Wacom Stylus Repair

[Domagoj T]

I have an Onyx Boox e-ink reader that is around seven years old. It has a Wacom's Penabled type display that is sensitive only to, I presume proprietary, stylus and not fingers (I like this behavior since it means you can grab the device any way you want without the pages jumping around). Vast majority of features I use can be accessed with physical keys, but I find the touchscreen very useful for dictionary lookup - just tap the word and the dictionary opens on that entry. Unfortunately, dictionary can not be used without the touchscreen, which leads me to the problem I have.
Recently I noticed that the stylus doesn't work. Initially I suspected the touchscreen was at fault, but further tinkering with it leads me to believe that it's the stylus itself. While I was testing it, I noticed that if I angle the stylus almost parallel to the screen I get response, so I tried using a stylus I got with a Wacom Bamboo drawing tablet. Unfortunately it does not work, however, while I was testing I noticed that if I put both styluses together the Bamboo stylus sort of boosts the Boox one and it works.

The replacement is rather expensive, so before I dish out $40, plus about $40 more for shipping, I decided to have a look inside. I opened the stylus and this is what I found (image in attachment).

I was fiddling with the trim pot, but I get no perceivable change in response throughout my testing. What did produce a dramatic change was the part marked with B. The one in the picture is a replacement I made, since the original jumped from my tweezer into some other dimension. Before I lost it, I got enough testing with the original piece that I am thoroughly confused regarding its purpose. The original piece was made from material visually similar to the long cylinder with coil wrapped around it (marked C). It was ferromagnetic, just like the cylinder and had a dimple to accommodate the small protrusion on the nib (marked A). I tried replicating it as closely as my memory allows me, but I can't replicate the material, so I used a piece of steel. In any case, both the original and my replacement part have the same effect on the stylus. If I place it between the nib and the coil there is a total loss of functionality.
The weird part is that if I remove it and put nothing between (nib and cylinder with coil in direct contact), or a small brass cylinder (I tried similar dimension, and about three times as long), or a piece of paper, or a piece of PVC wire insulation (no wire inside) I get the functionality on the touchscreen. However, the display starts recognizing the stylus at around 10 mm, which is far too far away for usability. The original behavior was recognition only on direct contact (no pressure needed, but not with air gap).

There is one more thing that could be in play, and that is the oval port on the side of the housing. The plug marked E had two thin rods (one is still visible on the plug, the other broke off) leading roughly (or perhaps exactly) towards the two large pads just above the letter D. They were glued with silicone or something similar to the PCB. I tested the silicone looking stuff for conductivity, but I get nothing. No part of the white plastic (which looks like styrene) exhibits any electrical conductivity.

So before I buy a new stylus for much more than I would consider a reasonable price, does anybody have any idea if there is anything I can do?

[jmelson]

Could there have been a spring between parts B and C, that also got lost?  it might make sense that the stylus tip would press B against C when touched to the tablet, altering the magnetic properties.  I'm guessing the coil and circuitry form a tuned circuit, and pressing B and C together alters the resonant frequency.

Jon

[Domagoj T]

That certainly could be the case. The parts are tiny and I won't claim my eyesight is perfect, but when I pulled the PCB out and dumped the contents of the housing on a clean piece of paper I didn't see anything other than the A and B. That being said, I don't remember the nib A being springy. The spring action (if there was any) might have been to short to notice, though.
However, the lack of spring in the current situation does not explain the initial failure, since this is the first time I disassembled it.
I Forgot to mention this, but I visually examined the board and couldn't see anything wrong with it (bad solder joints, cracked laminate etc). The energies involved are minuscule so no charring, or exploded caps. The coil wire is a very fine braid, not solid core. It might have been damaged (some strands broken) due to vibration, but the device was not exposed to a lot of rough handling.

[booyeah]

I don't know anything about the Onyx ebook reader, but I have in the past used a number of Wacom EMR penabled windows tablet pc/laptops. Which seem to be the same internally as the pen that you posted. Generally I found that I'd drop one and the ferrite in the core would be what cracked or fractured, causing the pen to be much less sensitive/ holding it close to the screen to work etc.

I don't think there's a spring missing. Usually the nibs are made of plastic/ replaceable and sit loosely in the pen. Even if they were conductive the laptop/tablet screens usually had an extra plastic layer in front of the lcd panel so that you could apply pressure when writing without actually writing on the lcd panel and damaging it.

A good article about the pens or at least the Wacom Emr penabled tablet ones
https://blog.tabletpc.com.au/2013/07/19/how-does-the-wacom-tablet-pc-pen-work/

One thing I don't understand though is the hover distance, with all the penabled laptops the greater the hover distance (distance from the screen before the pen would be recognised) the better. The cursor would follow the pen, and only be clicked when you actually pushed the pen into the screen.
Maybe the e-reader implemented it differently.

[Domagoj T]

One thing I don't understand though is the hover distance, with all the penabled laptops the greater the hover distance (distance from the screen before the pen would be recognised) the better. The cursor would follow the pen, and only be clicked when you actually pushed the pen into the screen.
Maybe the e-reader implemented it differently.

That is the behavior on the Bamboo drawing tablet too, and for a drawing tablet without integrated screen it is useful, since the tablet and the monitor are separate and you need a feedback on where your cursor is. With the ereader the cursor is wherever you tap the screen. You don't need to see it moving all the time and since the refresh rate e-ink display is very slow compared to LCD it would also not look good. With ereader (especially one with e-ink) actual touch to achieve a click is perfectly fine.

From the article you posted:

Unfortunately these pens generally aren’t serviceable and this one is certainly, to quote Taylor Swift, “never, ever, ever getting back together!”

That's a shame, however while messing around with the various bits and pieces I stumbled upon a combination of spacers that might just prove to be usable for my case. The functionality is not as the original, but beggars can't be choosers, right? Anyway, one of the combinations (don't ask me what just yet) had the effect of distant activation, but it would go out when the pen touched the screen. When the pen was removed it would again briefly register, which results in double registering. This means that if I tap the screen just once it acts like a double click. Believe it or not that sort of works for me, since the main use of the pen (in my case) is the dictionary, which requires a double click on a word. Unfortunately, just as I found out this behavior I was distracted and had to leave my bench, so I'll have to rediscover it.
I'll update you guys when I finish the project.

I'll also need to rebuild part of the casing, since disassembly required a bit of persuasion.

[Domagoj T]

Just a little update on the project, and a solution should somebody find this thread in the future.
I tried various materials and shapes and finally got the desired functionality by inserting a piece of PVC wire insulation between the coil and the nib (no wire inside). The length of insulation ended up being fairly important for functionality and I settled for 19 mm long.

It's not exactly as new, but close enough to be usable without much trouble.
Cheers.