Tuesday 17 March 2020

KDF: Kingfisher Digital Foundry

In my day job I'm a high school computer engineering teacher.  The subject is still relatively new and covers a ridiculous range of differing technologies, all lumped under 'computer technology' because they all have some vague connection to it.  While teaching it in class can be demanding, the upside is that in the decade I've been working away at it I've become familiar with technology well beyond my background in IT.  From electronics engineering to 3d modelling and 3d printing, I've had a front row seat to the evolution of digital technology as it rapidly expands and evolves in the 21st Century.

The other benefit of this front row seat is that I've gotten to work with some skilled teachers with diverse technical backgrounds.  From Jeff who was working in CAD and 3d industrial design when most people didn't have a computer to Katy the engineer who was one of the only people in our school board who could get the first generation of 3d printers dependable enough to actually use in class, I know some very technically adept adults.

Then there are my graduates, who have gone off to work in fields ranging from robotics and industrial engineering to electronics, IT and computer science; I know years of very technically advanced young adults who bring a staggering array of expertise and intelligence to the table.

On this cold and rainy March Break Tuesday in the middle of the COVID-19 epidemic, I'm wondering how we could leverage all that expertise and create a niche product that could serve a number of different markets.

I recently found myself frustrated by a lack of parts for the late '90s Honda Fireblade I was restoring.  Knowing that Katy has been using 3d printing to create prototypes in her class, including printing in flexible materials, I was wondering why no one has filled this gap as business proposition.  To replace one perished airbox rubber tube I would have had to pay to get four them shipped from a place in the UK at over €100; no one in North America had one.  I suspect Jeff could have made an accurate model in an hour and then Katy could have printed it in about 30 minutes.  We'd then have that model on hand if it was ever ordered again and could be filling individual orders for them in thirty minutes.  At $20 plus shipping for the part we'd be offering rare parts that meet specific needs for way less than the market is willing or able to at the moment.  For ten bucks more we could put initials or a symbol on a printed piece to satisfy the customizers.  Beyond the capital costs of getting the 3d printer needed, printing a part we had a model for is a quick and easy process and would only require maybe $5 in parts and power.

There are other angles to this besides micro-manufacturing old, out of production parts.  We could also create small batch bespoke parts for companies building prototypes.  By rapidly producing accurate, high tolerance parts, we'd also be creating a library of digital 3d model files that could also be part of the service. Those models could then go on be used in production.

Beyond the out-of-production parts market and assisting companies in with their prototyping needs, there is also the opportunity to pursue bespoke custom parts.  Within five minutes of the first time I saw 3d printed additive manufacturing technology, I started thinking about custom motorcycle fairings.  The default at the moment is to stamp out cheap copies of fairings, but it wouldn't cost much more to digitally redesign unique 3d variants of fairings and sell those, it would just need a large format printer.  Variations in ducting for people wanting to fit a turbocharger?  No problem.  Want to get really crazy?  A dragon scale fairing for a Game of Thrones fan?  This is a 3d printed fairing with scales that have depth and texture.  It would take custom motorcycle design to the next level, especially with a sympathetic paint job on top.

As far as the technology needed to create our digital foundry, I'm partial to the Formlabs 3d printers because they look like something out of Terminator.  They also produce very high resolution models. Their new 3L large format printer comes close to being able to make detailed, high resolution models almost up to a cubic foot in size.



The process of additive manufacturing (3d printing) is surging forward.  It isn't quite ready for the range of parts I have in mind, which would include being able to print 3d flexible parts that are fuel resistant, but this is more a chemical engineering bottleneck than anything else, and chemistry these days is rocketing ahead


I'm hoping that just as I'm ready to retire from teaching, micro-manufacturing will have caught up and I can retire right into another profession making locally developed and manufactured bespoke components in a micro-manufacturing facility of my own, Big Hero 6 style.


Front right is a holographic desktop and keyboard - not quite there yet, but I've got physical hardware that does the same job now.  The blue thing in the back right corner is a large format 3d printer - in the film he prints everything from carbon fibre armour to metal mechanical parts.  This kind of localized production will be the norm rather than the exception in the next couple of decades.  You can watch Big Hero 6 on Disney - I highly recommend it.
***

This isn't the first time I've kicked around the idea of applying emerging digital technologies to mechanics and manufacturing:

Yesterday I was out in the garage using a Structure Sensor to 3d model my motorcycles, there are so many things we could do digitally with mechanical engineering and manufacturing:

Tuesday 10 March 2020

'97 CBR900RR LED Indicators and Rubber Parts

The Fireblade project has the main bits (fuel leak leading to an engine drowned in gasoline) fixed with a carb rebuild and a new petcock.  But there are lots of bits and pieces to sort out before I go get it safetied and put on the road.

The weather warms up tomorrow so I'm hoping for a ride, but Saturday was a -20° March 7th kinda day, so into the garage I went to get the little details worked out.

I was initially going to hold off on the LED indicators, but dead bulbs and broken covers on a 23 year old motorcycle meant the LEDs were actually the cheaper, easier and more modern looking fix anyway.  I'd brought a second set when I got the ones for the Tiger, and also got an adjustable indicator relay, so I had all the parts on hand.






The rear indicators look like they were attached by a monkey.  I ended up pulling them all off and removing all the stripped, half installed wood screws that were holding them in.  I then drilled a hole in the rear plastic under-tray and mounted the second set of LEDs I had on hand from the Tiger upgrade.




The wiring was pretty straight forward with green being ground on both indicators (connected to black on the LED) an orange to LED power (yellow) on one side and light blue to LED power (yellow) on the other.  The LEDs also have brake light function where they strobe when you first press the brake.  Both red cables on the LEDs go together into the middle pin on the brake lights.  It works a trick:

Front LED indicators working:  https://photos.app.goo.gl/4psz7VfHg6mLnvcK6

Rear indicators working with brakes:  https://photos.app.goo.gl/AKJeo69DCpV4gzGAA

***

Some other odds and ends are also proving troublesome.  Living in rural Canada means everyone's still in love with imperial sized fasteners.  All our local hardware stores have rows of 'em, and maybe 2 metric bolts.  It was tricky figuring out the sizes of missing fasteners anyway, but some internet research into OEM parts supplier parts listings got me this far:



https://www.megazip.net/zapchasti-dlya-motocyklov/honda/cbr900rr-14743/cbr900rr-cbr954rr-fire-blade-28645/cbr900rrv-713455

Bolts to hold down the fuel tank:
96300-0604007
Bolt, flange (6x40)


Other metric sized bolts in the same assembly:
BOLT, FLANGE (6X20)

https://www.cmsnl.com/honda-cbr900rr-fireblade-1997-v-canada_model2523/bolt-flange-6x32_960010603200/#.Xlvt-KhKiHs

BOLT, FLANGE, 6X32 CBR900RR FIREBLADE 1997 (V) CANADA


A 6x40 metric bold means (as I understand it) a 6mm wide bolt that is 40mm long.  But metric sizing also looks like M6x40. I'm assuming their the same but don't understand why there isn't a consistent format for metric bolt sizing.

Some other resources around motorcycle fasteners:


https://www.cmsnl.com/honda-cbr900rr-fireblade-1997-v-canada_model2523/partslist/#.XlvtcahKiHs


https://www.webbikeworld.com/motorcycle-nuts-bolts-fasteners/

Lots of links to other websites.

https://www.boltdepot.com/fastener-information/Type-Chart.aspx

The bolts into the frame that hold down the fuel tank and the brace for the windshield are 6x40, but I couldn't get an M6 bolt in there.  I ended up cleaning out the thread by tapping it out again and then it went in nicely.

***

Rubber parts are particularly hard to find on this kind of bike - it's just old enough for existing stock to be gone and just young enough not to have a classic aftermarket parts ecosystem.

The cracked airbox rubber connector to the carbs is nearly impossible to find.  The only
online place I can find them is for a full set of four plus shipping from Europe and ends up at over 70 Euro ($110CAD)... and I only need one of the damned things!

I'm looking into some cunning fabrication options.  Some people have tried plumbing PBC couplings as a kind of secure bandage over the original rubber tube.  Though if I can find one that is the right diameter and well put together I might simply be able to trim it and substitute it.

Others have tried various fixes like heat shrink or inner tubes.  I'll do some more research and figure out next steps...



These guys suggest silicon hose as a temporary measure.  Since what I need is on the airbox side (I was easily able to get carb side rubber replacements), it shouldn't see too much gasoline so the silicon fix might be the quickest way forward.

These guys suggest a sealant.  Again, because it's not on the carb side this might do the trick.  In terms of cost this is the cheapest fix, so I'll probably start here and see how it goes.

I'm wondering if I can't 3d print the correct part.  There are flexible materials we can use to print objects using a 3d printer.  I suspect we're only a few years away from having a 3d printer in our garages that can print these rare parts out of flexible filaments: