Left Alt fixed with pics - CM Storm QuickFire Rapid

[khaangaaroo]

I ordered a QFR with reds from Amazon Warehouse Deals and it had a non-working left alt. Instead of RMAing it, I decided to open it up and tackle the problem, for the sake of science! Just as some people have guessed, it was caused by a bad solder joint.

I'm sure a lot of bad boards have already been returned to Cooler Master and newer batches have had this issue resolved by now. But I decided to take some pictures and make a post anyways in case people run into an intermittent alt key in the future outside of warranty, or live overseas where RMA shipping would be ridiculous.

Here's a look at the PCB underneath the left alt.


A closeup of the bad joint. You can see the line from the bottom left breaks before it reaches the joint.


I first tried to add a little bit of solder to see if i could close the gap, but it came clean off.


I scratched off the green coating to expose the tips for resoldering.


The gap was too big, so I had to use a bit of copper wire to bridge the pins.


Soldered and fixed.

[tp4tissue]

HURRAY!

Glad to see a DIYer.. rather than the usual whiny lil' B%&ch$

these problems are so incredibly prevalent in low tech electronics that manufacturer should ASSUME the responsibility of the fix will go to the end user.

Consumers who are babied turn into iSheep... they can't do anything... besides whine...

[npgatech]

Awesome fix....the hard part was to spot the dodgy soldering :-).

Just shows that CM folks are cutting costs by using inferior soldering process. Wondering if they have anywhere close to 500 DPM (Defects per million) which is considered to be extremely good and difficult to achieve in a production line. Most companies like Apple, Intel and Logitech probably have 0.005% with a confidence level of 95% (or 50 DPM) defect rate or better. I am guessing CM has over 10k DPM quality control by just the looks of this problem. Also their solder joints don't look shiny...a clue that it was probably a drifting soldering temperature in their machines (probably around 650 F).

[tp4tissue]

Awesome fix....the hard part was to spot the dodgy soldering :-).

Just shows that CM folks are cutting costs by using inferior soldering process. Wondering if they have anywhere close to 500 DPM (Defects per million) which is considered to be extremely good and difficult to achieve in a production line. Most companies like Apple, Intel and Logitech probably have 0.005% with a confidence level of 95% (or 50 DPM) defect rate or better. I am guessing CM has over 10k DPM quality control by just the looks of this problem. Also their solder joints don't look shiny...a clue that it was probably a drifting soldering temperature in their machines (probably around 650 F).

For HALF the price of a filco.. I'll fix it myself thanks..

[TAdams]

Good to see you got it fixed!

[npgatech]

No ****, tp4tissue.

But for regular joe without know-how & soldering quipment...it would be a huge disappointment if the soldering joints are compromised over the life of the keyboard.

[tp4tissue]

No ****, tp4tissue.

But for regular joe without know-how & soldering quipment...it would be a huge disappointment if the soldering joints are compromised over the life of the keyboard.

Any man who can not solder, must surrender his man card.

If little kids in 3rd world countries can do it, the average-joe has no excuse. 

[npgatech]

Any man who can not solder, must surrender his man card.

LOL, I can't say I disagree with you. A man must know how to solder... but what about her:



haha...

[bazemk1979]

Any man who can not solder, must surrender his man card.

LOL, I can't say I disagree with you. A man must know how to solder... but what about her:

Show Image

haha...

hahaha

[uJalled]

Nice job with the solder

[jabar]

How did you determine that the gap was "too big" instead of doing a massive solder blob?

[khaangaaroo]

How did you determine that the gap was "too big" instead of doing a massive solder blob?

I actually tried the massive solder blob first...haha. You can see some leftover solder from it on the pins in the picture #5. I couldn't get it to hold its shape to connect to both pins at once.

[rowdy]

Nice fix

[KangarooZombies]

Another kangaroo! :O

[khaangaaroo]

hey! and in California too!

[KangarooZombies]

We're basically twins! 

[rowdy]

Expats?

[rknize]

The reason the trace cracked was probably because the switch was not fully-seated before is was soldered.  I noticed a bunch of switches were not seated on my Leopold when I modded it.  Your typical, cheap, single-sided PCB does not have through-hole solder pads.  The pad is only on the PCB surface and can't handle lifted by the switch from behind.

Nice find.  Glad it's fixed!

[WhiteFireDragon]

Great job. From the first pic though, it looks like there were no pads to begin with. Both of the ALT switch pin has no pads there. It came off really easily because solder cannot stick to solder mask.

[limmy]

Nice fix!

I noticed plastic of the switch melted away as you work on them. Your soldering iron must be pretty hot.

Also, learning from your experience, I think we can go directly to last two steps whenever we see a broken joint.
1. Scratch the coating around the broken connection.
2. Place a little piece of wire between the gap and solder.

[khaangaaroo]

Nice fix!

I noticed plastic of the switch melted away as you work on them. Your soldering iron must be pretty hot.

Thanks!

Yeah, I don't have a temp adjustable iron. I'm using a battery powered Hakko.

Great job. From the first pic though, it looks like there were no pads to begin with. Both of the ALT switch pin has no pads there. It came off really easily because solder cannot stick to solder mask.

The pad was there, just broken. It came off along with the original solder. I saw a little round disc when I wiped the tip clean.

Also, learning from your experience, I think we can go directly to last two steps whenever we see a broken joint.
1. Scratch the coating around the broken connection.
2. Place a little piece of wire between the gap and solder.

Sometimes the pad lifts up but doesn't break off completely. It's possible for it to still be attached to one of the two lines. In that case, you would only need to solder the gap between the pad and the broken line.

Also, if there's only one line going to the pin to begin with, you can easily do the big solder blob method on it and skip the wire.

[rknize]

The pad was there, just broken. It came off along with the original solder. I saw a little round disc when I wiped the tip clean.

The pad was lifted from the board and the traces broken by the switch itself.  Like I said, it was soldered with a gap between the switch and the PCB.  The stress of pressing the key seated the switch, which broke the trace.

[tp4tissue]

The pad was there, just broken. It came off along with the original solder. I saw a little round disc when I wiped the tip clean.

The pad was lifted from the board and the traces broken by the switch itself.  Like I said, it was soldered with a gap between the switch and the PCB.  The stress of pressing the key seated the switch, which broke the trace.

Hrmm... did the key switch move by "THAT" much to lift the pad? I was under the assumption that the plate mounting would keep these things pretty taught.

[tp4tissue]

Any man who can not solder, must surrender his man card.

LOL, I can't say I disagree with you. A man must know how to solder... but what about her:

Show Image

haha...

Oh come on... A girl like that can get any nearby guy to do ANYTHING....  Yea, the nose looks weird up-close, but from 1 meter distance, she's a goddess..

[rknize]

Hrmm... did the key switch move by "THAT" much to lift the pad? I was under the assumption that the plate mounting would keep these things pretty taught.

It doesn't have to move that much to lift the pad.  Less than a millimeter is more than enough to cause fatigue.  Once the bond under pad fails, the trace will fail at the nearest stress point, usually the edge of the solder.  The plate won't help if the switch isn't seated all the way down.  I would look carefully at all of the switches.  If it was my keyboard, I'd reflow all of the joints (at least in that corner) while push the switch down.

[xDezor]

Thanks for the post khaangaaroo,

I too just noticed that my new QFR (brown) has a non functional ALT key. However, I think I am just going to RMA it.

[Element2k3]

had the exact same problem with the QFR. and since the alt left key is the most important key for me, I returned it and decided to stick to my Natural 4000. looking forward to the stealth and roccat ryos and hope that one of the two will be able to replace my aging keyboard

[firebt]

That's so weird.  I recently did the same on my Ducky Shine II to clean out some ginger ale and honestly those solder joints that come stock on the QFR...look quite frankly terrible. 

See how they are not shiny and glazed over?  That means those are cold joints.  Cold joints generally have higher resistance and also are more brittle, due to the amount of dislocations that occur in the solder from incomplete phase change. 

Also, I wanted to ask, did you use flux?  Your solder joint is very pooled..meaning that the solder is bunched up and not more "liquified" looking if you will?  It seems like they never hit the target temp, which may be understandable with the larger blob of solder and the copper wire dispersing your heat rather quickly.



The solder should be very shiny and flat, meaning that it should seem like it was a pure liquid at one point with lower surface tension.

Also interesting how all of your pins are already bent.

[alaricljs]

Those are jumper or diode leads that are bent to retain them in the PCB.  It's single sided which will have an effect on the appearance of the 'blob' since it doesn't get sucked into the thru-hole.

[Soarer]

Hrmm... did the key switch move by "THAT" much to lift the pad? I was under the assumption that the plate mounting would keep these things pretty taught.

It doesn't have to move that much to lift the pad.  Less than a millimeter is more than enough to cause fatigue.  Once the bond under pad fails, the trace will fail at the nearest stress point, usually the edge of the solder.  The plate won't help if the switch isn't seated all the way down.  I would look carefully at all of the switches.  If it was my keyboard, I'd reflow all of the joints (at least in that corner) while push the switch down.

Exactly right. ANY movement is bad!

Also, I do wonder if many of the plates on modern keyboards are basically just for show.

[alaricljs]

I think there are a number of plate KBs that only function for as long as they do because of the plate.  Assembly quality seems to be declining across the board.

[khaangaaroo]

Also, I wanted to ask, did you use flux?  Your solder joint is very pooled..meaning that the solder is bunched up and not more "liquified" looking if you will?  It seems like they never hit the target temp, which may be understandable with the larger blob of solder and the copper wire dispersing your heat rather quickly.

The solder should be very shiny and flat, meaning that it should seem like it was a pure liquid at one point with lower surface tension.

Also interesting how all of your pins are already bent.

I'm no soldering pro. It works fine, and if it breaks again I'll fix it again

I did use flux. All the solder joints that I have fixed in the past that were missing a copper pad have never laid flat. There's nothing on the surface to adhere too.

There was definitely enough heat. It took some time to get the solder into the right shape that would connect all three points. I was afraid it was getting too hot at times.

And what alaricljs said about the bent pins. Just jumpers.

Also, keep in mind that these photos were taken with a macro lens. Things always seem bigger and more detailed. The copper wire is probably about as thick as 0.5mm pencil lead. And the joints look more shiny to the naked eye than they do in the photos.

[firebt]

Also, I wanted to ask, did you use flux?  Your solder joint is very pooled..meaning that the solder is bunched up and not more "liquified" looking if you will?  It seems like they never hit the target temp, which may be understandable with the larger blob of solder and the copper wire dispersing your heat rather quickly.

The solder should be very shiny and flat, meaning that it should seem like it was a pure liquid at one point with lower surface tension.

Also interesting how all of your pins are already bent.

I'm no soldering pro. It works fine, and if it breaks again I'll fix it again

I did use flux. All the solder joints that I have fixed in the past that were missing a copper pad have never laid flat. There's nothing on the surface to adhere too.

There was definitely enough heat. It took some time to get the solder into the right shape that would connect all three points. I was afraid it was getting too hot at times.

And what alaricljs said about the bent pins. Just jumpers.

Also, keep in mind that these photos were taken with a macro lens. Things always seem bigger and more detailed. The copper wire is probably about as thick as 0.5mm pencil lead. And the joints look more shiny to the naked eye than they do in the photos.

Ah, I see.  Sounds good then.  I don't know, just seemed a little different, but as you said, if it breaks, we have the power to just fix it again lol.  In the end keyboards are not complex at all.  Just don't break your SSD now.

[krojas]

Hey,

Thanks for the guide and information about the bad soldering of the keyboard.

I got the same issue, but wondering if you could give pics as well on how you opened up the casing without us doing any unecessary scratches while figuring out:). Already RMAd my old one but got the keep it and got a new one, but I would like to revive the one with the issue 

Thanks!

[tp4tissue]

Hey,

Thanks for the guide and information about the bad soldering of the keyboard.

I got the same issue, but wondering if you could give pics as well on how you opened up the casing without us doing any unecessary scratches while figuring out:). Already RMAd my old one but got the keep it and got a new one, but I would like to revive the one with the issue 

Thanks!

ah yes,,

Ok, basically what you do is,

first remove the screw..

then pull the front lip from the bottom VERY slowly, and when there's enough space, slide a credit card into the spot.

Then slide the card along the edge,

[firebt]

http://geekhack.org/index.php?topic=39292.msg767305#msg767305

Here is the link of my walkthrough of how to take apart a plate mounted board and change switches.  It's a bit general, but feel free to PM me or reply here for questions.

[tp4tissue]

http://geekhack.org/index.php?topic=39292.msg767305#msg767305

Here is the link of my walkthrough of how to take apart a plate mounted board and change switches.  It's a bit general, but feel free to PM me or reply here for questions.

walk-throughs are nice...

Now if only you could come up with a cheat code that lets the keyboard take itself apart.. 

[firebt]

http://geekhack.org/index.php?topic=39292.msg767305#msg767305

Here is the link of my walkthrough of how to take apart a plate mounted board and change switches.  It's a bit general, but feel free to PM me or reply here for questions.

walk-throughs are nice...

Now if only you could come up with a cheat code that lets the keyboard take itself apart.. 

What? lol...

btw tp4tissue opening the cover is simply not sliding a credit card.. my fingernails man...my fingernails bro...

oh and its 3 screws, not just one.  Don't try to pry it off without removing all of them LOL

[khaangaaroo]

What? lol...

btw tp4tissue opening the cover is simply not sliding a credit card.. my fingernails man...my fingernails bro...

oh and its 3 screws, not just one.  Don't try to pry it off without removing all of them LOL

This thread is for the QFR, not the Ducky Shine...haha

One screw and a credit card is all you need.

[firebt]

What? lol...

btw tp4tissue opening the cover is simply not sliding a credit card.. my fingernails man...my fingernails bro...

oh and its 3 screws, not just one.  Don't try to pry it off without removing all of them LOL

This thread is for the QFR, not the Ducky Shine...haha

One screw and a credit card is all you need.

Touche...lol

[tp4tissue]

What? lol...

btw tp4tissue opening the cover is simply not sliding a credit card.. my fingernails man...my fingernails bro...

oh and its 3 screws, not just one.  Don't try to pry it off without removing all of them LOL

This thread is for the QFR, not the Ducky Shine...haha

One screw and a credit card is all you need.

Touche...lol

yea, don't be come up in here and all like hizzel....

[firebt]

What? lol...

btw tp4tissue opening the cover is simply not sliding a credit card.. my fingernails man...my fingernails bro...

oh and its 3 screws, not just one.  Don't try to pry it off without removing all of them LOL

This thread is for the QFR, not the Ducky Shine...haha

One screw and a credit card is all you need.

Touche...lol

yea, don't be come up in here and all like hizzel....

...what the f does that even mean

[tp4tissue]

What? lol...

btw tp4tissue opening the cover is simply not sliding a credit card.. my fingernails man...my fingernails bro...

oh and its 3 screws, not just one.  Don't try to pry it off without removing all of them LOL

This thread is for the QFR, not the Ducky Shine...haha

One screw and a credit card is all you need.

Touche...lol

yea, don't be come up in here and all like hizzel....

...what the f does that even mean

f... idk  //strikes gangster pose

[domoaligato]

No ****, tp4tissue.

But for regular joe without know-how & soldering quipment...it would be a huge disappointment if the soldering joints are compromised over the life of the keyboard.

Any man who can not solder, must surrender his man card.

If little kids in 3rd world countries can do it, the average-joe has no excuse. 

I rma'd the first cm qfr after 2 weeks of usage for this issue/defect.
I just got the replacement (brand new never opened) left alt key doesn't work.
@ $18.00 shipping to send this thing back again the $90.00 USD keyboard is already over $108.00 before the second rma.

I should have just bought a filco in the first place.
I also have 3 KBT Race's w/reds. the led's on the kbt's are cr@p and prone to having entire rows of led's stop working.

I love the hobby but getting defective keyboards when I buy them @ full price is starting to upset me.

[aviphysics]

or you could spend $15 on a soldering iron.

[domoaligato]

or you could spend $15 on a soldering iron.

Or I could get a working keyboard out of the box. terrible QA is no excuse.

[noons]

or you could spend $15 on a soldering iron.

Or I could get a working keyboard out of the box. terrible QA is no excuse.

Agreed, I would have just returned or RMA'd it..

[BucklingSpring]

Glad to see a DIYer.. rather than the usual whiny lil' B%&ch$

The usual whiny lil' B%&ch$ -> That would be me :-)

[tp4tissue]

what's wit all the necro-in' again..

[noons]

Glad to see a DIYer.. rather than the usual whiny lil' B%&ch$

The usual whiny lil' B%&ch$ -> That would be me :-)

LMAO same! IDK I just feel like if you sink that the type of money you do for a quality mechanical keyboard all the keys should at least work...