Carbon fiber keyboard base - Updated 9/4/15

[heedpantsnow]

2015.9.4- a small update here.


Continue below for background.  Enjoy!


Hi, guys, I'm always looking for anything I can do to lighten my load when I'm out and about.  I have this quasi-bipolar taste of leather/canvas and carbon fiber/brush metal.  Don't know what to do with that sometimes.

Anyway, many years ago I built a carbon fiber shell for my Macbook that you can read about here.  Man, that was a long time ago.  I've been thinking that it would be cool to have a carbon-fiber base (not shell on an existing base, mind you) for my Poker.  It will also have a titanium plate (thanks mkawa and The_Beast), so that goes along with the "light but awesome" theme.

My question for you guys is this:  would a plate make up for the crappy feeling that often accompanies plastic/lightweight bases?  Would I be shooting myself in the foot by going through all this trouble (and working with CF is a LOT of trouble) just to end up with a keyboard that sucks?  Could I do something like a foam fill or a silicone insert that would counteract that feeling?

What say you, geekhackers?

[SpAmRaY]

Depending on clearance between the PCB and plate you could line the case with something like this

http://www.amazon.com/gp/product/B002KQY92W

[heedpantsnow]

I appreciate the advice, but I'd prefer to take out the cheap feel of a lightweight base without adding much weight.  I feel like the steel weights would add too much.

[SpAmRaY]

Ah sorry was just thinking about adding weight to make it heavier.

For stability maybe a dampening mat? Either shelf liner or something similar.

http://geekhack.org/index.php?topic=53792.0

[CPTBadAss]

I like this idea. If you can figure out how to marry the plate to the case bottom or create a top and a bottom out of carbon fiber that'd be really cool.

Based on your last mod, it seems like you know your way around CF though.

[heedpantsnow]

Thanks, that's definitely an option.  At one time I remember people still filling their case inside with expanding foam.  Do people still do that?  Does it work?

[SpAmRaY]

Thanks, that's definitely an option.  At one time I remember people still filling their case inside with expanding foam.  Do people still do that?  Does it work?

I thought that was done almost as a joke to counter the 'ping'. Personally that expanding foam gets messy fast, I don't think I'd trust myself using it near a keyboard.

[CPTBadAss]

Thanks, that's definitely an option.  At one time I remember people still filling their case inside with expanding foam.  Do people still do that?  Does it work?

It went ALL OVER the place. It seemed to be hard to control and far messier than it was worth. You can try using EDPM sheets or drawer liner.

[heedpantsnow]

OK, thanks for the suggestions.  At one time I had thought about making a carbon fiber plate to go with the bottom, but cutting all the holes and perfecting filing all of them is dangerous work due to the way tiny carbon fiber shards work their way under your skin...and have to be surgically removed   But the Ti plate will be just as cool I think. 

Sounds like I definitely ought to steer clear of foaming the inside and just use sorbithane or shelf liners.  You guys are great!  Thanks for the advice! 

Any other comments or suggestions?

[luis911]

OK, thanks for the suggestions.  At one time I had thought about making a carbon fiber plate to go with the bottom, but cutting all the holes and perfecting filing all of them is dangerous work due to the way tiny carbon fiber shards work their way under your skin...and have to be surgically removed   But the Ti plate will be just as cool I think. 

Sounds like I definitely ought to steer clear of foaming the inside and just use sorbithane or shelf liners.  You guys are great!  Thanks for the advice! 

Any other comments or suggestions?

Are carbon fiber shards harmful?

[heedpantsnow]

...
Are carbon fiber shards harmful?

Dude, they are seriously bad.  Inhaled even worse!  Masheth forth:
http://www.rcuniverse.com/forum/composites-fabrication-repair-97/634298-carbon-fiber-hazard.html#post634298
http://www.rcgroups.com/forums/showthread.php?t=311845

[Zeal]

...
Are carbon fiber shards harmful?

Dude, they are seriously bad.  Inhaled even worse!  Masheth forth:
http://www.rcuniverse.com/forum/composites-fabrication-repair-97/634298-carbon-fiber-hazard.html#post634298
http://www.rcgroups.com/forums/showthread.php?t=311845

  TIL that one does not mess with carbon fiber...

[mkawa]

Hi, guys, I'm always looking for anything I can do to lighten my load when I'm out and about.  I have this quasi-bipolar taste of leather/canvas and carbon fiber/brush metal.  Don't know what to do with that sometimes.

Anyway, many years ago I built a carbon fiber shell for my Macbook that you can read about here.  Man, that was a long time ago.  I've been thinking that it would be cool to have a carbon-fiber base (not shell on an existing base, mind you) for my Poker.  It will also have a titanium plate (thanks mkawa and The_Beast), so that goes along with the "light but awesome" theme.

My question for you guys is this:  would a plate make up for the crappy feeling that often accompanies plastic/lightweight bases?  Would I be shooting myself in the foot by going through all this trouble (and working with CF is a LOT of trouble) just to end up with a keyboard that sucks?  Could I do something like a foam fill or a silicone insert that would counteract that feeling?

What say you, geekhackers?

hmmmm....  why don't you start with a fiberglass phenolic G10 or G11 base and see what it feels like? otoh the pcb will probably already be G10/FR4, but it's going to be very thin compared to the 1/8"+ you could make yourself (and YES, wear a respirator WHENEVER you're cutting glass or carbon fiber composites!!! the particles are exactly the worst size for lung cilia (the cell formations that pull oxygen out of the air). they're too small to eject via the bronchial tubes, but too large to pass into the blood stream. the result is damage to lung cilia, and lung cilia, like corneal cells, are very fast growers. what that means is that when you damage them, they duplicate and grow back very quickly. it also means that when you damage the RNA carriage mechanisms or base dna they're carrying and they become malignant cancer, they grow _fast_. late stage lunch cancer is almost always terminal.

_always_ be careful around materials with small hard fibrous formations. lung cancer and emphysema is NO JOKE. whatever the work is, it's not worth your life.

[mkawa]

wait, sorry, i was talking about g10 bases.

yah, why not pick up some GPO3 or G10 or G11 for a nice thick, but lightweight bottom plate.

UHMW and delrin/acetal GF20 or GF30 are also really good materials that are lightweight but very strong and abrasion resistant.

the titanium is actually going to be a little more springy than stainless (400 series anyway) at the same thickness, so the main benefit is lightweight, and all the materials above keep the weight down while maintaining about the same tensile and young's modulus.

the terrible feeling mostly comes with very weak plastics like ABS formulations. you don't _have_ to have woven carbon to get a strong light composite. in fact, glass fibered thermoplastics tend to be stronger than anything but very high end carbon fiber because they maintain about the same isometric tensile, while a single layer carbon weave will only have high tensile along the grain of the weave.

[heedpantsnow]

hmmmm....  why don't you start with a fiberglass phenolic G10 or G11 base and see what it feels like? otoh the pcb will probably already be G10/FR4, but it's going to be very thin compared to the 1/8"+ you could make yourself (and YES, wear a respirator WHENEVER you're cutting glass or carbon fiber composites!!! the particles are exactly the worst size for lung cilia (the cell formations that pull oxygen out of the air). they're too small to eject via the bronchial tubes, but too large to pass into the blood stream. the result is damage to lung cilia, and lung cilia, like corneal cells, are very fast growers. what that means is that when you damage them, they duplicate and grow back very quickly. it also means that when you damage the RNA carriage mechanisms or base dna they're carrying and they become malignant cancer, they grow _fast_. late stage lunch cancer is almost always terminal.

_always_ be careful around materials with small hard fibrous formations. lung cancer and emphysema is NO JOKE. whatever the work is, it's not worth your life.

Thank you so much for the warnings; I usually wear a mask, goggles, and a rain jacket with the hood up and cinched around my face, and exam gloves.  I've heard stories, and they almost never end up okay.

wait, sorry, i was talking about g10 bases.

yah, why not pick up some GPO3 or G10 or G11 for a nice thick, but lightweight bottom plate.

UHMW and delrin/acetal GF20 or GF30 are also really good materials that are lightweight but very strong and abrasion resistant.

the titanium is actually going to be a little more springy than stainless (400 series anyway) at the same thickness, so the main benefit is lightweight, and all the materials above keep the weight down while maintaining about the same tensile and young's modulus.

the terrible feeling mostly comes with very weak plastics like ABS formulations. you don't _have_ to have woven carbon to get a strong light composite. in fact, glass fibered thermoplastics tend to be stronger than anything but very high end carbon fiber because they maintain about the same isometric tensile, while a single layer carbon weave will only have high tensile along the grain of the weave.

A machined-out G10 base would be DOPE, but I don't think I have access to the proper equipment.  I have a G10 handled Benchmade (knife) and I love it.  The problem is that I don't just want a flat lower plate as a base, I want sides on it.  Can G10 (or the other materials) be glued?  I could glue sides on it then use a router to bevel the edges.  How hard is it to cut?

I'm not familiar with any of the other materials.  And really the only thing I know about G10 is that (I think) it is reinforced aramid fiber in epoxy.  Is that right?

How else could I do it with premade plates and have sides on it?

[Melvang]

I love the idea of a cf case.  But if you are going light weight plate mount an aluminium plate would actually be lighter than titanium and depending on the alloy could be stiffer.  That is if I remember my material properties right.

[heedpantsnow]

I love the idea of a cf case.  But if you are going light weight plate mount an aluminium plate would actually be lighter than titanium and depending on the alloy could be stiffer.  That is if I remember my material properties right.

I'm not totally convinced of that.  Anyhow, an aluminum plate wouldn't protect the side of the PCB so that's a no go.

One issue I'll have to deal with is that CF is electrically conductive. What other issues can you guys think of?

[Melvang]

I love the idea of a cf case.  But if you are going light weight plate mount an aluminium plate would actually be lighter than titanium and depending on the alloy could be stiffer.  That is if I remember my material properties right.

I'm not totally convinced of that.  Anyhow, an aluminum plate wouldn't protect the side of the PCB so that's a no go.

One issue I'll have to deal with is that CF is electrically conductive. What other issues can you guys think of?

I was speaking for the mounting plate.  Not for the case.  But titanium is heavier than aluminum in this case since the part dimensions are not going to be altered here like they are in other industries.  Aluminum is about 168 pounds per cubic foot depending on alloy, and titanium is about 235 pounds per cubic foot.

http://mobilitymgmt.com/Articles/2010/03/01/Aluminum-vs-Titanium.aspx?Page=2

Also here is  web page relating to wheelchairs comparing aluminum to titanium.  A lot of tubing tech in wheel chairs comes from the trickle down concept from the bicycle industry.  In that industry aluminum, titanium, and carbon fiber have been used for a number of years.  To get the same stiffness in a titanum frame that you would get with an aluminum frame you have to make the tubing a lot larger and go with a thinner wall to increase the surface area enough to get to the same stiffness.  A number of years ago I was lusting after a specific MTB frame from a company named Ibis.  The frame was the Bow Ti.  It was a unified rear triangle setup with a simple air shock for rear suspension damping only.  To actually hold the weight of the rider it had a pair of tubes on either side of the frame that went from the headset to the rear dropouts uninterrupted that acted as two leaf springs. 

Plus, I have experience in bending aluminum, and a bit with titanium.  This comes from working as an Aviation Structural Mechanic in the Navy, working on H-60 Seahawks.  These are the Navy's version of the Blackhawk.  The only major differences being the addition of a rescue hoist and a lot of avionics packages available.

One other nice advantage of aluminum is price.  Check mcmastercarr.com and compare some raw material prices.

[heedpantsnow]

I love the idea of a cf case.  But if you are going light weight plate mount an aluminium plate would actually be lighter than titanium and depending on the alloy could be stiffer.  That is if I remember my material properties right.

I'm not totally convinced of that.  Anyhow, an aluminum plate wouldn't protect the side of the PCB so that's a no go.

One issue I'll have to deal with is that CF is electrically conductive. What other issues can you guys think of?

I was speaking for the mounting plate.  Not for the case.  But titanium is heavier than aluminum in this case since the part dimensions are not going to be altered here like they are in other industries.  Aluminum is about 168 pounds per cubic foot depending on alloy, and titanium is about 235 pounds per cubic foot.

http://mobilitymgmt.com/Articles/2010/03/01/Aluminum-vs-Titanium.aspx?Page=2

Also here is  web page relating to wheelchairs comparing aluminum to titanium.  A lot of tubing tech in wheel chairs comes from the trickle down concept from the bicycle industry.  In that industry aluminum, titanium, and carbon fiber have been used for a number of years.  To get the same stiffness in a titanum frame that you would get with an aluminum frame you have to make the tubing a lot larger and go with a thinner wall to increase the surface area enough to get to the same stiffness.  A number of years ago I was lusting after a specific MTB frame from a company named Ibis.  The frame was the Bow Ti.  It was a unified rear triangle setup with a simple air shock for rear suspension damping only.  To actually hold the weight of the rider it had a pair of tubes on either side of the frame that went from the headset to the rear dropouts uninterrupted that acted as two leaf springs. 

Plus, I have experience in bending aluminum, and a bit with titanium.  This comes from working as an Aviation Structural Mechanic in the Navy, working on H-60 Seahawks.  These are the Navy's version of the Blackhawk.  The only major differences being the addition of a rescue hoist and a lot of avionics packages available.

One other nice advantage of aluminum is price.  Check mcmastercarr.com and compare some raw material prices.

Ahh, okay, thanks for taking the time to explain all of that.  You definitely know more than I do.  So the aluminum would be cheaper, lighter, and stronger.  Yet I still find myself wanting to do it in carbon fiber.       So I don't know what I will end up doing.  I'll have to see how much carbon fiber and epoxy I have left over from my other project.  If I can do it in CF without purchasing anything  I may end up going in that direction.

Regardless of what I end up doing I will post a build log.

[Melvang]

Not a problem man.  I don't have an issue with either material.  But titanium is a massive pain in the ass to work with unless you have good high quality tooling.  But stronger isn't really the word that I would use here.  Depending on the shape and if you can reengineer the part for the physical properties, neither really is stronger.  But given the same shape and dimensions aluminum is generally stiffer.  I wouldn't hesitate to make a case out of CF if you have to tooling and the experience with the stuff.  But in the op you mentioned that you wanted to make it as light as possible and was thinking of going with titanium for the plate.  I just wanted to correct you on some very wide spread misconceptions.

[heedpantsnow]

Not a problem man.  I don't have an issue with either material.  But titanium is a massive pain in the ass to work with unless you have good high quality tooling.  But stronger isn't really the word that I would use here.  Depending on the shape and if you can reengineer the part for the physical properties, neither really is stronger.  But given the same shape and dimensions aluminum is generally stiffer.  I wouldn't hesitate to make a case out of CF if you have to tooling and the experience with the stuff.  But in the op you mentioned that you wanted to make it as light as possible and was thinking of going with titanium for the plate.  I just wanted to correct you on some very wide spread misconceptions.

Thanks, dude!

[Jixr]

I'm confused by this, if you're just doing all this work to lighten the keyboard, how much of a difference will all that work really make.

I'm sure the few ounces could be saved elsewhere, removing things from you bag, changing the thick stock poker cable for a shorter and lighter weight one, get a smaller bag, a lighter one, or even a more comfortable one to manage the weight.....


but doing it for the looks is a completely different option, in which case why not just get an alluminum case and make some CF accents for it. Seems much easier than all this, or just shell the stock poker case.

[Findecanor]

I think, get a Poker case where
- the PCB makes as much contact with the case bottom as possible
- the case does not flex: some fibreglass tape and epoxy might reinforce the bottom of a plastic case enough, without having to trim afterwards.
- the feet provide a good grip against the desk so that the keyboard does not float around

[JaccoW]

How about two CF plates with an aluminum middle layer, similar to what you see in acrylic cases?
Or maybe something similar to your previous adventures where the top plate is CF and the bottom part is a stiffer material.

[C5Allroad]

Hydrographics comes to mind...

[heedpantsnow]

I have a plan, and I'm planning on it being awesome.  I won't be able to work on it until June, though.  Will probably go with a mdf mold with channels for interior ribs, kevlar first (inside of case and electrically inert), then short-cut fiberglass (to absorb a lot of resin and give it some rigidity, then 2-3 layers of CF.  I'll post here when I start on it.

[C5Allroad]

That. Sounds. Awesome lol.

[JaccoW]

Sounds pretty awesome indeed. Looking forward to seeing that.

[mashby]

I have a plan, and I'm planning on it being awesome.  I won't be able to work on it until June, though.  Will probably go with a mdf mold with channels for interior ribs, kevlar first (inside of case and electrically inert), then short-cut fiberglass (to absorb a lot of resin and give it some rigidity, then 2-3 layers of CF.  I'll post here when I start on it.

Sounds very, very interesting. Now I just need to find that "stalk" button so I can track every post you make between now and then. ;-)

[heedpantsnow]

Finally started. Spent about 30 hours making a plug to mold. A mdf mold might have been easier in the end, but making a urethane mold allows me to get the mold out of the base easier.

[JaccoW]

Excellent. Looking forward to some pictures.

[heedpantsnow]

Here's the plug (model) that I made for the base. I spent a lot of time looking at pics and specs of different bases to try to ensure as much future compatibility as possible.  The plug looks really messy, and I'm not 100% satisfied with it, but that's life...

I used a specific plastilene clay to form the inner ribs and supports. With urethane rubber, you can use any clay containing sulfur. Most of them don't say on the package whether or not, so you have to stand in the craft store and google each brand.   

I used cardboard coated with wax for the sides. You want the plug to be as perfect as possible because fixing problems becomes harder and harder as you go from plug to mold to piece to finished piece. But, I will still have to do some trimming of the mold because I couldn't get the clay formed in the outer corners because there wasn't room for a knife blade.

I hope you guys enjoy the process!

[JaccoW]

I can see you have put some thought into it (or a whole load of trial and error ).

Keep those updates coming!

[heedpantsnow]

Alrighty, here's the mold. It turned out pretty well. It has some bubbles deep inside, but should still be usable enough (the bubbles only affect long-term strength like if I was manufacturing these day in and out).



I had to do a bit of trimming to make sure the sides are nice and smooth and defined.

The support ridges and the screw mounts are obviously very important, and they all turned out fine. Looking forward to putting this thing to use!

[mashby]

Sweet! I can't wait to see what the first one looks like. Nice work. 

[heedpantsnow]

Alrighty. The first layer is "in the bag" LOL. I went with a first layer of Kevlar because it shows and follows contours well (and I can see problems), is easy to work with, and most importantly it is electrically non-conductive so it won't interfere with the KB pcb.

So I sprayed the mold with release, then laid the Kevlar, then mixed up my medium-speed low viscosity epoxy resin. I used a cheap brush to lay it on and work it into the fabric.

I then laid down perforated release film (makes a nice finish but allows excess resin to seep through), followed by poly batting, and finally the vacuum bagging material.

Lastly, I bagged it down to ~25 psi. Now the waiting begins!

Here's a quick pic:

[Melvang]

Thats almost as sexy as machining porn.

[heedpantsnow]

Oh, and a few tips:
When working with epoxy I double glove my hands so I can just pull off a glove if it gets too messy.

When you use these cheap paintbrushes you MUST pull all the excess fibers before use. They shed like crazy, and you don't want that embedded in your project!

Always pre-cut and lay everything out so you can grab it in order before you mix the epoxy.

[heedpantsnow]

Thats almost as sexy as machining porn.

LOL

[JaccoW]

It is sexy.

[heedpantsnow]

Well, a bit of tragedy struck...the vacuum compression didn't push the kevlar down in the support ridges in the mold well enough.  I'll have to try again    More updates later.

[Demetrium]

Bummer, keep us posted!

[heedpantsnow]

So, learned a few lessons:  I need to do something to force the fabric down into the channels. So I found some scrap plastic rod (cut up a hanger, said sorry to my wife). I cut it to length of each channel, and will put it on top of the release film to force the fabric into the contours.



I also didn't like the flimsy feeling of the threaded inserts on the first try (they are made of brass and will be used to screw the PCB to the base). They were just stuck to the base with a bit of epoxy. Like I learned back in uni:  anything worth engineering is worth over-engineering!  So now I used a bit of superglue and wrapped each one with Kevlar tow (basically a bundle of fibers pulled from the fabric). This should absorb the epoxy better and create a much stronger mount for the inserts.



Will let you know how this try turns out!

[heedpantsnow]

By the way, cutting Kevlar (even with nice fabric shears) is a pain in the butt!  It takes me about 10 mins to cut across my 6 inch wide strip.

I've found that it's actually easier once it's in resin and partially cured; the epoxy holds the fibers so the shears can cut them (using angled serrated medical shears).

I'll be opening up the vacuum bag in a little while; it's important to not let each layer cure fully otherwise you won't get good adhesion layer-to-layer (that's for epoxy resin; poly resin is the other way around).

[Badwrench]

Nice work.  I love using kevlar, but totally agree on how hard it is to cut. 

Curious, how are you going to finish the top edge of the case? 

[heedpantsnow]

Nice work.  I love using kevlar, but totally agree on how hard it is to cut. 

Curious, how are you going to finish the top edge of the case? 

My first choice is to fold over and clamp the last layer of carbon, using superglue to hold it before the epoxy.  It's a rather light 5.7 oz 3k tow 2x2 twill.  The tiny bend radius can work to my advantage by making a little rounded edge, as long as it doesn't splinter.  What do you think?

If that doesn't work, I can either round the edges off with a dremel or last resort use some tiny c-channel molding from McMaster:  http://www.mcmaster.com/#rubber-molding/=sun8vb (I would prefer to avoid that as it won't be as clean).

I'd love any feedback or opinion, I'm totally not an expert at this but enjoy learning.

[Badwrench]

Nice work.  I love using kevlar, but totally agree on how hard it is to cut. 

Curious, how are you going to finish the top edge of the case? 

My first choice is to fold over and clamp the last layer of carbon, using superglue to hold it before the epoxy.  It's a rather light 5.7 oz 3k tow 2x2 twill.  The tiny bend radius can work to my advantage by making a little rounded edge, as long as it doesn't splinter.  What do you think?

If that doesn't work, I can either round the edges off with a dremel or last resort use some tiny c-channel molding from McMaster:  http://www.mcmaster.com/#rubber-molding/=sun8vb (I would prefer to avoid that as it won't be as clean).

I'd love any feedback or opinion, I'm totally not an expert at this but enjoy learning.

If the weave is small enough, definitely try to wrap the finishing layer over the edge to get a nice rounded edge. 

[heedpantsnow]

Nice work.  I love using kevlar, but totally agree on how hard it is to cut. 

Curious, how are you going to finish the top edge of the case? 

My first choice is to fold over and clamp the last layer of carbon, using superglue to hold it before the epoxy.  It's a rather light 5.7 oz 3k tow 2x2 twill.  The tiny bend radius can work to my advantage by making a little rounded edge, as long as it doesn't splinter.  What do you think?

If that doesn't work, I can either round the edges off with a dremel or last resort use some tiny c-channel molding from McMaster:  http://www.mcmaster.com/#rubber-molding/=sun8vb (I would prefer to avoid that as it won't be as clean).

I'd love any feedback or opinion, I'm totally not an expert at this but enjoy learning.

If the weave is small enough, definitely try to wrap the finishing layer over the edge to get a nice rounded edge. 

Thanks. Any concerns over the superglue (dried) interacting with the epoxy resin?

[Badwrench]

Nice work.  I love using kevlar, but totally agree on how hard it is to cut. 

Curious, how are you going to finish the top edge of the case? 

My first choice is to fold over and clamp the last layer of carbon, using superglue to hold it before the epoxy.  It's a rather light 5.7 oz 3k tow 2x2 twill.  The tiny bend radius can work to my advantage by making a little rounded edge, as long as it doesn't splinter.  What do you think?

If that doesn't work, I can either round the edges off with a dremel or last resort use some tiny c-channel molding from McMaster:  http://www.mcmaster.com/#rubber-molding/=sun8vb (I would prefer to avoid that as it won't be as clean).

I'd love any feedback or opinion, I'm totally not an expert at this but enjoy learning.

If the weave is small enough, definitely try to wrap the finishing layer over the edge to get a nice rounded edge. 

Thanks. Any concerns over the superglue (dried) interacting with the epoxy resin?

Nope.  Should have no effect.

[heedpantsnow]

Double cool.