Alright, it seems like I managed to create a prototype for inductively charging the N9 via the Palm Touchstone stuff.


First of all the usual disclaimer and words of caution:
The things shown here are highly dangerous.
By trying to do anything as described here you will for sure void your warranty.
Furhtermore, you might destroy your precious N9 or even worse cause other severe damage or even harm people.
I am not responsible for anything you do based on the things written and described here. You do everything on your own full responsibility.


Having said that, lets move on.
I currently do not have the time to write a nice, full-fledged article about this so I'll just introduce the very basics. I will upload pictures and more descriptions later on when I find the time.

The general idea is to have the inductive sender deliver ~ +5V. This will then be fed into the N9 via the USB connector. This way all the internal charging logic etc. should be able to do its job. Well, we'll see how this works out.

First of all, the good news is that there is enough room in the N9 to place the Touchstone inductive charging receiver, the cabling, and even a simple diode in it. The diode is for "protecting" the receiver circuit from receiving current from the USB port when an external charger is attached or the N9 is attached to a computer via USB. This way it is possible to inductively charge, charge via an external, wired charger, and use the USB connectivity to tranfer data.
Another good news is that the N9 detects when there is sufficient supply voltage at +5V of the USB connector. Thus, it is not necessary to shorten the data pins to trigger charging anymore (as was needed for the N900). This eases the whole effort a lot. However, it is still a very very delicate undertaking. See post #3 below.
And yet another good news is that there are acutally some maintenance(?) pads on the N9 mainboard that ease the soldering part at least a little bit. This way it is not required to mess with the tiny micro USB connector but it is possible to use these pads instead. The pads have a diameter of approximately 1-1.5mm.


There is some video that shows how to inductively charge the Lumia 800. In this video it is shown that some of the metal part of the inner case is removed for the charger to work correctly.
However, I did not want to remove parts of the case that are potentially needed for maintaining the structural stability. Instead I chose to completely remove the inner metal casing, put the receiver coil beneath it and then insert the metal casing again.
On another note, I also put the four, small metal discs that are used for attaching the cell phone to the sender underneath the interior metal casing.

For a very good and detailed guide on how to disassemble the N9 see the following video:
http://mynokiablog.com/2012/05/22/vi...e-replacement/
Note that I used a hot air gun/fan with a temperature of 180°C to heaten the internal metal casing in order to remove it. This is needed because the internal metal casing is glued into the polycarbonate external case. Also note that this process might take some time and is best done in several steps in order to avoid damage to the internal and external casing as well as the flat ribbon cables.



Edit: Please see post #3. There I explain how I could solve the low current issue. I am leaving the following paragraphs in this post for the sake of completeness.

The placement of the coil in relation to the four small metal discs seems to be very important. Here, it showed that there is only very few space for variation in order to assure the charging works at all.

Generally, I am currently facing the issue that the current delivered by the receiver is very low. Placing the receiver on the sender without any load attached shows a more or less constant voltage of 5.6V (note that the drop off voltage of the diode reduces this volatage to a "more USB friendly" ~5.1V). With no attached load the placement also seems not to be very critical.
However, once a load (in this case the N9) is attached the whole system seems to be more fragile. Generally, somehow the receiver seems to deliver a very low current. I am currently investigating this issue.


This issue of the receiver only delivering very low current has some undesired consequences that affect the overall practicability.
Firstly, the charging is very very slow. At the moment it seems like that this is much too slow for the inductive charging being of any practical use.
Secondly, the supplied current seems to depend on the placement of the receiver on the sender. When the sender is not placed correctly, the power draw by the N9 seems to trigger a failsafe shutdown in the receiver such that charging stops immediately after it had started. This then results in a cycle of the USB being reported attached, removed, attached, etc.
Thirdly, even when the charging current is a little bit more stable the N9 reports that the power supplied by the USB port is too low. However, it continues charging but very slowly.
Finally (for now), even when the placement seems to be good and the device charges very slowly it does disconnect from time to time in intervals of about 5 or 10 minutes.


So far, as you can see this is still very very experimental. While there are some good news that illustrate the general practicability there are still some issues that render the daily usage impractical for now.
Comments and feedback are highly appreciated. Especially hints on how to solve the issue with the low current.
Are there any people that have more experience with the Touchstone charging kit and its specifications?

Very interesting indeed, which type of touchstone are you using? TouchPad or Pre?

As a TouchPad and TP touchstone owner, I have learned that USB charging is considerably faster than the touchstone, however it is designed to be left standing on charge throughout the day. The Touchpad will also activate its charge cycle once battery drops below 95% charge.

I can't speak for any of the pre stones tbh, so I don't know if they have different voltages to compensate for varying batteries?

Just another thought.. This little project of yours would also open the doors for various Exhibition mode apps too.

Just a small update:

Seems like shortening the data wires is still needed for the whole shebang to work correctly.
The good news is that shortening the data wires solves the low current issue. However, this complicates the whole endeavor.


Also thanks MINKIN2 for the feedback.
I am using the palm pre touchstone kit.
I made some more tests with this kit. I could measure currents up to ~500 mA. I didn't go for higher values as this is sufficient for me at least. As a side note: I used resistors with different values for testing and measuring the output current.

Still, I made some interesting observation with respect to the behavior of the touchstone kit when confronted with different loads. Using a load of approximately 100 mA I could reproduce the phenomenon of the receiver cyclically being active and inactive. Interestingly this issue only showed for that particular load. Loads below and above (tested up to 500mA) did not show this issue.

It seems the N9 charging circuits/algorithms only draw very low current when only the power wires are connected. As mentioned above, shortening the data wires results in a much higher charging current. I tested this by observing the charging time as well as the estimated charging time left. The results with the data wire shortened indicate a very much higher charging current.


So far.. pictures are still yet to come. I will also update the first post once I have the time.

As promised here are some pictures. I will upload high-resolution versions to my blog later.

Note that the "internal" pictures are taken from a different N9. Reason is that I messed with the "inductive charging prototype" so much and I don't want to stress it too much.

This is amazing. Good work

Very cool. I was planning to do the same thing once I got the N9. but alas, it seems I can't bring myself to fork money over to Microkia, now even less than before. Moreover, ironically the SGS3 seems to be the more open platform, even in this regard: additional charging connectors were placed easily accessible right next to the battery, so modding on the touchstone wireless charging coil should be quite easy. (Or wait for the official wireless charging solution in autumn)

Btw, just found the youtube video for the equivalent SGS3 "hack" (when comparing it with above procedure it doesn't really deserve that label ).

Alright, attached I uploaded some more pictures. These pictures show the placement of the touchstone receiver and the small metal discs. Please note that this is, again, not the actual prototype. This time I simply used a left over case I had ordered some time ago. The placement of the small metal discs should be roughly the same as in the prototype. However, the placement depends on the placement of the receiver coil and thus is very likely to vary.

The main reason why I uploaded these pictures is that you get an idea of the inner bits of this mod. Note that I used a small piece of paper to avoid the soldering points where the coil is attached to the receiver logic being short circuit by the internal metal housing. In the actual prototype I also put some paper between the metal foil "shielding" of the receiver and the metal housing.

Hey I was thinking, do u think that is possible to put those plates inside those dual layer cases and make a mini usb plug that attach to the N9 like a wall charger using a case like this:

http://www.ebay.com/itm/Dual-layers-...ht_1689wt_1162...

this one even cover the usb port and u can make the shortcut in the cable... so if u want to use the usb just disconect the case...

Hello Wonko,

Your post i a little bit old, I hope You made some (hopefully) positive experience in this time ;-)

I made this mod with my Nokia N9 and Palm Pre TouchStone.
After 6 assembly-disassembly cycles, i placed the inductive charging unit inside my N9 and try to charge it.
Tocuhstone charge it, but I got the message: "Low power from the USB port to charge"

the positive (+) wire I soldered to the +5V of the USB Port, the negative (-) - to the ground of the usb port.

You write, I must shorten the data ports to charge with a higher rates than 100 maH?

If i made it, it would not be possible to use the usb port?

Can you tell, how You solved this problem?