WIP: Designing and building the new Resolute with the help of CNC.

[aarongough]

It might be better to ask when you're in full production mode, but how are you finding tooling costs? It didn't bother me at the time, but I feel pretty guilty breaking all those little end mills in machine shop now that I know how much they cost.

Cost for actual cutting tools so far has been very reasonable, but as you said I won't have a real answer until I've done a few batches. I get the feeling most of the tools I'm running should last a long time, with only a few needing to be replaced regularly, the tools I use for roughing essentially.

The real cost is machinery, toolholders and so on. Pretty crazy how much money I've sunk into the shop trying to get everything all lined up.. Gettting pretty close now though!

 

[aarongough]

Thank you Aaron,
yes the thing i had in mind was if you had to do something "simple" like making i.e. longer fixtures and longer blades, altering just the lenght of the blade, leaving all the other parameters pretty much invariated. I was wondering if the software would easily allow for a command like "insert 1 inch more of X axis in this point" and run the fixture and then the blade programs in just a few hours since your decision.

In a lot of cases it would work out something like what you describe, but I'm sure it would still be a day of just CAD work, then several days of making new fixtures and so on... It really does all take quite a while!

 

[HSC ///]

Aaron this is so interesting to watch you go through this process.
As someone who has been around CAD/CAM since 1992, managed machine shops and tool shops for most of my professional career, I'm trying to get away from all that and do things by hand (relative to being a knife-maker hobbyist). And here you are trying to automate everything!
I have access to all the CNC equipment you could ever want. and yet I have no desire to use any of it in knifemaking.
But I get what you are trying to do and commend you for jumping in and doing it.

 

[rustyrazor]

At some point I really want to make a 'getting started with CNC' video... Likely still more than a few months away unfortunately.

at some point I would love it if you did this, lol. I have wanted to look into CNC since I first heard of it, but with the army taking up more of my time than I would like and trying to get a few hand made knives out and build my reputation with the rest of my time, it's never came to be as of yet... I have my user knives to 80% where I want them but I can't keep up with demand and I honestly think CNC will be my fix at some point, I'm just stunned as to where to start.

 

[aarongough]

Aaron this is so interesting to watch you go through this process.
As someone who has been around CAD/CAM since 1992, managed machine shops and tool shops for most of my professional career, I'm trying to get away from all that and do things by hand (relative to being a knife-maker hobbyist). And here you are trying to automate everything!
I have access to all the CNC equipment you could ever want. and yet I have no desire to use any of it in knifemaking.
But I get what you are trying to do and commend you for jumping in and doing it.

Hey mate!
I understand the desire to do things by hand for sure! For me it's not so much about doing them by hand as it is about producing an actual tangible physical thing... I worked in software for so long, go tired of the 'intangible house of cards' that you're constantly 'renovating'... That's basically what large-scale software feels like to me, an invisible house that's never finished.

Automating the knifemaking stuff really puts the focus on what I enjoy which is developing processes and tooling! The last 6 weeks have been a lot of fun... Bit stressful because I'm running short on money and time, but a lot of fun none the less. Soon I'll be into a rhythm with production and then I'll be able to spend more relaxed time working on improved tooling and so on, looking forward to that very much!

the other goal with CNC was simply to finally get to the quality and consistency that I wanted and could never quite hit by hand. The new knives look so far like the best ones I've ever made, and I'm *really* stoked about that!

-Aaron

 

[rustyrazor]

"the other goal with CNC was simply to finally get to the quality and consistency that I wanted and could never quite hit by hand"

-Aaron

That is a big factor for me as well. every knife I make in a particular model can have a visible difference from another of the exact same model. I know this adds to the "handmade and Unique" part of it, but when I get something right and that's the product I want to get into the hands of the end user then it would be nice to be able to get it exact.

 

[12345678910]

Aaron

Can you go into detail about the Cerakote ?

 

[AVigil]

So how many Resolutes are you going to need to sell to break even?

 

[aarongough]

So how many Resolutes are you going to need to sell to break even?

Quite a few... like a lot! But that's fine, I get to make knives! Right now things are a little dicey financially, but I think once the next batch is finished that should put me on a good footing and I'll just work my way back from there. Hopefully by the end of the year I'll be much more relaxed

 

[aarongough]

Aaron

Can you go into detail about the Cerakote ?

Hey Count!
Sure!

The main equipment needed to apply Cerakote is:
* A sandblasting cabinet, with 100 grit or 120 grit aluminum oxide sand.
* An air compressor (needs to be pretty decent sized to run the blasting cabinet)
* HVLP spray gun with 0.8mm nozzle
* An oven for baking the blades at 250ºF, an old kitchen oven would probably work ok, bonus for convection.
* A GOOD respirator, you really don't want to breathe this stuff. I recommend combination Organic Vapor and P100 cartridges on a full-face respirator. Mine is a North 5400 series respirator.
* A very clean place to spray.
* A dessicant drier for your air-line when spraying, the small disposable ones work great: http://www.mcmaster.com/#7793t1/=yh7gcz
* Graduated glass cylinders for mixing the paint
* Strainer for straining the mixed paint
* Large syringes to help dispense the paint from the bottle when mixing
* Polypropylene funnels for pouring paint (dollar store ones are fine)
* Nitrile disposable gloves for handling parts
* Wire for hanging parts (I use mechanics wire, very cheap and pliable)
* Acetone for cleaning the spray gun
* Neoprene disposable gloves for use with acetone when cleaning your spray gun (acetone eats nitrile gloves instantly)
* Really, really, good lighting.

I buy my Cerakote straight from NIC industries in the USA. They're pleasant to deal with and shipping isn't too bad. I bought my HVLP gun from them too, it's an Anest Iwata LPH-80 and seems like an excellent spray gun so far, here's the link: http://www.cerakoteguncoatings.com/finishes/SE-138/iwata-spray-gun-kit-lph80/

They have a less expensive spray gun now, when I bought mine a month or so back they only sold the Iwata gun. Technically speaking you can use any HVLP gun with a 0.8mm nozzle, but I wanted to buy what they recommend to avoid issues.

I use the Graphite Black cerakote, and I mix it in a 19:1 ratio.

The coating overall is kind of fussy to apply. If you have any dust floating round your shop it will ruin the finish, if you have water in your air-lines it will ruin the finish. Same for dirty blasting media, incorrect blasting media, etc... It's important to nail everything down to get the finish you want. There's no opportunity to do touch-ups with this stuff, and you can't sand it or anything so it takes a bit of work to get it to go on right.

Application steps:

First the parts are finished to an appropriate grit by hand or on the grinder. 180 or 220 grit is fine, no need to go finer than that. Then the parts are degreased in Acetone for about 15 minutes. After this point it's important not to touch the blades with bare hands, nitrile gloves only from here on. Don't get any oil of any kind on the blade.

I then blast them with 120 grit sand at 90PSI. Make sure every bit of the blade has a matte finish after you're done. If there's any shinyness that part needs to be re-blasted. Good visibility into the blasting cabinet is a must.

The parts are then baked for 30 minutes or so at 250ºF to flash off any water or solvent left on them. If you see any 'wet patches' on the blade after the baking this indicates the presence of oil and the parts need to be degreased and baked again.

The parts should then be wired up using thin wire so you can handle them without touching them with your gloves!

You're now ready to paint!

I set my gun for about 15PSI (with the trigger down) and full airflow (the knob at the bottom of the gun). I then set it up so I'm spraying about a 3" wide fan at 6" away, and putting down quite a bit of paint.

I then hold the part up in front of the little spray booth I made and spray the hardest to reach parts first. The plunge line, the choil, then the butt of the knife, then the inside of the skeletonization, then the spine, then the edge, then the flats. I'll generally spray the spine and the flats twice each (two coats) as it's really, really, hard to tell when you've got full coverage.

The part should 'dry' after about 30 seconds or so. If it still looks wet after 40 seconds (tops) then you're applying the paint too thick.

The part should then be hung up for about 10 minutes to let all the solvents in the paint flash off before it goes in the oven.

Note: even though at this stage the paint will look dry, IT'S NOT! You absolutely cannot handle it or knock it at all before it's been baked, it will smudge and you'll have to strip the paint off and re-apply it.

Now the parts are ready to be baked! Bake them in the oven for 2 hours at 250ºF. My oven is pretty small so I 'flash cure' them in lots of 3 at a time for 15 minutes at 250ºF, after the flash cure they can be handled without marring the paint, which means I can pack them in tightly for one final long cure. I'm planning on getting a bigger oven soon!

After that the results should look very much like 'black steel', rather than 'black paint'. The coating shouldn't have any visible 'thickness' to it, and should look more like DLC than paint.

At this point you're done! Make sure to clean the spray gun really well. Empty any excess paint, then put some acetone in the cup and spray it through, then disassemble the gun and clean the individual parts in a container of acetone.

The coating really is a bit fussy to apply, but the results are pretty awesome.

I don't get my blades 100% right every time. So far I'd say I have to re-coat at least 5-10% of the blades I spray, usually because of incomplete coverage.

Hope that's helpful mate!
-A

 

[aarongough]

Hey guys!
Been a while since my last update.. I've been head-down barging through issues and getting lots of work done!

As I said previously that I was going to I had a shot at making a vacuum form for the sheaths. First I made a vacuum spreader to go under the mold:

Then I made the mold itself, here it is mounted on top of the spreader:

As Nathan warned it's a bit tricker than it appears, and this unfortunately did not work out. In this case it seems that because the Kydex that's in contact with the aluminum cools faster than the side that's not, the whole piece slowly warps as it cools! The ends of the shape are raised about 1/8" after it's cooled to room temperature. The slight bend made for terrible gaps in between the rivets when assembling the sheaths...

I want to try new, more consistent, ways of making sheaths going forward but after spending several days working on this mold and the spreader I decided that for the time being I had to go back to what I knew worked and make the sheaths for this batch by hand.

The difficulty with this though is the fact that I don't want the inside of the sheaths to match up to the knives! I want the sheath only to touch the handle and the very tip of the blade, with space in the sheath everywhere else to help prevent scratching, and to prevent any sticky residue on the blade (sap, blood, etc...) from being able to freeze the blade into the sheath.

To help with this I made a 'master' that I use instead of the blades to form the kydex sheaths when they're in the press:

You can see the 'master' on the right, the template for the outline and rivet positions on the left, and the final sheath in the middle! The production sheaths will all be black, but I made my prototypes with grey as I had a lot of it kicking around that I wasn't using for anything.

It took several days of tweaking and hard work next to a hot griddle, but in the end I got the sheaths for the batch all made up:

In addition to the sheath body itself I also had to make the webbing belt-loops and the kydex plate that allows them to be attached to the sheaths... During this process I found out the hard way that the fumes from melting/burning nylon are a) very bad for you, and b) not stopped by OV/P100 respirator cartridges... I was part-way through burning the rivets holes when all of a sudden chest pains and a wicked headache appeared... I ventilated the workshop and removed myself to fresh air in a hurry.

After that little issue I had to work out a better way of making the holes through the webbing. Standard drill bits won't cut them cleanly, and neither will anything else I've seen. So I doodled about a bit and ended up making this:

It's shaped very similarly to a standard punch, except it's sharpened on the inside edge which allows it to be run through a drilling guide. Using it while it's rotating in a hand drill produces reasonably clean holes very quickly, than can then be heat-sealed with a quick wipe of a blowtorch. No fumes and it's much faster than burning the holes through...

Once those were done I started making the plates to attach the loops to the sheaths. I had done only a few by hand when I realized I was being silly and that it would be much faster to spend a few minutes programming the machine to do them. I wasn't sure if the holding the kydex in the vise would work, but for drilling them it worked fine. I then made a quick fixture out of MDF that allowed me to bolt them down through the holes so the periphery of the part could be milled with a straight-flute router bit. Worked very well and I was able to turn out the little plates for all the sheaths quite quickly!

Here you can see all the parts that go together to make the sheath:

After all the sheath parts were done it was finally time to start assembling the knives! I've used epoxy when doing all my glue-ups in the past, but I've also found that the viscosity of the epoxy can cause issues with slow squeeze out and even gaps. To remedy this I was interested in trying a water-thin CA wicked in between the handle scales and the tang. I've used this technique on prototype kitchen knives in the past with great success! I only really treat the adhesive in my handles as a sealant, as the corby bolts I use provide plenty of mechanical strength.

In order to see if it would work for the production blades I put together a scrap blade and scales using CA wicked in, then ground the handle in half to see how far the glue had penetrated:

On one side the penetration was quite good, on the other not so much. Also some of the guys on Instagram warned me off CA telling stories of slow deterioration...

You can also see that I would benefit from Corby bolts with longer shafts... These ones are the longest that I've found so far, I'm planning to have JephCo make me some custom ones soon.

Given the results of the cutaway I decided to stick with the West Systems epoxy that I've always used, however I changed to the 206 'slow hardener' to get some more working time. With that sorted I got the batch glued up in reasonably short order:

Next batch I'm going to try the 209 'extra slow hardener' as it supposed to have much longer open time and also lower viscosity, which should help prevent the slow squeeze out that's such a pain.

So this all brings us up to today! Today I was able to start grinding off the heads of the corby bolts on the handles and doing my final finishing and sharpening... The very first customer knife was assembled and ready to go today!







Very, very, pleased with how the MkIII knives have come out so far! There's definitely still some areas where I'd like to get quicker/better (sheaths being a big one) but I'm really very happy with how everything has come together!

Thanks for following along so far guys, and for all your help and advice!

-Aaron

 

[HSC ///]

You can also see that I would benefit from Corby bolts with longer shafts... These ones are the longest that I've found so far, I'm planning to have JephCo make me some custom ones soon.

-Aaron

Aaron you probably know about fastening principles so while longer Corby bolts wouldn't hurt, I believe you only need threads engagement equal to the diameter of the fastener and that gives you the max holding power. 1.5 x would be very sufficient

 

[Nathan the Machinist]

I don't think he's worried about thread engagement. I think he wants more of the barrel going into the second scale.

 

[Nathan the Machinist]

That mold was a cool project. You need to heat the tool and put a fan blowing on the work to get a more even cooling to reduce part warping. Also consider pulling the part off hot and cooling in a fixture. Thermoforming is a dark art...

Edit
A metal tool like that can make many thousands of parts, that's a production tool. For lower volume consider a soft tool. Ren or wood or even mdf can form hundreds of parts and don't chill the part so drastically. You can pull a vacuum through mdf too.

 

[aarongough]

Aaron you probably know about fastening principles so while longer Corby bolts wouldn't hurt, I believe you only need threads engagement equal to the diameter of the fastener and that gives you the max holding power. 1.5 x would be very sufficient

I don't think he's worried about thread engagement. I think he wants more of the barrel going into the second scale.

Yep, as Nathan said I'd like to get a bit more of the barrel into the second scale for strength and alignment purposes... I actually may move away from corby bolts and to a completely custom fastener to maximize strength, will be looking at this more in a few months.

 

[aarongough]

That mold was a cool project. You need to heat the tool and put a fan blowing on the work to get a more even cooling to reduce part warping. Also consider pulling the part off hot and cooling in a fixture. Thermoforming is a dark art...

Edit
A metal tool like that can make many thousands of parts, that's a production tool. For lower volume consider a soft tool. Ren or wood or even mdf can form hundreds of parts and don't chill the part so drastically. You can pull a vacuum through mdf too.

Thanks mate! I'm actually considering whether I should skip the thermoforming and go straight to having the sheath halves injection molded. It would be nice to switch to a less temperature sensitive material, like say glass reinforced nylon. Apparently Protolabs can do low volume stuff fairly cost effectively...

I may also experiment with reaction molding using glass reinforced urethane or perhaps compression molding of fibre reinforced nylon pellets, as that's something I'd be do in my shop... If you have thoughts on which of the process I should chase down I would love to hear them! Thanks again for the help!

 

[AVigil]

You could just have some custom Loveless bolts made.

That is about as strong as they come.

If you need a stronger fastner then it is time to use a crow bar instead.

 

[aarongough]

You could just have some custom Loveless bolts made.

That is about as strong as they come.

If you need a stronger fastner then it is time to use a crow bar instead.

Yes, I agree they're very strong! There are two issues though:

1) I don't like the way they finish out...

2) My handles and tang require *very* close location to one another because they're made separately... The threaded portion of a loveless bolt, or acorn bolt is not close-tolerance enough to provide the repeatability I need. On the corby bolts I'm using now, the shaft is 0.1231" in diameter (give or take about 0.0002") and I ream my tang and scale holes to exactly 0.1235" to provide a close fit. Whichever fastener I switch to needs to have a close tolerance smooth central section to act as a locating pin...

So given those requirements I have plans to get some custom fasteners made. They'll likely look like small 'shoulder bolts' and will be made from stainless steel. The female half will basically be a small blind nut with 5/32" outside diameter.

 

[AVigil]

Might be easier to just put in a few locating pins to insure perfect fit.

That would give you more leeway in the fastener dept.

 

[aarongough]

Might be easier to just put in a few locating pins to insure perfect fit.

That would give you more leeway in the fastener dept.

Yes, I had considered that in the past. Problem is that then you'd have to drill/ream blind locating holes on the backside of the handle scales, and the fixture for machining the front of the scales would have to locate off that. It's not impossible by any means but having that critical relationship split up over two operations introduces quite a bit more opportunity for mismatch. The way I make them now all the operations including drilling the pin holes are done in one setup which ensure the relationship between the pin holes and the profile of the handle scales is as close as possible...