Bongofish
February 24, 2018, 03:00:48 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News: You can attach photos hosted by the forum rather than using an external image hosting site, this means they will stay forever and not disappear after a year or two.
 
   Home   Help Search Login Register  
Pages: 1 2 [3] 4 5 ... 14
  Print  
Author Topic: Experimenting with the HP 2710p Tablet PC digitizer (SU-12W18A-01X)  (Read 69041 times)
bernard
Administrator
Hero Member
*****
Posts: 2588


pato mania


View Profile
« Reply #30 on: August 19, 2013, 03:55:04 AM »

Hum. Wow. What a ride! Amps at 3v is not the same at 5v. I might have given you the 5v numbers, but 1.1amp is 1100 mA which is really over the roof. No wonder something was hot. The 3.3v regulator must have been limiting (switching on and off) or plain running hot.

If you see more than  250mA disconnect immediately and I would be a bit worried over 150mA. If pulling a lot of current, this might even harm your computer if you are connected to it - typically it will protect itself by either shutting down the power or limiting the current (for this USB port or all) or doing an immediate emergency power off of the whole computer.

The cable outer shell might be "conductive" (for EMI insulation and static protection). Each bluish wire seem to have its own insulation. While powered off, test all connections with your multimeter. Test for "shorts" between seemingly unconnected wires.

Have you touched the chips to see if any was hot? Regulator and all cpus and Wacom chips?
« Last Edit: August 19, 2013, 03:57:08 AM by bernard » Logged
ThrowingChicken
Full Member
***
Posts: 112



View Profile WWW
« Reply #31 on: August 19, 2013, 04:04:23 AM »

I don't think the digitizer ever got hot, and while the Teensy may have, I've since tested it and it runs fine.  I can't test the digitizer again since I clipped the cable, so I'll have to move forward assuming the digitizer still works.  If push comes to shove I can test the set up on the second digitizer that I am saving for the final.

So looking at these wire, they look like enamel.  I was talking to a friend and he says that a lot of times enamel is used to build up capacitance, but because of that it could potentially short if it is next to the ground wire.  Does that sound right to you?

My uncle came by to look over this with me; he worked in electronics before retiring.  He took a look at the connector and he thinks he can see a metal rod connecting pin 14 to the metal part of the connector, which is connected to the conductive sleeve.  So he thinks the sleeve acts as the ground.  The inside of this sleeve is coated with enamel or something so it isn't conductive on the inside. 

Yeah I don't know.  I can try going directly to pin 14 and if that keeps resulting in high current then maybe I should try pin 11? 
Logged
bernard
Administrator
Hero Member
*****
Posts: 2588


pato mania


View Profile
« Reply #32 on: August 19, 2013, 04:10:26 AM »

Another trick of the trade: if you have a can of dust buster (compressed air) you can turn it upside down while slightly pressing the nozzle to let a few drop of the liquid get out. This will freeze instantly anything (don't put your finger!) and you can put that on a circuit to cool the chips to attempt to protect them. Of course, if you have a can of "freeze" then it is better. It is okay to pour while powered, (low voltage only).

Test with the cable unconnected to the digitizer. (I understand you clipped it but it is still good advice). This would check for shorts in the cable alone).

Kudos to your uncle.
Logged
bernard
Administrator
Hero Member
*****
Posts: 2588


pato mania


View Profile
« Reply #33 on: August 19, 2013, 04:19:41 AM »

Also start by connecting the ground and power alone. Test at every step in any way you can. You are on a mine field.
« Last Edit: August 19, 2013, 04:26:13 AM by bernard » Logged
bernard
Administrator
Hero Member
*****
Posts: 2588


pato mania


View Profile
« Reply #34 on: August 19, 2013, 05:04:02 AM »

Enamel? No clue there. These custom cables are built to be inexpensive. As always for mass production, they use weird (but proven) tricks to save money and time while ensuring maximal EMI protection for CE/ULC conformance. My guess is that the blue thing is just an ultra-thin insulator. The 'cable' looks like a sandwich assembly of some sort of transparent plastic insulator where the wires run inside cut to shape overlaid with an electrically conductive/sticky sheet cut slightly bigger to connect to the connector case metal portion.
Logged
ThrowingChicken
Full Member
***
Posts: 112



View Profile WWW
« Reply #35 on: August 19, 2013, 05:39:23 AM »

I'll keep working at it until it's either dead or working. 

Assuming I start getting reasonable current levels, what do I need to load onto the Teensy to test the conversion? 
Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #36 on: August 19, 2013, 12:26:05 PM »

Hey there ThrowingChicken, any chance of getting some photos of how you have it wired so that we can see if you have made a mistake somewhere?

Also I've been looking at the digitizer circuit from your photos but there's bits i can't see properly, any chance of getting a close shot from above and a close shot of the underside (as some of the traces go through vias).

I have to second what Bernard said about exceeding 250mA, there's no way that the teensy and circuit should use more than that, 1100mA is far to high and extremely likely to cause permanent damage.

Also, I'm a bit confused about what you are doing here, am i right in thinking that you have a screen with a digitizer built in that you wish to put in a case and use as a diy Cintiq? I presume if this is the case that you want to power the digitizer from the USB of your laptop, but how will you power the LCD? Or is this going to be a laptop Cintiq all in one?

Wait, you mentioned getting an lvds controller so I'm guessing my first thought is right.

So my next question follows that, if you are using the cable with the thin blue wires depicted at the beginning of this thread to power the screen, did you cut the wires which branch off to the digitizer before powering up the teensy? My thought being that you are shorting the teensy through the motherboard of the laptop which is why the current is high and the wire gets hot.
Logged
ThrowingChicken
Full Member
***
Posts: 112



View Profile WWW
« Reply #37 on: August 19, 2013, 06:53:18 PM »

Hey there ThrowingChicken, any chance of getting some photos of how you have it wired so that we can see if you have made a mistake somewhere?

Sure thing!

Teensy Side:



Digitizer Side:

Ground line coming from the metal housing.


Red = Pin 13 = VCC.  Two unconnected lines, pins 12 & 11, taped at the end.  White = Pins 10 & 9 = TX/RX.  Green = Ground.  



I had my multimeter in line on the VCC line when  I was getting such high current ratings.  


Quote
Also I've been looking at the digitizer circuit from your photos but there's bits i can't see properly, any chance of getting a close shot from above and a close shot of the underside (as some of the traces go through vias).

Sure.  I can't get the bottom side, though, as it is glued down to the backing.  I am afraid I will break it trying to pry it up.  If you think it will be okay separating it I will try to give it a go but it is on there pretty well.




Quote
I have to second what Bernard said about exceeding 250mA, there's no way that the teensy and circuit should use more than that, 1100mA is far to high and extremely likely to cause permanent damage.
Because we didn't have a successful test when it drew that much current, I am inclined to believe 1100mA is too high too.  I think a USB is only supposed to pull under 500, it just doesn't add up to me.  There must have been a short somewhere, right?  

Quote
Also, I'm a bit confused about what you are doing here, am i right in thinking that you have a screen with a digitizer built in that you wish to put in a case and use as a diy Cintiq? I presume if this is the case that you want to power the digitizer from the USB of your laptop, but how will you power the LCD? Or is this going to be a laptop Cintiq all in one?

Wait, you mentioned getting an lvds controller so I'm guessing my first thought is right.
DIY Cintiq, yes.  Assuming I can get the digitizer to work through USB, I will be purchasing the LVDS controller and packaging it all in a case so I can use it on my desktop.  

Quote
So my next question follows that, if you are using the cable with the thin blue wires depicted at the beginning of this thread to power the screen, did you cut the wires which branch off to the digitizer before powering up the teensy? My thought being that you are shorting the teensy through the motherboard of the laptop which is why the current is high and the wire gets hot.
The wires were cut, as shown above, but I have never run the Teensy and the laptop at the same time.  None of my tests with the Teensy involve the LCD cables.  I have tested the digitizer & LCD on my laptop using an extra cable.  In the final test my uncle and I wanted to figure out the current that the digitizer draws natively, so we opened the wire and put the multimeter on the line going to pin 13, but I believe we caused a short somehow since it ran high again, caused tape to melt, and messed up my laptop.  But the Teensy was not involved in that test.  

I hope this helps.  If you have any other questions or suggestions please let me know.  
« Last Edit: August 19, 2013, 06:55:57 PM by ThrowingChicken » Logged
bernard
Administrator
Hero Member
*****
Posts: 2588


pato mania


View Profile
« Reply #38 on: August 20, 2013, 05:50:13 AM »

What is your next step now?

I'll repeat that you need to do baby steps or test the smallest, most separated parts that you can. Disconnect as much wires as possible.

Is your meter functionning correctly when testing DCmAmps? It is not introducing a short?

I see that metallic-looking foil under the digitizer connector -- is this conductive?


Logged
ThrowingChicken
Full Member
***
Posts: 112



View Profile WWW
« Reply #39 on: August 20, 2013, 06:49:19 AM »

What is your next step now?

I'll repeat that you need to do baby steps or test the smallest, most separated parts that you can. Disconnect as much wires as possible.
Pretty much this.  I might even try seeing what happens when the Teensy is hooked to the cable, but the cable isn't plugged in.  Pending on the results of that...

 - Shrink tubing over all the wires, entire length of the wires.  

 - Replacing the wires, soldering directly to the pins.

*shrug*

Would there be an issue if I soldered a wire directly to the DGND pad?    

Quote
Is your meter functionning correctly when testing DCmAmps? It is not introducing a short?
I get high readings with two different meters.  Plus the meter wasn't hooked up when the wire got so hot it melted electric tape.  

Quote
I see that metallic-looking foil under the digitizer connector -- is this conductive?
No, though I suppose there could be a coating over it.  

I'll probably have a chance to play around with it tomorrow. If I can get reasonable mAs, which Waxbee hex file do I need to add to the Teensy to test function?  The  USB virtual serial port 2011?
« Last Edit: August 20, 2013, 07:22:59 AM by ThrowingChicken » Logged
bernard
Administrator
Hero Member
*****
Posts: 2588


pato mania


View Profile
« Reply #40 on: August 20, 2013, 02:48:41 PM »

Teensy: yeah, use the virtual serial port loaded with the Teensy Loader. Follow host setup http://www.pjrc.com/teensy/usb_serial.html and get a "terminal emulator". Realterm, putty in serial mode or teraterm. None is perfect.

No  parity, 8 bits, 1 stop bit. Half duplex or echo on to see what you are typing. Possible baud rates: 38400 or 19200. Note that this virtual serial port only supports 3 speeds (9600 being the other one). The virtual serial port running on the Teensy will "respond" a number upon baud rate change to confirm it understood the baud rate.

Soldering on the pad is no problem. Watch out not to put too much heat - it can lift pads and traces off the PCB. If out of pads you can try to "scratch" a connecting trace to expose the metal to solder on.

Do you have a looking glass? You could inspect the board for unwanted "jumpers" or crossover connections.
Logged
ThrowingChicken
Full Member
***
Posts: 112



View Profile WWW
« Reply #41 on: August 20, 2013, 07:27:50 PM »

Quick update:  When bypass the connector and go straight for the pins I'm getting around 50mA.  Sounds reasonable. 
Logged
ThrowingChicken
Full Member
***
Posts: 112



View Profile WWW
« Reply #42 on: August 20, 2013, 10:14:53 PM »

I am soldered directly to the pins, steady around 50mA.  When I put the pin close or on the digitizer it jumps to around 130mA and holds there for a second before going back to around 50mA.

I might be totally lost about the terminal stuff.  Tera Term seems to be the most basic.  I've figured out how to set up all of your options for that except for half duplex or echo (not finding the option).  But I will set up what I can and hit OK, the baud number I entered will show up on the display.  That's about as far as I have gotten with it.  Am I supposed to send it files, or tell it to send me something? 
Logged
ThrowingChicken
Full Member
***
Posts: 112



View Profile WWW
« Reply #43 on: August 21, 2013, 01:26:34 AM »

I was able to get the board up.  When I looked down the small gap with a flashlight I could see it was just tape holding it down.  Pretty strong stuff, but I took it slowly and used a business card to slice through the tape fibers until it was apart. 



Here is the shot of the solder job onto the pins.  A little rough, but I checked and continuity is fine. 


Based on how the amps are in a stable place and changing when the pen comes into contact I feel like we are on the right track, but as mentioned before I'm having trouble making sense of the virtual terminal stuff. 
Logged
bernard
Administrator
Hero Member
*****
Posts: 2588


pato mania


View Profile
« Reply #44 on: August 21, 2013, 01:34:06 AM »

Send the command * .  (star). I mean just press star and potentially return. It should output a single line of text with important info. This is the ISDV4 protocol.
Logged
Pages: 1 2 [3] 4 5 ... 14
  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!