The CNC'd Slipjoint

I was hungry and excited to dive in, so I pulled out my pocket knife, and sliced in. Upon doing so, one of our directors stopped what he was doing to ask if I was really using that knife to cut into my lunch. I panicked internally and blurted out "Oh yea, I totally cleaned this before sitting down" a blatant lie.

He had a point though, I work as a prototype fabricator and that same pocket knife gets used on packages, chemical containers, and who knows what else. "I'm going to make a pocket knife just for food" was the next thought through my mind.

looking at this from a product design perspective, the knife should be something that can be brought to a high end restaurant and not feel out of place. It should use natural materials, be slow and simple to open and close. It should feel like a fine watch. The blade should be small enough to not look threatening but large enough to effectively cut food.

Over the next month and a half I went from concept design to machined prototype and will likely keep working on refining this design so I can start to sell these.

In this instructable I won't be sharing my files, however I will take you through my process and hopefully you will learn a thing or two!

Ideation Process:

• Pen/pencil
• Paper/notebook (personally prefer MUJI notebooks)
• X acto knife
• Cardboard/chip board

3d Modeling Process:

• Computer
• 3d modeling program (I prefer Fusion 360)
• Calipers

CAM Process:

• Cam Package (I prefer Fusion 360 cam)
• CNC Mill
• (Optional CNC router)

Machining:

• CNC Mill (mine is a heavily modified one of these)
  • If you have a similar setup, get a pendant! makes a world of difference)
• Lathe (optional)
• Vise with removable jaws
• Drills / Endmills (yours may be different in size but try and get 3-10 of each endmill/drill you use. All endmills I used are carbide, all drill bits I used are cobalt)
  • #1 spot drill
  • #2 spot drill
  • 1/16" cobalt stub-ish drill bit (amazon)
  • 1/8" cobalt stub length drill bit (amazon)
  • 4mm cobalt drill bit (amazon)
  • 1/4" 4 flute carbide flat endmill (pretty happy with speed tiger end mills off amazon)
  • 1/8" 4 flute carbide flat endmill (pretty happy with speed tiger end mills off amazon)
  • 1/16" 4 flute carbide flat endmill (pretty happy with speed tiger end mills off amazon)
  • 1/4" 4 flute ball endmill (pretty happy with kodiak end mills off amazon
  • 8-32 drill/tap
• Edge finder
• Noga base and dial indicator (amazon)
• Mist setup
  • Air compressor (amazon)
  • Blower block? (amazon)
  • Kool Mist 77 (amazon)
• Collets (got a general pack of R8 when i got my mill, but it was missing a 4mm and a 1/16")
• Parallel set
• Strap Clamp set (those black steel things that you can find everywhere)
• Micrometer
• 123 block
• Masking tape
• Paper towels
• Chip Brush
• Ear protection
• Eye protection
• Respirator
• Belt Sander
• Hand files
• Deburring tools
• Super Glue

Assembly/Finishing:

As with anything in product design I began by looking around for inspiration and found a lot of french and european knives that had some nice qualities to them.

Pulling from these I began to make some paper mockups to get the size about right. The nice thing about paper/cardboard is its so cheap its practically free. This makes the cost of errors incredibly low. This is a really important idea in developing a product or a prototype, If you only take one thing away from this instructable it should be to think "how I can decrease the cost of errors". This can look like buying 2 taps instead of one just in case you break one inside your part. It could be buying extra material so you dont have to wait another week for more material to show up. Or even packing an extra shirt on a trip just in case you spill something on yourself. Thinking this way will make it ok for you to fail which in turn means you can move faster and take more risks.

Life advice aside, after making a cardboard mockup I took this design to fusion 360 in order to make a CAD model.

I based some measurements off the sotck I had laying around, and made sure nothing was wider than the stock i'd be using.

On the subject of keeping the cost of your errors low, I decided to get some 3d prints made via Polyjet. This style of printing is insanely accurate, but the machines cost a good deal of money to own and maintain. Parts that should be flat will print flat, parts that should be round, will come out round. Having these prints in hand will help for 2 reasons, first I can make sure that I'm not spending a month working on a design that I don't actually like. Second, since they are so accurate I can use these to validate that hole locations are where they need to be, and that the size of the part is correct. This is the second real project I've used my mill for, so i'm still learning to trust it.

It's a good thing that I used these prints because after seeing them, I realized that some of the surfacing I did on the handle would feel really sharp and uncomfortable so I went in and made some tweaks.

I'm not sure exactly what the "right" approach for doing this is, but what I ended up doing was copying the full cad model into multiple separate assemblies. In those assemblies I would make a block with oversized dimensions of the stock I had laying around. Since these blocks will define how my tool paths run I would rather assume that the material is larger than it actually is so I spend some time cutting air rather than plunging right into tool steel that the machine didn't expect to be there.

With these blocks set up, i imported them all into a master file. This one file will hold every cam operation I do, from milling soft jaws to milling each component. This makes it really easy to copy toolpaths from one part to another since the parameters would largely be the same. If you use this approach take the time to lay everything out so it's neat and orderly, label everything. This will prevent errors from popping up down the line because you were using the wrong fixture setup.

For the primary set of soft jaws I'm essentially making a set of parallels with an endstop. Using aluminum for these jaws has the benefit that if a toolpath cuts into the jaws it's not a big deal.

For the flipped operation of the handle, I'll be using the negative profile of the handle to clamp and locate the part. incredibly repeatable, and simple.

Since I don't have any cam l validated at this point, I chose to use soft jaws to fixture everything. Quick, cheap, and effective

On the note of having nothing validated at this point, the tool paths, speeds, feeds will change around a bunch as I work through this project. One of these days I might be a good enough machinist to get everything right on my first try.

With cad/cam stuff loosely sorted at this point, lets start cutting material.

A while ago I bought a ton of material from an auction website which is why I have soooo much 440c stainless in weird sizes. Makes it easy to play around and experement!

I'll be using this little horizontal bandsaw to cut up stock, you can get one off amazon for somewhere around $270, while its not automated, it is easy to store especially in a small shop, and cuts through just about every material ive thrown at it.

When using a horizontal bandsaw, or any kind of cutting tool with a clamp be sure to pay attention to how you are clamping your material. For angle iron, cut down the mountain /\ For flat stock, if you place the widest dimension down, you will reduce how much the blade can drift and therefore will get a more square cut. For multiples try and sandwhich everything together rather than clamping from the sides where the width of bars may vary which could result in only one of your multiple items actually getting clamped.

If you are cutting something a bit wider and the jaws are only able to clamp onto a small portion of the stock, try to place another piece of stock over on the other edge of the vise jaws. This will help balance out the pressure of the jaws and will hopefully allow you to clamp more securely.

Another trick is for longer stock, use a c clamp to keep all the bars constantly clamped together. You should be able to scootch stock forward while keeping the cut ends together, leaving you with more consistent lengths across all your parts.

Make sure to be aware of what edge of your line you are cutting on. This is vital if you want to

get a really accurate cut. If you tried to cut directly on the line, you could end up 1/16th on either side of you line, that could mean up to an 1/8th of size difference between the largest piece you cut and the smallest! Of course, that might be totally fine for your project, so it all depends.

Since these parts are going into a vise and need to be perfectly flat, adding a chamfer to all edges and corners will greatly help with this. It will prevent the part from lifting up just a hair because it was resting on a super small burr. May seem like a minor detail, but it's incredibly important to do!

Have you ever been in the middle of a project and been wondering where on earth you put that one thing you need? Well it doesn't have to be that way! I'm really good at being messy, but I've found if every tool, drill bit, endmill or whatever has a home, ideally something I can just toss the tool in without thinking about, the amount of searching I have to do drops substantially. So take a moment before you start a project, lay out all your tools, and give them homes.

Over at the mill, lets set up the vise. Some people think its really hard to tram a vise, but you just need to know the right approach. Tighten down one side of the vise until its tight, but can still move if you really try. The other side needs to be about finger tight, so it should be able to move a bit more freely.

With the bolts set up this way any time you tap the vise, it will rotate around the tighter bolt instead of sliding around freely.

As you sweep your indicator across the fixed back jaw of the vise you will find a low side and a high side. Do a little mental math to figure out which side you need to hit to get the difference to decrease.

With a little bit of practice, tramming a vise should take about 3-5 minutes.

I used to have a hard time finding these matte black wrenches that came with my mill. Whats helped a ton is to add a couple strips of reflective tape. The multiple strips help them stand out substantially more than If the whole tool was a uniform shine.

Additionally, the collet wrench only works in one direction. Instead of flipping it back and forth to figure out which side tightens and which side looses, I just etched into the tool and labeled each side as eithe TIghten or Loosen. This makes it so much less frustrating!

In order to get the cad model of the soft jaws I need to know exactly how big the current jaws are. With the use of calipers, I took some measurements and made a technical drawing of the vise jaw that I could reference.

In my line of work these kinds of drawings are vital and tell some subtle information to whoever is reading it. It's a pet peeve of mine to see bad drawings because its really easy to make good ones. Just keep these things in mind.

• Can your features actually be measured? Think about holding the part and taking calipers to where you call out a dimension, could you do it without bending space-time?
  
• When possible, keep one constant reference edge from feature to feature. So from side A to the edge of the first hole, then from side A to the edge of the next hole, then side A to the furthest edge of the part. This reduces the likelihood of whoever is measuring this part to make errots.
  
• Try to keep your widest dimensions furthest out, and keep your smallest ones closest to the outside of the part. This just visually cleans things up, you could also stagger them a little bit in height. At the end of the day, someone should be able to take your drawing, take a part that was made, and effortlessly check to see if features are the right size.
  
• Be sure to keep all of your dimensions outside of the lines of your part! This one kills me when the dimension you need is nested somewhere inside the lines of a complex part, it could block other critical features or just plainly get overlooked.
  
• Lastly, when you write dimensions, or list them in a drawing, every decimal place you add to a feature adds cost. While 1.000000" and 1" may look the same, one I can measure with a tape measure (1") and the other I need to have a very expensive machine in a climate controlled room to take a very accurate reading of your part. On average, most machinists can hit 0.001" without much trouble, more decimal places than that, things get increasingly more complicated and expensive.

In CAM I set up my WCS (Work Coordinate System) off of the top of the back jaw of the vise, so everything will be relative to that position.

At the machine I use an edge finder to locate the face and side of the back jaw, making sure to account for half the diameter of the edge finder tip since that is the center of the spindle.

Using a set of paralells to lift the aluminum stock off of the floor of the vise. I use another parallel to make sure the edge of the material is on the same plane as the vise jaws.

I make a seperate instructable on this part, but to touch off the bit I used a manual touch plate that I made. This manual touch plate eventually became annoying to set up so i just slid a parallel under the bit, and lowered the Z height until i could feel a little resistance on the plate. Then I know the bit is exactly 0.1308" (thickness of parallel ) off the surface of the surface i'm trying to reference.

Once that setup is complete I run the tool path, and start cycling through the stock I have so i'd have more than enough soft jaw blanks.

I unbolted the hard jaws that came with the vise, loaded in the soft jaw blanks. I tightened them up against some parallels that were positioned as high as I could get away with. By clamping against something high up in the jaws in theory will cause the jaws to bend outward a little bit. They will be cut in this bent position, which will actually cause them to clamp the part more effectively under pressure. HAAS did a great tool tip video detailing this.

Next we indicate off the same corner of the vise jaw that we have been using. We only use this corner for this initial operation because none of the faces we are using are machined surfaces, if anything weird was to happen we would not reliably be able to pick up the same locations again.

Added a bit of mist/air blast to blow out chips from the cut. Helps prevent chips from welding to the endmill, prevents you from recutting chips, and as a result of all of that you get a much better surface finish.

Now that we have machined features, lets wipe down the jaws (deburr them too) and touch off on the inside corner we just machined.

This is another tiny thing that will save a lot of headache down the line. I know this won't work for every cnc, but when saving your files, try to set up a folder structure that allows you to check a couple things every step of the way. The way I do this is to have a file/folder tree that looks like :

-Date (this is the highest level folder in your cam folder)

-part 1 (date)

-op1

-tool size / tool type / rpm / maybe a brief toolpath description /part 1/ op1/ date

-op2

-tool size / tool type / rpm / maybe a brief toolpath description /part 1/ op2/ date

I cannot control rpm via CAM on my cnc so i need to add that detail. I don't have a tool changer either so i end up with a lot of different files.

Find something that works for you, the main idea is that when you load a file, you should know its up to date, its using the right tools, and its working on the correct component. Getting into the habit of this will greatly reduce pretty preventable errors.

If you are new to cnc it may be worth making a checklist of all of the little and big things that you need to do both in cam and in at the machine before hitting go. I've been machining things off and on over the past 6 or 7 years now and If its been about 2 or more weeks since i've run a part, i will whip out a physical checklist I made just to be certain that everything is set up correctly. It may take a few minutes more to get started, but id rather spend a few minutes setting myself up for success than an hour or more fixing the mistakes I made in haste.

Ok real quick, for anyone with a hobby/diy mill, this setup cost just north of $200 in total, but my word its nice!

Air compressor pushes air up through a tube into that little black box. Then by using a venturi effect it pulls the coolant through a second tube, it mixes these and jets out a pretty substantial amount of air and coolant. Its given great surface finishes, and the coolant is non toxic so its hopefully pretty safe to have in the air. It bugs my lungs a little bit when I am inside the mill enclosure, but a basic half mask respirator makes it a non issue.

The air compressor is pretty quiet at about 60-65db, even so I may build an enclosure for mine since that was the loudest part of the entire process.

So I designed the soft jaw less than ideal because the corner of the jaw does not have enough "free room" for the corner of the stock to hang out in. To mitigate this, and to clamp the stock better I just added a bevel to the corner of my stock. Simple to fix, but important to be aware of this.

First thing I did in every single part was to face the stock, this way once I flip it, I know that the material will rest perfectly flat in the soft jaws. I also made sure to add a pass that would clean up the "hanging" edge of the stock. This edge will be what bumps up against the back corner of the soft jaw after flipping. As I've been saying, any features that you use to locate your stock should be machined!

Next up I used an adaptive clearing pass to remove some stock from where the bevel will be, this should reduce axial loading when I use the ball mill to form the bevels. A bigger machine may be able to handle that amount of loading without blinking, but my machine didn't seem very happy when plunging with a ball mill.

At this point in the project, the bevels were designed as a hollow grind which brings up an interesting attribute of milling bevels, you can define EXACTLY what kind of bevel you want on your knife! This means that with some time and experimentation you could fine tune your bevel geometry to make it absolutely perfect for whatever task you are trying to do with it!

Something that's a little weird though is that the mill seems to have created some stepping along the the bevel. Presumably those steps are just the end points of the splines used to create the bevel. They shouldn't really be there, so that will take some adjustments and some further attempts to clear up.

At this point in the project I was likely running my tools way too slow which caused some burrs to form on the edges of the cuts. Its not a big deal, since deburring only takes a couple quick passes with a grinding stone. However if you see something like that happen in your own parts, it may be a sign that your tools are dull, or that your speed/feed rates are less than ideal.

Side note, these little stones are AMAZING, even the cheap ones! If anyone has used spray adhesive to mount sandpaper to a sanding block, this does effectively the same thing, but you don't have to keep re-applying the sandpaper! They are often used in the tool and die industry to polish molds, and a set of these will cost a whopping 12$ on amazon. The other awesome thing is that since they are rigid you can get really nice sharp crisp lines while sanding parts which can be difficult to do with sandpaper even when it is adhered to a backing block.

One of these endmills is now dull while the other is nice and fresh. A dull bit will take more force to cut, leave a worse surface finish, and since there is material missing from the diameter of the bit, it is no longer the nice precise dimension you had input into your CAM setup.

What do we do with this bit? Well, its not completely dead, there is still a bit of life in it, so we will just use this for roughing operations! For roughing, just use the "stock to leave" feature in solidworks, add in 0.020" (0.5mm) or so then come back in with your sharp fresh endmill to take care of that last little bit.

SInce i'm writing this after ive already made the part, i know I was running the endmills too slow. Most stainless alloys do this really weird thing where they surface harden almost instantly. That means that as you are cutting it, the outside of the material becomes harder than the inside of the material. A weird analogy would be to think of stainless as a coconut. The husk (surface hardened area) is substantially harder than the inner meat (inside of the stainless). When you use your machete (endmill) to cut into the coconut (stainless) if you just tried to take small shavings (low chip per tooth/ low feed, high rprm) of the husk with your machete your edge would get dull pretty quick. instead if you took a big chunk (correct chip per tooth for your tool), you would pierce the husk and get into the meat of the coconut with minimal damage to your machete. Lastly, to conclude this analogy imagine that under the coconut that you are chopping, there is just a massive rock, if you swing too hard (too high chip load) and cut straight through the coconut your machete will hit that rock full force and will break in half (your endmill breaks).

That's probably the best way that I can possibly describe what its like cutting stainless. Some alloys are worse than others, but overall with stainless sometimes its a lot better to push harder than it is to slow down.

In this set up I used tabs and an "onion skin" or a really thin layer of stock at the bottom of the profiling pass in order to hold the part in place after profiling. Otherwise I use the same approach as before to tackle the machining.

Now that the part is done being machined I used a combination of jewelers saw, and hacksaw to remove the part from the base stock.

In this project I use hand files a bunch, here are some things that I think help a lot with filing.

First off, I zoomed into this file to show the teeth of the file. As you can see they only cut in one direction. If you haven't already been told this, or been yelled at by some salty old metal worker, files only cut on the push stroke. Its fine if you drag the file back on the pull but do not apply force. All that will do is accelerate the dulling of your file.

On that note, files wear out! Eventually files get dull, there are some ways to sharpen them, and sometimes you just need to get the grit out of the file either with a file card or with a piece of brass/copper that is pushed parallel to the teeth of the file. If you tilt your file back and forth in the light and you see the teeth shimmer, those teeth are blunt, and its time to retire that file. Turn the old one into a knife, or just use it on weird gunky materials. Lots of things can be done with old files.

Filing position is important! You back hand (the dominant one) will control the forward force applied, and the hand at the tip of the file will control its downward force. As you file pay attention to the feel of the metal beneath you, are you cutting at an angle or straight on? When you finish a stroke are you rolling the edge and creating a small radius?


Lastly, this is something I'm still figuring out, but in the 3rd image is how I like to position the file cut relative to my body. With your arms relaxed at your side, the workpiece should be just below your elbows. I've noticed that with long filing sessions where the workpiece is above the height of my elbows, my arms get sore, and I start feeling my elbow flair up a bit. This position leaves me feeling pretty good after a long session! Give it a shot, let me know what you think.

The tang of the the knife has 3 faces that all need to be very precise. When working on precise features like this, pay close attention to how everything feels on each pass and inbetween passes look at your workpiece to see where the material has been removed. While working on this feature I would do one stroke, check, stroke, check, stroke, check, until I got everything perfectly flush.

I added tiny chamfers to the edges of the tang in hopes that it would eliminate any burrs which may disrupt how the blade opens and closes.

I also used a jewlers saw in order to cut open this internal corner because the endmill that I used to cut it wouldn't be able to reach in there. It was then cleaned up with a small hand file. With that area cleaned up, the lockbar won't be resting on any tiny bumps or burrs and should seat properly.

Ok, mach 3 is weird, at least my version is. I accidentally exported the cam for the handle liner operations with this tab set to "document units" which was MM (I'm bilingual in measurement standards so I flip back and forth). When mach 3 opened this file, it read all the MM measurements as inches, and suddenly screwed with my WCS location. I had to relocate everything. Good news is that I could do that off machined faces!

So as a word of warning, if you use mach 3, make sure you post process your code as whatever measurement standard your machine is set up with!

For this i used an Adaptive clearing tool path both face the stock and to remove all the material around the profile of the part. The raised profile will be what the next set of soft jaws will grab onto for op2.

Holes for the lockbar pins were drilled, the pivot hole was also drilled and I counterbored an area around the pivot so I could set a phosphor bronze washer there.

The part had a bit of a burr which was easily removed by scraping it off with the tip of a file.

After roughly mounting the blade, handle and liner I found that the holes were off...

Distance from the center of the pivot to the 3 faces of the tang of the blade should all be with in 0.010" (0.25mm) Ideally they should be spot on, but for now, thats a fine tolerance. When measured, one face was almost 0.050" (1.25mm) undersized, while other faces were around 0.020" over. The faces that were over sized are fine since I can always go in there with a file and remove material. The undersized face really messed everything up.

Whats worse is i know exactly what caused it...I didn't spot drill my holes. For those that don't know, any time you drill a precise hole, you never "just" drill a hole, you always always always spot drill (or center drill, effectively the same term) then drill.

A spot drill is a super short drill with a small tip that tapers into a much stouter shank. You start a hole with them because they are too rigid to "walk" or drift around on your material. Longer standard bits may just wander around on your material leaving you will holes in the wrong location, and holes that may not be perfectly round.

ALWAYS SPOT DRILL BEFORE DRILLING

I had also skipped spot drilling on one of the handles, so the holes there are off. The one piece that was drilled correctly ran into a weird thing when it was being cut, where something in the machine moved, and a small section of the face was a few thou proud of the rest of the face.

Oh and I looked outside that night, and someone set a car on fire right outside of our building, and right in front of a pallet factory. You know...just a reasonably logical place to set a fire....(nobody was hurt)

This is Toshi, he is extremely cute.

Learning from last time, I cut out a little trench around the bevel of the knife, this way the ball mill would not be axially loaded so aggressively. By doing that I got substantially better surface finishes on the bevel. Also I swapped from a hollow grind to a flat bevel. and now the stepping is substantially better!

Flipped the part, repeated the same trench approach and way way better results! With the exception of this mysterious blip. I think my mill had a literal hiccup and cut too deep in this one spot.

Last time I did this, I left tabs and a web. If you are machining steel, just pick one or the other, not both. Having both made it wayyy to much of a pain to remove. This time I went with just a web. I took a rough profile pass around the part leaving 0.030" radial, and 0.030" above the bottom of the part. On the finishing pass I went from 0.030" to 0.005" above the bottom of the part...but it turns out "0.005" was too small of a web. If you try to leave that much material, there is a good chance you may have a part fly out. Not that it happened when I was taking the final clean up pass on the lockbar or anything. That would be absurd. (Fortunately I had left about 0.030" around the perimeter of the lockbar which I could just file back)

I really wanted to use a jewelers saw to free the blade from the web, but it was just taking too much time for it to be enjoyable (I really love jewelers saws)

Instead I found that using a hacksaw works really really well for cutting through the thicker portions.

Since I was already going to recut the handle liners I decided to change some stuff up. Instead of using an adaptive clearing tool path to clear out the material around the part I opted to use a 2d profile tool path leaving 0.020" webbing. This means in the next op there is a lot less material I need to remove!

When cutting parts free from webbing, I sectioned out the thick parts with the hacksaw, and then bent them in a vise until the webbing just snapped off. Worked great!

Quick trick! If you need to remove a lot of stock really fast, you can just use the corner of your file to cut notches in the material. The sides of the file are often far more aggressive than the big flat face. Once you have a bunch of these little notches you can clean them up with the large flat section of the file.

Since the fitup is not dependent on op2 of the handles being done I can start the process of fitting everything up!

For some reason that I have yet to measure, the 1/16" hole was too small or the diameter of the pin was too big to fit into the hole I had drilled. Fortunately it looks like the location of the hole was correct so I just had to go in there and carefully open up the hole using one of the drill bits I had been using, a 1/16th in end mill (try to avoid using an endmill in anything that isn't a milling machine) , and a spot drill.

So there is a pretty big gap there, lets clean that up. Otherwise everything looks like its in the right place!

When I had the knife assembled like in that last photo, I brought it out to our buildings outside seating area and was just pondering about it. A buddy of mine came up and i happily showed him, mentioning that its assembled but im trying to not close it right now because there is some webbing in the way. He nodded like he understood, fiddled with it a little bit, and then closed it... I freaked out a little bit, and freaked out a little bit more when I saw that the spring was now bent proud of the lines of the handle! So much slop could be found now...

Turns out he did not hear me at all, he is a machinist and puts a ton of care into the things he builds and felt awful about it. Stuff happens.

This taught me a great lesson though, and solved a problem for me though! I realized that the lock bar could be bent which means I could add more preload to the spring if I bent it the other way! Once I got inside I plopped it in a vise, bent it a little bit, and bam, now I had a really nice amount of tension in the spring.

I'll need to go into fusion 360 to see if I can edit the spring geometry so once its done being milled it already has the desired preload.

Back to filing the tang to get a really nice fit. One trick you can use to check the progress of your filing, especially if you are trying to get something super precise, is to add some sharpie to the section you are filing between ever 1-3 passes. This way you get a very very very clear picture of where you are removing material.

All assembled, the lockup is looking really nice, could still use a bit of finesse, but its good enough for now!

Much like before, pull off the jaws of the vise, then install the new ones indicating off the back corner. This time I wanted about 8mm of space between the jaws to give the profile the right amount of space. So I gathered up some stock that measured about 8mm, and tightened it between the jaws.

When indicating these parts I set my offsets slightly inward of where the true position of the edge actually was. This was so that I could machine in features that would allow me to indicate where the jaw was in case anything bad happened.

Ok I was really annoyed with how I was touching tools off of the top face of the vise, just didnt work. To solve this issue, I set up a 123 block off to the side of my vise and placed my gauge block on top of it. I then used a dial indicator mounted on the head of the mill to poke the Z zero location of the soft jaws. Once the dial read 0.050" (could be any number) I zeroed the Z axis on the mach 3 DRO. I moved the head of the mill up until I cleared the soft jaws, then moved over to touch the gauge block that was off to the side. Once the dial read 0.050" then I could write down the z position.

If you do this, you can now zero your tool off of the gauge block/123 block combo by setting the dro to that z position number you wrote down once the tip of the tool touches the gauge block.

I can already tell i'll be using this trick alllll the time!

Once the handle stock was loaded, I touched off my tool, did some adaptive clearing with a 1/4in endmill to remove of the bulk of the stock, then went through and cleaned it up with a 1/8 endmill.

Alright! Finally getting to cut out the bolster! I used a parallel 3d surfacing tool path on those faces and got it looking pretty clean! It helped to do a rouging pass with a large stepover using the ball mill 0.030" above the finish pass.

Turns out nobody makes the size washer I needed so I figured I could just machine my own, should be simple enough!

First scoring the sheet with calipers at a desired size.

Cutting a strip and shoving it around until it stopped being warped.

Then cutting out smaller squares

I've seen a bunch of machinists on the internet use superglue to do crazy things, so this should work. Just glue them all to a mandril

One by one the squares pop free from the super glue connection. I even tried to add some groves to the mandril to give the glue more surface area to adhere to. No success was had with this method,

Instead what worked great was just drilling and tapping that scrap aluminium mandril for an 8-32 screw, which oddly enough clears a 4mm hole perfectly!

Once the parts were screwed down, i just turned them to dimension sanded off the burrs, and bam, fit perfectly. This is an area that may need adjustment in the future, not sure yet.

Oh you thought I was just going to use a vertical mill this whole time? Nope! Routers are great especially if you don't need to worry about flipping your parts!

For some initial roughing, ill be using this down cut bit, notice how the helix will force chips down instead of puling them out of the cut like you would with a standard metal cutting bit. This is amazing for plywood and other laminates. For stabilized wood, it will work just fine.

If you have a router, I highly recommend getting an air blast, same setup as the mister, just no coolant. Gets chips out of the cut, prevents melting, and other weird things that happen when you use a downcut endmill.

Since ill be super gluing the wood down to mill bed using 2 layers of masking tape, thats what ill be using to touch off my z height. This way I don't really need to worry about entering offsets. At least thats what im telling myself.

While routing out the area for the wood blocks, I thought it would be a good idea to create a trench around the perimeter of the pocket so I wouldn't need to worry about the tape lifting or moving the part away from locating walls. In hindsight not really that big of a deal, but here we are. Anyway I deburred the trench a little bit, then wiped down the milled area with a very lightly damp cloth (also not sure if this was a good idea or not, but it worked) then used some alcohol wipes to clean off the surfaces of the wood blocks to be machined

Next up I added tape, trimmed off the excess and made sure get rid of any and all bubbles. I also made sure to rub the tape onto the surfaces to make sure I was getting a really good bond.

I then cleaned off the inside of the tape with more alcohol wipes.

Each of these two blocks will hold 2 handle scales, one block has the left scales, the other has the right ones. I tried my best to make sure that the scales will be cut from mirrored positions from the wood. When you look side to side on the knife, you should see almost exactly the same pattern. This is one of those little details that can make a knife (or any product for that matter) look incredibly well executed.

To make sure I have the positions right, I made some markings on the face to indicate which slot and which orientation the blocks go in.

I then added super glue to the tape and nested the corner into the corner of the fixture. Carefully lowering it so that the reference edges are butted up as close as possible to the sides of the wood.

Then I added some weight on top of the wood just to maximize the likelihood i'd get a good bond.

Sadly I don't have a really good way to cut these blocks into slices to maximize the amount of parts I can get out of one block. The blade on my bandsaw drifts a little, at best I could get 3 slices out of one block of wood, and I know its wasteful, but its more important to me that parts look perfect than it is to not waste material. Bums me out a little that I have that outlook.

Anyway, in that first image you should see just how much material I'm clearing.

Once all that material is cleared out, I took a tapered endmill with a 1/16" nose and surface the handles with a super fine step over. Each scale only took 15 minutes to machine, which is honestly pretty great!

At this point I carefully peeled off the material from the bed of the cnc. Doing it too aggressively would result in the parts snapping in half.

Once removed, the tape peels off easily.

Gotta love super glue and masking tape!

Saddly one of the bolsters did not have a great surface finish. After adding a roughing parallel pass before the finishing pass, and using a fresh ball mill with much higher feed rates, the second bolster looks pretty great right off the miil. However, these parts need a consistent finish, so i'll be polishing both of them.

I started by clamping the parts to the table between some leather. I then did my best to follow the contours and used a flat cut file then a smooth cut, then a #2 cut and finished with a #4 cut file. Even fine files move material faster than sand paper will, so I usually will go from a number 4 straight to 220 grit sandpaper.

When sanding stuff it can be tricky to keep your edges nice and crisp, but these are some techniques I've had good success with!

First, use a rod as a backing material for sandpaper, this will allow you to cut material faster, just be sure to rotate the paper after every pass or 2 so you are constantly using fresh grit.

Next you can use masking tape to protect certain areas from being sanded. This alone will allow you to get some really incredible results!

Also could have used those sanding stones I mentioned a while ago, not sure why I didn't...

The bolster with the bad surface finish is now sanded, and I used this ultra fine sanding pad to brighten it up a bit. Grit progression was files, 220, 320, 800, ultra fine pad, polishing compound, 5000 grit polishing pad.

Ideally I could pull parts out of the mill and they won't need to be sanded at all. I'm also pretty happy about how nice it looks polished!

I was a little concerned about how agressive the corners would be

With the handle lightly assembled and held in a vice (using leather strip to prevent marring) I began filing the lockbar to match the contours of the handle, and then sanded everything so that all the pieces would blend together.

It was really important to do this before heat treatment! Had I done this afterwords, it would have been substantially more difficult to remove that much material, even though it really isn't that much material in the first place.

When heat treating stainless you will be subjecting the steel to incredibly high temperatures. Some weird chemical reactions start happening when it gets that hot, and the atmospheric air can cause some damage to the steel. To prevent that, we are going to seal the spring and blade in a stainless foil envelope. Fold over the edges and crimp them to make sure that air cannot enter the bag.

Ideally there would also be a piece of paper that you would place in the bag which would eat up the remaining oxygen in the bag, giving you a really pure environment for the steel to be in.

So far I really really like this recipe for heat treating 440c stainless. I found it on a knife related forum (link) and it goes as such:

"Here is my HT specs for 440C:

The blade must be enclosed in a stainless foil packet to keep all oxygen away from the blade.

Pre-heat at 1400F ( 760C) and hold for 10 minutes.

Raise temperature to austenitize at 1900F (1040C) and hold for 30 minutes.

Air cool or use aluminum quench plates. ( quench plates are best to prevent warp)

Cool in water to room temperature.Immediately chill to at least -95F (-70C) for 30 minutes.

Allow to warm to room temperature.Temper at 375F (190C) for two hours, cool in water to room temperature, and repeat the temper for a second two hours."
-by Stacy E. Apelt - Bladesmith


Pictured is the stainless bag fresh out of the kiln, sandwhiched between 2 plates of aluminum until it reached room temp. The foil was cut open and now we can see that the parts got some oxides that we will have to sand off.

Right now these parts are incredibly hard and can snap with pretty minimal force.

Before we temper the parts, we are going to soak them in liquid nitrogen as previously mentioned in that recipe.

At a really high level, steel is basically a bunch of crystals stuck together. As you heat it up, and cool it down, those crystals change shape and they change their properties. Once we pulled the steel out of the kiln, almost all of the crystals turned into our desired form. However, there are still a few bits that didn't quite shift right. By soaking them in the extreme cold of liquid nitrogen we force those last remaining crystals to pop over into the same state as all the others giving us a really nice consistent internal structure.

This process is not really needed on simple steels like 1095 or 1080, but when you add a bunch of alloying ingredients to the mix, it becomes more necessary to do.

Once the steel come out of the kiln and is super hard, all of those crystals become really tense, so much like going to the spa, the tempering process relaxes them, and lets them become springy and durable. Just like a spa.

Even though there were not many tool marks, I still went back in and lightly touched up the handle scales, bringing them from a routed finish to 220grit , 320grit, 800 grit, and then used the same ultra fine pads I used for the bolsters.

This is my favorite wax to use to finish up handles. Even though you don't technically need to do this for stabilized wood, it makes the handles pop.

Just rub a little on, then buff it in!

As we start working on the pins, it will make life easier to taper the ends of the pins. This will allow them to self align as we start assembly.

Also I had to remove a little bit of material from the center of one pin. I was planning on using that one to prevent the knife from over traveling when it closes, but I don't really think its needed, and I'll be removing it from future versions.

Using a delightfully small hammer, and holding the pin in the vise between two pieces of leather I begin using the ball end of the hammer to strike around the perimeter of the of the pin progressively working my way in towards the center and back out towards the edge. As you do this use medium to light hits. If you hit too hard it can cause the deformation to happen closer to the center of the pin which will make your life hell when you try to assemble things.

Once you get a uniform mushrooming happening you are done with this step.

The pre peened area of the pins will now act as a stop for the pin. Once everything is assembled it's then time to start cutting off the excess pin material. A jewelers saw works great for this because of how fine the teeth are. Take your time, and cut carefully.

Once the pins are cut, place some masking tape over the remaining nub of a pin and file it down until it's maybe 0.020" or 0.5mm above the surface. When you do this, be sure that the pin is pushed as flush as possible from the backside. The masking tape will prevent you from scuffing up the handle scales if you miss with your file.

When you do this, you will likely need to remove, and reapply the tape to check your progress. Take your time with this.

This is it, the most nerve wracking, part of the whole process. Hit the pin too hard and it will bow out from the center which will cause binding as you open and close the knife. Peen too tight and the blade will not close. If its not peened enough, the blade will rock back and forth. If you peen one of the handle pins too tight that can cause binding near the pivot. Too tight around the scales can cause the scale to fracture.

Using a hard backing to brace the backside of the pin, start mushrooming out the heads of all the pins. Take your time with this. As you see the shaft of the pin slowly disappear test the action of the knife by opening and closing it. Pay attention and look for any places where a pin may be loose, or if anything feels like it's binding.

Ihad initially planned on adding a colored washer in the counter bored area on the bolster, not having done this the head of the pin started to get harder and harder to address, so I made a tool out of some spare steel I had laying around. The thin neck of the tool allowed me to access into those hard to reach areas. I made sure to round and smooth off the tool to avoid creating features that would embed themselves in the brass.

Full disclosure, I ended up making the pivot too tight. To loosen it up, I stuck a metal wedge in between the handle liners and lightly pried them apart. This opened the rivet up just enough to open and close nicely.

Sharpening seems to be an art form that people are scared of or overwhelmed by. Its not terribly hard, you just need to get a muscle memory developed for it. The quick and dirty on how you can sharpen a knife that you could shave with is this.

Get a diamond stone in corse and / or fine grit. True whetstones are cool and all, but they can develop an uneven surface that will make things generally harder than necessary. If you use the wrong technique you will cut into the stone and gouge it.

With a diamond stone, it will be consistently flat, it works great on every kind of super steel out there, and its a lot more forgiving when it comes to technique.

The diamond stone is your first stop, this is for coarse to medium work. Take your knife and cut into the stone as if you are trying to slice a sticker off the surface of the stone. This is about the angle that you want to be at. If you are working with a factory blade you can close your eyes, rock the edge of the blade back and forth until you feel the small secondary beve lay flat on the stone. That is the "ideal" angle of the knife. you can make that angle steeper for a longer lasting but blunter edge, or lower that angle to make a much finer, much sharper edge that will need to be touched up more frequently.

Once you get the correct angle and take a few passes on one side, lets say just 10 strokes, check the edge by rolling your thumb downward over the edge, you are not moving your thumb into the sharp end of the blade so its very very unlikely that you will cut yourself. You are trying to feel a burr form, (see the super zoomed in photo of the micro burr) and once you feel that burr all the way across the complete edge of the knife you will flip sides.and begin working on the opposing side until you feel that full burr form.

Now you can step up in grit, if you started with a coarse grit, move to fine. If you were at a fine grit move up to a ceramic rod. Same principle applies to the ceramic rod, try and slice a sticker off the surface.

When you are at your final grit and you have gotten a burr on both sides, start going from 10 passes on each side to 8, then 6, then 4, then 3,2 and 1. You know you are done when you can no longer feel a burr along the edge.

To close I though i'd make a note of showing what a photo setup could look like for knives. When you are shooting metallic thing pointing a light directly at the metal does nothing but create a hotspot. Try and diffuse the light with anything from frosted acrylic, or even paper towels. To make the different surfaces pop you need to realize that each metallic surface is reflecting its surroundings. Add white panels around the object and you will see the surfaces become more uniform and nice. If you don't know how to light the shot, just look through your camera lens and wave your hand around the object. Eventually you will see the surfaces react, and that's where you should make adjustments.

Lastly, get a piece of black acrylic, and white acrylic. Placing objects on these will make a subtle mirror of your object which ends up looking really slick. Keep in mind these shiny surfaces are also going to react to the things they are surrounded by.

This has been such an awesome project, and is absolutely not the end of this knife. If you would like to see more of what I'm doing with this knife, and some other projects that ill be starting up soon, come find me on instagram @studiosharps !

Thank you so much for taking the time to read this, hopefully you learned a thing or two! Feel free to ask me anything about the project here or on instagram!