Multi Jet Fusion Report – A look under the hood

In the office we’ve been quite excited for some time about the upcoming RAPID conference that’s been going on this week in Orlando, Florida. RAPID is always a big affair and this year was no different with much speculation that HP might finally release their much awaited Multi Jet Fusion, MJF machine. As well as XJet demoing their NanoParticle Jetting technology for the first time.

Multi Jet Fusion is a powder bed process that uses inkjet technology and infrared radiation to build parts with very similar if not better resolution and mechanical properties to laser sintering. What HP brings to the market is revolutionary as a machine package, but is in itself actually nothing new in terms of technology. A closer look at the Multi Jet Fusion  process reveals the bringing together of two AM concepts that have existed since around 2003, namely High Speed Sintering and Selective Inhibition Sintering.

High Speed Sintering, HSS uses a carbon black ink sprayed onto each layer to effectively target a flood infrared light source onto specific areas of the build bed. The blacker an area of white powder, the more heat it will absorb.

Selective Inhibition Sintering, SIS is effectively the reverse of HSS and can work by various different methods. For example a liquid such as water can be sprayed onto the powder to keep it cool when exposed to IR, thus thermally preventing sintering. Or a chemical such as silicon or light oil can be sprayed over an area of the powder bed to mechanically inhibit sintering. This works by coating the individual grains of powder preventing them from sintering together despite being hot enough to do so.

So no prizes here for guessing which processes make up HP’s ‘Fusion Agent’ and which one the ‘Detailing Agent’

So this approach may not be original, but it is certainly valuable. These two processes together are genuinely faster than other polymer powder bed processes. This is due to the fact that the whole layers can be marked out and sintered in two passes of the build bed rather than waiting for a laser to be steered across the whole outline and cross section of the part. From this Multi Jet Fusion is able to make its claim of being 10 times faster than laser sintering. or course this claim is highly geometric dependent as a laser system can scan a thin wall section object very quickly.

The fact layers can be sintered with a more gradual and consistent heat input rather than the short sharp localised shot provided by a laser beam means it produces parts with marginally better UTS and elongation at break than SLS – High Speed Sintering – Continuing research into a new Rapid Manufacturing process, Helen R Thomas, Neil Hopkinson, and Poonjolai Erasenthiran

Finally there is the price which is quite agreeable considering the capacity and resolution of this machine. Here again HP is able to benefit from HSS and SIS due to a lack of expensive laser optical systems required in standard SLS systems.

However some of the HP’s marketing material has perhaps set the bar little too high and they could face challenges achieving their future aspirations.

After about the 3 minute mark we learn HP’s aspirations for future printers which can control the colour, surface properties, translucency, elasticity & strength, conductivity etc. at a voxel level on the part.

It’s worth noting that the current machine HP are releasing only prints in black, but HP does have a prototype machine running colour parts, in fact they’ve been showing them off for the last year or so, we also know from material jetting technologies that it is possible to control material properties at a voxel level. But the general feeling is that HP are perhaps being a little too sensationalist especially with regards to this suite of extra controllable properties. A brief analysis of each reveals some potential issues.

Colour:

The reason we see different colours is because of how materials absorb and remit more of certain wavelengths then others. HP use this very fact to control their sintering, a black body absorbs infrared radiation much better than a white one. The obvious difficultly being that when different colours are printed onto a white powder bed; how do you ensure that the whole part is evenly sintered? Subtle changes in heat delivered to the powder bed have been shown to effect the mechanical properties of the final part. Without careful management very light areas in close proximity to very dark areas could cause enough of a thermal stress that parts warp and cause build failures. So perhaps this will create a slight trade off between the use of full colour and mechanical strength or require a much more complicated thermal management system in the machine.

On an interesting side note HSS parts produced using dithered grey scale rather than 100% black are actually significantly stronger. But don’t look as good for juicy marketing shots! The HSS Sweet Spot.

Surface Properties:

By varying the amount and mixture of IR inhibitor and IR absorbent it is conceivable how you could produce parts with very smooth areas in one region and much rougher sections in another. It’s even possible to control surface hardness but your bulk material is still always going to be the same powder so there will be limits on what is possible. Not forgetting that these mechanical properties are always interlinked. You need a soft surface to protect your component, what if that compromises the ultimate strength of that area of the part?

Translucency:

When it comes to powders suitable for SLS, SIS & HSS generally only semi-crystalline or totally crystalline materials are fit for purpose. This enables much smaller targeted amounts of energies to be used to finally sinter the material as its sharp melting point means build chamber heaters can hold the material just below its tipping point. We also need a material with a higher melting point than its recrystallization temperature otherwise if the material is allowed to recyrstalise too quickly or before the part is finished printing it can warp causing build failures. It may be a real struggle to find a material which can be both opaque and transparent to any great degree at the command of an inkjetted agent. Multi crystalline structures also need to exhibit long range order before they can become transparent, something that this process is unlikely to lend itself too.

Elasticity & Strength: 

The following paper, Effect of greyscale/print density on the properties of high speed sintered nylon 12, shows us how by varying the grey-scale used to sinter HSS parts does have an effect on the UTS, elongation at break and the Young’s modulus of the parts. However again due to the fact that the bulk material is always going to be the same material there are limits on what is truly achievable here. We can make a part more elastic by reducing the degree to which it is sintered, however the elongation at break and the UTS also suffer somewhat limiting the practical application of this.

Conductivity:

Again it is conceivable that a conductive ink could be used to creative conductive pathways through a part. How much current and voltage these inkjet thin traces would have capacity for is another question. Whilst this process gives the user a great deal of flexibility in 2D it may prove quite difficult for a conductive ink to penetrate subsequent layers of the insulating powder used to build the part presenting issues when designing 3D circuity.

 

In summary: HP’s Multi Jet Fusion looks like a great machine, really interesting stuff and definitely worth following the developments to follow. Who will be the first to capitalise on a machine with such a high production capacity for end use parts and not just prototypes?