Bongofish
August 19, 2019, 08:37:12 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] 3
  Print  
Author Topic: My "Cintia", a work in progress  (Read 16356 times)
Drewid
Administrator
Hero Member
*****
Posts: 2290


wielder of electric crayons


View Profile
« Reply #15 on: August 19, 2011, 01:07:50 PM »

Depending on how your board is programmed you should be able to try out a lower resolution.  Everything does 640 x 480.
Logged

Games work site:  Drewnorthcott.co.uk
Personal site: Bongofish.co.uk
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #16 on: August 21, 2011, 05:06:50 AM »

hello all!  yes, I am a bit more into electronics, but really, I am learning the ins and outs of these beast at the same time as you guys.

Try unplugging the DVI/VGA off the PC and try to summon the OSD menu (that's On Screen Display menu).  You said the led turns green. Does it turn to another color as well?  (like when no VGA/DVI is connected or when you press the buttons?)

Are you sure your board was programmed for your LCD specs? Or: how did you know they should work together in the first place?

On the LCD Panel itself, I am not sure if your connector can be put in reverse -- is it well seated?  Is your cable broken maybe? 

Do you see any sort of glitches on the screen at any moment?  (if yes, describe the glitches and when it occurs).

Logged
marceloaguiar
Jr. Member
**
Posts: 33



View Profile WWW
« Reply #17 on: August 21, 2011, 02:24:14 PM »

Hey!

You're right, you can't invert the connector on the panel.
As for being sure about compatibility - I bought the controller from tabletmod (this one) and bought the screen for which they state it's programmed for (here). To be honest, there's nothing on the panel that indicates that this is a boehydis HV150UX1-101 screen.

I connected everything (without connecting it to the pc). Pressed "power" and the led light lits red. Backlight turns on, too. I press "menu" and nothing happens - no osd menu.

I tried switching to the panel cable that came with the first controller and ...... "voilá", that microchip started burning again. The problem was the cable. Now I'm not sure I still have a working controller.

Well, even after this, if I connect the pc to the controller, the led light turns green.
I extended the desktop to the panel and the light remains green. Windows suggests 1600x1200 as recommended resolution, so it must be detecting the screen, even after the 1 second burning incident.
However, I still get no picture and no osd menu.

I'm getting to the conclusion that my screen is either faulty or not a HV150UX1-101. What do you guys think?

edit: I forgot to tell you that the screen flickers when I press "power", but it seems that is just the backlight turning on and off.
« Last Edit: August 21, 2011, 02:27:25 PM by marceloaguiar » Logged

I love my tablet... that's why I'm going to tear it apart.
marceloaguiar
Jr. Member
**
Posts: 33



View Profile WWW
« Reply #18 on: January 06, 2012, 01:02:04 PM »

Hey guys,
After a long period of silence, here I am again.... with the same problem.

I returned the screen - as it turned out, it was not an actual boe-hydis. Its serial number, when googled, only returned IBM screens.
I received a new screen today. This time it has a sticker on the back that reads "boe"...

Anyway, I have two different controllers - one I ordered from tabletmod and the other from Luly.

The screen is deteced by the pc, but....
When the latter is connected to the screen, nothing happens. I can't even summon the on-screen menu.
When the former is connected, I get a thousand multi-coloured vertical stripes that are swept vertically with a difference in brightness. Eventually that fades away and the screen turns black.
Here's a pic:

* screen.jpg (22.15 KB. 800x600 - viewed 423 times.)


guys... what's happening?  Huh

Logged

I love my tablet... that's why I'm going to tear it apart.
bernard
Administrator
Hero Member
*****
Posts: 2590


pato mania


View Profile
« Reply #19 on: January 06, 2012, 04:39:18 PM »

The programming in the controller(s), I assume, does not fit with your screen specifications.

What is the model number of the screen?  You said it mentions BOE --- but no model number?  How can you know it is the same -- and that the specs are the same too?

Logged
marceloaguiar
Jr. Member
**
Posts: 33



View Profile WWW
« Reply #20 on: January 06, 2012, 05:50:21 PM »

Well, the screen model number is HV150UX1-101, that's on the back, too.

I wrote to Luly describing the problem of her controller (the one that does not do anything, all black) and she replied that the board she sent was not 3.3V input, so it's not compatible. She is sending another that hopefully will work.

The one from tabletmod was for this specific screen - that was stated in their website...  Undecided
Logged

I love my tablet... that's why I'm going to tear it apart.
marceloaguiar
Jr. Member
**
Posts: 33



View Profile WWW
« Reply #21 on: May 23, 2014, 11:52:44 PM »

So it's been a while since I've updated this topic......

I've finished it. It was not a super-advanced mod - it was basically stacking a screen on top of a tablet and build a casing around it but it turned out pretty nice! So here it is:

About the screen problems, Luly sent me another controller and everything turned out great.

Once the screen issues were solved, I started doing a casing for the cintiq. My idea was to build a casing in plywood with a sloped base so that it looked smaller when viewed from the side.
I did a 3D model of the thing I had in mind and printed out the several sheets to be cut.
The 3D model:

* 01.jpg (84.94 KB. 800x441 - viewed 349 times.)

* 02.jpg (81.58 KB. 800x441 - viewed 341 times.)


Cutting the pieces....

* IMG_20120326_111931 (Copy).jpg (82.87 KB. 1300x975 - viewed 334 times.)

* IMG_20120326_125534 (Copy).jpg (137.24 KB. 1300x975 - viewed 317 times.)

* IMG_20120326_105525 (Copy).jpg (110.33 KB. 1300x975 - viewed 320 times.)


And painted all of them white:

* IMG_20120402_105705 (Copy).jpg (128.64 KB. 1300x975 - viewed 346 times.)


For the monitor buttons I did the following: I got a plastic rectangular sheet and drilled the holes in the same position of the button board. Then I did a half-cut in the plywood so that there would be a recess for the plastic sheet. I glued it and then filled the gap with plaster, sanded it and painted it again. The result was an almost seamless surface.

* IMG_20120414_160212 (Copy).jpg (155.92 KB. 1300x975 - viewed 336 times.)

* IMG_20120414_160222 (Copy).jpg (176.04 KB. 1300x975 - viewed 328 times.)

* IMG_20120414_160234 (Copy).jpg (155.63 KB. 1300x975 - viewed 322 times.)


Frame and acrylic sheet:
I bought wooden strips, 4mm thick to build the frame. I build some supports for the hinges (the top would open up if I needed to do something in the "Cintia's" interior... that didn't sound good at all, did it?). I painted the acrylic sheet black on the inside so that it would remain glossy on the outside and glued it to the frame.

* IMG_20120409_182344 (Copy).jpg (162.13 KB. 1300x975 - viewed 329 times.)

* IMG_20120414_160156 (Copy).jpg (127.27 KB. 1300x975 - viewed 313 times.)

* IMG_20120414_164945 (Copy).jpg (135.43 KB. 1300x975 - viewed 338 times.)

* IMG_20120418_085752 (Copy).jpg (116.8 KB. 1300x975 - viewed 343 times.)


next post, the base....


Logged

I love my tablet... that's why I'm going to tear it apart.
DaBotz
Sr. Member
****
Posts: 243



View Profile WWW
« Reply #22 on: May 24, 2014, 03:48:39 PM »

"(the top would open up if I needed to do something in the "Cintia's" interior... that didn't sound good at all, did it?)"

Well, you may also come with the idea of doing some modification, after a while, on the machine when you start working with it and realize you could do with something different...

Nice idea the sloped case (I'm, too, a plywood fan).
Logged

The most incredible artist of... Barbanza?
marceloaguiar
Jr. Member
**
Posts: 33



View Profile WWW
« Reply #23 on: June 10, 2014, 09:20:22 PM »

Thanks, DaBotz!

For the base I started glueing all the parts together and in the process it ocurred to me that it would be nice to have board and power outlet accessible, i.e. without the need to open the whole thing up just to change a cable or something.
So when it was all glued together I sketched on the plywood a cutout for cable access.

* IMG_20120424_130948 (Copy).jpg (151.31 KB. 1300x975 - viewed 317 times.)

(the "beams" that go from side to side were for rigidity, but now that I look at them, it feels like overkill)

Then I built little walls along the cutout to accomodate the power adapter and the pcb.

* IMG_20120601_125141 (Copy).jpg (199 KB. 1300x975 - viewed 341 times.)


* IMG_20120601_125203 (Copy).jpg (104.15 KB. 1300x975 - viewed 319 times.)


Here we can see all the components already in place. The power adapter is fixed by a rigid plastic strip so that it stays put when I push the plug in or pull it out.

* IMG_20120829_102716 (Copy).jpg (201.33 KB. 1300x975 - viewed 343 times.)


On the outside it looks like this:

* IMG_20120831_095644 (Copy).jpg (104.95 KB. 1300x975 - viewed 336 times.)

The recess in the base gives me space to organize the cables and choose to which side they go.
I used plaster to make nice round edges. Then I sanded the whole thing and painted it!

And here are some photos of the finished Cintia!

* IMG_20120831_100447 (Copy).jpg (79.14 KB. 1300x975 - viewed 316 times.)

* IMG_20120831_100514 (Copy).jpg (81.04 KB. 1300x975 - viewed 334 times.)

* IMG_20120831_100619 (Copy).jpg (100.62 KB. 1300x975 - viewed 345 times.)

* IMG_20120916_151653 (Copy).jpg (121.52 KB. 1300x975 - viewed 353 times.)

* IMG_20120916_151835 (Copy).jpg (112.18 KB. 1300x975 - viewed 362 times.)


As you can see, I left the intuos buttons accessible, they're handy even though they are few!
When I want to draw almost horizontal, I just place it on the table and have a nice angle. When I use it as a monitor, I place it on the stand and it's almost vertical.

It works great. Jitter is not always an issue, although sometimes it becomes a little annoying. It comes and goes, I don't really know why.


« Last Edit: June 10, 2014, 09:23:12 PM by marceloaguiar » Logged

I love my tablet... that's why I'm going to tear it apart.
DaBotz
Sr. Member
****
Posts: 243



View Profile WWW
« Reply #24 on: July 04, 2014, 11:38:22 PM »

About the "vanishing" jitter, I had much of the same issue till my power brick fried.

When it arrived the new one, I used the ferrite toroidal nucleus from its transformer (and from a previous brick that fried) to build a common mode choker .

One on the dc power in and one on the usb cord, they cured completely both jitter and the false clicks issues of my Cabinetiq.

And made the bloody mess that the UBiQ was - at th -first tests -  a functional tool.

Common mode choker is a tecnical way to say: get the broken transformer, open it, get rid of its coils, and coil the usb cable (if you have two, the power cable too) as many times you can around its ferrite nucleus, then close the transformer back together, with the cable in the place of the original coils.
 
As you can do it on the cable going in, you don't even need to open The Cintia to see how it goes.

(P.S. I hope I didn't already write you this)
Logged

The most incredible artist of... Barbanza?
marceloaguiar
Jr. Member
**
Posts: 33



View Profile WWW
« Reply #25 on: July 11, 2014, 06:47:38 PM »

Thanks for you suggestions DaBotz.

Actually since I switched to my new laptop jitter and random click problem has become really severe.

My new laptop has no VGA output, only HDMI, DisplayPort and mini DisplayPort. So I got an HDMI-DVI cable to connect my Cintia and now I get 5mm jitter (yeah, it looks like it has parkinson's) and frequent random clicks.
This happens more on the edges (3cm) of the screen - on the center it's ok.

I was trying to learn about the common mode choker you mentioned and decided to open up an old adapter I had here:


* IMG_20140711_181944.jpg (217.34 KB. 1300x813 - viewed 335 times.)


I noticed the choke is super tiny compared to the usb cable. How did you manage to wind it around the core? Or am I misunderstanding something?
Do you have photos of that part of the UBiQ?
Does this photo show the usb cable around the core? How many times did you wind?

Logged

I love my tablet... that's why I'm going to tear it apart.
DaBotz
Sr. Member
****
Posts: 243



View Profile WWW
« Reply #26 on: July 17, 2014, 09:38:58 PM »

Common mode choke is a pretty generic term...

They go from small IC circuits that ferret out the common mode noise by simply ignoring it and reporting only the differences between signal in and out, to the ferrite torus around the termination of many cables, to the one you see in your adapter. Actually, to do "my trick", I would destroy the adapter main transformer. That's the "right" size.

If you take a ferrite bar out of an old AM radio, coil the USB cable around it, the assembly already chokes common mode currents (and/or, maybe, some signal) on the internal data and the power pairs.

At a base level, it's as simple as that: two identical coils, one for the signal in, one for its return, possibly on a ferromagnetic nucleus (ferrite ones hold better higher frequencies and are easier to handle than laminated iron) for improved efficiency. The induced magnetic fields from each coil cancel each other on balanced signals, but sums up on common mode noise.

By coiling the full USB cables, you are actually making 4 parallel coils (unfortunately, only ALMOST identical, which may become an issue if you need to add a lot of coils because you realize you have a severe common mode noise trying to pass through your system).

In my case, lacking '70s AM radios, I scavenged transformer nuclei out of burnt power adapters (the firsts two, on the C..Q, from burnt power bricks I had around, for the UBiQ  I asked to a place where they fix electric appliances if they had some PC power units to throw away... actually, they had some burnt transformers too, but were too big to place in the 3 cm tall space inside the UBiQ), opened them up, got rid of the original coil and coiled  my USB cable in the place left by it.

The result was something like this:



or like this


* 00-[1024px]-DSC01345.JPG (207.46 KB. 1572x1132 - viewed 321 times.)


One of my two "Q"s only needed two of the first type to behave (in all, 4 coils - the end result is an almost unnoticeable jitter, though 45º lines wobbles a bit), on the other I use five bigger ones (in all, around 20 coils... yes, I should take the High Route and buy two or more  "proper" common chokes, like the one in your adapter, and replace the whole unholy mess - it could be a good idea... but it would also mean to make at least 8 good solders, and I don't really have anything to debug circuits).

USB and power cables seldom use fancy things like twisted pairs or coaxials, so it may happen that one of the cables of the pair ends up "cnostantly" inside the other, which would mean for one of the two lines to have a greater inductance.

In this case, the choke ceases to be "transparent" - the induced magnetic fields doesn't cancel each other - for balanced signals, and dampens them too.

To the extent that USB signals may be "killed".

Just to stay on the safe side, place an un-powered usb hub (if powered, the power line to the hub will need its own choke!) between the Wacom and the "doctored cable".

Before connecting the wacom, if when connecting the hub you see that Windows protests because he can't recognize it, it probably means that the USB signal is degraded too  much to work.

Also, hopefully, if there are odd transients (everything is possible, though the inductances associated to so few coils must not be very high), better for a 5$ hub to take the brunt of it than our adored Wacoms.

 ____________________________________________________________________________

At the very crude level I used, mine is a cheap-dirt trick.  

With the exception of the hub - that, given my propensity for short-cuts dedicated keyboards stacked vertically, would be part of my builds anyway - I, literally, used junk to do it.

So if it does not work, you lose just some of your time, and not much of it.

In my case, it was just the time to open up the nuclei - the shop where I got them is on the side of my favoured bakery, so the acquisition time for the material was negligible.

____________________________________________________________________________

As I said, in my two cases, it did work, but it is not a given, and there is some arguing about the fact that it should work at all (I may be self-deluding myself), or how and why it does when it works, between me and Aerendraca.

However, while "UBiQ" is still a bit jittery, ( +- 2 px on the X axis, +-  px on 0.5px on the y axis;  it may also be that I need to retune the pen RF system; Intuos2s are a bit temperamental on that side, that I have hitten on the hard rock of the EMI generated jitter or simply that I have no more ferrite nuclei - this latter is true), for "CabinetiQ" it is down to some fraction of pixel - I can see it if I work at zooms below 25%, and then I zoom in back to 100%.

Neither machine shows any sign of false clicks, whereas on Cabinetiq - my first - I had to switch the second button off every other week, before "the cure".

To be honest, right before "the cure", I was being strongly seduced by the dark powers of YiYnova and its MSP19U - when it was in his worse days (the severity of jitters and false clicks was on a "daily" basis, and intensive Hard disks operations made it positively crazy!!! Yes, if I copied a bunch of movies to master a dvd, or my antivirus did a scan, I couldn't draw), Cabinetiq drove me mad.

After the cure, the 3.6" (19" vs. CabinetiQ's 15.4" ) more of crappy 1440x900 screen with washout colours (same specs of the CabinetiQ, these...) in the YiYnova didn't justify its 700 Euros investment anymore - battery pen pressure curve apart - so I built the UBiQ (it's a 22" with a good screen - TN, but good; I like drawing big.) with half of that.  
______________________________________________

A good thing is also that you can test this trick without really altering Cintia, just applying the chokes on the cables outside the enclosure... get your hands on some burnt transformer with ferrite nucleus (the more the better) of the "right" size, "open" them and wind the usb cable on it as many coils as you can while closing the nucleus (open, the magnetic fild isdisrupted and the choke lose efficiency), and see if something change.

Ideally, it should be done also for the power and the video cables... this latter can't really be coiled too strictly, but hopefully it is not needed too much either.

If the one you opened is Cintia's power adapter, you can spare the effort on the power cable as it already have a choke, probably way better than my little trick (maybe... experiment is experiment is experiment).

If not, you should take a peek... if Cintia uses the same adapter from NYJTouch that I fried first (plenty of electric storms, around here), it had no trace of an internal common mode choke.  Just a small thorus at the DC cable's end (ok, already better than the 3mm thick ring at the start of the Intuos2 USB, but still way insufficient) .

A bad thing is that, if it works, this is a "linear" trick... there is not a magic threshold after which the jitter disappears, even hypothesizing that the common mode noises were the only source of it (highly unlikely).

It is below one pixel, but CabinetiQ has still casual jitter!

And I don't think it can do anything for the "topographical" jitter (drawing a 45º line with a ruler, it comes out slightly "wobbly"; drawing a second line, parallel, at ten pixel, it comes out identical to the first... they "wobbles" together,  even if both horizontal lines both vertical lines shows no imperfections. )

_________________________________________________________________________________________________________


Finally, about stripping the transformers nuclei.

The "boxy" ones usually have a "top bar" kept in place by a faint glue, so it is enough to gently hammering sideways on the top of them, after having removed the tape that closes them, to loosen it up; the monolithic ring ones need to be broken in half, by gently hammering on a big screwdriver - or drill point - laid flatly across them after having removed the coiling (the most time consuming phase), while they rest on a semi-rigid surface (I use plywood).

I am not kidding about gently hammering - though one can easily glue back the ferrite together, if shit happens, it is still better when you have 2 neat pieces to glue, instead of 7 irregular ones.



P.S. I beg your pardon if I have been too detailed with these last notes... it's just that I assume that even most electronic guys don't go around stripping the guts of very basic components, and I believe that you are way more of an artist than a capacitors cruncher.


If you manage to try it, let me know if you experience good or bad results.

« Last Edit: July 18, 2014, 03:39:59 AM by DaBotz » Logged

The most incredible artist of... Barbanza?
DaBotz
Sr. Member
****
Posts: 243



View Profile WWW
« Reply #27 on: July 19, 2014, 02:30:21 PM »

Alternatively, you may try with some of these:

http://www.amazon.co.uk/Diameter-Black-Noise-Ferrite-Filter/dp/B007Q93APA

I do not know how many of these are equivalent to one of my rounds of coi... the mathematics is not that simple, ad I do not remember it anyway.

On a side note, after reading a bit more on the EMI filters and common mode chokes, I realize that they may add a "danger"  if the cables they are placed on are home to dc common  currents.

It should not be a problem on the single USB cable that goes to the WACOM tablet (unless the physic laws of the universe does not change, DC currents can't flow through a purely capacitive coupling, which is what wpould allow the noise to flow), may be something to keep in consideration on the video and power adapter cables.

Logged

The most incredible artist of... Barbanza?
marceloaguiar
Jr. Member
**
Posts: 33



View Profile WWW
« Reply #28 on: August 16, 2014, 08:54:40 PM »

DaBotz, thank you so much for your tips.
I tried to use the nucleus in my transformer, but it didn't work, so I got some filters like the ones in your last post.

* IMG_20140806_195223_.jpg (175.44 KB. 900x563 - viewed 295 times.)

(actually the photo shows only 3 cores but I ended up placing 4)

I also applied cores to the dvi cable and 1 additional for the ac adapter.

It still didn't work. Jitter improved like 10% but random clicking, especially in the upper zone of the screen, remains desastrous. Sad

I'm selling my set, if anyone is interested. >>> http://forum.bongofish.co.uk/index.php?topic=2442.msg20362#msg20362

« Last Edit: August 17, 2014, 11:12:12 PM by marceloaguiar » Logged

I love my tablet... that's why I'm going to tear it apart.
Aerendraca
Administrator
Hero Member
*****
Posts: 1069


View Profile
« Reply #29 on: August 18, 2014, 08:31:51 AM »

Sorry to hear you're selling up, but I think I can understand your reasons. I'm sure had you managed to complete the build with next to zero jitter and random clicks you would have kept it and tweaked it a little, but there's only so far you can tolerate a project that is not working as you would like. Stick around the forums though, we could do with a few experienced people around here recently - especially those that have come across the pitfalls of building a DIY Cintiq.

Logged
Pages: 1 [2] 3
  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!