Re: Automatic Tool Path Generation

From: PnRMolds@aol.com
Date: Tue Feb 03 1998 - 16:56:35 EET


Hmmmm... Some interesting points of view here on this topic. I must say first that I agree with Mark Bannister.. STL files have terrible resolution! Use the original geometry data if at all possible. I must tell you as an injection molder and mold builder STL files are the bain of my existence. I would certainly prefer to work with the original part data before any STL file. While it is not disputed that the accuracy of the STL file may be increased, this step itself makes working with the part more cumbersome. In addition no matter how much you increase the accuracy, the STL model will still not be a very good representation of the of the part. Consider that you will end up approximating ("posted" NC code sent to NC machine) an approximation (generated tool path) of an approximation (STL file) of the original part. If you are planning on NC machining an RP part to improve the accuracy I would question the use of the STL file vs. the original part data - even if it has been translated through IGES, VDA, STEP etc. Of course if you are working on a Cray perhaps you can run your approximations until they approach infiinity..... ;-) Personally I have been in the mold making end of this biz for about 10 years. In that time I have seen dramatic improvements in the quality of imported data files which for us arrive in IGES format unless the customer is working in Pro/Engineer or Cimatron -what we utilize for tool path generation. Over the last few years I have seen VERY FEW problems with IGES files including those with complete (trimmed) surface and wireframe data. I am not saying that we do not see anomalies but in general things have improved a great deal in this regard. See below..... ****************************************************************************** ****** In a message dated 98-02-03 07:10:58 EST, it was written: << Subj: Re: Automatic Tool Path Generation Date: 98-02-03 07:10:58 EST From: info@spline.nl (Delft Spline Systems) Sender: owner-rp-ml@bart.lpt.fi To: rp-ml@bart.lpt.fi ('RPML') Brent Stucker asked: >I have heard about software packages that use STL files to generate tool >paths automatically. Is anyone familiar with commercially available software >that can be used for that purpose? > Mark Bannister replied: > AlphaCam Bert VandenBerg wrote: > Delcam's PowerMill/CopyCad programs > Cimitron's CAM systems > VERO's VisiCAM > Tebis AG's Tebis 3.0 > MatraDataVision CAD/CAM system Hi Brent, So sorry (not quite) to send the following self-promotion to this list, however, you did ask for it. All the above CAM packages can indeed do the job that you require. However, they are aimed at professional toolmakers (CAM specialists), not at (rapid) prototype builders. This means that they include many milling parameters that must be set, and that they are quite expensive. The advantage is that they offer many possibilities, the disadvantage that much training is needed. In any case no automatic toolpath generation. While some of this is true, in my opinion this can be a misleading statment. The comment above seems to suggest a complicated routine requiring tons of experience is needed to use such CAM applications. I do have a fair amount of experience in modeling but virtually none in tool pathing. However, I spent a few hours by myself on a seat of Cimatron and was able to figure it our all by my lonesome even though I have had no training in NC programming and am not a mold maker or CAM specialist in any way. As far as setting parameters is concerned, Cimatron has the functionality to store tool path "templates" for future use. Thereby parameters can be recycled time after time without the repetitive task of changing the parameters maually everytime. Even so, since every toolmaker/programmer that I have seen has their own ideas ot speeds and feeds for cutting tools I would consider having control over the machining paramters to be a requirement. As far as the cost goes... ten times $1500 is not as high as it can get but it seems to me that you still get what you pay for. Note that I will be downloading the DeskProto demo. In the final analysis.. I would like to know how things turn out. As far as we know our DeskProto software is the only package that does indeed offer (almost) automatich toolpath generation based on STL geometry. The almost indicates that some user interaction is needed: you have to enter the tool to be used and the distance between the toolpaths (which is in fact the accuracy). The rest is done automatically. Price of the software is ca USD 1,500.00 where the above packages will cost about ten times as much. Please download a demo-version from our Web-Site and test the software with your own STL files. Mark Bannister also wrote: > Again as has been mentioned many times here before, STL files have > terrible resolution! The resolution in most cases is not the problem as it can be freely chosen. The big advantage of STL is the data-transmission without any problems. This in contrast to other formats like IGES. Because of this oven some professional injection mould toolmakers do indeed use STL. Best Regards, Lex Lennings. ------------------------------------------------------------------------- Delft Spline Systems, The Netherlands. We offer DeskProto: affordable Rapid Prototyping using CNC milling mailto:info@spline.nl website:http://www.spline.nl >> ****************************************************************************** ****** ****************************************************************************** ****** Best Regards to all, Ron Ptaszek P&R Precision, Inc. - InjectionMolds.com T: 810-254-6720 Direct Email: Ron@InjectionMolds.com http://InjectionMolds.com and coming soon.... http://Prototyping.com ************************************************************************* Plastic Injection Molds direct from your database for 10 years. Rapid deliveries and the highest quality ************************************************************************* For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/



This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:44:49 EEST