From: Liquid Plastic Solutions (firstname.lastname@example.org)
Date: Thu May 30 2002 - 19:22:16 EEST
Without knowing what kind of resin you use or the manner in which you
post-cured and measured the part I can offer a partial explanation as to
what happened. You are dealing with a number of issues when building a
1. Shrink Compensation. This I would think is your culprit. If you build a
part that is say 5 inches long and you have a shrink compensation number
that is off by say only 0.003"per inch your cumulative shrink is .015" for
the length of the part. If you extrapolate it to 18 inches you have a
cumulative shrink of .054". As the part gets longer the error is magnified.
I would check you X and Y Shrink comp. numbers against reality
2. Beam Width Compensation. If you don't periodically check your beam for
roundness and size you can add to dimensional error. You need to furnish
your slicing program with the correct beam width information otherwise you
might be building walls too thick or too thin. The best way to check for
beam width accuracy is to measure wall thickness in x and y against cad
values. If they are larger or smaller you have a beam comp issue. I don't
think that this was your major problem, but it may contribute by up to
0.004-.010 on part length.
3. Resin and Vector Length. Depending on what resin you are using, there
may be an issue with shrinkage in longer vectors that you don't see in parts
built in diagonal orientations. With the exception of drawing boarders,
your SLA machine only draws vectors in two directions (X and Y). At the
same time, whenever a beam draws a vector in your vat of resin it changes
your resin from a liquid to a solid and contracts on itself due to it's new
physical state. In combination this means that parts built in strict X and
Y orientations have longer vectors than do those built in in diagonal
orientations and as a result have greater "cumulative shrink" due to the
reorganization of the material due to phase change. This factor becomes
less of an issue as the thickness in Z increases. The practical
demonstration of this phenomenon can be seen by building a solid rectangular
box say 6"long by 0.5"wide and 0.3"thick. Upon examination of the long side
in profile you will find that the lowest layers shrank more than the highest
layers giving a sort of reverse draft appearance. The lowest layers were
unsupported by previous layers and only tokenly by the supports.
Again, by not knowing the geometry of the part I'm not sure of the effect of
4. Post Cure. SLA parts continue to shrink even during the post-cure phase.
I don't have any data on how much, but I'm told that it is very measurable.
There are other effects I'm not mentioning either deliberately or not, but I
think that these are the most likely suspects in your case. A good rule of
thumb for buildable tolorence is +/- .005" per inch. So in your case your
part was still within tolerence.
Liquid Plastic Solutions
----- Original Message -----
From: Rafael Santillan <email@example.com>
Sent: Wednesday, May 29, 2002 11:48 AM
Subject: SLA 500 tolerances problem !! I need a clue...
> hi listers.
> I need your help ... again.
> we just built a large part 18in long in a SLA 500, since the size is
> than the base we built it diagonally to fit in to the machine envelope.
> te parts is with good finishing but the of the dimensions "length" came
> be 0.0768in longer than the actual dimension we have on the STL file.!!
> any clue why the tolerances change from 0.009inches we had on the other
> dimensions of the part to 0.0768inches on the lenght ??
> I need to give some explanations to my client who is quite upset.
> Is it normal ? and if so what tolerance should I offer to a client for a
> long part ?
> any feedback will be very appreciated.
> as usually trully yours
> * Dr. Rafael Santillan, Phone: 52-442-2230745 *
> * Director 3D RAPID Fax: 52-442-2226774 *
> * Mobile: 52-442-2366015 *
> * Email: firstname.lastname@example.org *
> * MOVIL Email: email@example.com *
> * *
> * Address: Priv. Martin Carrera 105, *
> * Colonia Presidentes *
> * City: Queretaro,Qro. MEXICO. Zip. 76080 *
> * WEB: http://www.3drapid.com *
> Este correo fue enviado utilizando el servicio IMP para CIATEQ
> For more information about the rp-ml, see http://rapid.lpt.fi/rp-ml/
For more information about the rp-ml, see http://rapid.lpt.fi/rp-ml/
This archive was generated by hypermail 2.1.4 : Tue Jan 21 2003 - 20:13:50 EET