[Kickstarter Update] - Kossel Pro's Delivery Schedule

Hello everyone!

At long last, we have enough information to put out a meaningful update to the Kossel's schedule. 

Current projected schedule for Kossel Pro Delivery

First, let's start with the BOM of the Kossel.  For those new to manufacturing, the BOM is the bill of materials - a list of all the components that go into making up the Kossel.  The current BOM is a work in progress; it is reasonably complete in the sense that it captures all the custom made components to build one of the configurations of the Kossel (in this case, the full size Kossel Pro, 750mm tall, 360mm triangular structs, with heated build platform).   What it does not include are all the fasteners that go into the Kossel; adding the fasteners into the model properly actually will take 2-3 times as long as adding the main design components, so this level of BOM completion is a reasonably acceptable compromise.  (It will become complete, down to the last screw, in a matter of days; this then gets sent to our packaging supplier to help with kitting the subassemblies).  When complete, we estimate close to 1000 individual parts going into the final BOM (we count each individual fastener as a part, since we are selling a kit, the expectation is that everything you need to build the machine is in the box).  

A very big spreadsheet of items.  This Spreadsheet is CAD generated and list all the major components of the Kossel.

The items highlighted in yellow are custom manufactured items for the Kossel.  As you can see, this comprises of a significant portion of the BOM.  In this group are made-to-order components such as our linear rails, carbon fiber structs and borosilicate glass plate (which are cut to order), as well as OpenBeam designed and manufactured parts, such as the actual OpenBeam extrusions themselves, the corner vertexes, and all the injection molded parts.  Also in this category is the Brainwave Pro (we decided to change the name to Brainwave Pro after consulting with Matthew Wilson, the designer of the original Brainwave).  Each of these items have been in manufacturing for a very long time.  Let's take a look at each in detail.

Injection Molds

Injection molding is what sets the OpenBeam Kossel Pro family of printers apart from the rest of the printers out there.  It is a painful process to get started on, but the resultant part is much stronger than anything laser cutting (DeltaPrintr, Kossel Clear) can produce.  The material we are using for injection molding is EMS Grivory GVX5H.  It is a high performance, 47% fiberglass reinforced nylon copolymer that was created by the Germans and Swiss to replace die cast aluminum parts.  We have been using EMS Grivory for our joining brackets now and we have had great results with it.  It is a good compromise on part cost and stiffness and it will allow us to ramp production to meet future demands.  

OpenBeam signed off on the injection mold designs on March 5th.  For those interested in what goes on in the construction of an injection mold tool, please take a look at our old blog entry here covering the original OpenBeam mold.  This time around, we commissioned 4 molds to manufacture the 15 custom parts for the OpenBeam Kossel.  

The current projected end date on the injection molding is April 19th.  This is the "T0" date, where the first shots of plastic are shot into the mold and parts are made.  We have to approve the T0 parts (or make modifications as the case often is) to tune the process to get good parts.  Our injection molder has been really good about getting usable T0 shots on past projects, but we will never release the full 150+ set order in one go without the sampling.  Due to the project running late, however, I will risk buy (ie, buy sight unseen) the T0s for the early bird kits, so that if the T0s are successful, we can ship early bird mechanical kits right away, instead of waiting for our turn again on the injection molding presses.    

Based on this schedule we expect to be able to ship early bird mechanical kits on or about April 28th, and the balance of the mechanical kits on May 15th.  We will hand kit the early birds to get them to backers ASAP and we will go through our normal kitting channel for the mechanical kits.

Relevant links:  

Status update from Western Tools

Mold CAD files - viewable with free eDrawings viewer:

Vertex Extrusions

Our production quantity of vertex extrusions have arrived at our machinist in Shoreline, WA and we are finishing up the production fixture.  We switched to a CNC 4th axis setup for machining these parts to eliminate having to reload the parts.

As a side note:  12 years ago, when I was a student-mentor for FIRST Robotics, Sherline Tools sold us a CNC mill and CNC lathe to get our robotics lab started at a really steep discount.  That was my first real intro into engineering.  I was more than happy to buy one of their CNC tables for this project.  :-D

We do not expect the vertex extrusions to be a gating item in kit delivery.

Electronics and Software Integration:

Electronics and software integration have been giving us a lot more trouble than we'd like. As of Thursday of this past week we've validated the board electrically; turns out, older variants of Marlin broke the bed levelling switch logic, and newer variants of Marlin fixed it, but Johann's implementation of G29 auto levelling code had been taken out of Marlin head.  Johann will be working on porting this back into the most current version of Marlin, as well as testing FSR bed levelling and SD Card printing capabilities of the new Brainwave Pro.

The OpenBeam Kossel Pro will NOT ship with FSR support (as the configuration has already been locked for manufacturing), but we will be testing FSR support and releasing documentation on how to retrofit for FSR bed levelling.

The boards that we are currently testing with are hand assembled by PCA Corporation in Bellevue, WA.  While their worksmanship is superb, hand assembly is way too expensive a process to use for mass production.  Our next set of boards will be built by our contract manufacturer overseas using robotic pick and place machinery, and will feature "design for test" features on the board.  We will be performing 100% testing on all boards before they leave the factory; to aid in this, critical key test points are bought out to the bottom of the PCB, which are then contacted by a custom designed (and 3D Printed) test fixture with a bed of pogo pins.  With the pogo pins in place, we can energize the board and "exercise" all the components on it, before flashing on the production boot loader.

The catch is, with over $20,000 on the line for the production run, we want to get about 2 weeks of solid testing on multiple printers on these prototype boards before launching the production run.   Even after launching the run with the contract manufacturer, we will need to do what is called a manufacturing pilot, where we verify that the test procedure put in place, as well as the contract manufacturer's tooling (such as solder paste stencil, pick and place programming, etc) before releasing the rest of the boards to the build.  Remember, it is awfully expensive to rework boards.  

We are building a total of 450 Brainwave Pros for launch.  We arrived at this figure in order to hit the price break for the TI DRV8825 motor driver chips, and that also happens to be how many Atmel processors are in a tray.  We have already purchased the long lead time components and they are sitting at the CM.  This is important, because we can be looking at 20+ week lead time on components.

Based on the above projections, we expect to ship early bird full kits on or about May 16th and the balance on or above June 6th.  

Kitting and fulfillment:

As mentioned above, the early birds (as well as manufacturing pilot units, built for validating the instructions) will be kitted by hand here in the US.  The final assembly will always be kitted in the US, due to the high amount of domestically produced parts, but sub-assemblies will be kitted overseas.  We will be air freighting the kitted sub-assemblies in to get kits into people's hands as quick as possible.

We plan on using Amazon Fulfillment for kickstarter reward fulfillment in the US.  We are still finalizing our international shipping arrangements at this time and I will have more on this in a few weeks.  

That's it for this update, back to work for all of us here on the Kossel team.

-=- Terence

 

 

HUGE OpenBeam discounts at Seattle Mini Maker Faire

We are proud to be sponsoring our local Maker Faire again.  This year's Seattle Mini Maker Faire will be held on March 22nd and 23rd at the EMP Museum.  

Printing Totoros at the 2013 Seattle Mini Maker Faire

Printing Totoros at the 2013 Seattle Mini Maker Faire

Seattle is our home town, and we like to reserve our best Maker Faire promotions for our home town makers.  We will be running Maker Faire specials (about 20% discount) on our extrusions, brackets and precut kits (subject to availability).  We will also be selling our extrusions made by our old manufacturer at half price.  Although we have sorted through and rejected the worst of the bunch, the remaining inventory, while usable, isn't something I'm particularly proud to sell.  (My philosophy on this is simple: If I won't be happy receiving the product as a paying customer, it doesn't go onto the web store).  

I will therefore have a bundle of non-anodized extrusions (good for making custom colored OpenBeam) as well as various cut lengths of our old stock.  In some of these instances, the bars may be warped up to 3-5mm per meter (this is 2-3 times the allowable industry standard for extrusions, FYI), or may have surface cosmetic blemishes, or feel dirty from mishandling at the anodizer.  They are still likely ok for construction projects, just not good enough for me to sell full price on the web store.  So, come on down and support our maker faire and load up on OpenBeam loot.  

Our new overseas vendor is amazing, by the way.  They have been handling our cutting and kitting, and stuff that doesn't pass inspection doesn't get shipped.  As a result, we no longer have manufacturing yield fall outs.  I plan on donating what I doesn't sell to local schools, so this will be the only time where we will be running such an aggressive special.

Thanks,

-=- Terence, Rachel and the furry monster puppy

[Designer Deep Dive] - The Auto Leveling Probe

In today’s blog post, let’s take a look at the development history of Johann’s Auto Levelling bed probe.

Automated bed leveling and calibration, in my opinion, is one of the most important design improvements in recent Reprap / OHW 3D Printer development.  Anyone that’s ever operated a non “professional” 3D printer will tell you that perhaps the biggest challenge faced by the user is ensuring that the first layer height is set correctly and that the bed is parallel with respect to the X, Y motion plane of the machine.   I’ve always explained to curious observers that a 3D printer is, essentially, a computer controlled hot glue gun.  To build a good model, the foundation layer must offer good adhesion.  If the nozzle is too high, the bead of plastic being applied won’t stick, and if it’s too low, the bead of plastic will squish out too flat, deforming the part, or in worse case situations, crash the head into the print bed and jam the nozzle.

An auto-levelling software routine, such as Johann’s Marlin G29 implementation, will therefore do the following:

1)      Deploy a sensor of some sort that can register contact with the print surface.  The registration point (where the sensor trips) shall be a known, repeatable distance from the tip of the hot end in the Z-axis.

2)      Over an array of points, determine the Z coordinates of the print bed.

3)      Perform a linear extrapolation to create a Z map of the print bed’s X, Y coordinates.  We can do this by assuming that the print bed is perfectly flat; this assumption is safe to make with a borosilicate glass print bed.

4)      By definition, the probe must physically extend past the tip of the print head in –Z direction.  This needs to be fixed by retracting the probe above the print head so that printing may start.

5)      Magic happens in the firmware that offsets each point on each layer of the print by the Z-map determined in point #3.  The results?  Within reasonable amount of tilt, the first layer and subsequent layers will come out correctly.

It is worth also pointing out that Delta printers are traditionally harder to calibrate.  Unlike their Cartesian counterparts, the X, Y and Z motions on a Deltabot are generated by mixing input from the 3 drive towers.  The geometry of the linkages and simple trigonometry are used to compute the necessary tower motions to generate the proper motion in Cartesian space.  However, ball joint center-to-center distance, critical to the calculations, are rather difficult to measure without specialized fixtures.  The Delta Marlin G29 routine compensates for errors in this ball joint center-to-center measurement, which traditionally will result in a parabolic plane being generated.  Prior to the introduction of the Delta Marlin G29 routine, machine calibration required iterative changing and recompiling of the firmware to experimentally determine the correct values.

Das Original

Johann's auto probe: The Original since 2013 :-)

Johann's auto probe: The Original since 2013 :-)

Johann’s original test probe is the classic definition of McGyver ingenuity:  Comprised of a couple of springs salvaged from ball point pens, an allen key, a safety pin, a terminal block (used as a shaft clamp!), a microswitch and a 3D Printed part, the probe is a shining example of a Reprap developer’s resourcefulness.

However, no sooner than the design been published did questions started popping up.  What size safety pin do I use?  Where do I clamp the terminal block?  What’s the digikey part number of the terminal block?  And more importantly, why are all the local hardware stores suddenly sold out of 1.5mm ball end allen keys? :-D

All kidding aside, the original auto-levelling probe was hard to replicate.  Not only that, there were a few issues:

1)  The print's "grain" direction is normal to the probe's travel.  On a not-perfectly calibrated printer, this causes a very rough deployment on the probe and can adversely affect accuracy.

2)  The 1.5mm long arm allen key is actually pretty flimsy.  The contact area with the probe body is quite small, so the tip has a nasty tendency to wobble and drift.  Combined with the grain in the probe body, it is very prone to jamming.  And finally,   in order for the 1.5mm probe to reliably engage the belts, heat shrink had to be added to the short arm.  This proved to be a very delicate and difficult to replicate process.

Terence's contraption

At the same time, Terence was working on the probe for the Kossel Pro.  After being relentlessly teased by the folks at Metrix to keep shrinking the probe, this was the design:

Terence's original design featured an adjustable deployment point, in keeping with OpenBeam's design philosophy of designing parts that are flexible in end-use case and inexpensive to manufacture.  However, adjustable deployment point proved to be very problematic in testing; vibration on the end effector tends to shake things loose, but adding lock nuts made the probe very difficult to adjust.  Additionally, owing to the compact nature of the probe, the spring force often had to be readjusted after adjusting the deployment point.  Good idea, but sometimes simpler is better.

Terence's original design featured an adjustable deployment point, in keeping with OpenBeam's design philosophy of designing parts that are flexible in end-use case and inexpensive to manufacture.  However, adjustable deployment point proved to be very problematic in testing; vibration on the end effector tends to shake things loose, but adding lock nuts made the probe very difficult to adjust.  Additionally, owing to the compact nature of the probe, the spring force often had to be readjusted after adjusting the deployment point.  Good idea, but sometimes simpler is better.

Turns out, it was a bad idea, for a few reasons:

1)  The prototype parts did not hold up well under real use, and real life longevity of the plastic latch was difficult to determine without additional thousands of dollars in machined prototype testing.

2)  The adjustability turned out to be a liability more than an asset.  With Johann's probe, you perform a one-time calibration and you're done.  With an adjustable probe, the adjustment can actually drift over time.

3)  Holy part count, Batman!  

Back to the drawing board, for the Reprap design:

With the OpenBeam Mini Kossel installed at an art gallery after being returned from Make Magazine's 3D Printer Shootout, it was apparently that the issues with Das Original had to be fixed - quickly.  So, Terence drew this up pretty quickly, and after a few quick iterations, arrived at the final design released reprap probe:

Green is the original design.  Minor improvements were made, such as moving the spring pivot point on the torsion spring, and adding features to guide and capture the spring leg to prevent it from overshooting, to arrive at the red part.  We also added a safety position to prevent accidental deployment of the probe during long prints; vibration can shake the probe loose and an accidental deployment pretty much guarantees a destroyed print.

Green is the original design.  Minor improvements were made, such as moving the spring pivot point on the torsion spring, and adding features to guide and capture the spring leg to prevent it from overshooting, to arrive at the red part.  We also added a safety position to prevent accidental deployment of the probe during long prints; vibration can shake the probe loose and an accidental deployment pretty much guarantees a destroyed print.

Porting over to the Kossel Pro

The new Reprap Kossel probe worked so well, Terence decided to port it back to the OpenBeam Kossel Pro:

The decision was made to replace the allen key with a custom injection molded part.  This had the advantage of combining the shaft clamp and the allen key, placing the shaft clamp in a consistent position (thus allowing for consistent spring force), and also allowing the nose of the probe to be made bullet-shaped to prevent scratching / marring the print surface.

The "A-ha" moment was figuring how to install the probe.  By making the shaft diameter big enough to hold a M2 flat socket head screw, it became possible to install the probe from below and lock it into position with a screw.  This also works to our advantage, as a larger diameter probe pin will be more stable during deployment (less likely to flex).

Taking an inspiration from Legos, we core out the plastic pin into an "X" cross section to keep the wall thickness uniform, we arrive at the following parts:

Auto Probe, mass production, early prototype.  We beefed up the shaft after a few of these snapped after repeated deployment.

Auto Probe, mass production, early prototype.  We beefed up the shaft after a few of these snapped after repeated deployment.

It's been a long journey to get here (and for my very patient kickstarter backers, probably *too* long).  But for such a critical piece of the printer, we hope that the time spent was worth it - remember, this probe is also meant to be an inexpensive retrofit option onto the thousands of cartesian printers out there as well.  

The nature of Reprap and Open Source 3D Printers is that we are always evolving the ecosystem.  Our friend, Steve Graber, was onto something good when he designed the Cerebus prototype that used the hot end itself for probing.  In doing this, you eliminate any systematic error in setting the probe offset from the actual print head itself.  Johann recently demonstrated the use of force sensitive resistors (FSR) in the print bed mount for auto probing.  The next generation of Kossel Multi-Material printers that will be developed after this will likely feature some form of force sensitive resistor probing setup, to eliminate calibration error on different hot ends.  The engineering challenge there will be reliably and heated bed compatibility (FSRs are heat sensitive, so mounting them onto the build platform will be problematic on a heated bed).  

But, in the meantime, I have kickstarter backers to take care of, and overdue printer kits to ship.  So, while we dream about the next thing, here are some more pictures from the development of the probes - including the injection mold drawings for the auto probe that is going into production.  We just received the mold design from our molder late last night, and I have already approved the production of the steel molds.  The end of the tunnel is finally in sight!





Engineering Quickie: Testing the new auto calibration routine in Marlin

Johann and I met up at Metrix: Create Space tonight for some brainstorming, and to work on projects.  I'm working on kitting and packaging in preparation for a engineering validation and verification (V&V) build, while Johann is verifying his auto levelling code.  

One complaint about the original auto-levelling probe is a systematic error that occurs when the probed position is different from what is in the probe level offset.  Since I had my thermal printer on me for printing kitting labels, we came up with a really awesome way to verify probe offset:

(We heated up the hot end and had it touch off on the bed.  Since label printers uses thermal labels that change color upon exposure to heat, each hit registered with a very crisp black dot on white paper.  We then ran the G29 routine to observe where the probe actually hits vs the coordinate system of the printer.  

In the final shot, you can see there is still systematic tilt in the bed, and the line traces for an over-compressed layer appeared darker / wider.  The debugging and tuning continues).


 

Kickstarter Update - The logistics nightmare that is Chinese New Year

TL; DR:  We will be running an additional 3-4 weeks late at this point; final projected ship date to be updated by the end of Feb.

I blogged about the impacts of Chinese New Year last year briefly, especially to a new business owner.  This year, taxes were a bit easier, thanks to my awesome wife and household CFO helping with doing the taxes. 

As you may have read in the last update, we released our design to our injection molder about 4 weeks ago; that was the last major hurdle for us in engineering this and bringing it to production.

Unfortunately, we are not the only ones trying to beat "Spring Migration".  Spring Migration, or Chunyun, is the largest human migration and happens every year leading up to Chinese New Year and up to a month after Chinese New Year.  Chinese New Year is based on the lunar calendar and this year landed about a week and a half early.  No doubt, every major western engineering firm scrambled to try to beat it, and the resulting work load overloaded just about every good shop in China.  This, coupled with abnormally bad weather causing severe travel delays, meant that most factories are still NOT back to operating at their full capacity.  Even for our vendors that are not based in China, Chunyun is still a major logistics headache, as so many of the downstream suppliers (such as mold base suppliers, etc) are based in China.

(To give you an idea how much of an impact this is:  The number of passenger journeys during Chunyun exceed the population of China, hitting 2 BILLION PEOPLE in 2006.  Now imagine that this travel is done by train and by other modes of public transport that is suddenly handling 50-100 times their normal traffic loads.  Now throw in a Noreaster or a major storm event.  There are no winners in this logistics challenge).

We are seeing across-the-board delays as a result of this, unfortunately - not just for OpenBeam projects, but projects at our day job as well.  Our injection molder only now starting work on mold designs, even though the files were submitted a few weeks ago.  (In their defense - they did NOT accept our initial PO and confirm the delivery dates, and we have full faith that they will take care of us - they do AWESOME work).  Our Rev 2 Brainwave Pro (formerly known as the Brainwave II) boards were supposed to be delivered the week of Feb 1st.  We are now told that we may see the boards at the end of Feb.  (This is then followed by a week or two weeks of testing, before we can release the production run of the 450 boards used for Kickstarter fulfillment).

So, there's the long version of the story.  We will not be able to make a firm delivery promise for all the units until A) we receive a confirmed T0 date for our injection molded parts and B) receive our rev 2 test electronics board and have a chance to verify functionality of the entire board.  Currently based on our best guesses, the early birds should ship the week of April 15th, but this is an estimate subjected to change until confirmation of manufacturing schedule.

However, this does not mean that we've been idling the past 4 weeks.  This is what we have been up to:

A)  Moving ahead with machined aluminum vertexes

We are moving full speed ahead with our aluminum vertexes.  The manufacturer for these parts is based in South Elgin, Illinois, and is not affected by Chinese New Year at all.  We used 3D Printing to create the first fixtures to test the machine programming and design of the part; now we are making the aluminum fixtures required to machine the components on a production scale.

B)  Test printer fleet

We are really reaping the rewards of designing the Kossel and Kossel Pro to have fully interchangeable parts at the subassembly level (and having almost identical firmware).  With our plastic parts delayed we are turning to the Kossel Reprap branch to get printer test time in. 

We will be releasing the Kossel Reprap design in the next 7-14 days.  The idea is that by the time the Brainwave Pros arrive, we'll have a fleet of about 10 printers and we'll push those machines hard to see if there are any show stopping bugs. The lack of control boards to perform full testing is a disappointment, but the original Brainwaves are still perfectly capable boards when paired with matching motors.  We will make another announcement when we design the OpenBeam Kossel Reprap printer.

(Below:  Test output from the OpenBeam Kossel Reprap, using the Brainwave 1 with properly matched motors.  Printed from this thingiverse object, 100 micron layer height.  Print sample courtesy Metrix Create:Space

C)  Documentation efforts

We have purchased a Stratasys Mojo, a low end commercial FDM system, for printing injection molded Kossel Pro parts for prototyping.  The parts you see on the Kossel Pro below are all built on this machine.  This is not a cheap investment, but we are at the point in the company's growth where such an investment can be justified.  (Another way to look at this is that if I had purchased this machine when the Kickstarter was funded, we would have been able to close that painful prototyping cycle on the development of the Kossel Pro and we wouldn't be this late!  We wouldn't have gotten as good of a deal on the machine though, but the amount we paid in outside prototyping costs way more than the steep "used machine" discount we got)  We expect that this machine will pay dividends when it comes to the work on the next generation multi-material Kossel, as well as expansion to the OpenBeam system - both slated for when Kickstarter delivery is complete.

It may seem a bit off for an open source 3D printer developer to buy something as proprietary as a Stratasys Mojo, but the truth is, these machines are the only ones capable of repeatedly and reliably printing parts that require soluble support material.  I am paying through the nose for the reliability of being able to hit "print", go to work and come back to functional parts.

For now, the Mojo is churning out copies of the injection molded parts, so that our manufacturing engineer can start writing the assembly documentation.  By the time this project is completed we estimate that we would have spent 15 minutes in testing and 15 minutes in documentation for every hour of engineering time that went into the final product.  

D)  Kitting of subassembly components

In parallel with our documentation efforts, kitting is already under way.  We have purchased, and received majority of the 'off the shelf" components for fulfillment of the early bird kits.  (Some of these are risk buy items for evaluating supplier quality, when we qualified the part the parts then transfer to inventory for the early birds).  Fastener kitting had been the most painful of this (t's not a kit if it's missing a single screw, and it would be highly unfortunate if we had to rework kits because a handful of screws were missing).  There is much work to be done, and everything is subjected to careful checking.

E)  Test, test, test.

As we continue down the development path, testing is paramount.  We test our own documentation and processes -  we sit people down at an empty table and hand them fastener packs that have been kitted according to our work instructions to verify fastener count.  We use weight scales to check the aforementioned fastener pack and we use statistical process control to check the kitting of individual fastener packs.  We gutted a toaster for the resistive heater coils to stress test our power supplies, to make sure that they will hold up under the use of a heated bed (Dad didn't want to pay $80.00 for power resistors big enough to do the job safely, so he found a $5.00 toaster from Goodwill and hacked something up in the garage).  We use purpose build fixtures to control the cutting of all stock that have to be cut to length, from timing belts to PTFE insulators for the thermistor on the hot ends, to all the precrimped cable harnesses and bowden feed tubes.  

Oh, and one more thing...


When the OpenBeam Kossel Reprap is released in the upcoming 2 weeks, we are also releasing the tooling and calibration fixture designs as well, under open source license. We designed all the tooling and calibration fixtures (shown below) for building these printers to be printable on hobbyist Reprap class machines, to help spread the Delta love far and wide.  The tooling, no surprise, is also compatible with building Kossel Pros.

And, finally, the promised "More frequent updates"

We're been taking notes (see screenshot below).  Unfortunately, with all the supply chain fires, we have not had a lot of editorial time.  Priority one is still trying to not slip the delivery date too much.  Hopefully as the workload gets delegated off my shoulders, I'll have more time to write in the next few weeks.

 

-=- Terence and the Kossel Pro team

OpenBeam Applications - PCBGrip

One of the best things about creating an awesome construction system, is being able to see what other people use OpenBeam for.  While we have been focused on bringing the OpenBeam Kossel Pro to market and the OpenBeam Kossel Reprap to public release, we also use a lot of OpenBeam for fixture and tooling.  (A lot of the fixture applications at work I can't show, due to confidentiality reason - but I do eat my own dog food quite bit).

I've been corresponding with Jason out of Toronto, Canada, for a close to a year now.  A couple of weeks ago he finally revealed what he's been up to, and it's pretty cool!

I'll let them tell you more about the system, in the videos below, but needless to say, I have already backed them.  The parts look to be beautifully machined, and not very different from my much cruder bits that I've often clobbered together for work holding. 
 

Kickstarter Update - The Kossel Vertex

 A quick update:

We have received the first tooled parts for the Kossel Pro family.  These parts replace the 3D Printed Vertex that form the top and bottom of the printer.

 What are the orange parts?

You are looking at a set of soft jaws, designed to fit a Kurt 6" machining vise.  Kurt machining vises (and their numerous clones) are industry standard for a machine shop; just about any machine shop will have a few of these lying around.  The soft jaws are designed to replace the hardened steel jaws of the vise and have a contour that holds these extrusions in the proper orientation for the machining operations that have to be done on them.

The beautiful thing about these soft jaws is that it with the right tooling, the machining operations can be done with just a manual milling machine.  A Tormach CNC Mill is on the horizon for me as a purchase, but until now, I am limited to what I can get access to.  (A friend of Matthew and I is performing the machining work for us, but in a pinch, I wanted to be able to hop on a manual knee mill and be able to punch these things out).  In fact, most of the operations can now be done on a drill press, with these jaws.

We should be getting machined parts back later this week, but I thought it'd be fun to share the work-in-progress shots.

-=- Terence

2013 - Year in Review

What a year 2013 had been!

2013 started really badly for OpenBeam.  With the switch to closed loop profile OpenBeam in late 2012, our extrusion vendor had failed spectacularly to produce  acceptable parts.  We ended up having to fire the vendor, which also resulted in us losing our fulfillment capabilities and speeding up our switch to Amazon Fulfillment.

With our supply chain disrupted and a lackluster 2012 holiday season sales (due to supply chain limitations), our finances were further stretched since we had already committed to sponsoring our Seattle Mini Maker Faire.  Things didn't improve for us either when we failed to recoup our investment in Deltabot / Kossel research; a cloner tried to steal our trademark, came into Metrix, and copied what everybody was doing to stock his web store, undercutting our ability to sell Kossel extrusion kits to recover our R&D expenses.  

The fear at that time was that we would be beaten to Kickstarter as well; various other groups are preparing their individually developed printer for Kickstarter, and it would have been a financial bloodbath if someone copying the design in progress also managed to beat us to Kickstarter (at this time, we've already committed thousands of dollars in component cost alone, not to mention all the engineering time).  First to market is important, but it goes against my philosophy of releasing things prematurely and without proper testing.  After some deliberation with the crew at Metrix, the decision was made to launch the OpenBeam Kossel Pro Kickstarter at Maker Faire in San Mateo.  In hindsight, the kickstarter was launched prematurely and I take full responsibility for the project delays as a result of this.  

Fortunately, things started to turn around for us mid year.  In the 2 weeks leading up to the Kickstarter campaign's conclusion, we qualified our off-shore manufacturer on OpenBeam extrusions.  (As a historical note: if the offshore vendor had failed to meet quality control standards, we would have cancelled the Kickstarter campaign, refunded all money, and shut down the business.  Things were that dire).  We completed a long overdue overhaul of our website (which didn't make sense to do when we didn't have anything to sell - we found that we were just drumming up business for our competitor).  We completed the redesign of the Kossel Pro, and over the last few months, we've been doing testing and refinements to various injection molded components, culminating with the launch of capital expenditure tooling.  With our improved efficiency with our off-shore vendor, we are now price competitive in our black and clear Mini Kossel kit offerings.  And since we are fulfilled by Amazon, Amazon Affiliates works with all our offerings; the previous two links donate 8% towards Johann Rocholl with each purchase made by clicking through them, while offering Prime households free 2nd day shipping and free shipping to all buyers.  And with Amazon's distributed inventory system, generally 3 business days is all it takes even for ground shipping to cover most of the continental US.

2013 was also the first time we've seen derivative work on OpenBeam extrusions.  In South Africa, the folks at OpenHardware.co.za have been producing a local variant of OpenBeam that does not have the problematic center hole and anodized finish (since electricity is very expensive in South Africa).  They've already been doing really cool things with it.  I think it's great that they can serve a customer that I cannot serve, and what they do strengthens our ecosystem as a whole, as their projects are shared back with our community.  Check out their facebook page for some of the projects that they and their customers have been working on!

2014  is going to be an awesome year to look forward to.  In no particular order, this is what we have in the works:

  • Shipping all Kickstarter OpenBeam Kossel rewards, then making the Kossel series printer kits and parts available via Amazon fulfillment.  As we continue to make improvements to our Amazon listing and inventory, all sales made through the OpenBeam website will donate 8% either towards individual open hardware contributors (such as Johann for Kossel components) or into a fund set aside for donation back into the open source hardware community (such as sponsoring Slic3r).
  • The growth numbers and the sales figures over the past two month shows that we've traditionally been supply chain limited.  For the first time in the company's history, stock availability isn't as dire of a concern.  I ended up borrowing some money to increase the amount of inventory held in stock - hopefully we can fully repay the loan in the coming year.
  • With our supply chain  issue fixed, we look forward to growing our retail network.  We have experienced tremendous growth in our European Union distribution network, with the guys at MakerBeam.EU doing an awesome job as lead importer.
  • Other back-burnered OpenBeam projects will start coming back into development and release cycles.  Watch this space for updates.
  • We will, once again, be sponsoring our local Maker Faire.  This year's fair will be held at the EMP building in Seattle - a much bigger and more visible venue!  

We look forward to a fantastic 2014, and would like to thank all our customers who had believed in us, helped us grow, and help spread the ideals of Open Source Hardware. Please keep building and sharing awesome things.  Thanks,

-=- Terence, Rachel and the furry monster puppy

 

Kickstarter Update: Milestones unlocked...

Hello Kickstarter!

Rachel and I have returned from our European honeymoon and the crew and I have been hard at work on bringing the OpenBeam Kossel Pro to production.  I'm happy to announce that we've hit three major milestones this week;

1)  The tooling for our extruded aluminum vertex profile is done!  We are happy to locate an American manufacturer to manufacture this part.  I had an opportunity to tour the extrusion plant a few months ago and it was very reassuring to see that they do extrusion, cutting and anodizing all under one roof.  As many of you may know, we fired our primary aluminum extruder early this year due to consistent quality concerns.  Even at the very end, we still couldn't establish WHY the parts are bad with the extruder pointing fingers at the anodizer and me scratching my head wondering how the anodizer can produce straight bars the first time but consistently bend my extrusions 4-5mm / meter on the new orders.

We are currently awaiting samples to be shipped to us.  These parts are coming in cut to length and anodized (black), but not machined.  The features such as screw holes, motor mount face and access ports for screw drivers have to be machined in. We will be working with local small business machinists to perform the final machining steps to finish the part.

2)   We have finished our Proto-B design review process and incorporated all the feedback from testers into our parts to prepare the files for injection molding.  The injection molding process is by far the longest lead time item in this project; it clocks in at 7 weeks after mold design is complete.  It had been a major source of embarrassment to me to have to launch this almost 3 months behind schedule, but we only have one chance to get this right.  Launching tooling will be the subject of another blog post - it is a lot more than just sending our STP files to a vendor.  We've also been test building, modifying and rebuilding the end effector and auto level / bed probe assembly; both of these went through about 2-3 major design revisions before we are comfortable and happy with what we have.  (Yes, those are thermal camera images; we borrowed a FLIR thermal imaging camera to study the cooling for the hot end as well as to see how efficient our thermal cooling for our Brainwave II board is.  You can see just how sensitive the camera is - it picked up the thermal footprints left by the dog after she walked across the carpet.  With this baby we hope to be able to tune the cooling profiles of the micro-processor controlled fans in the end effector).

3)  We have been developing the Brainwave II in the background for the past 5 months; and tonight I'm happy to let you know that the Brainwave II had completed systems integration testing.  There were obviously some rework done to the board (as seen in the pictures) but we have verified all core functionality and successfully completed the Arduino Hardware bundle for the board (which greatly simplifies programming the board) and made the necessary modifications to Marlin to make the firmware work with the new board.  

We haven't been updating the project as frequently as we'd like; the focus on the past 6+ months had been to "keep head down and jam out the design" Now that we are on our way to launching all the long lead time tooling, we have more time to allocate towards blogging and social media outreach.  (We also have been growing the prototype fleet to get more combined test time and tuning time in).  We look forward to finally sharing our design process and trade-offs with the world!

-=- Terence, Mike, Matthew, Johann, and the crew of Metrix Create Space

Shipping update - Kossel Pro

First, the bad news. As some of you may already suspect, the delivery of Kossel Pro kits will be delayed, due to a few minor technical difficulties. We have faith in our engineering and design but we will not be able to make the original Christmas deadline. As someone who had delivered the last kickstarter ahead of schedule (and had padded this kickstarter schedule too to hope to achieve the same thing), I am obviously disappointed at having to make this announcement and take full responsibility for the delays. We understand that some of you may have purchased this as a Christmas present; we are going to create a set of downloadable engineering prints that people can print themselves as a token gift for Christmas, with the understanding that the real McCoy will follow shortly. We will also be looking into some form of compensation, likely in the form of a discount coupon to the OpenBeam store, to ship with our kits. We are currently projecting a ship date of early March, 2014, based on schedules outlined by our injection molding, extrusion, machining vendors as well as our lead times for electronics.  At the end of the day, given the scope of the project, and how tight funds can get, we did not feel it was prudent to launch the capital expenditure in tooling without sufficient testing and until the rest of the printer, including the electronics board, were fully developed and at least protototyped.

The Kossel Pro features very high degree of intergration; we specifically laid out our control board with delta geometry in mind and paid a lot of attention to connector placement and compatibility across the various sizes in the Kossel Family.  To give you an idea; the original OpenBeam project raised over $100,000, and less than $10,000 went into tooling. The rest of the expenses went into spinning up the company and support infrastructure, as well as merchandise for the initial system launch. We had 5 molded parts that didn’t really have to interact with each other, an extrusion and 2 stamped parts. We could afford to be more aggressive on the launch schedule for all the tooling; in fact, we purchased all the tooling “at risk” before we had money in hand from the Kickstarter to get a jump on shipping.

As a comparison, the OpenBeam Kossel Pro currently has over 20 unique injection molded parts, with suppliers in the EU, Asia/Pacific Rim, and the US. The budget for our tooling is well over $40,000.00. The rest of it had to be earmarked for the BOM of the printer. We simply cannot afford to make a mistake - a few thousand dollars here and there in a mold change will quickly make this project run into the red. As the great Shigeru Miyamoto once said: “A delayed game is eventually good, but a rushed game is forever bad”. We’d rather take our time to make sure things are done right and do the right amount of testing, before releasing the printer.

Now, for the good news:

We have received the first engineering prototypes of the new Brainwave IIs and although there were a few hiccups with initial testing, we believe that we are on a good path forward.

We have signed the final round of die approval prints and paid for our extrusion die. We hope to have first articles in about 4 weeks, and we were able to keep this part of the project on US soil by finding what appears to be a good, competent US based aluminum extruder.

We have started building the test fleet of printers to stress test the new Brainwave II boards. Part of this printer fleet will be deployed at Metrix Create Space in Seattle to print - you guessed it - more reprap Kossel parts.  The original Mini Kossel Prototype 2, featured in the 2014 Make Ultimate Guide to 3D printing, have now logged hundreds of print hours. We are looking at the different failure modes that have occured and making sure that these design issues are adequately addressed in the Kossel Pro.

We have received all the extrusions for kit fulfillment, and the quality is top notch. We are in the process of sourcing validation quantities of all the rest of the printer’s hardware - the power supply, bearings, screws, nuts, bolts, etc.

We have already received into inventory some of this material, such as the timing belt pulleys and ball bearings. The longest “tent pole” in the schedule is our injection molding.

We have released all parts, except for the auto-levelling probe, to injection molding. We will be launching the tooling this coming week on all parts except the touch probe; we hope to launch the touch probe tooling the week of Dec 2nd when I return to Seattle. Currently injection molding is sitting at 8 weeks lead time for T0 (first shots off the mold). These first sample parts have to be qualified - basically measured against the engineering prints and signed off, before mass production can begin.

There’s still plenty of work to do, even during this time when tooling is built; packaging needs to be defined, documentation needs to be written, machines needs to be tested, CNC machines (performing the second-operation machining after extrusion and cutting for the metal vertices) needs to be programmed, videos needs to be shot and edited. Our team will have a busy season ahead of us, but we will rise to the challenge. In fact, while I am writing this update in Germany on my honeymoon, the crew at Metrix is already busy putting together the printer test fleet, and Matthew Wilson is diligently working on the bootloader and patching Marlin for our new Brainwave II.

Finally, for a glimpse of the Kossel and how it stacked up against other machines, consider buying a copy of the Make 3D Printer Review. We are very proud of how the prototype did and what lies ahead for the project.

Thanks, The Kossel Development Team

 

And now, a guest post from Mrs. OpenBeam

Hi!

As some of you know, Terence and I got married a few weeks ago (October 26, to be exact!) The wedding was fantastic and we loved getting to see all our family and friends from all over the world. We figured this would be a good time to show off a little bit of our geeky side, and we integrated some laser cutting projects, some robotics, a 3D printer, and of course, some touches of OpenBeam.

The table numbers were something very simple I designed, cut out of 3 mm Baltic birch, and mounted in OpenBeam.

The card box was a simple box utilizing Baltic birch and OpenBeam again.

Our "ring bear wrangler," the daughter of one of Terence's friends/colleagues, is a budding geek herself, and an active Hackerscout, so we knew she would be up to the task of driving the ring bear robot down the aisle. It was controlled via Bluetooth on a Nexus 7, building upon the work done by Keith Baston, who won the OpenBeam Build Off a few months back.

The ring bear robot is powered by a Sparkfun Redboard, a Bluetooth Mate Gold Bluetooth modem, and 2 Hub-ee wheels.  Terence put the robot together in about 3 hours.  During rehearsal and testing though, one of the wheel's internals melted and caused the robot to swerve as our ring bear wrangler tried to navigate it down the aisle, adding to the challenge.  Fortunately Terence had bought a spare wheel, though he left it at home so we didn't get a change for a full practice run with the working robot.

3D printing was also featured at the wedding.  We 3D printed a few LED tealight holders and small vases and gave them away at the end of the night:

We were glad to have a number of Terence's 3D printer and geek buddies in attendance:

Last but not least, the laser cut guest book. The hinge is based on the living hinge used in Snijlab's folding wood booklet and I used binding posts from the scrapbook section of the local craft store to fasten about 30 pages into the book. The pages were printed with photos of us from the last 3 years, including many of our engagement photos, with plenty of room left for guest signatures.

I think that about covers the relevant bits of the wedding! Thanks for being patient with us as this has taken over a lot of our life for the last couple months. I think we are both looking forward to getting back to "normal," whatever that might mean!

Cheers,

Rachel, Terence and furry monster puppy

Kossel / Kossel Pro - In progress pictures

Hello! First, a personal note.  I'm getting married this Saturday!  I've been completely swamped.  The future Mrs. OpenBeam had commissioned me for a few OpenBeam related wedding decoration.   On top of all this is my day job, plus the usual logistics challenge of coordinating a typical wedding - Relatives from around the globe, transportation logistics, etc.  Come think about it, it's not very far off from supply chain management...

I just want to share a few pictures of the work-in-progress Kossel Pro and Kossel redesign:

Our ball bearing ball joints.  Each ball joint contains 4 bearings, for butter smooth, low friction, zero backlash motion.

We've managed to retain about the same dimensions as the popular Traxxis 5347 ball joints!  This means that carbon fiber tubes cut to length for the Traxxis ball end rods can also be used with the OpenBeam ball ends to make a rod with the same center-to-center distance.

A set of these babies on carbon fiber rods.

Here's the carriage, with our belt end clamp.  The belt loops back onto itself, you pull it to tighten it around the peg.  We've ported this geometry back into the RepRap branch as well, and we give you two options:  Build the unit with captive M3 nylock, or build it with a captive M3 x 37mm spacer:

The use of a spacer greatly improves rigidity, as it couples both ends of the ball joint together.

And here's a sneak peek at the new end effector, with integral cooling ducts.  The ring around the hot end is fed by two top mounted fans that are PWM controlled, while the hot end gets its own dedicated, always on ducted fan cooling system to prevent thermal jamming issues.  We are VERY proud of the work we've done on the end effector; as you can see it packs a LOT of technology into a very small package without growing the overall size by too much.

You can see more "in progress" pictures on the OpenBeam Flickr page.

I'll update more after my wedding!

-=- Terence

[In Depth] - The Kossel linear rails

Terence's note:  This blog entry is a work in progress.  We are releasing it ahead of our usual editorial process as our backers are now receiving their linear rail sets and the linear rails contains an instruction sheet that links to this article.  Photos and video will be forthcoming.  Thanks. Hello Kickstarter backers!

By now, all the "rails only" Kickstarter rewards have shipped via Fedex to our backers. You should have received a Fedex shipping notification email with your tracking number and information.

One of the things that separate the Kossel Pro from the rest of the machines on the market and at this price point is the use of linear recirculating ball rails on the machine chassis. There are a few great Open Source hardware options when it comes to linear bearings: Bart Dring's MakerSlide, Mark and Trish Carew's OpenRail and V-Slot, and Steve Graber's W wheels are three of the systems that makers are most likely to be familiar with. When designing the Kossel Pro, we also took into consideration patents that are existing in the field, namely this patent held by the 80/20 corporation. After all, we are a small start up challenging the status quo of expensive T-nuts with cheap extrusions in the T-slot framing space and we cannot afford any legal entanglements with a much larger corporation.

The decision to use linear rails become pretty apparent when you consider the following factors.  Being able to adjust the play between the carriage and the rail is good; not having to worry about that adjustment is even better.  Given that I speak both Cantonese and Mandarin, and given our investment in our trade and logistics capabilities, we are able to source these linear rails at a very competitive price.  In fact, our linear rail and ball bearing carriage subassembly BOM cost is actually lower than the retail cost of a comparable length of MakerSlide and its carriage and just slightly above the retail cost of 3 W-wheels from Mr. Graber.

The biggest advantage with the ball rails over other ball bearing on aluminum extrusion systems, however, is in the material selection. The rails are made from induction hardened tool steel.  In fact, the steel used in the rails is so hard, the vendor that we purchase the rails from uses a Wire EDM process to cut the rails to length. These rails are also precision ground to incredible levels of flatness. This, combined with our new extrusion vendor's awesome aluminum profiles, meant that the linear rail assemblies on our machines will be straight and true, and likely will stay that way unless the machine is put through serious abuse. Remember, Aluminum Extrusion Council's standard tolerances is about 1.1 mm per meter length and a competent vendor generally can be expected to deliver parts within half of commercial tolerances. That is still 0.25 to 0.5mm of deviation across the Kossel's travel on the linear axis, and we would like to do better than this. Our rails, being a ground metal product, is flat to within 0.02mm per meter - order of magnitude flatter than normal aluminum extrusions. They also add an incredible amount of rigidity to the machine, and that is the secret to how we can build such a large, rigid machine chassis on such a thin aluminum profile.

The rails we source are Chinese made clones of Thompson / THK / HiWin linear recirculating linear ball rails. Contrary to popular belief, they are NOT from HiWin! Early on in the Kossel's development, there was a cloner who would shamelessly come into Metrix, observe what people are working on, and proceed to stock his webstore based on his observations. As the cloner himself does not do any meaningful R&D work and we rely on future sales to recuperate our R&D cost and had already lost money due to his cloning of early legacy Kossel extrusion kits, we could not afford to lose the opportunity to sell the rails if the cloner beat us to the market. Therefore, we masked the true origin of the rails by calling it a (much more expensive) HiWin MGN-12C equivalent. In reality, our rails are the same generic ones that are sometimes sold on AliExpress; we just found a reliable supplier, qualified them, and verified that the rails that we are sourcing will work with our design.

There is a big cost difference between our rails and a genuine HiWin / THK rail. Although our rails are ground from hardened tool steel, the surface finish on our rails is rougher than a HiWin / THK rail, and as a result our rails run nosier. (Surface finish is measured in micro inches or microns. The rougher surface does not affect accuracy, just performance noise). The tolerance range on the grinding is also looser than what would be on a genuine HiWin rails and carts, and some carriages may be stiffer than others to move on the rails. None of these issues affect the accuracy or machine chassis rigidity of the rails and is not considered a defect, and a little bit of care and preparation can help improve the rail's performance drastically.

The rails ships from the factory with a retained ball carriage. This means that there is a wire cage running down the length of the rail to make it harder for the ball bearings from falling out when you remove the carriage. Note that it is still possible for individual ball bearings to fall out, but it is also possible to push the ball bearings back in. There is a gap in the chain of balls in the recirculating ball race - in a higher end carriage, there would be a special plastic cage piece between individual balls that contains a reservoir of lubrication grease. For maximum life, it is recommended that you replace the protective oil that the rails shipped from the factory with a proper grease in the ball bearings.

I've found it easiest to slide the carriage off by laying it on its back and sliding the rail out from the top. This way any displaced balls hopefully won't fall and get lost. A towel over the work area to capture the occasional lose ball is a good idea.

We are shipping all kickstarter rewards (and all Amazon rails) with a piece of lint free wipe to wipe off the preservation oil, as well as a light bearing grease for greasing the rails. The seals do a pretty good job at keeping debris and dirt out of the rails. Unfortunately, this also means that any grease applied to the outside of the carriage on the rail will likely be wiped off and will never make it to the ball race. We therefore recommend our users to remove the ball carriage carefully, and apply the grease directly to the line of ball bearings using the provided cocktail stick, moving the balls around and applying the grease. We have chosen a non-toxic grease that is rated for intermittent food contact that complies with the new regulations on VOCs in lubrication fluid.

i've taken a short video showing the cleaning process here. With a little bit of care, these rails can be a very high performance upgrade option to the 3D printer (or any other mechanical build). We hope you find this information useful for your project. Happy building!

-=- Terence, Rachel and the furry monster puppy

OpenBeam visiting Maker Faire NYC

Rachel, my lovely wife-to-be, was invited to be one of her childhood friend's bridesmaid. Unfortunately, the wedding got moved shortly after she signed up for the bridal party to Maker Faire weekend. Without her help a booth at MF:NYC would be impossible. So we compromised: I attended the wedding Saturday night, and early Sunday morning, I hopped onto a 5:30am flight from Bloomington, Illinois to La Guardia Airport in New York, conveniently located minutes away from the New York Hall of Science, where Maker Faire is held. It's funny how intermingled business and private life can become when you are a small business owner. And although a whirlwind 5 hours at Maker Faire doesn't sound like much, it is actually the most number of hours I've ever got to spend walking around and enjoying the event instead of being stuck behind a booth. With five uninterrupted hours, I charged up my camera batteries and loaded up on memory cards, and took pictures of anything remotely interesting. Here are the highlights:

The BioBus project is a converted school bus full of microscopes that does science and educational outreach.

What is really cool though, is that I designed one of the microscopes onboard the bus.  (Disclaimer:  I only design these microscopes.  I don't pick the names...)   I waited in line with everyone else for a bus tour to see my creation in action - it was REALLY well done.

Here's OpenBeam in the MakerShed! We'll need to improve our game on the packaging and packaging design, but this is an AWESOME start! We've been working very hard since the beginning to get into MakerShed and we are really happy that it finally happened. It also put a big smile on my face to think that both my work from my regular day job (FLoid) and my personal side business (OpenBeam) were represented here at Maker Faire.

Here's my friends Faye and Dawn, with BitWise eTextiles. I believe they are launching a kickstarter campaign soon.

 

Husband and wife team Matt and Michelle Hertel presented a desktop 5 Axis CNC machine, the PocketNC. They are hoping to be selling the machine at around $3k price point.

 

I came across a mechanical computer component in someone's antique collection.  I was told this was used to compute bomb trajectory in a World War II era bomber.  The gray cylinder is a mathematical formula graphed in 3D and the ball bearing arm reads the "solution" given the input variables.  It is mind boggling to think that free form mathematical shape was created before there were CNC milling machines!

In the 3D printer community, we have our friends at Matter Hackers:

Diego and WhosaWhatsit from DeezMaker, and Roy from Panucatt Devices:

John Oly and Party Daddy from SeeMeCNC:

Josef Prusa:

The Gigabot guys. I don't think I could ever get my wife-to-be's approval on getting a printer this big.

Eugene from Tinkerine Studios in Vancouver, BC:

Michael Joyce and his wife from B9 Creator - the Open Source DLP / Polymer based machine.  We look forward to seeing it go head to head against the Form1 in the upcoming Make 3D Printer special issue.

And finally, a few more delta robots:

Makerbot is on hand to show off their latest digitizer. Looks interesting, but close source vs open source aside I'm not sure if I can justify the price tag. Given the mechanical simplicity of the device, I sure hope it will be more reliable than their printers.

And finally, Roy's got some of his new toys to show off:

The Azteeg X5. This is a 32 bit controller card that runs Smoothieware. It will be a eagerly awaited upgrade, as we currently are maxing out the little Arduino microprocessor.

And a successor to the Viki:

Full set of photos from today's excursion here. Wondering through the museum, I came across a Tesla exhibit, so I snapped a bunch of pictures of that too. All images are licensed under Creative Commons BY-NC-SA.

Our new extrusions are in - and they are AMAZING.

Six months ago, things were looking pretty grim for OpenBeam.  Our old extrusion vendor had delivered 3 consecutive batches of bad extrusions.  We've received extrusions that passed their quality control that are warped, fork lift damaged, or corroded beyond recognition.  With our inventory depleted, most of our cash got tied up in bad product that can't be sold, and back orders piled up.   Truth be told, if we did not find a new, qualified vendor in time that we can source the extrusions at a reasonable enough price for, we would have had no choice but to go out of business.

Fortunately, this didn't happen.  About a year ago, over cups of Chinese tea, I laid out the direction of business expansion to my father.  We decided to develop our import and logistics capability to offer a cost advantage in sourcing commodity components such as stepper motors, pulleys and bearings that goes into 3D Printer kits, in direct response to our first 3D printer offering being cloned within weeks of launch.  That decision ended up saving our company, as gaining the ability to ship Less-than Container Load (LCL) meant that we were no longer restricted to buying from American suppliers for our extrusion.

In an LCL shipment, the cargo is collected and bought to a container freight station (CFS).  Depending on the terms of sales (Ex Works - where the supplier's responsibility ends in producing the product and leaving it on their shipping dock, or "Free On Board" where the supplier's responsibility includes getting it to the dock (traditionally, over the side of the cargo ship), either the supplier or the freight company provides a pickup service to bring the cargo to the CFS.  In our case, we are buying ExWorks, so Expeditors International, the company I am using, provides local trucking service overseas to bring the cargo to the CFS.  Once at the CFS, the cargo is consolidated with others into a container.  This is then loaded on the next available Seattle bound ship.  From most of the common ports in Asia, there is a weekly sailing, so in our case, we have to get our cargo to the CFS by noon on Tuesday to catch the boat sailing that next Sunday.  After about 3 weeks on the water, the container ship docks at port of Seattle or port of Tacoma, and the container gets unloaded, through customs, and the cargo is made available for pickup at the local CFS.  (I opt for local pickup, because I have a day job, and it's somewhat easier to take a long lunch to go pick up the cargo, than to have to take a day off and pay for delivery and wait for the delivery).

Setting up to ship LCL is not a short process; and it requires posting a $50,000 bond with US customs and paperwork to become an importer.  However, once in place, this is a very cheap and reasonable method to move large bulky cargo internationally.  In fact, it is cheaper to ship a pallet of aluminum extrusions via LCL than paying Fedex to drive it up from California, with a transit time from Asia of approximately 20-25 days.  Once you factor in that Asian mills usually are more responsive (the one we are using turns orders around in 2 weeks, instead of 8 weeks for my old supplier), however, the lead time becomes quite competitive too even factoring in a 3-4 week journey across the ocean.

Early this year, as I was standing on the loading dock listening to another excuse as to why I'm staring at another pallet of garbage extrusions, how the extrusion mill wouldn't be heartbroken if they never saw another order again, etc, I've already had multiple quotes from American and off-shore extrusion vendors in my back pocket.  The final "reject" rate was between 5-6%, but this is highly debatable.  In my haste to sort through the extrusions, I've missed quite a few on my first pass that was warped, only to find it jamming up the auto saw later.   But even at 5% rejection rates, the fact that someone would think that a 5% manufacturing defect constitutes acceptable is mind boggling.  Consider for example our pre-cut kit.  It contains extrusions cut from 8 pieces of 1 meter length bars.  If 1 out of 20 pieces (5%) are problematic, I will have to do a 100% inspection or 2 out of 5 kits I manufacture will have defects.  This is simply not a sustainable way to run a business.

One of the reasons our new extrusion vendor overseas took so long to spin up is because they are willing to do cutting and kitting of extrusions for us - and there were a lot of back and forth in communicating the cutting list and packing procedures.  Cutting and kitting had traditionally been the choke point in the OpenBeam business.  The old process consisted of me driving close to 100 mile round trip after work with as many pieces of aluminum extrusions I can fit into my Mazda5 "Urban Deployment Vehicle" (I am not at the stage in my life yet where I am willing to admit to people I drive a Minivan... ) to my friends at MicroRAX to run the extrusions on their autosaw.  OpenBeam had to absorb any defects in manufacturing as well as manufacturing losses; for example, 8 meters of OpenBeam goes into each pre-cut kit, but the actual sell able extrusions in the pre-cut kit is only 7.5 meters.  500mm of OpenBeam disappear in the form of little drops and a big pile of aluminum shavings per precut kit made - and that's assuming that the saw behaved.  If a badly warped piece of extrusion that I missed on first pass inspection jame the saw that extrusion piece along with the other three that are being cut simultaneously with it gets written off as scrap.  This is why we temporarily pulled out of the Kossel extrusion kit business - the labor involved for us was too high to compete against someone reselling a compatible extrusion that had been cut for them at the factory.

About two weeks ago, the cargo ship Akinada Bridge carrying the OpenBeam container docked in Tacoma, and after clearing customs, I hauled home my loot from the container freight station.

Given our sour experience with our previous extrusion vendor, we set out to inspect our extrusions for two things:  Flatness, and accuracy of cut.  The latter is extremely important for 3D printer kits.

For flatness, we inspected a random sampling of extrusions on a granite inspection plate.  Granite inspection plates are precision ground blocks of granite.  Their job is to set up a reference plane upon which measurements can be taken.  This particular plate we used is made by the L.S. Starrett company.  They, along with Brown & Sharpe, Mitutoyo, are considered to be some of the finest metrology instrument manufacturers in the world.  The plate we have is a Class A inspection grade pink granite table; the table is flat to within 10 microns across its 36" x 48" surface.

To inspect for flatness, we randomly pulled 5 samples of each color out from the different shipping containers.  (it is CRUCIAL that the sampling be as random as possible, and not pull the top 5 pieces of box 1.  Years ago on my engineering job, our inspection department is known to take short cuts.  They inspected a lot of medical battery cases from one box and accepted the shipment.  Little did they know the vendor had shut of the injection molding machine to go to lunch halfway through the production run and there was a batch of undersized parts from the machine warming up after lunch.  The parts didn't get caught and quarantined and the production line went down as a result.)  We place the bars onto the granite table and check for signs of the bar rocking.  Then we take feeler gauges to probe the maximum size feeler gauge that can slide between the granite and the aluminum without lifting the bar off the table.

What is simply amazing to me is that 3 of these 5 bars don't exhibit any rocking at all, and I was hard pressed to slide the 0.04mm feeler gauge - the thinnest in my set, by the way - in.  On the worst warped piece, I can slide the 0.1mm gauge in.  I repeated the test by putting a dial test indicator on the aluminum and pressing down onto the bar to see if I can measure any deflection.  On over 50% of the parts, the total deviation is only 0.02mm from when I press down on the extrusion.

Allowable warp on the extrusion is 1.1mm per meter, per Aluminum Extrusion Council's guidelines.  These extrusions are, at worst case, an order of magnitude flatter than what the AEC allows.  This is a far cry better than the 3-5mm warp that we were seeing from our other extrusion vendor!  3-5mm warp, in a 1 meter length, is very very visible to the naked eye.  It certainly makes me question the other supplier's quality control procedures.

Next, we check for accuracy of cut lengths.  To measure length, we stood the extrusions up and used a Mitutoyo Digimatic Height Gauge to measure the distance between the granite reference surface and the top of the extrusion.  This method is somewhat faster and less prone to parallax errors than measuring with a pair of digital calipers.  The cut lengths are most critical on our new Mini Kossel kit, so we pulled 3 sets randomly to give us 36 samples to measure.  (This was before we merged the design, at one point I still had 12 pieces of 240mm in my design for the Mini Kossel).  I've placed the metrology results in an Google spreadsheet here.

The results are incredible, once again.  The calculated standard deviation - a measurement of how tightly clustered the spread is, is a mere 0.03mm.  As a comparison, my hair clocked in at 0.08mm.  If you were to assume that the distribution of the cut lenghs is Gaussian  - and that's a pretty safe assumption to make - 99.7% of all samples fall within a six sigma (6x standard deviation) spread around the mathematical mean.  In other words; given the standard deviation of  0.04mm; 99.7% of the measured values will fall within 0.04*3 or 0.12mm of each side of average.  To be safe, I can claim +/-0.3mm tolerance on the cuts (which is damn good, by the way, that is approaching machine shop tolerance for machined parts), and in selling 1000 pieces of extrusions, I may get 3 returns for being out of specification on cut length tolerance.  That is seriously impressive, and certainly something that I can build a profitable business upon.

I measured the precut kit lengths too, but as you can see from the data, that was a strictly academic exercise.  In fact, I stopped half way through and just packed up to go home.

Armed with this data, I am happy to announce the following:

A)  Extrusions, in the form of 1 meter long bars, are now on their way to Amazon fulfillment centers, and will be back in stock shortly.  We can now guarantee the new extrusions to 0.5mm per meter flatness.

B)  We will be re-entering the Kossel extrusion supply business, with the following difference:  we are contributing members to the Open Source Hardware community, we offer free shipping on our extrusion kits (free second day shipping for Prime households), our extrusions are competitively priced, available in black anodized, and we guarantee their cut accuracy to +/- 0.3mm.

C)  We have removed our biggest road block for building pre-cut kits; dealer backorders are now being processed, and there will be a few kits showing up on our web store.  In a month or so, when our next shipment arrives, we will start selling precut lengths a-la-carte with Amazon Prime fulfillment as well.

D)  With our supply chain issues resolved, we have signed up with MakerShed; even though I will not have a booth at Maker Faire this year, OpenBeam will be.  Look for OpenBeam kits in MakerShed soon!

E)  With our confidence in our new supplier, we have already placed a very large order (bigger than ALL the previous OpenBeam orders, combined to the American supplier, including from our original kickstarter campaign).  This, along with the competence of our new supplier and the fact that they only need 2 weeks instead of 8 to turn an order around, hopefully resolves our supply chain problem and having to hang an "out of stock" sign for months at a time on our web store.

F)  We still have a handful of extrusions from our old vendor; we will be getting rid of these in a fire sale to recoup the money sunk into it.  Basically, these are extrusions that may have flatness or cosmetic issues.  We don't feel good selling them at full price, so we'll just sell it at a large discount to benefit our local maker community.  Details will be announced soon.

We are happy to say that after an incredibly long and dark few months, the light at the end of the tunnel wasn't from an oncoming freight train and we are happy to be on the right path again.  With the production issues behind us, our next order that is about to leave for its ocean journey is for a whopping 13km of extrusions.  This single order (and not including the 3+km that we just purchased as a "test run") is more than all the extrusion orders to the old vendor combined, including the material for our first kickstarter campaign, and we expect to be placing an order of similar size regularly to ensure that we stay in stock.  We are happy to be growing, and we thank you, our loyal customers, for supporting Open Source Hardware.

-=- Terence, Rachel and the furry monster puppy.

OpenBeam in South Africa - and the awesomeness that is Open Source Hardware

Presently I am sitting by my gate at Boston Logan Airport - homeward bound from the awesome Open Hardware Summit.  I can't think of a better time to announce that it had finally happened:  someone finally made a derivative of OpenBeam! Not only had OpenHardware.co.za done what looks to be a wonderful job with making OpenBeam extrusions (minus the problematic center hole and anodizing finish), they also significantly improved upon the system by lasercutting brackets out of stainless steel.  In fact, we are going to try to see if we can import some of the stainless brackets for our customers here in the US.

This is how Open Hardware is supposed to work.  Paraphrasing Nathan from Sparkfun at the OHSummit's forking panel yesterday:  "This is a customer I cannot serve.  And every person that gets interested in electronics, strengthens the community and some day will help me too".  In this case, even though we offered the South Africans access to our extrusion die, shipping and import logistics to South Africa was just too expensive.  We are very happy to see OpenBeam used around the world.

In the US, if you would like to carry OpenBeam, our preference is for you to work with us and share manufacturing expenses.  It takes an incredible amount of money to keep supply chain moving smoothly - we buy our extrusions at multiple kilometers at a time - and we'd prefer a collaborative environment to help market the product.  We've set OpenBeam up in such a way that we can have dealers, and our dealers make 30% gross margin on resale of OpenBeam.

Just don't be a dick and try to steal our trademark.

An Epic Update - Kossels past, present and future

I'm sure everyone's eager to know what the status of the OpenBeam Kossel Pro is. Well, we have one heck of a surprise for you. As you may know Johann and I started on the Kossel project and went down somewhat separate paths. Johann's designs have always been elegant and rooted in the Reprap world, where parts are printed. I did not have a 3D printer for most of the earlier development process so I focused on the RepStrap route: I used a laser cutter and designed around laser-cut geometry with the understanding that I would later replace the laser cut parts with stamped metal components.

The original Kossel Legacy had a LOT of 3D printed parts. Earlier on in the design the extrusions were rotated 45 degrees from each other to simplify the motor vertices.  From a design for manufacturing standpoint this was also a nightmare for assembly. Assembling the vertex required holding 3 different pieces in the right orientation and sliding them together. But the biggest problem came from extrusion wear. The original Kossel had custom designed sliders that held a set of 623 bearings at 45 degree angles.  These bearings then rode up and down on the Kossel extrusions.  Because ball bearings were typically case hardened steel, they wore down the aluminum extrusions pretty significantly.

Around this time, the Kossel Pro's design looked very different. I've always favored building the linear axis with OpenBeam's NEMA17 motor mount on one end, 608 bearings on the other side (mounted via our 608 to NEMA17 adapter) and a shoulder bolt for the shaft. As the OpenBeam system designer this makes for great business sense; the amount of parts tooled would be minimal and they can be approximated with lasercut parts for the most part. However the system wasn't without it's drawback; Johann quickly pointed out that we needed two long pieces of OpenBeam for each linear axis when he only needed one and on top of that there was no way to control the two pieces of extrusions being parallel and keeping even spacing throughout the travel span. (An extrusion that long can be expected to bow and twist, both under load and from allowed manufacturing tolerances).

Blog photos
Blog photos
Blog photos

Then there's the question of what the carriages would ride on. We looked at IGUS N17 rails but we weren't impressed with the amount of slop in them. Worse yet the cost was pretty significant, even with Pacific Bearings' N17 slider clones. We almost purchased a die for a 1530 extrusion with integrated N17 bearing rail geometry, but around this time our old extrusion supplier started having major problems producing our regular extrusion to specification and we quickly cancelled the order over quality concerns.

Blog photos

In early January, (armed with a computer loaded with Chinese IM software), we located a supplier for recirculating ball bearing linear rails. We immediately arranged to have a small engineering sample set sent over with one of our regular plastic part shipments. Initial testing was very positive.  The rails are made of high carbon tool steel, so hard that they had to be wire-EDM cut to length when we order them. Bolting this to the OpenBeam extrusion greatly stiffened the printer's frame and guarantee that the printer frame will be straight. The recirculating ball carriage works like a charm - there are no eccentric spacers to adjust, no tedious assembly process to worry about. With the linear rail in hand, we focused on the tri-wing / Y shaped design, rotating each of the linear axis 90 degrees. The result was the printer you saw in the video.

I also provided Johann a set of linear rails and asked him to consider rotating the vertical members 45 degrees back into the "normal" configuration.  Armed with these rails, Johann redesigned the Kossel's vertices by combining a few of the motor brackets into one. This design later evolved into what became the "Mini Kossel".  The redesign addressed a lot of my concerns with regards to ease-of-assembly, but the shape of the 3D printed vertexes posed a challenge to retool into injection molding.  We had designed some stamped metal parts to replace the Kossel's vertexes, but they lack the good looks of the 3D printed parts.

The breakthrough came when I started modelling the Mini Kossel into Solidworks to help with assembly instructions.  I was driving to MicroRAX to cut some aluminum extrusions one night after work, when it hit me like a ton of bricks: The shape of the lower and upper vertices are so similar they may as well be aluminum extrusions cut to length with holes drilled into them.  It turns out that the answer was in front of us all this time!

A frantic few nights of late night CAD and a few request-for-quotations later I am happy to present to you the newly redesigned OpenBeam Kossel Pro:

Blog photos

Not surprisingly this looks just like a Mini Kossel scaled up.  It maintains the same 250mm diameter x 250mm height build envelope and yet the foot print of the printer is slightly smaller than the original Kossel Pro with more usable space inside the machine frame for electronics  (although the printer did grow a little bit in height).  The parts will be 100% compatible with the Mini Kossel (except in extrusion and linear rail length).  The Reprap Mini Kossel's frame is exceptionally stiff and we expect this full metal frame construction to improve in stiffness over the Reprap counterpart.

As a design engineer who had sunk 6 months into prototyping / R&D of my own design, there had to be darn good reasons(and lots of pride swallowing) to mothball a design six months in the making.  So, allow me to walk you through my reasoning.

As a crowd funded project and an open source hardware company we feel that we have a responsibility to the community to do what is best for everyone. In this case what evolved into Johann's Mini Kossel is clearly the more elegant and minimalist of the two machines. Early on we had made the decision to try to make as many parts interchangeable between the Mini Kossel and Kossel Pro; the availability of well produced injection molded parts that can be reused in various configurations will help bring quality of the machines up and the cost down.

Today we are happy to announce that by merging the Kossel and Kossel Pro's design tree we have achieved 100% sub assembly level compatibility between the two branches. To support the community we will supply BOTH a mechanical "vitamins only" kit to the Reprap community (requiring 3D printed parts) as well as a turnkey mechanical kit.  We aim to release the Reprap Vitamins kit within the next month, and we will make the turnkey mechanical kit available for purchase after our Kickstarter rewards have been fulfilled.  Builders will be able to start with a Reprap Kossel and upgrade various components as needed with injection molded parts (available from our web store and fulfilled by Amazon.com). Our injection molded components will be slightly more expensive than 3D Printed parts but will offer more advantages thus still representing a good value proposition.

We also believe that Johann's Mini Kossel design is more elegant and minimalist.  If we were to bring the Kossel Pro as presented in the Kickstarter video to market, that design will likely face rapid obsolescence upon release.  By taking the efforts to redesign now and making the design 100% compatible to the more open Reprap Mini Kossel branch,  we stand a very good chance of establishing the Kossel / Mini Kossel as the benchmark printer for Delta 3D printers.  As 3D Printer designs come and go we certainly want ours to be produced and used 2-3 years down the road (which is an eternity in the Reprap world but then again people are still building threaded rod Prusa Mendels).

It has been a mad frantic crunch to redesign the printer; we are releasing manufacturing files to our manufacturing partners this week. With regards to the schedule we had padded the original schedule considerably but this safety padding is now gone.  Any additional major delays will now result in day-per-day slip in the manufacturing schedule. That being said we are hopeful that we can still honor the original delivery dates (before Christmas) and we will advise in a few weeks when our manufacturing partners have had a chance to plan their operations around the new parts.

So, let's take a closer look at the CAD screenshots!

Here is the redesigned end-effector with the auto-deploy probe. The end effector is only 10mm bigger than Johann's 3D printed end effector for the mini kossel, but it packs a LOT more features.  A ducted fan system provides uniform cooling to the part and a 3rd fan provides dedicated cooling to the hot end's thermal barrier.

Blog photos
Blog photos
Blog photos

Here is the auto-touch probe.  It is adjustable for deployed length. In Johann's design you set the probe off and measure the offset in height to the hot end tip.  Here you have the option of doing things the other way around:  Shim the hot-end tip up with a piece of material of known thickness then adjust the probe until contact is made.  I think this is an easier way to go.

Blog photos

The auto probe now requires just a simple bump to the side to deploy, greatly simplifying the deployment process.  It is modular and can be mounted onto Cartesian style printers as well.  We will be making the injection molded parts available to other 3D Printer kit manufacturer at reasonable prices to speed adaptation of this crucial improvement.

Blog photos

There are more CAD files; but for now these are the first two parts that we are going to be prototyping (as they require the most TLC to get correct). We expect to have prototype parts back in hand this coming week and we’ll be assembling it into our test bed printer.

We apologize for being behind in our updates; we needed to focus on getting the redesign completed so that we can alleviate any schedule slip.  Now that we are over the hump with the design work, we will be updating more frequently as prototype and first article parts start coming in.

-=- Terence, Rachel and the furry monster puppy

Kickstarter Update - Linear Rails shipping soon (and should be back in stock at Amazon.com soon)

Hello Kickstarter! About 3 weeks ago, we closed the fundraising part of the campaign successfully. Now the real journey begins. By now, all the credit cards have been processed, Amazon and Kickstarter have taken their respective ~10% slice, and the rest of the money had been safely transferred to OpenBeam’s bank accounts.

A project creator wears many hats. On a challenging project like this one, one of those hats is that of a project engineer / project coordinator. The sooner we design release engineering data and specifications to other engineers, the sooner we unleash them to do their work and the earlier we get into testing and systems integration. Towards that end, we have been working on solidifying the electronics specification; work has been progressing steadily on the Product Requirement Specification on Brainstorm - Brainwave’s big brother that we plan on shipping with our kickstarter printers. The Product Requirement Specification is a technical document that calls out the requirements of the product - from connector selection and population, to chipset selection, to cosmetic finishes and so on. Our electrical engineer, Mike Ziomkowski, has been working closely with Matthew Wilson and me on drafting the Product Requirement Specification and translating it into a schematic block diagram. The process is an iterative one with two way dialogue - and a careful tradeoff of adding features vs raising the retail cost of the board by too much. A good portion of this dialogue occurs at 3D Printer nights at Metrix:Create Space; if you are local to Seattle, feel free to drop in on Thursday nights and say hello to our team. We are not ready to share out the Brainstorm PRS yet, but we will at the completion of the board design.

On the hardware front, we have been working closely with Johann Rocholl, the originator of the Deltabot family, as well as the designer of the Reprap Rostock, Kossel and Mini Kossel, on trying to improve part compatibility. We believe that the more we make parts interchangeable between the Reprap branch and the “Pro” (manufactured using mass production processes) branch of the Kossel design tree, the better it will be for the 3D printer community in general. We are also putting considerable efforts into improving Johann’s auto levelling probe design by making it more robust, reliable, reproducible and modular.

On the marketing front, we recently helped Johann build and ship a Reprap Mini Kossel prototype to MakeZine for inclusion in their 2013 3D Printer Special Issue. From a user experience standpoint, the auto probe demonstrated its value - we were able to assemble the printer, flash the firmware with almost no modifications - and then proceed to crank out a perfect torture test print with zero calibration! (Johann’s personal machine and our design use different extruders, and we had to correct for the different E-Step value in firmware. That, along with compiling our software for Azteeg X3 instead of a Printrboard, were the only changes we made.) We are very excited about this; the Reprap Mini Kossel and even the low end Kossel Pro fall under the “entry level” classification from MAKE (based on a MSRP of less than US$1,000). We believe we are a very strong contender in this category given the print results.

We have initiated prototyping of some of the injection molded parts; we hope to complete the rest of the design (for injection molding) in approximately 2 weeks’ time and prepare for tooling launch after that. We look forward to, in about two weeks, sharing our completed redesign.

Finally, on the project administrative side, our vendor for linear rails should be finishing up the EDM operations on the linear rails. (The steel used in our linear rails are so hard, they have to be cut via a wire EDM - it is harder than the carbides used for cutting tools.) These rails will then leave our vendor in Shanghai for Hong Kong, and from Hong Kong go into consolidated air freight for their trip to Seattle. We’ve elected to bring the parts over by air freight because believe it or not, it is almost the same cost as ocean freight given the relatively small volume and weight, without the additional 3 week transit penalty or 1 week of waiting for a ship’s scheduled sailing. Along with our shipment will be samples of fasteners and bearings used on this project, along with a few power supply units for testing and evaluation. Our linear rail backers will be receiving a shipping survey when I receive confirmation of air cargo pickup. The rest of our backers can expect a shipping survey when we get closer to shipping their rewards.

That’s it for this update. We’ll be back in about two weeks with the completed Kossel Pro design. :-)

-=- Terence, Rachel and the furry monster puppy. :-)

Going Global - Expanding our EU presence

Hello OpenBeamers!

We have some very, very exciting news to share with you.

We have located and qualified an off-shore supplier for our flagship extrusions.  We had tried very hard to keep manufacturing in the US, but in the end, the numbers just did not work out.

It was bad enough that our US supplier screwed up three extrusion runs in a roll - but what made it fundamentally unacceptable to keep doing business with them was their belief that the 5+% reject rate we experienced in the last batch constitutes as “okay” - they seem to have a very different barometer for customer service and quality control.  Borrowing a quote from Marsellius Wallace from Pulp Fiction:  “We are pretty *beeping* far from Ok".  We have literally spent more time and money sorting good parts from bad than on our new extrusion die.  

In the end, it came down to a very simple business decision:  The offshore vendor is cheaper, the quality is better (although it would be hard to get worse in the quality department), the shipping rates to ship a pallet of OpenBeam across the world is comparable to Fedex ground freight from California.  Our off-shore vendor can manufacture good parts, load them onto a container ship and have the container ship dock here in Seattle before the extruder in California can deliver questionable quality parts.  Our off shore vendor is also more willing to provide value-added services such as kitting and cutting at a reasonable cost - something that our American supplier had been unwilling to even consider.  

We are working with MakerBeam.eu on the new extrusion supplier, and we have reached a joint agreement to produce these extrusions.  Under this arrangement, both MakerBeam.EU and OpenBeamUSA.com will be using the same extrusion vendor to manufacture the extrusions, and both companies committed to placing regularly timed orders together.  OpenBeamUSA also explicitly gave MakerBeam.EU the rights to use the OpenBeam trademark in the EU as well as to manufacture the extrusions without paying royalties, and extended special discount pricing for volume buys on the injection molded brackets from our supplier in Hong Kong.  In other words, we are setting up MakerBeam.EU to become the de-facto lead importer of OpenBeam to the European Union region while keeping prices reasonable.  So, for our European friends, you will soon be able to buy OpenBeam at very reasonable prices; and if you are interested in becoming a European reseller, you can contact MakerBeam.EU directly.   Shipping is something that benefits from economies of scale and by concentrating the orders this way we can amortize the freight across a much bigger shipment and get better pricing from our vendors due to the higher volumes.  

OpenBeam is also happy to announce that Technobots has also started carrying the entire range of OpenBeam products.  The really exciting thing is that Technobots has a brick and mortar presence as well, giving us our first walk-in retail outlet in the EU.  

For our Australian and New Zealand friends, we will be making an announcement in the coming month concerning a similar deal with regards to the AU/NZ territory.

With today’s announcement of a qualified, competant extrusion supplier going forward, we are also happy to release the new revised CAD geometry onto Thingiverse.  We have been a little hesitant to release the new CAD files before fixing our supply chain issues - remember, once you open source something, there is no going back - and generally speed of execution and established distribution network / community (and brand name recognition) is the only protection an open source company has to protect its market share.  

OpenBeam is still a very young company - we shipped our first extrusion only about 10 months ago.  We have been working very hard in the past 10 months to building out our supply chain and distribution network, so that you, the end user, can get your hands on the OpenBeam Construction System easily.  At times such improvements may not have been visible to the causal observer, but they were a very necessary step to build a strong foundation upon which we are going to continue to grow our business.  Now that the foundation is laid, stay tuned for more exciting updates from us.

-=- Terence, Rachel and the furry monster puppy