Socket design

I have a pretty big project ahead of me designing an arm that would get me back on a motorbike.  Its close to initial testing, so watch out for that.  Cost for the proposed motorbike arm is a little high, so I'll have to double down on paying work.

In the meantime, I'm improving the socket used with my residual limb.  Looking to improve my shifts in my car.  Currently, with the now 2-year-old socket design, getting fifth and second gear can be hard work.  I imagine that this wouldn't be such a big issue if my injury wasn't a brachial plexus injury.  I just don't have quite enough strength in some ranges of movement yet.

So, a new socket was required.  This time I leveraged 3D Scanning to generate a base part to model the socket from.  This little exercise is a testament to how easy the Einscan Pro+ is to use as I had my Dad scan my stump.  All over in less than five minutes and the scan data was impressive.  I'll spare you all the grisly detail and not share the original scan data - the doctors put in a lot of effort to save the arm and whats left has some impressive scars.

Scan data was imported directly to Autodesk Inventor and the mesh point projected to an array of sketch planes.  A raw (ugly) loft feature was then carried through to the socket design shown.

A quick and dirty PLA print proved to be a nice fit.  I'm really happy with both the fit and the fact the "free play" is greatly reduced, but it's yet to be tested.  This will mean drilling out the carbon fiber tube and assembling a new "hand" from spares.  Over the weekend maybe.

 

Multiframe L Plates

Here I am learning to use Multiframe.  Why?  Multiframe is used by my RPEQ engineer (Dr. Bill McWhinney) and we're working together to find a more efficient way to produce Multiframe models for analysis.

I think I have the modeling licked, and the only real time is spent dragging the data out of Autodesk Inventor and through AutoCAD to Multiframe.  Not too difficult, but requires an intimate knowledge of the assembly being modeled.

Once the data is imported correctly its a breeze, and the model pictured has only taken a couple of hours.

Might be some time before I can run any form of analysis, but the reality is that the analysis is Dr. Bill's race to run.

Flashforge Guider II

Larger build volume has been something I've been chasing for a while, and after a mixed experience with a Tevo Black Widow I've got myself this beast of a printer.  Big thanks to Nathan over at NP3D for the great service, price and short lead time.

Build volume is 250mm x 280mm x 300mm, perfect for larger test parts and panel/trim components.  Accuracy is great, bed adhesion and I are still working things out.  So far I've only run a few PLA parts through; one 26 hour print run and another 14-hour print run.  Perfect prints.  I believe, the printer and its cooling are best suited to ABS - the destructions suggest running the PLA prints with the top cover off and door open.

Einscan Pro+ 3D Scanner

A few years ago I bought a structure 3D scanner and started messing around leveraging what 3D scanning had to offer.  The structure scanner may not have been a good starting point, great scanner but the accuracy was pretty woeful.

From there I used 123D Catch from Autodesk when it was around before moving to Recap Pro.  Recap Pro works quite well if you have the time to take good photos.  The Ducati pipes I modelled a while ago were built from a Recap Pro photogrammetry scan, so not too shabby.  However, when I'm on site and I need a reference cloud of a HV substation I don't (usually) have the time to gather as many photos required. 

Photogrammetry and the structure scanner are "okay" for larger parts/locations, and great for building around a reference, but I needed something with greater accuracy for smaller parts, and for the location of components/mounting holes in mid-sized assemblies.

I'd have liked to have had the budget to afford a FAROarm, like the one I had used working with Ian Patterson, but I ended up purchasing the affordable Einscan Pro+.  Purchased through the 3D Printing Superstore.

The first 24 hours with the scanner has shown that the scanner is great, the user on the other hand...  Scanning an object requires a good amount of pre-planning and preparation.  I've got a fair way to go as I find the limits of the scanner.  Meanwhile, Nick over at 3D Scan Expert has given the Pro+ a good working over here if you're curious.

Successfully completing a scan is one thing, producing usable data is another.  The next steps will be dragging the scan dat through Inventor/Recap to produce accurate data sets.

Testing 1, 2, 3

To keep costs low Ontrak Diesel repairs has had me print a few prototype suspension mounts.  These are upper and lower spring mounts for various vehicles, so fit to chassis is critical.  Cutting and pressing even 6mm plate is a costly venture when you're trying to get things to fit.  A days worth of 3D printing PLA parts provides workable test parts at a fraction of the cost.

Lego engine headers

3D Printed header design kit - half of the batch, other half still printing.

Used to form a custom set of headers on a 351 Windsor powered hotrod.

I'll update this post with images of the build.

Ninja Stars

Drill spider plates from an unknown drill rig.  Models and drawings produced from cardboard cutouts, hand measurements and a 3D scan.

Original items are from 80mm steel plate and 2 x 25mm steel plates.

Pulling Teeth

My most recent project is to measure and model teeth for a crusher.  The details of the crusher are irrelevant for this blog, but lets just say its DAMN HUGE.  This will be the first time I've used surfaces to model a production part.

Preview shown below is the smaller of the teeth.

Image is of a 3D printed section for client review.