Nexus 5X, Hand Warmer

Power drain
Red rectangle marks the problematic hot area.

I’ve been pleasantly surprised at how far I was able to revive a Nexus 5X soaked in a public swimming pool for 30 minutes, but I think I’ve reached my limit.

All the phone functions I tested had worked, but using the phone for more than an hour uncovered a serious problem: something related to battery power management is very ill.

The first symptom was heat: the phone got far hotter than it used to, and trying to run the phone while plugged in to the charger resulted in a lot of heat and not a lot of increase in charge level.

I shut the phone down, unplugged it, and set it aside to wait for it to cool. It remained warm for many hours and never cooled. When I turned the phone back on briefly to check, I see the battery charge level has dropped dramatically. Something was draining battery power and turning it into heat even when the phone was turned off.

I knew most phones had two charging programs: One while the phone is on, running under the main operating system. And a second while the phone is off, run by a very minimal piece of code. I’ve already tried the first without success, so I turned off the phone and tried the second. Fortunately this was able to charge the battery though the phone still got pretty toasty.

After the battery was charged again, I disconnected the battery. Only then was the phone able to cool off. When I reconnected the battery, it was quickly obvious the heat source is under one of the pieces of metal shielding, marked in the picture by a red rectangle. I don’t know how to remove that metal shield without destroying the components underneath.

Two hypothesis are open: (1) this circuit was damaged earlier trying to run a ruined battery, or (2) the shield isn’t as watertight as it looks: there’s a lot of pool water chemicals in there corroding parts. Either way, it is beyond my current ability to address, so I have to stop here.

I can’t use this as a phone for more than an hour or two on battery. And I can’t store it with the battery connected because of the drain. So I’m going to disconnect the battery and put it aside in the hope I’ll have better ideas later.

Nexus 5X Reassembly after Cleaning

The Amazon Prime truck has delivered the aftermarket replacement battery and a syringe of thermal compound, so it’s time to put the phone back together.

The first order of business is applying a dollop of thermal compound on the frame where it accepts heat transferred from the CPU. Previously, there was a piece of pliable tape sitting in this square depression in the frame, but it was damaged by the alcohol and removed. The gray compound is intended for DIY PC builders to be applied between their computer CPU and its associated HSF (heat sink + fan) assembly. It is designed to bridge only a very small gap, since the CPU and the heat sink of a desktop PC are in physical contact and this compound only needed to fill the remaining tiny air gaps. Here it is asked to bridge a much large gap of almost 1 mm with no physical contact between parts.

Thermal compound

The reassembly went smoothly until I installed the aftermarket battery and found the connector cable is too long by a few millimeters. It is a large distance in the tight confines of a cell phone’s guts. Looking on the bright side, I guess it is better to have a connector cable that’s too long rather than too short.

Too long

With the connector plugged in to the appropriate location on the main circuit board, the cable is bent at an angle that won’t fit under the rest of the phone.

Bent

As much as I hated to do it, I pressed down the cable and made a sharp fold in it. This is a recipe for metal fatigue and the cable won’t last long if I keep doing this. Well, at least the battery is cheap.

Folded

Folding the battery cable allowed the rest of the pieces to be fully reassembled. The phone can power on and launch into Android, so the basics look OK and clear the way for more testing. Let’s see if the phone works well enough to justify replacing the screen assembly.

Powering Up the Waterlogged Nexus 5X

After waiting overnight for the re-cleaned parts to dry, I reassembled the phone and pressed power. The lack of response was no surprise. I then plugged the phone into the charger and was very encouraged to see the screen light up with the depleted battery icon. This tells me the some of the phone made it through the ordeal in at least partial working order.

But after a few hours, the situation did not improve. So I took the phone apart again to take a closer look at the battery. When I first took the phone apart to soak in distilled water, I measured the voltage across the connector terminals and got 3.0 volts. It was lower than healthy for a lithium battery but not necessarily fatal. I measured it now – after charging for several hours – and it read zero.

So, that’s not good.

I thought the battery cell might be dead, so I got another one to test my hypothesis. I wanted the connector from the existing battery so I cut apart the plastic wrap to extract it. I was surprised to find that there’s a tiny circuit board inside. I’m not sure what that circuit board does… but looking at its current condition, it’s not doing that job anymore. Since I didn’t know it was there, it didn’t receive the water + alcohol cleaning treatment received by the rest of the phone electronics. It has been under attack by swimming pool chemicals for the past few days.

Battery Board Toast

Well, I wanted the connector, and now I can access the connector. Let’s use it to wire up the lithium cell straight up. There were four soldering contacts on the connector, two with large conductors and two with small conductors. The two large ones were helpfully labeled “V+” and “V-” so that’s how I soldered the lithium cell.

Battery hackAnd it worked! This setup was sufficient to get into the Android OS. Now that the screen is showing more than the “depleted battery” icon, I could see that it was damaged in this adventure. Thankfully it was still legible, and the touchscreen still worked, so I could run the phone for about 40 minutes. Long enough to access the multi-factor authentication app so I could transfer my MFA security to another phone.

Since the phone appears to be running, I ordered a proper replacement battery. I don’t know if the corroded battery circuit board did anything bad to the rest of the phone. The charging circuit may have been damaged trying to charge a zero volt battery for hours. I’m going to see how the phone works with the replacement battery before spending money to address the damaged screen.

Drying Nexus 5X Off From Swim

My Nexus 5X phone took a 30-minute swim in a pool due to my negligence. It was unsurprisingly dark when retrieved from the pool. I’ve already ordered a replacement phone but I was curious: could it be brought back to life?

The first order of business was water removal. A public swimming pool has all sorts of chemicals unfriendly to electronics. The first thing I found upon return to home was a jug of distilled water originally intended for the car’s battery. Good enough for a starting point, I left the phone soaking in distilled water while I went online to read up on Nexus 5X disassembly on iFixit.

The information is promising – by modern phone standards, this model is very easy to disassemble and repair. Following the instructions, I disassembled the phone into its major components, performed a second round of rinsing, and laid the parts out to dry.

Disassembled

After drying overnight, it was obvious soaking in distilled water was not enough. There were enough chemicals remaining to leave a white residue on many surfaces and corrosion began eating many components. Here’s a close-up picture of the SIM slot and a few of the surrounding components. The brown stuff building up in the lower-right is especially worrisome.

Soak in Distilled

If gentle soak in distilled water wasn’t enough, it’s time to step things up. Isopropyl alcohol is easily available as a first aid disinfectant though at a lower concentration than ideal. First aid rubbing alcohol is 70% alcohol and electronics cleaning usually specifies alcohol content of 90% or higher. Since time is of the essence, the first aid stuff will have to do. Once the parts are soaking, I also ran a small plastic bristle brush over the surfaces to dislodge any remaining pool chemical and the corrosion that is accessible.

It’s not clear if the alcohol or the brushing was more useful, or if they were both required, but things look much better after the alcohol dried off overnight.

Alcohol and Brush

Some printed numbers were erased by the alcohol, which I wasn’t worried about. Some adhesives were dissolved by the alcohol, and I’m worried about the tape that used to sit over the CPU. I will need a replacement heat conductor to help transfer heat generated by the CPU to the chassis frame for dissipation.

Portable External Monitor 2.0: Stacking Plates

Enclosure2_CADPortable External Monitor version 2.0 (PEM2) explored a different construction technique from PEM1. Instead of building a box by assembling its six side pieces (top, bottom, left, right, front back) the box is built up by stacking sheets of acrylic.

With this construction technique, it is much quicker to place components in arbitrary locations in 3D space. Control along the X/Y laser cutting axis are trivial. Control in the Z axis takes a little more effort. The components can be aligned to the thickness of the sheet of acrylic, but if that’s not enough, it is possible to use engraving operations to precisely locate the component in Z.

In contrast, when we want to locate components inside a box at a specific coordinate, we’ll have to design additional pieces – supports and brackets – to mount the item at the appropriate location in the box.

It is also very easy to assure alignment between the parts of the box. Cut a few fastener holes at the same location across all the sheets. After they are stacked up, inserting the fasteners to align all the sheets.

The downside of this approach is that it is very wasteful of material. Each layer will consume an acrylic sheet of the overall X and Y dimensions. And if we only cut away the parts we need for the components, there is potential for a lot of unnecessary acrylic in the final assembly. They add weight without usefully contributing to the structure. Putting in the design time to cut away those parts reduces the time savings of this technique, as it starts approaching the work needed to design supports and brackets in an empty box.

If there’s an upside to the wasted material, it is the fact that this glue-less technique can be easily disassembled. When we’re done evaluating this prototype, every sheet of acrylic can be reused as material for future (necessarily smaller) projects.

Lesson learned: This “stacking plates” construction technique offer a trade off of reducing design time and effort at a cost of reduced material usage efficiency.

Enclosure2_Complete

Testing Heat-Set Inserts in Acrylic

As a beginner playing with plastic fabrication on a 3D printer, I hadn’t known about heat-set inserts for putting durable and reliable threads in plastic construction. In all my projects to date, I tapped threads into the plastic directly and made sure to be careful when tightening a screw threaded into plastic. The inserts look like a much, much better solution and they are easily available from hardware vendors like McMaster-Carr.

Before I put in an order, though, I wanted to do a quick experiment. I salvaged some M2.5 heat-set inserts from the dead Dell laptop, and I laser cut holes of various diameters into a scrap piece of 3 mm acrylic. When the hole is too large, the result seems to be obvious: insert will be unable to grip tightly. It’s less obvious to me what happens when the hole starts becoming too small. Recognizing the symptoms will help me determine proper diameter for future applications.

HeatSetTests

For their M2.5 inserts, McMaster-Carr recommends drilling a hole .152″ in diameter. This translates to about 3.86 mm. The largest hole in this test piece is nominally 3.75 mm, but with laser kerf will end up closer to 3.91 mm. The hole labelled 3.7 would, after laser kerf, end up right on the dot at 3.86 mm.

The experiment showed that they will all suffice to hold the insert into the acrylic, so in practice there is some amount of tolerance for the diameter precision. As the holes got smaller, more heating is required to install the insert, and more acrylic is visibly distorted around the insert due to the additional heat. Fortunately optical clarity seems to be mostly preserved, the distortion is barely visible in the above picture.

Once I got down to around “3.5” (actually ~3.66 mm with kerf) I started seeing the insert pushing plastic out of the way during installation. This results in a small ring of excess plastic around the base of the insert, which is undesirable. This is a good enough marker for “too small” and I stopped there. The holes smaller than “3.5” remain unused.

Experiment complete: In the future, the combination of optical distortion and excess plastic at the base will serve as my first warning sign that I’m installing heat-set inserts in too small of a hole.

Thread Tapping Failure and Heat-Set Threaded Inserts

Part of the design for PEM1 (portable external monitor version 1.0) was a VESA-standard 100 x 100mm pattern to be tapped with M5 thread. This way I can mount it on an existing monitor stand and avoid having to design a stand for it.

I had hand tapped many M5 threads in 3D printed plastic for the Luggable PC project, so I anticipated little difficulty here. I was surprised when I pulled the manual tapping tool away from one of the four mounting holes and realized I had destroyed the thread. Out of four holes in the mounting pattern, two were usable, one was marginal, one was unusable.

AcrylicTappedThreads
Right: usable #6-32 thread for circuit board standoff. Left: Unusable M5 thread for VESA 100 monitor mount.

A little debugging pointed to laser-cutting too small of a hole for the tapping tool. But still the fact remains tapping threads in plastic is time-consuming and error-prone. I think it is a good time to pause the project and learn: What can we do instead?

One answer was literally sitting right in front of me: the carcass of the laptop I had disassembled to extract the LCD panel. Dell laptop cases are made from plastic, and the case screws (mostly M2.5) fasten into small metal threaded inserts that were heat-set into the plastic.

Different plastics have different behavior, so I thought I should experiment with heat-set inserts in acrylic before buying them in quantity. It doesn’t have to be M5 – just something to get a feel of the behavior of the mechanism. Where can I get my hands on some inserts? The answer is again in the laptop carcass: well, there’s some right here!

Attempting to extract an insert by brute force instead served as an unplanned demonstration of the mechanical strength of a properly installed heat-set insert. That little thing put up quite a fight against departing from its assigned post.

But if heat helped soften the insert for installation, perhaps heat can help soften the plastic for extraction. And indeed, heat did. A soldering iron helped made it far easier to salvage the inserts from the laptop chassis for experimentation.

Portable External Monitor 1.0

LCD Enclosure 1 piecesOnce the LCD panel and matching frame had been salvaged from the laptop, it’s time to build an enclosure to hold it and the associated driver board together. Since this was only the first draft, I was not very aggressive about packing the components tightly. It’s merely a simple big box to hold all the bits checking to see if I have all the mounting dimensions for all the circuit boards correct.

It was also the first time I had the chance to try acrylic sheets in a color other than clear. There was a dusty stack of 6 mm green acrylic that I enlisted into this project. Since this is just an early draft project, I valued ease of construction over appearance or strength (6 mm is more than sufficient) and so I used the interlocking tab design for self-aligning assembly.

The resulting box was functional, but not very interesting from a design viewpoint. I just wanted to prove that all the components worked together before I proceeded to the next draft.

I did not design this enclosure to stand by itself. Instead, I had designed this enclosure with a VESA standard 100x100mm mounting pattern in the back and intended to tap those laser-cut holes to take M5 fasteners. Once so prepared, I can mount this enclosure on any existing stand that conforms to the standard. That little design detail – independent of the LCD panel and driver board – sent me off on a little side exploration of plastic construction techniques.

That is a story for the next update.

LCD Panel Frame From Laptop Lid

Drilling out plastic rivetsMy Luggable PC display was a LCD panel I had salvaged from an old laptop, which I’m doing again for this external monitor project. When I pulled the Luggable PC panel out of the old laptop, I left most of the associated mounting hardware behind. During the Luggable PC project I wished I had also preserved the old mounting hardware.

The first reason is dimension data. When I mounted the screen to my Luggable PC, I had to measure the panel and design my frame to match. A Dell engineer did this work years ago, and when I threw away the mounting hardware, I threw that away as well.

The second reason is strength. A LCD panel is fragile, but when backed with its sheet metal frame, it becomes quite a bit stronger. This is usually a worthwhile trade off against the increased size and weight.

The third reason, less obvious than the previous two, is to manage heat. The back light assembly across the bottom of the screen would get quite hot when the panel is just sitting by itself. However, when the panel is mounted in its frame, the frame served a secondary purpose as heat sink.

The metal frame I want to reuse is attached to the plastic outer cover of the laptop lid. The attachment is done via small plastic rivets: bits of the plastic lid cover melted into the metal frame. Pulling off the frame with brute force is likely to bend and damage the frame, so the assembly is put under the drill press. After cores of all of the plastic rivets were drilled out (above), the metal frame easily pops off the plastic lid cover (below).

Frame freed

The metal frame can now be used to build the rest of the enclosure. The frame can be cut, drilled, and generally manipulated in ways that I would never do to the LCD panel itself. And when I’m done with all the prep work, the panel itself will drop right in to the frame. This should be much easier than what I had to do for the Luggable PC screen.

Portable External Monitor Project

Panel and DriverThanks to a friend’s generous donation of a nonfunctional Dell Inspiron E1505, I have another LCD panel to play with. (And distract me from FreeNAS Box project.) The eventual ambition is to upgrade my Luggable PC to a multiple-monitor system but as a first step, I’ll learn to work with the new panel by turning it into a portable external monitor. If phase 1 is successful, it becomes an optional additional accessory I can lug alongside the Luggable PC. Then, if I’m feeling ambitions, I can move on to phase 2 of integrating everything into a multi-monitor Luggable PC.

The first order of business is to extract the panel and look at its specifications. Dell laptops around that vintage offer resolution as low as 1024×768, which wouldn’t even be worth the effort to resurrect. Fortunately, thisĀ LG Philips LP154W02 panel has a decently respectable 1680×1050 resolution.

Since the computer doesn’t work, next I have to see if the panel does. Off to Amazon to look for boards that claim to drive this panel. The first board (pictured here) was able to present all the right info to a computer, and it can power the back light, but no picture showed on screen. At this time I was worried – did I get a bum board, or is the display dead?

After some diagnostic chatter with the seller, I was pointed to another board they carried. The first board was computer-focused, the second board was more TV and media focused. The upside is that it worked. The downside is that it seems to have trouble preserving 1:1 pixel information at 1680×1050. I wonder if its media-focused nature meant it up scale all signals to 1080p and then down sample to the panel resolution. That would explain the minor visual artifacts.

The artifacts does take a bit away from the success. But it’s lit, it shows a picture, and that’s good enough to proceed.