My Winter break took an unexpected turn when two keys on my laptop stopped working, after a minor water “incident”. What initially seemed like a series of repeated typos, would eventually lead me on a multi-day journey to tear apart my laptop and its keyboard to find and repair the culprit.
This is the story of that quest and all it’s struggles, written entirely on the keyboard that’s been given a second chance at life.
- Getting Started
- Laptop Tear Down
- A Quick Note on Testing
- Getting to the Membrane
- Opening the Membrane
- Fixing the Bad Traces
- Reassembling the Keyboard
- Putting the Laptop Back Together
This post details my experience tearing down my Razer Blade 14” (2013) laptop, to repair a few keyboard keys that became unresponsive after some minor water damage.
Getting to the keyboard itself wasn’t very difficult, due to Razer’s repair friendly design. However, opening up the keyboard to repair the damaged circuit traces was extremely labor intensive and tedious. I spent several hours poking and prodding the keyboard’s housing to figure out the best way to open it up, so that I could actually diagnose and repair the broken traces. Putting everything back together was another struggle in discovery, but I eventually came up with some hacks to get it all reassembled so I would hopefully never have to open it all up again.
The challenges of the repair were greatly compounded by the fact that I had absolutely no guidance for much of it, and was forced to rely on experimentation and intuition as I went along. Hopefully, some of what I learned might assist others doing a similar repair, so that they may avoid some of the mistakes I made along the way.
Skill Level: Moderate
Duration: 5-10 hours
This was one of the more difficult electronics repair I’ve attempted, and definitely not a project I would recommend to very many people. However, if you have the patience, a free afternoon (or two or three in my case), and some basic electronic repair tools and knowledge, it should be possible to repair a few unresponsive keys on your Razer Blade keyboard.
- Precision screwdriver set (phillips and T5 heads)
- Multimeter (with continuity test)
- Conductive ink pen (ex. CircuitWriter)
- Hair dryer (heat gun)
- Small clamp
- Thermal paste (ex. Antec Formula 7 or Artic Silver)
- Compressed air
- Insulating Sealant (ex. CircuitSealer)
- Needle nose pliers
- Fine tipped tweezers (curved)
- External keyboard and mouse (for testing)
- External monitor (for testing)
This post is meant to detail my own experiences, and should not be considered as advice (sound or otherwise) for what you choose to do with your own laptop. I take no responsibility for any damage you might cause to your system while attempting this repair, or one like it.
The fine circuitry and components inside your laptop are delicate. requiring specialized tools and knowledge to properly disassemble without damage. Following any of the procedures discussed here will void any warranty and service agreement(s) you may have on your device.
Please exercise extreme caution ,if you attempt this repair or similar on your own laptop.
And as always, remember to have fun!
Please read through this entire guide before deciding to attempt any of these procedures, as it gets more complicated as it goes along.
Keyboard Replacement Options
Before starting the repair, I researched options for replacing the broken keyboard entirely.
It seems Razer’s official support offers to replace broken keyboards for somewhere between $250-$350, based on forum posts I found (shipping is likely extra). There doesn’t appear to be any option to purchase spare components directly from Razer support, so you’d have to send the entire system off for repairs, with a reported turn around of 2-3 weeks.
I also tried searching for any 3rd party dealers that sell Razer Blade components, but didn’t turn up much of anything. I did spot a couple closed Ebay auctions for the entire base (palm rest, keyboard, and touchpad) for around $100-$200 (like this one). A bit cheaper than the repair quotes from Razer Support, but still not much more appealing (and also not even an option as no auctions were active!).
In the end, the only valid replacement option seemed to be sending the entire laptop off to Razer’s support site for a few weeks and shell out a few hundred bucks for the repair. Not a terrible fall back, but a bit wasteful and the 2-3 week turn around was unappealing.
So, I decided to try my hand at doing the repair myself!
Laptop Tear Down
The repair starts by disassembling the laptop into it’s major components.
Either follow along below, or feel free to check out one of these alternate guides and then meat us back here to continue the repair.
Getting to the Keyboard
The only way to get into the laptop’s case is via the back panel, which means we’ll have to remove most of the major components to get the keyboard. Luckily, Razer’s engineers made this process about as easy as I’ve seen in such a thin laptop, so it’s not as bad as it might appear.
Remove Back Panel
Opening the laptop begins by flipping it over to remove the back panel. The panel is held in place with several T5 screws along the perimeter, that must be removed. With the screws removed, the back panel comes off freely (no latches or plastic clips to battle).
Disconnect Keyboard Ribbons
Once inside the case, the the next step is to disconnect the two laptop ribbon cables located above the battery pack (see Image). Using a small flathead screwdriver (or fingernail), gently pull up from the bottom of the black plastic clip to lift it up until it is perpendicular to the motherboard, then slide the cables out.
Be careful to only apply a light pressure, as plastic clips are known to snap easily if forced (especially the tiny plastic tabs holding the clip in place).
Test Point: Check for a possible faulty connection between the ribbon cables and connectors on the motherboard. Try cleaning out the connectors with a few blasts of compressed air. Then, using a clean rubber eraser (the regular pink kind), gently rub the contacts on the ribbons once or twice to remove any oxidation or other build up that may be present. Check if this fixes your issue before continuing the tear down.
Disconnect Power Cable to Motherboard
On the right side of the battery, behind the keyboard ribbon we just disconnected, is another connector attaching the battery to the motherboard. This connector has no clip, so just use a tiny bit of force to slide the cable out. Avoid pulling on the wires!
Disconnect Cable to Status Indicator
Next up, the white cable running over the battery connecting the Status indicator LED to the main board (see Image).
This connector has another simple latch like the keyboard ribbon cables, which will also “pop” up with a little bit of force from a flathead screwdriver. Once safely removed from the connector, slowly peel the cable away from the battery since it is taped down with a bit of double sided tape.
Remove the Battery from it’s Compartment
Along the perimeter of the battery, there are several small phillips head screws which need to be removed. With all the screws removed and cables disconnected, the battery should slide right out of it’s compartment.
Disconnect Keyboard Backlight Ribbon
Now that the battery is out of the way, we can get to the small ribbon cable for the keyboard backlight LEDs (see Image). The clasp for this cable is a bit different than the previous ones, and both brown side tabs get pulled down (or back) to release the cable.
Remove the Heatsink
The heatsink is attached to the motherboard with 6 screws located in the top center of the board (see Image).
Instead of removing each screw individually, one by one (like normal), alternate between all of them, unscrewing each only a single turn or two at a time, until all have been removed. This will help keep the pressure evenly distributed across the chips to avoid cracking or chipping them.
Once all 6 screws have been removed, the heatsink should pull straight up and off (see Image). The factory thermal paste can be a bit sticky, but just apply a constant upward pressure to gently pry the heatsink away from the chips. You may need to twist it ever so slightly to help separate the bond, but try and keep it to a minimum.
Eesh, what a mess!
I’m not sure why I expected the factory application of thermal compound to not be as messy as every other manufacturer I’ve ever seen (maybe because of the superb build quality thus far?), but I was a bit disappointed none the less. It does appear to be a higher quality compound than most manufacturers use, which likely explains several articles I’ve seen that saw disappointing improvements from redoing the thermal paste with after market compounds.
Note: The thermal compound on the chips and heatsink is conductive, so take care not to get any of it onto the circuitry or inside the case.
Unscrew Speakers and Fans
The speakers are located to the left and right sides of the motherboard, and are held in place with two screws and a small cable that connects them to the motherboard.
Next, remove the two fans at the top of the motherboard. Each fan is held in place with 3 screws hidden below little rubber domes that pop right off. I chose to leave the fan wires connected to the motherboard, since they will need to be connected during testing to keep the chips from overheating.
Remove the Motherboard
Now the only thing standing between us and the busted keyboard is the motherboard.
There are several screws around the perimeter and a few scattered throughout the middle of the PCB. Once all the screws are removed, the motherboard should pull up freely from the bottom edge, although I had a bit of resistance getting the audio jack free on the right side of the case.
I chose to leave the motherboard connected to the screen and just rotated the board up 90 degrees to access the keyboard underneath, but you may prefer removing it from the case entirely. Based on the number of tests I ended up doing, I would probably remove the motherboard entirely from the base, if I did it all over again.
Removing the Keyboard Assembly
Finally, we can get to the keyboard we’re after we came for.
Although be warned, it doesn’t come out as easily as the previous components.
Pull Away Back Plastic
Most of the backside of the keyboard (except for where the fans were located) is covered by a layer of protective black plastic. This plastic must be pried away from the keyboard in order to access the screws holding the keyboard in place.
There’s not really a good secret I found, just slowly and patiently pulling the plastic off, until it’s entirely removed from the keyboard.
Remove the Screws Holding Keyboard Down
Once the backing has been removed, you’ll see dozens of super tiny screws holding the keyboard in place. They all need to be removed (yes ALL of them).
Get to it.
Pull Out the Keyboard
The keyboard will now slide out of the case, although you have to lift it slightly to get the key caps out of their holes.
Take a moment to check it out, as it’s sort of cool to play with.
The next steps are very tedious, so take a moment to check out the keyboard (it’s sorta cool to play with) and take a break if you need one.
A Quick Note on Testing
Testing the keyboard while the case is disassembled requires some special attention.
Before powering the system on, the heatsink and fans must be be in place to cool the CPU and GPU. I found that I could get away with resting the heatsink back in place (with the old thermal compound still present) with the fans in their compartments. The thermal compound helped hold the heatsink in place, but it wouldn’t last for very long before the heatsink started to pull away.
While not a “recommonded” solution, I developed a system where I would push the heatsink into place and then perform my tests until I heard the fan motors kick into overdrive as the temps started rising. Then I’d stop and push the heatsink back down until the fans died down after a few seconds, before resuming testing. Again not ideal, but modern chips have throttling circuity to prevent permanent damage from high temps, that should keep things from getting out of hand.
Getting to the Membrane
The easy part is over.
From here on out, nothing is intended to be disassembled, repaired, or rebuilt. I had to chart my own path for opening the keyboard up, and I made a few mistakes along the way.
Hopefully this guide will make things a bit easier for anyone else out there attempting this. However, keep in mind that everything below was made up as I went and there are likely better ways to do some of this, so please don’t just follow along blindly.
Remove all the Key Caps
All the key caps on the keyboard must be removed. Yes, I’m aware there are a lot. :D
This part is tricky because the retaining clips underneath the keys that hold them in place have itty-bitty plastic hinges that are very easy to snap off (see Image).
Start with one of the normal sized alpha-numeric keys, as they are the easiest to remove.
I used a small flathead screwdriver to slowly lift up from the bottom of each key popping them free from their clasp, and then sliding them up and out of the top hinge (see Image).
The first one you may just need to force off a bit, so that you can inspect the retaining clip mechanism yourself to refine your technique. Once you’ve gotten through a few, you should develop a good feel for it and each cap will only take a second or two. Remember to be diligent though, because those hinges are really really easy to break.
I’d recommend leaving all the special keys (F-keys, Arrows Keys, Enter, Shift, Caps Lock, and Space) last, as they are a bit more challenging. The “long” keys are especially bad since they have retaining rod (or two) and some even have 2 retaining clips! (see Image)
Remove all the Retaining Clips from the Keyboard
Depending on your method for removing the keycaps, you may or may not still the black retaining clips attached to the keyboard. If the clips came off with the keys, then you can skip this step.
Otherwise, start carefully removing all those clips from the keyboard, trying not to break any of their tiny plastic hinges. I found a small set of curved tweezers quite handy for this job.
Remove the LED Backlight Sheet
With all the kepcaps removed, the black sheet with all the LEDs for the keyboard backlight should come right off, as it was only held in place by the key caps. You should now have a bare keyboard with squishy rubber domes for each button and a bunch of visible circuitry (see Image).
Test Point: At this point in my tear down I had no clue what the actual problem was, and spent a lot of time probing and examining the rubber dome buttons. I’d recommend trying to clean out under the domes corresponding to broken keys with compressed air, and anything else that comes to mind, as the next steps are invasive and will leave permanent changes (read: scars) to your still pristine (albeit disassembled) keyboard.
Opening the Membrane
I sat at this step for awhile during my repair, unsure of how to proceed and whether I should throw in the towel and just send off my laptop for repairs. After a long break, I concluded there was no real harm in attempting a fix, since I could always send it in to get the keyboard replaced if I caused irreparable (or just failed to fix it).
This section is invasive and will permanently alter the state of your keyboard’s inner membrane. While it’s necessary to open up the membrane to get at the traces for each key, it was clearly engineered to remain sealed, never to be opened after being assembled.
Please proceed at your own risk
The 3 Membrane Layers
The internal function of the Razer Blade keyboard is very standard for a rubber membrane keyboard. There are 3 layers (or sheets) that are stacked on top of one another to build the circuitry. (see Image)
The top layer contains the rubber membrane that is used to form the key domes on it’s top surface and a set of traces for transmitting the button signals down to the ribbon cable on the reverse side. The bottom layer contains a long trace that routes the power signal (Vcc) to every button.
To keep the top and bottom layers from being in constant contact, a middle layer is included to act as an insulator between the top and bottom, with just a small hole where each button can squeeze the top and bottom traces together to complete the circuit.
The challenge with this part of the repair is that unlike typical rubber membrane keyboards, Razer’s engineers decided to glue the 3 layers permanently into place.
This was likely a result of Razer support’s policy of replacing keyboards instead of repair and to help avoid moisture and other particles getting between the 3 layers and shorting the circuitry or block the signal (a lot of good that did ;) ).
Regardless their reasons, the layers are glued together and must be separated to get at the individual traces for the keys.
Peeling Away the Bottom Layer
The first layer I peeled away ended up being the bottom layer. At the time, I wasn’t even aware that there was a middle layer, it just happened to be where the peel naturally started.
I first attempted to separate the layers from the natural edge around the membrane, as it’s sealed together very well. After struggling for awhile, not sure if the membrane could be opened (or if there even were layers), I decided to make an invasive cut.
With a pair of wire cutters, I found an area of the membrane where there were no traces and nothing else of importance that appeared greyish in color (possibly an internal gap?), and gave it a snip. This proved successful as I was able to get my tiny tweezers into the small gap and start widening it. Once I had my foothold, it was only a matter of time before I had pulled away enough of the bottom layer to start peeling it away by hand, and it went pretty smoothing from there.
With the bottom layer removed, you can now clearly see the two different traces (see Image). The traces on the top layer go from each button to one of the contacts on the ribbon, while the traces on the bottom layer are really just one long trace all leading to ground V+ (Vcc).
Test Point: At this point, the traces for V+ (Vcc) are exposed on the bottom layer. After an initial inspection for broken or oxidized traces, I did a continuity test with my multimeter from the junction (see Image) to each of the pads that sit beneath all the buttons (paying special attention to the traces leading to the unresponsive keys). In my case, all these traces checked out fine. (see First Examining a Working Trace for more detailed explanation of testing continuity)
The Second (Harder) Peel
For me, the bottom trace checked out, so my journey continued onward towards the upper traces.
Exposing the upper traces required peeling away the blue plastic of the insulating middle layer (see Image). Again with tiny tweezers, I focused on a corner and carefully worked the blue plastic of the middle layer away from the top layer. It took awhile to get started, but I was eventually able to peel enough away to grab on and start peeling by hand.
For me, the two bad keys where both on the left side of the keyboard (right side in most images because everything’s upside down), so I only peeled the middle layer away from the one side to the innermost broken key. The whole process was slow going, as the middle layer was quite thin and brittle.
Fixing the Bad Traces
With the keyboard split open and all the circuitry exposed, we can finally begin the actual repair on the bad trace.
First Examining a Working Trace
I started by examining a working button on the top layer, switching my multimeter to continuity mode.
With one multimeter probe touching the button’s pad on the data line trace on the top layer, I touched the other probe to a nearby location on the same trace, “Beep!”. Continuing with the first probe on the pad, I continued testing points further along the trace until reaching the ribbon cable, waiting for a “beep” each time.
I then slowly moved the second probe across all the contacts on the tip of the ribbon cable, searching for the contact that led back to the button’s pad. I’m not sure why it took a few tries, but I finally heard a beep during my sweeps across the ribbon connector at a consistent spot, indicating the trace was continuous from button to the contact on the cable.
Note: Remember to be mindful of your multimeter’s speed at determining continuity, while testing. To check the speed, simply touch the two probes together, and see how long before it beeps. My nice Fluke meter is nearly instantaneous, but many off the cheaper meters you’re likely to find off the shelf can take half a second or more to detect continuity.
Locating the Breaks
Confident that my multimeter was correctly detecting continuity along the entire data line trace for a working button, I turned my sights to my 2 broken keys. And sure enough, both of them failed the continuity test from pad to ribbon contact, showing the first glimpse into the root problem.
To locate exactly where the break occurred, I started testing continuity along various sections of each trace. I quickly discovered the entire trace from the pad to where the ribbon cable begins was good, since that section of the trace was easy to probe. I also found that I could not get a signal from where the ribbon cable begins to the ribbon cable contacts, indicating the break was somewhere “inside” the ribbon cable. Eek!
However, upon closer visual inspection, I noticed a bit a discoloration (brown with a touch a blue) right at the juncture where the ribbon is attached for both of the faulty traces. I figured this could be some oxidation that was triggered when I spilled water into the keyboard. I wasn’t sure, but the prospect of the break being so close to the end of the ribbon cable was great news, since I’d had enough peeling already.
Exposing Traces Along the Ribbon Cable
Both breaks seemed to occur right where the ribbon cable begins, so the next step was to peel away a bit pf the ribbon cable’s insulation to expose more of the traces.
The two sides of the ribbon were held together with a pretty strong adhesive, so it was less of a “peel” and more of a “scrape” in actuality. I focused exclusively on the section of the ribbon where the bad traces where. Luckily, both of the traces where on the outside of a trace “grouping”, allowing me to only worry about a neighbor trace on one side.
After exposing a bit more of the trace, I quickly saw what seemed to be good trace again just a millimeter or so below the discolored area. I pulled out my multimeter again and did a continuity test touching one probe to the newly exposed trace while running the other across the ribbon contacts. Nothing on the first sweep, and then “BEEEEP”!
Moving to the second faulty trace, I found nearly identical results, with the break also being located right at the junction where the ribbon cable is attached (see Image).
It seemed the discoloration was a bit of oxidation after all. and I wouldn’t have to completely tear the ribbon apart to locate the break.
Repairing the Broken Traces
I tried cleaning the oxidation from the bad traces with my eraser, but it didn’t seem to have any positive effect. So, I decided to just make a new trace over the broken area.
I bought a conductive ink pen (CircuitWriter) from my local Fry’s Electronics store, and started writing a new trace over the bad area with it. The pen makes the application very easy, although you still need to be very careful to avoid touching any neighboring traces.
At first I thought the pen didn’t work, as my continuity tests were failing, but it was because the ink isn’t very conductive until the agent that makes it liquid evaporates away to leave a deposit of silver particles. It seemed to dry in about 10-15m, but I was able to speed up the drying time to a minute or so using a hair dryer.
With the conductive ink in place for the two broken buttons, I did my continuity test again over the entire length of the data trace, and it beeping worked!
Testing the Fix
To test the repair, I had to partially reassemble the keyboard.
I laid the blue middle layer back down enough to put the back layer in place, and just just sort of held everything together with my hands as best as I could. It was a bit awkward to get the ribbon cables reattached to the motherboard with everything loose, but I managed to get the keyboard reconnected and fire up the system.
Annnd… No response from the keyboard at all. :’(
Confused, disappointed, and frustrated I started to frantically try and figure out what went wrong.
A-ha, the junction between the top and bottom layers needs to be in contact, since the V+ (Vcc) comes through the ribbon cable and must be routed from the top layer to the bottom layer. I found that simply pushing the two layers together between my fingers (see Image) while testing the keyboard buttons was enough to maintain the connection and test the trace repairs.
To test each key, I found a handy website keyboardtester.com that displays each key as it’s pressed. One by one, I tested every single key, including the broken “C” and “3” keys, and one by one they each passed. Except for my “F7” key, which was unresponsive.
Shit, I made a rookie mistake and didn’t test for additional broken keys outside of the 2 I knew were broken when I started.
A quick check on the circuit after powering down the system, and the “F7” trace had the same brown discoloration at the same area where the ribbon cable is attached. I repeated the my previous fix for the “C” and “3” keys, and “F7” joined the rest of my keys in the land of the living.
Sealing the New Traces
To help prevent additional shorts and breaks, I decided to be be responsible and add a bit of insulation over my new traces.
While at the Fry’s to pick up a conducting ink pen (CircuitWriter), I also grabbed an insulating sealant by the same manufacturer on the shelf next to it (CircuitSealer). It worked exactly like the first pen, except instead of a conducting silver material it writes with a clear sealant.
A few dabs over the exposed trace and a blast from the hair dryer to dry it, and it’s like I never picked away part of the ribbon’s top layer (well, electrically at least).
Reassembling the Keyboard
With the traces repaired, the only remaining task was figuring out how the hell to put everything back together. The keyboard internals were definitely not designed for being opened up (especially by an end user), and my lack of experience had definitely left a toll on some of the membrane layers I had to forcibly separate.
Membrane Layers Back Together Again
I started reassembling the membrane layers by rolling the blue middle layer back down as nicely as possible. The adhesive left behind from did a reasonable job of holding it in place and taming some of the curl it had developed during the peel, but it was still far from laying flat.
Once I got the middle layer as good as it was going to get, I carefully placed the bottom layer on where it belongs and again spent some time smoothing everything out by hand. Since the glue residue seemed to be sort of holding everything in place, I got the grand idea to blast it with the hair dryer for a bit to try and reactive some of the adhesive. A few minutes with the hot air while smoothing everything down, and everything was holding together much better than I had imagined.
As a final step, I decided to add some masking tape around the edges to help keep everything from sliding around once back in the case (see Image)
Reconnecting Junction Between Top and Bottom Layers
The old adhesive seemed to be doing a good job of holding the keyboard back together so that the button switches functioned, but the junction between the top and bottom layers was not making a solid connection.
I decided to try and “meld” the junction back together by setting the membrane in a clamp overnight (see Image).
My first attempt at getting the junction worked for awhile, but the connection broke during my final testing. The second time I applied the clamp after heating the area extensively with a hair dryer, hoping to reactivate whatever used to hold the layers together, and it seems to hold a lot better.
Putting the Keys Back on the Keyboard
With the membrane successfully rebuilt, we can start reattaching all the key caps back to the keyboard. Just don’t forget to first put the LED backlight sheet back in place between the membrane and keys like I did, or you’ll need to take the caps off again. :/
The process is basically the reverse of how the caps were removed, and you’ll quickly find a good technique to put the standard keys on without much fuss. The long “special” keys are still a pain in the ass, but there aren’t that many of those.
Final Tests Before Rebuilding the Laptop
Before reassembling the laptop, I performed one last final test to ensure all the keyboard keys were functioning before sealing everything back up, and the keys were still all functional.
All systems are go!
Putting the Laptop Back Together
Having completed the hard part of repairing the faulty traces inside the keyboard, the only thing left before enjoying the fruits of my labor is putting the laptop back together.
Keyboard Goes Back in the Case
The keyboard slides back into place under the motherboard, just as it came out.
Next, screw in all of the super tiny screws to hold it in place nice and tight. Once the screws are in place, put the black plastic backing on as best as possible. I gave the backing a blast with my hair dryer once it was down to help reactivate the glue and smooth it out.
Reattach Motherboard and Accessories
Screw the motherboard, speakers, and fan back into place, just how you found them during the tear down.
Remounting the Heatsink
If you’re comfortable mounting heatsinks, you can safely skip over this section, as there’s nothing out of the ordinary with this one.
Remove the Old Thermal Compound
Armed with some Q-tips (or paper towel) dipped in acetone, the old compound should remove easily with a few passes. Be careful while removing the compound, as most pastes are conductive, and you don’t want to get it on any of the fine circuitry around the chips.
Make sure to clean each surface entirely, removing all the old compound and any other bits of dirt and dust you see. The final result should be a nice mirror finish on the chip heat spreaders and shinny copper on the heatsink.
Apply new Thermal Compound
I decided to try out a new thermal compound, and ended up with buying a tube of Antec Formula 7. It has diamonds, and all signs seem to indicate this new wave of compounds with nano diamond particles are a bit better than the always reliable Artic Silver compound and the harder to apply ceramic ones.
Most people have a preferred method for applying thermal paste, and I’m a stickler for spreading a razer thin layer across the heat spreader. If you’re doing the razer thin layer technique, you’ll need some sort of straight edged spreader, such as an old credit card.
Begin by squirting a small dot of compound (size of a grain of rice) onto one of the copper heatsinks. Then, using your spreader, grab a small amount of compound and transfer it to one of the chip’s metallic heat spreader. Start out with less compound than you’ll need (you can always go back for more) and spread it across the heat spreader on the chip. The goal is to cover the entire surface of the chip’s heat spreader with a thin continuous layer of compound (thinner the better, but don’t make it “streaky”).
After finishing one chip, do the same for the other one, so that both have a nice, thin, smooth layer of compound.
Lastly, smear the remaining compound around the surface of the copper pads with your finger covered with some plastic wrap (don’t want your sweaty finger greases mixing into the paste). The goal is to get the compound into the micro crevasses and imperfections on the copper surface.
Once you’re satisfied that you’ve coated the heatsinks well, lightly wipe the compound away from the copper heatsink, leaving behind just enough compound to give the copper with a dull or greasy appearance.
Gently Reattach the Heatsink
With new thermal compound in place, it’s time to reattach the heatsink.
The goal is to place it gently onto the chips as straight up and down as possible, so as to not disturb the nice layer of thermal compound we just spent so long smoothing across the entire heat spreader on the chips. Just a nice gentle “plop” right onto the chips, after aligning the screw holes, trying to avoid all “twisting” motion.
Once the heatsink is “down” start screwing the 6 mounting screws back into place, trying not to move or disturb the heatsink. You’ll want to alternate between each screw, turning each only a a single turn or two at a time, to reseat the heatsink with constant pressure.
Then repeat this process until each screw lightly “bottomed out”, before backing them up a quarter turn to let the springs apply the correct pressure.
Reinstall the Battery
Just like it came out, put the battery back into it’s compartment and screw it down
Reconnect all the Cables
All the components are in place and screwed down, so we can now connect everything back together. Reattach the battery connector, the 2 data ribbon cables to the keyboard, the ribbon powering the keyboard backlight, and the ribbon cable connecting the status LED over the backlight.
Put the Back Panel Back
We’re almost there!
Just put the back panel back on with the screws from before and we should be done with all assembly. A final test to make sure we didn’t screw anything up, and we can call this a success!
This was quite the journey, especially considering I thought I could knock this out in a single afternoon. It still find it a bit humorous how a small piece of plastic and a simple button (which together can’t cost more than a buck or so) can nearly render a thousand dollars worth of electronics worthless.
Overall my Razer Blade 14” laptop seems to be holding up quite nicely, with several days of usage since the rebuild. A few keys aren’t quite as crisp as they were, since I damaged several of the retaining clip hinges while removing and reattaching the key caps. Although it’s really not noticeable while typing unless you hit the key “just right”. I think if I had this guide when removing the caps, they would have faired a little better, but it’s really not a huge issue for me.
You could make the economic argument that it would have been cheaper to ship it off to be repaired by Razer based on how long it took to fix myself, but I found the project itself deeply satisfying and educational.
It was interesting to find that all 3 unresponsive keys had a broken trace right where the keyboard’s ribbon cable attaches. It’s possible the water was localized to that area, but I suspect the keyboard’s membrane just had a better seal and the area where the ribbon attaches was partially exposed. Perhaps Razer’s engineers would be interested in these findings (or perhaps they’re already aware of the issue)?
In the end, I’m just thrilled to have a working keyboard again.
Half way done with the chess pieces, and now I’m ready to just wrap it up. Well, except for the knight… I’m looking forward to that one, which is why it’s saved for last. There’s nothing particularlly special about today’s rook, besides it being a required part out the set.
As this was more an excersise of previous techniques, I turned it into a mini time challange for myself. At just over half an hour, it’s one of the quicker models I’ve made to date, and the shortest from the chess set.
Total Time: 33 minutes
[ download: chess_rook.blend ]
Following the chess theme, next up is the queen.
If you’ve been following the progression from the pawn and bishop, you may have noticed the pattern of progression is going from the simple pieces to the ones with the most challenging shapes to build. The complexity in the queen comes from her crown, and it was a fun challenge to tackle. I had to try a couple techniques because the first one resulting in points that were.. well uhm, too pointy I guess.
Speaking of redoing things, remember to bump your undo limit all the way up in your “user preferences”, because 32 is insufficient for when you mess up your crown (and mess it up you will). Also, save, save, save!
Total Time: ~1 hour
[ download: chess_queen.blend ]
Happy 4th of July! I’m celebrating over here by venturing into VFX with absolutely no guidance or proper supervision (ahh, found memories of lighting of fireworks in the neighborhood :-Þ). While I managed to not mess things up horribly, I should probably find some tutorials before playing around with particles again.
Total Time: ~45 minutes
[ download: sparkler.blend ]