test1
March 30, 2017, 05:49:40 AM *
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 ... 5
  Print  
Author Topic: Finally Ready; A Newbie Build for a Newbie Maker - Syntic  (Read 7245 times)
comfychair
Jr. Member
**
Posts: 41


View Profile
« on: October 15, 2015, 06:10:21 PM »

Greetings Bongofish

I'm really excited about this build and its been a lot of fun so far. Fun and Hard.

Seeing as how I'm new to Making, I'm starting slow and doing a "Sandwich Build" first. The original inspiration came from TabletMod.com

Here is the equipment:

Intuous4 PTK-840
LP154WE3
VS-TY2662-V1 (controller)
A re-purposed cd-rom enclosure and brass stand-offs (to mount the controller)


As for the build log, there is not much to say. I had to strip the aluminium off the back of the display, and am using electrical tape to hold the display together.

* screen1.jpg (143.02 KB. 1288x724 - viewed 368 times.)


* screen2.jpg (117.67 KB. 1288x724 - viewed 326 times.)


On the first test, I got the usual jitters and false clicks, nothing too serious. One problem I did have earlier was the read height of the pen was greatly diminished.

The TCon board is at the bottom of the screen (pictured below), I covered it in plastic and wrapped in in copper tape to help with both of those. (It did Cheesy)

* tcon.jpg (153.39 KB. 1288x724 - viewed 297 times.)


* tcon2.jpg (120.75 KB. 1288x724 - viewed 272 times.)


* tcon3.jpg (117.48 KB. 1288x724 - viewed 271 times.)


You will notice that there is copper around the screw mounting holes on the Tcon board. (so these are grounding points?)

Here is the controller:

* controller.jpg (148.24 KB. 1288x724 - viewed 318 times.)


Initially I'm getting very good results. Except for the false clicks! They are everywhere and random.

As you can see, I've only done some shielding but no grounding.

So I have some questions:

1. A lot of you here are using DVI, is this preferable to HDMI?
I will be ordering ferrite cores as soon as I know what cable I'm going to settle on

2. In order to ground the tablet and screen, I have to have them on the same ground.
I'm quoting DaBotz here:
Quote
False clicks usually can be ameliorated through grounding together tablet and screen, and using ferrites to choke common mode currents, so they should not be too worrysome.
I have no idea what common mode currents are, but am I going to have to take my tablet apart to do this or is there a way around that?


Also of note, You will notice that there is no copper rings around the screw holes on the controller, so there are no grounding points on the controller. (Am I right about this?)


Really, False clicks are the only real problem, Jitter is almost non-existent, but I do have a confession: I unknowingly messed with the calibration in the pen before reading that I should make note of where it was initially Undecided

So there it is, I look forward to everyone's input!

« Last Edit: October 15, 2015, 06:25:22 PM by comfychair » Logged
DaBotz
Sr. Member
****
Posts: 171



View Profile WWW
« Reply #1 on: October 16, 2015, 12:43:36 PM »

Common current noises are a kind of electric interference in which the sum of the currents (in and out) of a balanced input (like the USB cable should be) is different from 0... i.e. the whole sets of cables behave as ne cable tucking some a.c. current it (that goes out of the others).

If it happens, you have an unaccounted alternating current flowing through your system... depending on you set-up, it may or may not be an issue.

I am not a fan of grounding things together (done wrong, you fry something, and I am not that courageous... also, my first build was a Tabletmod thing, with an untouched Intuos4, and I wasn't smart enough to peel the usb cable up, find its ground thread, and connect it to the tCON ground - which is, by the way, connected to those coppered holes you mentioned...), so I tend to coil my usb cables (up to two meters of it in the UBIQ) around toroidal transformers nuclei (scavenged from burnt PC power units that I "steal" from a local computer shop), to reduce said noises to a lower level.

If you kept the tcon cover from the screen original construction, you can find some metallic base for it, modify it a bit and encase the whole tcon in a faraday cage(ideally, well attached to the screen, so that you shan't worry about the flexible cables from the TCon to the TFT getting damaged).

I did it in my second build, and it helped too.
It worked for me (the noisiest and jitterest of my two builds has a +/- 0.2mm horizontal jitter, and a couple of false clicks a day, once you are away from the screen's perimeter frame)

Calibrating is not all that difficult (maybe)

https://www.youtube.com/watch?v=rRXBb_hafAY
« Last Edit: October 16, 2015, 02:05:18 PM by DaBotz » Logged

The most incredible artist of... Barbanza?
woodguy32
Jr. Member
**
Posts: 50


View Profile
« Reply #2 on: October 16, 2015, 07:23:14 PM »

Calibration is very simple and you know you got your build almost there for creation time..

 I am seeing and hearing a bunch on USB 3 issues with wireless keyboards and other stuff plugged directly plug in them..its fine for your hard drive speeds....

But having an powered external USB 2 hub does wonders..  I haven't had any jitter issues with the last 6-8 builds..with the new monitors..( ISP Led and power supplies ) its pretty much foolproof to build one now..  Gad.. the days of the old tube lcd screens caused all sorts of jitter..

 wood..
« Last Edit: October 16, 2015, 07:25:17 PM by woodguy32 » Logged
DaBotz
Sr. Member
****
Posts: 171



View Profile WWW
« Reply #3 on: October 17, 2015, 12:33:51 PM »

Common mode chokes, the fast, easy and greasy way: https://www.youtube.com/watch?v=1v0zFVrv5fE , at 3:00 minutes.

And, yes, every one of that things helped making damn UBiQ work. When I first placed the screen over the tablet, I just wanted to cry.
a good 1cm jitter with enough false clicks to give you epilepsy.


Common mode chokes, the serious electronics way:



You'd need two of these on the USB line (one on the pair +5V GND, and  one for the pair  D+ D- ), while it is debatable that they may help things to add more on the video input and power lines (if my idea is correct, they are not needed, once the USB is protected enough... if you have a brick power unit and a separated video board, adding one on the videoboard input is a question of an hour of work).

Unfortunately, one of the provisions of the USB standard is a maximum latency of 1 ns for the signal between a host and a client, I.E. 5 meters of cable between the tablet and a hub or the PC.

So, before firing up the soldering iron, you need to know the effective length of cable coiled around the ferrite in the choke you want to use.

If it is 4 mt, and you plan to hide a hub inside the case of your build - never to worry about breaking the damn Wacom's cables or plugs -
you are left with 1 mt to route things in the case.

Note that a common mode noise jumping from tablet to screen is just one of the possible Jitter and click sources, that often is already cured when you ground together screen and the "antenna plane" of the Wacom - the other classes, starting with the horizontal frequency of the screen synching with the Wacom's main - couldn't care less for chokes.

So, if you have a burnt transformer laying around (I had some), you can make a try with the fast and greasy, to see IF it helps () before committing money to buying components.

A choke of this type inserts an impedance in the path of this kind of noise, but can't shut it out completely - an impedance whose value is  - if I remember well  -         2 * Pi * Frequency * Inductance of the coil .

The trick is reducing it's entity below the point where the Wacom manages to ignore it most of the time, while avoid crashing on the USB latency limit....







  
 
« Last Edit: October 17, 2015, 12:45:02 PM by DaBotz » Logged

The most incredible artist of... Barbanza?
comfychair
Jr. Member
**
Posts: 41


View Profile
« Reply #4 on: October 17, 2015, 06:44:53 PM »

thanks daBotz
I do have some ferrite cores from a power supply from a computer. The torus is too small to fit the usb cable through, but can I wrap the cable around the outside of the torus?

alternatively, will something like this work?
http://www.ebay.com/itm/10-Pcs-Black-Plastic-Clip-On-Noise-Suppressor-5mm-Cable-Ferrite-Core-Filters-/191573995144?hash=item2c9ab32e88:g:2WUAAOSwNSxVSeI~


also, thanks woodguy- ill get me a hub and let you know how it goes

So I will have to remove the tablet from its enclosure in order to ground it?

Also, will using DVI instead of HDMI help with interference


Thanks again everybody
Logged
DaBotz
Sr. Member
****
Posts: 171



View Profile WWW
« Reply #5 on: October 21, 2015, 12:36:25 PM »

You can try, and see if you see an improvement. It will not be as efficient as actually winding the cable inside the thorus, but a coil with a ferromagnetic nucleus should be already do something.

(the inside of the torus should have a relatively homogeneous magnetic flux, so that the cable in and out lines should have the same inductance).

As I said, it is just one type of interference, and grounding together the panel and the wacom should already cut most of it.
Should.

Personally, I haunted a local electric shop, as well as the computer shop that sold my family all our computers (we are now at a count of 1.2 each), to get my hands on all mines...



 
Logged

The most incredible artist of... Barbanza?
comfychair
Jr. Member
**
Posts: 41


View Profile
« Reply #6 on: October 22, 2015, 03:49:07 AM »

I would love to ground the wacom and panel together. Can you reference a post here that you think would make great instruction for this?
Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #7 on: October 26, 2015, 08:17:39 AM »

Best thing to do is to look for relatively large exposed pads of copper on both boards. I recommend that if you're unsure, post a picture on the forum and we'll check out our for you.

« Last Edit: October 27, 2015, 07:39:15 AM by Aerendraca » Logged
comfychair
Jr. Member
**
Posts: 41


View Profile
« Reply #8 on: October 27, 2015, 05:52:32 AM »

So I stripped my wacom table down,

 Smiley

* wacomstripped (2).JPG (138.11 KB. 1288x966 - viewed 290 times.)


The question is, do I have to attach the grounding wire to anything else? Reading over the Build log for the Anntiq I saw that he attached the grounding wire to the aluminum backplate. Will I have to do something similar?

http://forum.bongofish.co.uk/index.php?topic=2185.30

Here is where I'm Thinking of putting the ground wire (underneath the shielding)



* wacomstripped (1).JPG (200.94 KB. 1288x966 - viewed 286 times.)



(Also Aerendraca, I'm a big fan of your work here)

Cheers

Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #9 on: October 27, 2015, 07:42:53 AM »

Thanks, that's very kind of you to say.

That copper piece you've highlighted is the perfect spot for joining grounds. Do you know where you will attaching to at the other end?
Logged
comfychair
Jr. Member
**
Posts: 41


View Profile
« Reply #10 on: October 27, 2015, 03:35:11 PM »

Yes, I think I do, the screw holes on the TCON board on the panel are grounding points, I was planning to use a screw and nut to affix the grounding wire.

Will I have to connect the panel and digitizer to some metal like what was done in the Anntiq build?
Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #11 on: October 27, 2015, 07:40:16 PM »

Screw holes on the tcon sounds good to me. It's been a long time since I read through that build, but I imagine that the piece of metal was just used for shielding. If you use some metal to shield parts this should also be grounded, but this is not always necessary.
Logged
comfychair
Jr. Member
**
Posts: 41


View Profile
« Reply #12 on: October 28, 2015, 06:45:22 PM »

Alright I attached the grounding wire to the panel and the panel does still work Cheesy

The following pics are how I attached the grounding wire to the TCON board


* grounding (5).JPG (98.89 KB. 1288x966 - viewed 233 times.)



* grounding (4).JPG (144.18 KB. 1288x966 - viewed 275 times.)



* grounding (1).JPG (105.92 KB. 1288x966 - viewed 261 times.)


I soldered that ground wire to the wacom digitizer,


* grounding (2).JPG (160.93 KB. 1288x966 - viewed 269 times.)



* grounding (3).JPG (140.57 KB. 1288x966 - viewed 259 times.)


I have not tested it yet. If anyone sees something wrong or has questions or concerns, I want to hear them before I potentially destroy something.

So bongofish, how does it look?


Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #13 on: October 28, 2015, 09:29:16 PM »

Looks ok to me. Was the copper tape wrapped the tcon board when you got it? Or is this something you've implemented?

So long as you don't have any metal contacting any part of either circuits that shouldn't be there, I say switch it on.
Logged
comfychair
Jr. Member
**
Posts: 41


View Profile
« Reply #14 on: October 28, 2015, 11:06:58 PM »

I did implement that, My way of shielding the TCON without a TEKO box. Wink It did help with the jitters, but not the false clicks.

Alright, I'll run my first test when my shift ends, wish me luck.
Logged
Pages: [1] 2 3 ... 5
  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!