Alpha Pens are 97% Shipped!
about 11 years ago – Wed, Nov 13, 2013 at 04:18:04 AM

Hey Backers! 

Hope everyone is doing well. I just got back from Hong Kong, and it might seem like I've been hiding out...because it's true! I've been hiding out getting (most) of the remaining pens assembled and shipped! As of tomorrow, I'll be 97% shipped and I'm pretty excited about that; I hope you are too :) I also wanted to thank everyone for their patience and sticking with me through the trials of creating and manufacturing a new product. It's obviously taken much longer than expected, but I'm looking forward to the next 100% shipped update. 

This chart actually represents tomorrow. I'll be putting together and shipping another 30 pens, which will leave 30 rewards that remain outstanding. Those will ship (hopefully) late next week. 

Continuing clip problems and possible solutions

It's funny, this was the only part of the project that I wasn't worried about at all. I've made thousands of clips over the last couple of years and was not expecting any sort of surprise here. Well, as Murphy's Law would have it: SURPRISE! I think there is a cautionary manufacturing lesson to be learned here...

As reported last time: I ended up with a batch of 500 clips that were breaking 100% of the time when they were bent by the spring company. Not a very good "yield rate" as we call it in the industry. After trying a couple of work-arounds, we did some empirical testing and discovered the Rockwell hardness was over 45. The spec for 6Al4V titanium is 33-35 Rockwell. It was basically a really bad batch of material. The company that waterjets the clips also provides the material. They dug around in their scrap pile and found a sheet of the "old" material that I've been using for the last year and a half. They had enough for 265 clips...but that's not enough to finish all the pens...and there is no more of that material. 

So, my solution was to order in some "certified" 6/4 Titanium. This is produced to exacting aerospace standards, and normally used for air and spacecraft. I mean that literally, and not in the "marketing-hype" sort of way that most people talk about "aerospace" material. As such, it's 50% more expensive than non-certified 6/4 titanium, but the purpose was to have material "guaranteed" to be 33-35 Rockwell. 

I had 350 clips cut and set to my local spring company for bending. That was just before I left for Hong Kong. I planned to pick up about 600 finished clips the day after I got back from Hong Kong. Er, nope. I picked up the 265 clips cut from the old material because these were fine, and this is what I have shipped out over the last three days. 

However, the new "certified" material has also been exhibiting a hairline fracture after bending, even though it is within specification for hardness. It was not discovered until about 150 clips had been bent. So, I spent about an hour at the spring company today working with the production manager and their tooling engineer to figure out what the heck the problem was. At this time the general consensus is that the material is a slightly different thickness than the previous material, making the precision forming tool the wrong size, and leading to excessive force at one part of the bend, instead of even force over the entire radius. 

It will be amazingly good if this is actually the problem, because it's easily solved by making new tooling. I'm hopeful at this point, but it's entire possible that this new fix will also fail...just setting expectations. The breakage issue is a particularly bad problem because it not only affects the final fulfillment of the KS project, but my entire business. I use Ti clips on all my lights, and I make a significant number of "aftermarket" clips for other flashlights. So, if this fix does not work, I have no idea what the next step is. 

Before crossing that very problematic bridge, the plan is for them to re-tool and try to bend the remainder of the clips on Tuesday. Assuming the corrective action actually fixes the problem, I will pick up on Wednesday and ship out the remaining pens on Thursday and Friday. My fingers are certainly crossed, are yours? 

Cheers and thanks again for your infinite patience and unfailing support. You guys rock! 

Jason

Pocket clips are now delaying delivery (slightly)
about 11 years ago – Tue, Oct 15, 2013 at 09:53:46 PM

Bad news first...

Hi Folks,

Well, as is traditional, let's start with the bad news. As you heard in my last update, there was a problem with the new batch of pocket clips. The material is much harder than normal, and causing the pocket clips to break, instead of bend. As of this moment, there is no way to repair or re-work the last batch of 500 clips. The vendor has a small sheet of (hopefully) good material; enough for about 100 clips. They are running those in the next couple of days. They are trying to get more material on order, but I don't yet have any information on how long that will take. 

If anyone was wondering, I'm hoping this incident is this project's "major malfunction." Other than being a little slow (working on a clone for my next KS project) things have gone fairly smoothly. 

(Above) In total I've shipped rewards to about 750 out of 1063 backers. A total of 70% if anyone is counting :) The Ti and Twice charts have creeped up a bit since the last update. I shipped about 100 Ti pens this week and about 25 Twice is nice orders. I only have about 25 pocket clips left and those will find their way onto pens, and the pens into envelopes this week. 

On to the good news! 

Machining is proceeding well at this point! I'm actually hoping to finish all of the machining this week. Then I'll just be waiting for pocket clips, and pens will ship very quickly after that. I'll also be in Hong Kong (Oct 18 - Nov 2) visiting family and meeting with the factory who is manufacturing my Beta-QR flashlight...my new campaign that is wrapping up in a few days. So it's actually "good" timing for a delay because I'll be gone anyway...right?...hey...are you with me?...guys?...hello?

Drilling the cap remains the most challenging operation out of all the machining. I've learned it's difficult not because of the material, but because of the machining conditions. If you drill a hole with .4063" diameter in a part that is .4900" in diameter, it means the walls are "very" thin. A fine thickness for a pen cap, plenty strong and all, but if you are a drill...you really really don't like that situation. The cutting pressure generated by the drill forces the walls of the cap to bulge microscopically, and then contract, over and over. This is basically vibration and vibration eats the very sharp corners of the drill super fast. 

Abnormal point wear on a drill is most often an indication of too much RPM. My first attempt to alter the machining condition was to slow down the RPM of the drilling, below the manufacturer's recommendation for this specific drill, in titanium. After that change my drill ran about 100 parts, instead of 50. Sweet right? The next drill ran 12 parts before failing. The next about 8. Must be two bad drills right? I mean the first one worked great. The third drill made about 10 parts. Well, an hour and $750 later I decided it was time to try something else. 

(Above) Just in case you ever wondered what $1,600 dollars in used drills looks like. These can be resharpened once or twice. The three gold drills have already been resharpened. Most drills can be resharpened more times, but the coolant-through drills cannot; because grinding the drill changes the relative location of the coolant hole. If you look at the drill on the far right, you can see the hole is entirely on the flat. Each time you re-grind the hole moves slightly closer to the cutting edge. The far right drill is now scrap material. 

To make a long story short, my next strategy was to "pilot drill" the hole in the cap. This means you run a smaller diameter drill first, and then enlarge the hole to the final size with the big drill. This is a standard machining practice with normal drills on manual machines, but specifically not recommended for coolant-through drills. Drilling into a pilot hole causes the large drill to wear more quickly in "normal" machining conditions. However, it also reduces the thrust force required for the drill to cut. The thrust is what's causing the material to bulge...and wear the drill out in ~10 parts. So far pilot drilling seems to be working! The drill is still wearing much faster than "it should," but much slower than before. I think that's about the best I can hope for at this point. 

Well, I'm tired of typing, and if you made it this far, you are tired of reading :) Thanks for your patience. Sorry for the delay, and I'll let you know as soon as I have more info on the clip delivery! So close to 100% shipped! 

Cheers,

Jason

Shipping Update and "Potential" Delay (fingers crossed!)
about 11 years ago – Tue, Oct 15, 2013 at 11:18:12 AM

Hi all, 

No big news to report but I wanted to give you my current shipping update. I also wanted to let people know about a possible delay. My last batch of 500 clips was supposed to be finished last week. Unfortunately we discovered the material it was cut from was much harder than it should be, and is unable to be bent. I'm currently working with the vendor to see if we can save the batch and get them finished ASAP. It's possible that they are not salvageable and we may have to start over. This would lead to a delay on the last couple hundred pens, and push back my goal to be 100% delivered by October 18.

The final remaining challenge is getting the Ti caps finished without running out of drills. Right now I have 7 drills (@ $300 each), and if you remember, I was getting about 50 parts out of each drill. The math says I'll only get about 350 parts, and that's not good enough. So, I'm going to spend a couple of days trying two new machining strategies that (I'm really really hoping) will increase the drill life. 

Over the course of the next two weeks, I'll be alternating making Ti caps and finishing the Ti "Body 2" machining. The "Body 2" operation finishes the tail end of the bodies that have already been machined on the front side. The good news is, the cycle time of the tail is much shorter because I don't have to drill the body or cut the internal threads for the grip. The "alternating" strategy will allow me to ship the final Ti pens in batches, rather than waiting till I have 100% of the remaining parts and shipping all at once. 

(Above) Slowly but surely getting there! Tomorrow I'm starting the final aluminum caps. Right now I "believe" I'm 100% shipped on Aluminum pens that are not combined with a Ti pen. These "combo" tiers include the "twice is nice" reward and people that may have pledged for an Aluminum pen and added a Ti pen after the campaign. There may be 5-10 Aluminum pens (single) that have not shipped, and they will be sent out with the next batch in a week or two. 

Thanks again for your patience and support! 

Jason

Shipping Continues, and Time Travel!
about 11 years ago – Tue, Oct 15, 2013 at 10:16:55 AM

Hi all,  

Just a quick progress update. Unfortunately I've been out of commission for three days with a 102 fever. I've basically lost about a week of production time, but we are still getting there! My hope was to have all the titanium "body 1" parts done this week...but that is looking like the end of next week. According to my spreadsheet, I have 10 full days of machining remaining. Probably smart to call that three weeks, but the end is in sight! 

(Above) I also have a "correction" to publish. I realized I made an error in my spreadsheet for the % machining completed. The current and "correct" percentage is listed above. The bad news is, we moved backwards in time. The good news is, there is still the same amount of work left to do :) 

(Above) So I have 200 aluminum pens on my desk waiting to be stuffed into envelopes. To continue the time travel theme...Looking into the future to Monday, September 23rd, I should be 100% fulfilled on aluminum pens once I get this batch shipped! That, of course, assumes I haven't made any gross errors in my calculations. My temperature is still about 99.6 so I don't make any guarantees :) 

Oh, and I also started an instagram account if anyone is interested :) 

Cheers,

Jason

"Failure to Cap" & Production Updates
about 11 years ago – Tue, Oct 15, 2013 at 06:51:39 AM

Hi Backers,

Hope all of you are well! Things are plugging along in the shop. Let's get right to the top story. 

Failure to Cap: 

Properly capping any 3 Pc pen requires a little bit of torque! 

My friend sent Wilson me the photo above. A very sad photo indeed. I don't want Alphas to ruin any more clothes so I need to point out: You must tighten the cap if you don't want it to come loose! This also applies to the grip where it threads into the body. Just a little bit of torque will cause enough friction in the threads to prevent them from coming loose. You don't need a lot of torque, just some. 

This applies to any object that threads into another, like the screw and bolt in the photo above. No pressure between the threads, no holding ability :) How much is enough? Just try to unscrew the parts...if they come apart too easily it wasn't enough torque. If they take too much force to separate, then you used too much torque :) 

Production Update! 

---------------------------------------------------------------------------------------------

The short version: the next batch of aluminum pens will start shipping next Wednesday.  Ti and "Twice is nice" will start the next round of shipping the week after. At that point, I should have about as much inventory on hand as I am able to ship in the evenings. Right now my best guess is 80% fulfillment by the end of September. 

---------------------------------------------------------------------------------------------

Nothing earth shaking above, steady progress, but it's still a long slog to the end. So I've finished all of the "Body 1" parts in aluminum. I've been working through the second op on the body. The cycle time is just over a minute (really fast) but each part must be loaded and unloaded by hand...roughly 1500 times. Since the cycle time is only a minute, that doesn't leave me much time to do anything else but sit there. I've had a buddy in the shop helping load the machine so I can do other things like buff threads, load/unload the tumbler, wash and dry parts, run to the plating company, pick up more material, etc.

I did make one change to the production schedule. I decided to run the "Body 2" operation next instead of just switching over to the Ti "Body 1" parts. "Body 2" is finishing the second op on the body; cutting the end where the tailcap "posts." Each tray above holds about 50 pcs. These are Tobiko (fish egg) trays from the Japanese restaurant my friend works at. Turns out they are the perfect size for pen parts :)

So the image above shows why I decided to run "Body 2" in aluminum. I have lots of Alu caps and grips that are finished and waiting for bodies. This way I can start shipping aluminum pens (which I do at home after work) and get a bunch out right away.

100% the aluminum grips are machined and finished, but they all need to be plated. I'll be taking these to they plating company in batches of 300 pcs, along with 120 pcs of aluminum bodies. The reason I'm taking fewer aluminum bodies at the same time is to avoid losing a lot of parts if the thread tolerance is off. If a batch comes back too tight, I can always take down the diameter of the threads on the next batch with the buffing wheel, even though the parts are already machined. Because each body has two threaded parts, that doubles the chance of an error. The body and cap also have larger threads, and larger threads require more force to machine, and this increased force wears the tools faster...compared to a smaller thread. 

A technical digression...

So far the Haas ST10-Y has 206 hour on it (cycle start time). This is the amount of the the machine has spent actively running a program. The "feed cutting time" is the actual time a tool has been engaged with the workpiece. This number is much lower because a CNC spends most of it's time changing tools or getting from one place to another...and not actually cutting material. 

New CNC machines are a lot faster than old ones in terms of how quickly they can physically move during "rapid" motion. "Rapids" generally occur when the machine is traveling, not cutting. A rapid takes place at the full speed the machine is capable of moving. This lathe can rapid at 1400 inches/minute. My old mill can only rapid at 250 inches/minute. 

Machine shops spend a lot of time trying to shorten the cycle time in order to lower cost and increase profit. This is an interesting example where investing in a new machine might actually "save" more money than trying to make an old machine more efficient...especially since about 2/3 of machine time is spent moving around and not actually making the part. Imagine if my lathe could only do 250 inches/minute? That means it would take about 6x as long to make the same parts! 

Okay folks, ta ta for now! 

Oh and please don't forget to post some photos on Tumblr when you get your pens! :) 

Jason