Close



Results 1 to 10 of 10
  1. #1
    Banned
    Join Date
    Jan 2015
    Posts
    306

    Newb with 3D scanning questions.

    Hey everyone, I am sorry in advance if these questions have been answered. But the questions are so general I don't think I would be able to find a concise answer, to what I'm looking for.

    I want to get into 3D scanning on top of 3D printing as part of my business. I was approached by a person who needed me to scan something and then 3D print it at a different scale, sounds simple enough.

    The question I have is, with my limited knowledge, does a relatively inexpensive 3D scanner such as a the FUEL3D scanner have the ability to scan and then create a workable STL file or STEP file for me to manipulate? Or must there be serious file processing after the very first scan operation takes place?

    For instance, is it just point and shoot and then the file requires some prep work? Or is it a lengthy healing process that requires the geometry to be rebuilt, meshed, processed, etc... or whatever?

    I was told by a friend in the scanning business that the $70k scanners are expensive because they do all the work for you, like the creaform handyscan series. How does that compare to the FUEL3D?

    Thanks

  2. #2
    Technologist
    Join Date
    Jan 2015
    Location
    Waterloo, ON, Canada
    Posts
    159
    Add truly_bent on Shapeways
    DBFIU;
    I don't know a great deal about 3D scanners, but i had a look at the FUEL3D you mentioned. It looks like their SCANIFY software is designed to create a single 3D face (read open mesh) from a stereoscopic view. In their technology overview they state;

    SCANIFY has a fixed focus and captures a maximum size of about 40cm/16” diagonal in a single scan, approximately the size of a sheet of letter sized/A4 paper. Acquiring a larger 3D model requires multiple scans to be stitched together.

    I didn't see anything further about stitching multiple scans together, so i expect you'd have to do that outside of their SCANIFY software. That could turn out to be pretty challenging. I'd ask them more about that before laying down my money.

  3. #3
    Banned
    Join Date
    Jan 2015
    Posts
    306
    open mesh is bad.. very bad.. for what I want to do. I need close manifold solids. What do you guys think is a good affordable scanner that I could use to start scanning 3D objects and printing them with or without processing time? (i dont care, I will learn)

  4. #4
    Engineer-in-Training iDig3Dprinting's Avatar
    Join Date
    Jan 2015
    Location
    UK
    Posts
    287
    Follow iDig3Dprinting On Twitter Add iDig3Dprinting on Facebook Add iDig3Dprinting on Google+ Add iDig3Dprinting on Thingiverse
    We have a couple of 3D scanners ( see https://www.idig3dprinting.co.uk/pro...s/3d-scanners/ ), namely 3D systems Sense and iSense (designed to connect to your iPad) . These are both relatively affordable coming in at £336 for the sense and £435 for the iSense.

    As to the level of post-processing, that is going to depend on what you what to 3D print. The Sense, iSense and Fuel3D scanners are hand held so the amount of post processing that you do is going to be determined to some extent by your technique. You will have to practice your scanning procedure so that you collect the best images off the bat. The next thing you are going to want to think about is resolution. For instance the sense has a resolution of approximately 1mm at 0.5m distance from the object. Note the resolution of the HandyScan series is 0.05-0.1mm, its those sorts of specs which determine the price difference. You will need to check the level of precision you need. Lastly, 3D scanner technology is great for solid objects, anything else is going to require more work.

    Currently there is always going to be some processing involved, whether it is post processing for the model file or post processing of the actual printed object. But, obviously, if you have a contract to print lots of an item, you only need to get the model files up to scratch once as once these have been set up and proven they are then good to go over and over again. If it is a contract for a single or short run of objects then you will have to increase your rates accordingly to the proportionate amount of time for each object. The alternative is to outsource the 3D modelling work.

    If you want to investigate using 3D scanners as a means to increase your 3D printing service offerings then there is only one way for it, you will have to experiment. For experimentation I would not suggest spending £20,000 on a scanner. My advice would be to choose a contract where the object lends itself to 3D scanning and get an affordable 3d scanner and start playing. The process for using a £20,000 scanner and a £340 scanner are going to be, conceptually, roughly the same so once you have the experience and you have found that it does have potential as a service then you can think of investing in top end equipment.

  5. #5
    Banned
    Join Date
    Jan 2015
    Posts
    306
    thank you idig, very helpful info. So is the fuel3D scanner really limited for what I want to do?

    What If I want to scan an engine block and print it? I understand the basic limitations of all scanners such as not being able to scan the internal passages of such an item, but for this purpose, a metal block with surface intricacies and tight tolerances will a fuel3d work? I dont want to get a scanner and then I find out it is useless for my application.

  6. #6
    Staff Engineer
    Join Date
    Jan 2014
    Location
    Oakland, CA
    Posts
    935
    Even a much more expensive scanner than the Fuel3D is not going to be able to scan an engine block and give you a model that will produce a working engine. What they can do is simplify the process of reverse-engineering a mechanical part like that, versus having to measure each feature by hand and recreate it in a CAD program.

    If you bought a scanner capable of capturing features at the level of detail specified in your tolerances, you'd be able to recover the "design intent" of the part in question, and reconstruct a clean CAD model of the engine or whatever, using a reverse-engineering program like Geomagic Design X. We sell this program, along with the Capture scanner, which can resolve details down to .11mm (the Capture Mini resolves even finer, but in smaller patches).

    Even then, an engine block printed in metal probably would need some machining, at least in the cylinders, before it would actually work.

    Andrew Werby
    www.computersculpture.com

  7. #7
    Banned
    Join Date
    Jan 2015
    Posts
    306
    Quote Originally Posted by awerby View Post
    Even a much more expensive scanner than the Fuel3D is not going to be able to scan an engine block and give you a model that will produce a working engine. What they can do is simplify the process of reverse-engineering a mechanical part like that, versus having to measure each feature by hand and recreate it in a CAD program.

    If you bought a scanner capable of capturing features at the level of detail specified in your tolerances, you'd be able to recover the "design intent" of the part in question, and reconstruct a clean CAD model of the engine or whatever, using a reverse-engineering program like Geomagic Design X. We sell this program, along with the Capture scanner, which can resolve details down to .11mm (the Capture Mini resolves even finer, but in smaller patches).

    Even then, an engine block printed in metal probably would need some machining, at least in the cylinders, before it would actually work.

    Andrew Werby
    www.computersculpture.com
    As much as I would love this program. I am on a budget of about 2000-2500 for the scanning equipment and software. I need something to get started. Do you have any suggestions?

  8. #8
    Engineer-in-Training iDig3Dprinting's Avatar
    Join Date
    Jan 2015
    Location
    UK
    Posts
    287
    Follow iDig3Dprinting On Twitter Add iDig3Dprinting on Facebook Add iDig3Dprinting on Google+ Add iDig3Dprinting on Thingiverse
    I think if your plan is to scan an engine block and produce something sensible it might be a long shot. the Fule3D says it scans to about 0.35mm which is OK but again that is going to depend on the size of the object your scanning. The accuracy is going to also be dependent upon your scanning technique.

    If you want to produce a production level 3D model using 3D printing you are going to have 2 sources of error, 1. the scanning process and subsequent model creation 2. the 3D printer. My guess is that the 3D printing is going to result in quite a lot of error depending upon your 3D printer. The scanning is also going to produce a lot of error, or rather low resolution. If you want to 3D print an engine block I would suggest creating the model from scratch in a 3D modelling/CAD suite (e.g. Cubify Design ). The 3D scanner may help you capture the essential elements but you would need to be very precise with your post-processing.

    What are you trying to do exactly?

  9. #9
    Technologist
    Join Date
    Jan 2015
    Location
    Waterloo, ON, Canada
    Posts
    159
    Add truly_bent on Shapeways
    Coincidentally, Brian just posted a piece on the Occipital Skanect v1.8 hardware/software solution for 3D scanning. This might just fit the bill for our friend DBFIU. It looks like their Structure Sensor requires an iPad, but their software accepts input from other "3D aware" cameras as well.

    They maintain that their system isn't good for "complex structures", but i'm not sure that an engine block qualifies as complex.

  10. #10
    Engineer-in-Training iDig3Dprinting's Avatar
    Join Date
    Jan 2015
    Location
    UK
    Posts
    287
    Follow iDig3Dprinting On Twitter Add iDig3Dprinting on Facebook Add iDig3Dprinting on Google+ Add iDig3Dprinting on Thingiverse
    What sort of resolution would this give. It seems very similar in concept to the iSense, Cubify's ipad 3D scanner. I think all of these scanners are going to result in the same issues to do with can scanning your engine block reduce your model processing time sufficiently and with enough accuracy to justify the expense. The only way to find out is to try it out. Start small as a test program and then look to refine it further if you think you can get it too work.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •