Bongofish
August 18, 2019, 05:10:28 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 ... 4 5 [6] 7 8 ... 10
  Print  
Author Topic: Playing with the tx2000 Tablet-PC digitizer (SU5R-12W12AU-01X)  (Read 82034 times)
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #75 on: April 13, 2011, 07:22:45 AM »

3.3v->5v AVR: Yes. Like in my previous post about voltage levels, the AVR chip (or arduino) can read very low voltage levels as "logic input".  Meaning that a 3.3v TTL/CMOS input signal will work even if the AVR is powered at 5V. I am pretty sure that going in the other direction is not as simple: I think you should not feed higher voltage signals than the VCC.
Logged
Trashie
Full Member
***
Posts: 120


View Profile
« Reply #76 on: April 13, 2011, 10:18:02 AM »

Ok, well, after setting all possible speeds on the serial interface of my Arduino and all, and getting nothing of it, i observed closely the way pin 6 was acting...
And looks like it's something way simpler.
It's acting like a sensor.If the pencil is near or touching the digitizer surface, i see 3.15V in the multimeter.That doesnt change.Even if i move the pencil, as long as it's near or touching the surface, there's 3.15 V there.
If i take the pencil further from the surface, there's a point where the V begins to go down, until it reaches 0 if i move a small distance away.
So, looks like pin6 is some sort of hardware flag to signal the PC about when the pencil is in the active area of the digitizer.

I was also able to insert wires for the 4 "unused" pins in the connector (digitizer-side).All i can say of those, is that pin 10 is at 3.0V.Confirmed by measuring the pin directly (behind the connector).
Grounding it, grounding any of the other 3, or setting them to +3V, showed no difference.Also, this behaviour didnt change after setting pin5 (the other unknown pin) to +3V , or grounding pin 12 (the "reset" pin), or moving the pencil over the surface.
If the "reset" pin wasnt grounded, the USB device was working in Windows, so none of those pins looked to disable it (except, again, pin 12).
Conclusions: Those "unused" pins arent completely disabled (pin 10 has 3V in it), but how do they work...no clue.
« Last Edit: April 13, 2011, 12:13:08 PM by Trashie » Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #77 on: April 13, 2011, 01:53:27 PM »

The 4 "unused" pins are supposed to be the serial signals! (check the wiki) -- According to what we know the behavior should be the following: the DTR/RTS (pins 7 & 8 ) are serial handshakes input pins -- nothing will come out of pin 9 (RXD) (output pin) if the RTS and DTR are not grounded.  some of them might have pullups to force them in the high state.  Pin 10 is a TXD (input pin) this is where you send the commands like "*" to receive the max values, "0" to stop sending coordinates and "1" to start sending coordinates again.

The baud rate is most probably 19200 bauds.
« Last Edit: April 13, 2011, 01:57:07 PM by bernard » Logged
Trashie
Full Member
***
Posts: 120


View Profile
« Reply #78 on: April 13, 2011, 03:19:48 PM »

I checked the wiki...That's why i tried to see what there was on those pins..
And the results are those... 3 V in pin 10, and 0 in pin 9,8 and 7.
Grounding 8 and 7, or setting them both to 3V, or any combination of those, only raised (in all combinations) voltage in pin 9 to 0.04V. In all these tests, pin 10 was at 3V.

Later i'll try grounding pin10 too...In any case, makes sense for pin10 to be Serial TX, and stay at 3V??
Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #79 on: April 13, 2011, 05:33:14 PM »

to add to the complexity, maybe there is a pin that toggle the serial/usb mode as well   -- have you tried unconnecting everything else?
Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #80 on: April 13, 2011, 05:41:05 PM »

TXD output: yes, I think it makes good sense. I am trying to remember if the serial port is always "high" by default. I think yes. That's because when idle, I believe it is in the "MARK" state. In TTL-Serial: MARK is VCC (high), SPACE is GND (low).

EDIT: Sorry, I am mixing myself:  the TXD is the input pin, right?  Well, there might be a pull-up in there, but not sure. If really this is an input pin, "reading" an input pin can be a bit random if it is floating I think. It could act weirdly.

« Last Edit: April 13, 2011, 05:47:53 PM by bernard » Logged
UnderSampled
Jr. Member
**
Posts: 54



View Profile
« Reply #81 on: April 13, 2011, 09:27:03 PM »

TxD means "Transmitted Data", so is actually an output, but because those names were from the reference of the computer (it was in the laptop's service manual), it's the output from the computer to the digitizer. That means that it's really the input (RxD, or "Received Data") serial port from the digitizer's perspective. Perhaps we should reverse the names so that everything is from the reference of the digitizer.
Logged

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


pato mania


View Profile
« Reply #82 on: April 14, 2011, 12:39:05 AM »

yeah -- That is why we have that "direction" information which is very important to understand what the signal is doing really. 

This is btw, the same problem with DTR and RTS -- the signal name have a meaning and also have a counterparts (like CTS / DCD).  In RS-232, the signal names are _supposed_ to be from the side of the DTE (computer) and not the DCE (device). http://en.wikipedia.org/wiki/RS-232#Pinouts

This is also why the sparkfun FTDI breakout board puts a "I" and a "O" next to the RX and TX pin names, I would have been very confused if they didn't do that. Clever them. Is the FTDI a considered the computer or the device?   

So according to wikipedia, TXD is  DTE (computer) -> DCE (device)  and RXD is:  DTE (computer) <- DCE (device). So the wiki is "right". 

I think there is no way to not make this non-confusing: it all depends from which point of view you are thinking about it.

Logged
UnderSampled
Jr. Member
**
Posts: 54



View Profile
« Reply #83 on: April 14, 2011, 12:53:34 AM »

The FTDI is a virtual com port, so it's basically just a serial port on the computer pushed through usb.
Logged

Trashie
Full Member
***
Posts: 120


View Profile
« Reply #84 on: April 14, 2011, 10:45:52 AM »

Nothing found on the 4 unconnected pins.Also, the cable connections in these spots are very weak, and it's difficult to keep the cables in place (they keep disconnecting).So, i'm not looking those pins any more.In the case there's a serial interface there, i'm not really sure if that'd be of any (general) use, as the USB connection is way simpler.I'll try now to gather pieces to build my first cintiq!
Logged
Trashie
Full Member
***
Posts: 120


View Profile
« Reply #85 on: April 17, 2011, 09:53:31 PM »

As i'd really like to use the serial interface of this board, i've spent a few hours more looking for it.And still, no results.
Summing up what i've seen in this card.Pins mean "pins in the babyboard" connector:
Pins 1-2 Must be set to ground.
Pin 3 USB D-
Pin 4 USB D+
Pin 5 Unknown
Pin 6 The board sets this pin to +3.3V if the digitizer is detecting the pencil.
Pin 7-8-9-10 Those are the supposed pins for serial output.Not originally connected in my board.
Of those, pin 7,8 and 9 are at 0V.Pin 10 is at 3V.
Pin 11Must be set to 3.3V.If this pin is unconnected, the USB device is found, but no pen movement is registered.This is the reason i think this input is a separate power for the digitizer surface,or something like that.
Pin 12Set by the board at 3.3V.If this pin is grounded, the USB device disconnects (you hear the USB device disconnected sound in windows).Looks like a reset pin
Pin 13Must be set to 3.3V.If this pin is unconnected,the USB device is not found
Pin 14Must be set to ground

Thinking there is an active serial interface there, assumes the following:
a) A babyboard with 14 pins have both interfaces (USB and Serial)
b) Both interfaces are active at the same time, and, if not, the "not-default" interface may be enabled using a certain pin.
Along with that, some more assumptions done are:
c) All boards with 14 pins, have the same signals in the same pins.


About b), and looking at the above info, the possible pins that may be the ones enabling the serial interface, would be 5 and,maybe,12.To test this interface, i used an Arduino program opening the port, sending an '1', waiting a bit,then reading, all in the main loop (so it tried to open the port each time).
No combination of those pins (along with other combinations with the Vin in pins 11 and 13) activated the serial port.I also tried underpowering the Arduino to 3.3V, so the voltage on its pins were even lower than what i was measuring in pin 10, so there was no problem with babyboard voltages being too low for the Arduino to detect.
I've tested both with 19200 and 38400.

It's difficult to measure the serial pins because they're not used by the connector.I inserted cables in the connector holes, and was able to test with this setup, but it's very weak and, eventually, the connector broke in those pins, so i had to test by "touching" with cables the pins on the back side of the connector on the babyboard.

So, after all that, i think the "penabled developer kit" shown in wacom-components.com may be the way those boards are set-up, maybe with different firmwares, depending on each OEM.What does that "developer kit", i dont know, but, for sure, it means each OEM has certain control, or can modify things in the board.That goes against the idea about all 14-pin boards have the same signals in their pins.Maybe, each OEM can set-up the digitizer so there are pins that may change meaning, or be enabled-disabled.

Obviously, i can also be doing something wrong.But to verify this, or, at least, having hopes that both interfaces are "visible" at the same time, we would need that somebody having a serial-only 14-pins digitizer, tries to activate the USB interface.The connector most probably doesnt have cables for those pins, and there are 2 pins to connect (D- and D+).But to test just that, holding 2 cables against those pins in the back of the connector on the babyboard (ie, where the pins are soldered to the babyboard),is more than enough.

Any volunteer?
Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #86 on: April 18, 2011, 07:16:10 PM »

Nice summary!  I will update the link of the wiki to point to this precise post. 

farinasa is selling me one of his 14 pin board. I will be in a position to do some tests to try to elucidate the remaining pieces of mystery (with the oscilloscope hooked on).

I won't have a wire/connector -- so I hope I will be able to solder jumpers either on the connector pads or somewhere else on the board.  Since the baby board is literally "glued" to the sensor and there are very fragile super-high density connections all around, this might be tricky. Would be nice to find the appropriate connector and supporting wire(s) so I avoid messing up with the baby board when possible.

One way to infer more info if something is a reset/enabled/power pin or whatever, is to hook a multimeter in series on the power source reading the mAmps the board is drawing. It is a good way to "see" in which "mode" the board is in.

Pin 12:  Reset vs. Enable
Your "potential" reset pin is the same pin than what looked like an enable (active high) or *reset (active low) -- with a pullup. I would think an "Enable" but it could be reset as well since the "name" of the pin (on the laptop-side connector) was:  DBGEN-P3N + PLTRS1-E3N  -- DeBuG-Enable. This looks like PLT-"ReSet"1.  The difference between a reset and an enable would be that you do not keep one device in reset-state forever (I think?). Also in reset state, the power is not necessarily reduced to a minimum.  I believe two devices shared the same serial port -- some sort of "debug port".  The same "enable" signal was shared by two devices. When high, it would activate one and deactivate the other, the reverse when false. The "P" and "E" I think had something to do with the "Active High or Positive" or "Active Low - or Negative" (deduced by looking at other pins with similar names). This is my theory at least.
« Last Edit: April 18, 2011, 07:35:28 PM by bernard » Logged
Trashie
Full Member
***
Posts: 120


View Profile
« Reply #87 on: April 18, 2011, 09:41:38 PM »

Bernard,
If you're thinking in ordering a connector cable, check this one out:
http://www.pchub.com/uph/laptop/72-56462-16838/HP-TouchSmart-tm2-1000-Series-Various-Item.html?
That one looks like it has the full set of cables, or, at least, that can be asked to the pchub people.

About soldering cables there...I've thought about it, but i really dont know how to solder, and less,in such a small surface, and with so little space between the pins.

About the serial connector..I may have news..But need more testing...

About the debug pin : Earlier in this thread i posted that you only connect pin 12 to a LED, and then set any other pin to 3.3V, the LED glows.This works even for GNDs.Can this be considered "debug"?
« Last Edit: April 18, 2011, 09:42:32 PM by bernard » Logged
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #88 on: April 18, 2011, 09:49:11 PM »

If the picture is right, I counted 14 "things" -- and it appears to be a "straightforward" cable. Might not be easy to solder directly to that cable, but I prefer to destroy a cable than the baby board. Smiley 

Eager to hear about your "serial news".   

LED: hum... This led thingy does not tell me much. There are many ways to make a led glow -- including a lot of invalid ways of driving power -- like connecting the 3.3v to the ground plane (ouch!) and connecting the led to ground pin and the other side of the led to the USB ground! That would essentially power the led from the 3.3v source using the "board ground plane" as a plain "wire". Smiley  I am sure there many (much weirder) combinations like these that we could come up with.  Is this what they call "circuit bending"?
« Last Edit: April 18, 2011, 09:56:30 PM by bernard » Logged
UnderSampled
Jr. Member
**
Posts: 54



View Profile
« Reply #89 on: April 19, 2011, 03:57:44 AM »

@Bernard: What was wrong with this connector? It's looks exactly like the one I have, is the correct pitch, and has the correct number of positions.


for comparison (the connector I have):
http://forum.bongofish.co.uk/index.php?topic=1659.msg11731#msg11731
Logged

Pages: 1 ... 4 5 [6] 7 8 ... 10
  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!