Gold plated SS - one particular problem

Hello All,

I recently received my first gold plated SS model and the goal of this model was to evaluate whether or not it may be possible that this method of manufacturing could successfully produce a particular class of parts. Generally, I'm encouraged by the print. There are a number of features that were very important and were produced accurately enough for functional prototyping use. In particular, the bolting pattern on the flanges as seen in the first attached photo were particularly impressive considering the method of production.

However, I ran into one defect that renders the part unusable for the desired application. As seen in the first picture, this structure has two flanges. These flanges need to be flat. In the received print, one of the two flanges is perfect. It's perfectly flat. However, the second has a huge amount of curvature. Please refer to the second picture, labeled as "Flange 1.jpg". As seen when placed next to a ruler, there is ~ 2 mm of curvature deviation across the face (the middle touches the rule, both edges are ~ 0.5-1 mm away).

Since one of the flanges is perfect, I would tend to think that this effect could only be caused during the brass infusion part of the process. I.E. when the part is being packed with powder perhaps it wasn't properly supported? As if the powder was just dumped on top of it as opposed to being pushed in from the sides. I could forsee such a method to cause this part warping.

All design rules as far as I could tell were upheld. All walls were at least 3 mm thick.

Can anyone think of a reason for why this deviation would occur? Any thoughts on how I should address this in the model?

It appears to me that it could be because it's a flat unsupported area. I'm thinking either in the build or infusion, the flat side was flat on the surface, and the curved side, because of the weight and heat bowed. I could be wrong though. But i think triangular supports would help in keeping both faces flat. Something like I've drawn here but on all four sides.

I was thinking of adding triangle supports as well for the next revision. We'll go at this one more time.

Obtaining dead flat surfaces with this 3dmp system is a constant challenge. The warpage generally occurs in the furnace, somewhere during the sintering/infusion cycle. Since visually monitoring these furnace cycles is not a practical option, the actual cause remains an open debate.
Prevention of warpage often involves design tweaks, custom appliances or furnace setups to better support the parts during furnace runs. Because of the large volume of SW part production custom setups are not possible.
Where tolerances can not be held by a given production method, allowances need to be made for post production finishing. In the case of this flange allowance for machining might be the only option. This is maybe not the answer you are seeking but producing absolutely net parts is rarely possible via a single technology.
-G

I know this is a frustrating problem. It drives the production crew absolutely crazy sometimes! Before you spend time adding a truss system you should be aware that sagging can and sometimes does occur between the bridges. The time involved in removing a truss system might be more costly than simply adding a few thousandths of material to the ends, then machining the faces dead flat and parallel.
-G

Hello Glen,

Thanks for the detailed replies. I'm pretty sure the sagging is caused by not enough ceramic material being placed under the overhangs.

After the printing process, it's my understanding that the model is removed from the stainless steel after being heated to harden the binder. Next it is placed into a new box and ceramic powder is placed around it. However, from all of the videos I've seen of the process the ceramic powder is basically dumped in from the top. For anything with an overhang, this would pretty much guarantee void space below overhangs (or at the very least a downward pressure on an unsupported surface. This box is then heated. At a certain temperature, the binder burns off and the stainless steel powder begins to sinter. I believe that during this part of the process the models would able to sag if there is any void space.

To correct this problem, you need to distribute ceramic powder uniformly around the part. The easiest way to do this would be with a mechanical vibrator. I would bet a concrete vibrator would do the trick:

http://www.toolking.com/1/dewalt_dc530b_18_volt_cordless_xrp _concrete_vibrator_bare_tool_1051640867.php

If you vibrate everything in the volume the powder will fill the volume to a uniform density and properly support the models.

I use a similar piece of equipment at work to eliminate a similar problem except that we're working at much smaller scales.

I went ahead and added some support structures and had the model reprinted. The support structures appear to have corrected the warping problem.

Thanks for the help!

Glad this worked out.
-G