Search results for:

@Gordon is the pin numbering on this page: https://infocenter.nordicsemi.com/topic/ps_nrf52840/pin.html?cp=5_0_0_6_0_1#qfn48 the same as on https://www.espruino.com/Bangle.js2+Technical#gpio?
If so 42 and 43 seem to be analog inputs and I could attach multiple buttons to just one of those pads with resistor dividers to detect which one is pressed. So all the buttons I want on one pin and software controllable flashlight as well.

Edit: It seems those pins are not usable as analog pins. At least analogRead only returns NaN while digitalRead works fine. So short of an I²C IO expander or something like that the easy thing to do is 2 additional buttons.

The flashlight is not connected to some controllable output, it currently just is powered by a 3.3V rail on the mainboard. Maybe a light connected in parallel to the buzzer would be cool for notifications. Since there are only two unused GPIO I will probably use those for buttons.

I have tested the bigger battery while wearing the prototype. GPS and HRM on always on got to well over 15 hours runtime. That's actually a bit better than I expected from the volume increase of the battery.
The next version will have space for two additional buttons. And somehow more space for the LEDs, since I botched the install by glueing the case shut before the resistors arrived. Those things are tiny... but at least it now works with usable brightness.
Trying to sand the surface of the nylon made it look grey and wouldn't get better by blasting with hot air to melt the surface. So it's probably better to not sand if not planning on painting.

That's awesome @devsnd. We might be getting close to needing to change from "protecting the Bangle" to "how to protect stuff from the Bangle" :D What is the diameter on these holes? In nylon SLS 1.5mm is actually a tiny bit to big. Still functional but a bit wobbly.

Thanks for the infos @Gordon, lots of stuff to think about and experiment with. Especially the two free pads. Could I just put a button on each one to 3.3V without further hardware and do the rest in software or are there pull-up/down or current limiting resistors needed? Probably that's all integrated in the chip anyway but I don't want to fry it. I imagine using something similar to the three button control scheme of the first Bangle would be awesome.

The botched button flat flex could not take the abuse of soldering thick solid core wire to it, so I had to make my own... Thankfully that wasn't too hard, some kapton style tape and a bit of copper tape does the job just fine.

@Gordon is there a better way to get power to the LED than the 3.3V "testpoint" on the board? The plan is to use one of the LEDs with a resistor in series to get to 20mA. I noticed LED+ and LED- written on some points, can those be controlled in software and maybe deliver a current without needing a resistor in series?

Sadly this will always need disassembly of the original watch. There are 3 main points which are common to all my versions and were solved by soldering:

• USB charging -> Connect USB port to charging pads
  
• Other button -> Connect new button to mainboard
  
• Bigger battery -> Replace the original lithium cell
  
  

The button is actually soldered to a bit of flat flex cable taken from the backlight LEDs of a broken Bangle display.

I would say what I am doing has a very high risk of killing the donor bangle. Not quite sure if it is actually worth doing. It currently is more of an experiment just because I (maybe) can ;)

I have never designed flat flex cables, so maybe that would be a way to at least remove the most difficult parts of the soldering job, but currently it is definitely at the absolute limit of my abilities.

The design by @devsnd will get you the best way to get a more rugged Bangle. I will order one as well since my design is not close to be a daily driver.

As for progress:

• Everything can be mashed in the new case
  
• Button works
  
• HRM fits perfectly
  
• Bigger battery fits and charges via USB-C
  
• Flashlight dimly works (two LEDs driven directly by 3.3V from board because I do not have a matching value resistor and one LED burns out instantly)
  
• Not yet glued together
  
• No way this holds together in daily use, at least the solder joints will probably shake apart in a short time
  
  

Since I could not get the watch built in the mark 2 case I have ordered the third revision. Mark 2 had the problem that it completely turned the order of things on its head. The display goes first and then everything need to be assembled (and soldered) the "wrong" way around.

To alleviate that mark 3 is designed to allow building on the base plate and then put the top including the display over it to be more similar to the original construction. I ordered in nylon SLS in to get similar precision as metal SLM but much cheaper prototypes. Surface is a bit rougher than SLM but overall quite good. Even very thin walls worked out mostly fine. Still some fine tuning needed but it seems get close.

Now I "just" need to assembly it into a working Bangle :)

The stainless is relatively soft and thus sensitive to scratches. I think the somewhat rustic optics you have now should withstand use pretty well. Could be nice combined with maybe band from wooden elements.

How is GPS, BT and touch for you?

That's a really elegant solution to the ruggedness problem. If it weren't for USB-C and a bigger battery that I want I would definitely go down your route. Doing the full transplant is very time consuming and the risk of killing the donor bangle is considerable.

Ebay has conductive glue/paint containing silver in tiny syringes for PCB repair and model building starting at 2,50€. Thats probably worth a try.

I have given the border around the display an additional 0,5mm in x and y direction in my drawing and it fits loosely in the case. I did not try to measure that with a feeler gauge (lost those somewhere) but there seem to be about 0,3mm space all around the display so I would suspect you can be closer around 100% size for your tries. Maybe 101% and 102% to leave a bit of room for glue if you have your current CAD at a perfect fit. Since the turnaround is just about one week (JLCPCB->Germany) and postage is not that bad a second try is not a big problem.

If this fits well and works nicely with Bluetooth and GPS you could think about removing the original watchband holders and modeling them on the case. There are some people having broken those off and having them in steel is probably more rugged.

You can find the firmware builds here: https://github.com/espruino/Espruino/actions. I don't know if there is an easy way to match them to versions, but you can see the git revision in each individual build.

That is the "BT dead on arrival" for parts mainboard that you sent me a while ago. Since the board is unusable anyway I decided to use it as the first victim for trying to solder the USB-C port to. Good idea, since I ripped a pad of it straight away... The screen is also an old one which is somewhat broken so no real info regarding that until I use one of my actually working ones :) Will try that when there is less shoehorning going on.

It is here and it is awesome! Ordered last Sunday, reviewed and payed for on Monday, delivered Saturday using the second cheapest delivery option.

Some things I have discovered:

• The contact pads for the push pins lift of the board really easy when soldered to -> use very flexible wire and support with some epoxy or something after soldering
  
• Stranded wire from an ethernet cable is nowhere near flexible enough
  
• My fingers are very big in comparison to a watch case
  
• Stainless SLM is awesome, width should have been 40mm and actually was 40.15mm
  
• Tiny holes in the drawing get even tinier in the end result. Planned 0.8mm for the wrist band mounting pins, got about 0.3mm. Widening them with a 1mm drill is no fun...
  
• Do not forget to plan tolerances for wrist band mounting stuff
  
• 1mm wall thickness seems absolutely fine
  
  

Watch body is 316L SLM and bottom plate is just 0.8mm stainless sheet, cut with tin snips and filed to match the body.

Bangle 2 has the slightly rounded button soldered to a tiny flat flex cable plugged into the mainboard. The pictures in this thread are of a Magic3 watch.

Did you find a fully waterproof button?

I plan to use normal momentary switches and cover them in silicone. My design has 2mm deep cavities which should make for a few tenths of silicone over the button. That will hopefully be flexible enough to use the button and still hold up to daily use.

useable without a button

I had tried this by listening to the accel events and calling everything in the pin watches array for the button. Worked halfway decent but I did not want to modify the bootloader to make it work there as well so I decided to keep the button.

Did you already dissassemble a bangle js 2?

Multiple Bangles, some of them multiple times :) My current way of choice for warming the Bangle up is a pot with a bit of simmering water and a flat top. My cookware has flat glass lids which get close to 100° with the steam from the simmering water. I just place the Bangle display down on the glass until it is hot enough. I have a heat gun with temperature settings but that has not worked as well for me.

Is there a reason you decided to use usb charging?

Multiple:

• More sturdy connection for charging in a bag
  
• One less cable to forget while packing for a trip
  
• Corrosion of the contact pads
  
• No idea how to get push-pins and magnets etc. aligned properly and waterproof :)
  
  

did you consider trying to make an outer case for the Bangle to waterproof it?

Maybe there is ABS as a very thin sheet or even film which could be "glued" over the button with acetone?

Cobbled a design together for the Backplate with space for the HRM using the file of the original one as reference. Seems to be thicker than I expected but I will see when I try to disassemble an HRM.
I placed the HRM slightly off center to account for the additional battery thickness. I do not think extension of the flatflex cable is doable in reasonable way.
I will try the case first without HRM and order the backplate when the next revision of the case itself is done after testing the current one. Production seems to be done and it should be shipped soon. It is even possible it arrives in time to tinker with on the weekend.

I plan to glue the watch together with either low melt hot glue or B-7010 double sided tape to be able to take it apart relatively easily. The surface roughness might play to my advantage there.
PCBWay will do aluminium and stainless SLM for about 3 times the price. Even titanium for a bit over 100$. JLC3DP has nylon SLS options for even less money than metal SLM so that would solve the faraday cage problems. I'm currently hoping to just connect the whole case as aerial and maybe it works well enough to be usable. Probably not because the whole tuning will be completely out of whack, but we will see.
I will take some measurements and compare with the CAD file to post here when the thing arrives. There are some videos [*1] where the tolerances seem to be pretty nice. Better than 0.3mm at least.
If I loose GPS anyway because of the metal case, could I just remove the "antenna block" for more space inside the case or a slightly smaller case design? Or is it needed for other stuff as well? The barometer hole is on the same side in the original case.

@Ivor: My mill is a set of old stuff I bought used for relatively cheap, Proxxon KT150 and BFB2000 (filled the pipe in the back with concrete) as the base and a Kress 1050 milling motor. It just about does aluminium well enough but it takes a lot of time doing this manually and it is very easy to make mistakes. If I had known how cheap 3D printing services had become I probably never would have tried to mill the case myself.

*1:
https://youtu.be/mrp28hZ_Qu0?si=7nVoD4aG9clbomw2&t=335

The 3D print service (JLC3DP) accepted the file (warned me of thin walls but I took the risk) and it seems to be in production now. Scheduled to be sent out in 4 days. I have no idea what to expect but hoping for the best :) The whole thing was surprisingly affordable. About 13€ shipped in 316L stainless steel SLM.

Thanks to @Gordon for the detailed watch body file in https://forum.espruino.com/comments/17414317/. That will help me with designing a backplate that can house the HRM sensor. If I can get it out of the original case without destroying it...

OK... after having broken this attempt while trying to drill and tap the screw holes the next one will not be manufactured by me. While I had fun doing this in the end my capabilities do not match my plans :D.
I have done a new design for some 3D print contractor to print (SLS/SLM) for me. No idea if this can work since I have absolutely zero experience with this but the custom case dream is not yet dead.
Features are still USB-C, more protection for the display and a bit more internal space for the battery while losing the HRM and a having decidedly chunkier watch. Additionally I will try to add an LED and an additional button to have a (dim) flashlight function for finding the way in absolute darkness.

I do have a 3D file for the body, but I'm afraid it won't be any use to you - the shell contains the bluetooth aerial, so you couldn't use a 3D printed one easily I'm afraid.

Is that the file on github with display and HRM included or one of only the body? I would be interested in something that I can take as a reference for the cutout needed for the HRM.

I had glued a Bangle with unsuitable glue and the first contact with water forced me to order a replacement screen since it killed the touchscreen and very nearly the board as well (had to remove surprising amounts of corrosion for the mere minutes the water was in there).
Usually I used B-7000 glue for phones etc. which worked well enough (stays somewhat flexible) but that had dried out in the tube and I decided to use some generic Uhu glue. Bad idea.
The new display is glued with B-7010 1mm double sided tape. It is sticky as hell, slightly foamy/thick and allegedly waterproof. Let's hope it fares better :) At the very least it was nice to apply with tweezers and the protective film on the future display side of the tape.

Found it... the wrap command that runs before in the demo test did not have a \n. Seems to work now. So I assume \x10 only works if at the start of a line?

The emu.tx line has the \x10 though. It is exactly the same code that works the second time. Is there maybe something that can happen before which cancels out the echo off for the following line?

It might be possible to fill the button board cavity with silicone and have it malleable enough to still press the button and get a reaction. Would probably be a total loss if it does not work or the button needs to be replaced.