of what the board is currently "doing". As you try things you eventually "remember" the different "amp" readings and eventually map it to various "states" -- for example, you can see that when you approach a pen the amp reading goes up. If you remove the pen the amp still stays up for small amount of time before returning to some lower "idle" state.  So even if there is nothing being outputed to a "serial port" you can testify that the board is active reading the pen.

For the power and ground, sometimes the lines on the PCB are wider -- and in particular on the cable (or it uses more than one connection to share the amperage).
Logged
Wilcorp70
Jr. Member
**
Posts: 63


View Profile
« Reply #21 on: May 17, 2011, 05:03:29 AM »

Yeah, I tried to find out what my multimeter was outputting.  I should have mentioned voltage was just my best guess based on some quick internet searching.  I actually have the ribbon cable for this connector already.  I'm not sure about the pitch on these pins I'll have to find something to measure it.  I'm not sure if I can solder that well, the ribbon cable has a couple of 90 degree bends and at those points the spacing is a little wider. 

I figure the digitizer probably runs off of 3.3v.  Couldn't I just use a couple of AA batteries?  I would think .3v would be within the acceptable margin.  I don't have a second multimeter, Ill see if I can scrounge one up.  None of the connections seem wider, I think it might share the power lines.  One test I might try is taking the meter to the motherboard while its powered, I just have to make sure to have the digitizer plugged in when windows starts and then unplug it otherwise it wont power it up in the first place.  Then Ill try by powering the digitizer alone.
Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #22 on: May 17, 2011, 05:27:57 AM »

Did you say to unplug it while being powered? I wouldn't risk doing this on a working OQO!  Can't you just probe the signals when it is powered (and connected)?  Have you found a reliable ground to hook the black multi-meter probe (to poke voltage levels with the red probe)?  I often try to find a way to hook the black probe so I do not need to "hold it" with my hand. When poking with the red probe to see voltage levels on a live PCB, you do not want to create shorts by touching two pins at the same time -- you thus must keep the full attention to that one probe without worrying about the other one.

3.0V ?  yeah - that might work out, but I would try not to go under 2.8v. For the wacom ICs I suspect that 4.5v could work as well (Undersampled saw a TabletPC babyboard fully working while being fed 5V by error). You could also use 4.5v and find some way of dropping the voltage (building a voltage divider with a couple of resistances or whatever?).

Again, I am still thinking that (other than probing a live PCB), trying to "match" the Wacom ICs and other parts with a TabletPC babyboard that we know the pinout might help out.  So the idea here is to find where the power, TXD,  RXD, D+, D-, pins are hooked on those wacom ICs and find that back on yours.  Also on the motherboard side, try to follow the signals (if you have the luxury to do so) and find which IC is hooked to it (and to which pin number) -- it might be a standard IC with a datasheet that might ease the pain a bit.

Oh! and did I say this was exciting?  Smiley

EDIT:  my tabletPC babyboard (10 pin connector; serial interface) has the w8001 wacom chip and a TQFP chip. Brand: NEC, number: D78F0034BGK  & 0331KP006  -- 64 pins.  This is where the serial signals are connected to. In particular, the RXD line (pin 2 of 10) is connected to pin #22 , the +3v (pin 1 of 10) to pin #24, the TXD line (pin 3 of 10) to pin #21.
« Last Edit: May 17, 2011, 05:54:52 AM by bernard » Logged
Wilcorp70
Jr. Member
**
Posts: 63


View Profile
« Reply #23 on: May 17, 2011, 05:44:36 AM »

Ive unplugged the whole screen assembly before while the machine was on, since they are all client devices its no more dangerous than unplugging a usb cable.  I would just unplug the end of the cable that attaches to the daughterboard.  Unless there is some danger to probing a live board with a multimeter other than shorting that I am unaware of I would also probably probe the ribbon directly as the traces are separated by a much greater distance and I can tape off all the traces but the ones Im testing that way I can eliminate the chance of shorting.  Maybe Im just not thinking it through hard enough (it is late right now), but do I need to find the ground first, or could I just ground it to a metal object away from the assembly?

Im glad Im not the only one who thinks this is exiting.  I was just happy when I figured out which 7 pins were used instead of having to go through 34.
Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #24 on: May 17, 2011, 06:03:00 AM »

I would not unplug it live -- it is dangerous -- and no, it is not like USB which was "meant" to be hot-plugged/unplugged!  (With USB not all wires connects/unconnects at the same time (check connector carefuly!) and circuitry is built to sustain the spikes this will cause to the signals.  It does not mean it will break, but it is still dangerous. I wouldn't do it unless there is a really good incentive to do so -- especially if your wife could be mad if the OQO broke.

Exciting it is!!  I did a bit of pin mapping (see EDIT in previous post).
Logged
Trashie
Full Member
***
Posts: 120


View Profile
« Reply #25 on: May 17, 2011, 11:26:40 PM »

One test I might try is taking the meter to the motherboard while its powered, I just have to make sure to have the digitizer plugged in when windows starts and then unplug it otherwise it wont power it up in the first place.  Then Ill try by powering the digitizer alone.
That's a chicken-egg problem :-].For the digitizer to power up, the motherboard must activate the voltage first.I'm pretty sure if you just power it up without the screen, you'll be able to read voltages.
You'll not know the GND's, but at least, if you later apply voltage to the screen (using batteries/power source) you'll be able to look for GND's just testing the other pins (you'll not be reverse-powering anything, you already know 0V in those pins is a valid value, and  only when testing a GND the circuit will actually close;until then, no current will exist on the device).
Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #26 on: May 18, 2011, 12:54:07 AM »

I got my hands on another tablet pc (14 pin connector) -- the TQFP chip is 48 pins -- number: 87F16C8A  & 7MA00 .  I will do some pin mapping later.
Logged
Pages: 1 [2]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!