Claes Lundstrom

Flickering cursors have been known to be clearly caused by Photoshop being active in the background, for some unknown reason. I have seen it in several apps using user interactive cursors (such as VW). It sometimes helps to turn off PS or, in worst cases, to restart the computer. It's extremely annoying.

My now slightly old Creatility 3 Pro costed about 230 $ if I remember it correctly. Works very well (much better than expected) and bigger than 8" in all directions. There is however a new generation out now that is much faster (claiming up to ten times). The Anycubic Kobra 2 is an example and I have seen ads for it at $479, and with a volume of 420 x 420 x 500 mm. Creatility has similar models, and it seems as the competition is strong in this segment.

I'm into my third printer now. First was purchased in the spring of 2011, and it was pretty useless. Nowadays, even cheap China machines are pretty good and they cost like an office printer. It's not difficult either generating models from VW, though you need a fundamental understanding how to think when preparing models. Some skills in solid modeling helps too.

A few suggestions regarding Ladar scanning: Stitching definitely helps. Practice scanning. It's a craft and you soon learn that slow gentle movements helps. Plan your scans so that you move the camera as little as possible. Avoid thin object like for example shrubs and bicycles, which seldom works well. Avoid shiny objects such as glass and cars, as reflections never works well (spray it with something dull if you have to). Dump scans where you get weird offsets. Better to try again. At best, you can expect an accuracy of about 1%, so don't expect miracles. You can use it with DTM with decent results, but I suggest that you clean away obvious errors and excessive measurements. Each point may be less accurate than traditional measuring, but on the other hand you don't get thousands of measurements from traditional measuring. I have found that using a combination of traditional measuring and scans to be quite useful.

I double-checked it with TouchCAD and it is indeed incorrect. I suspect that what happens is that the unfolder does not place the folding lines correctly, generating a slight compound shape where the shape is not really a compound shape at all. The red lines are my reference unfolds. wrong arch 2.vwx

Knowing how to model from scratch in not an easer option, but knowing how to means that you can take things far beyond the limitations of the existing tools. I have seen it so many times, where users avoid certain things that should have been there simply because they don't know how to do it. Knowing how to model anything from scratch therefore means that you have effectively reached a much higher level of know-how.

Sounds like 3D Studio Max has work to do to catch up. Surprisingly primitive if you ask me. Admittedly, VW could do with quite a lot of improvements as well in the fields of mesh and NURBS surface editing.

Try grouping the mesh. At least you can scale if from above. Otherwise, try the menu command Modify-> Scale Objects, where you can scale in all directions, though only by a scale factor. You can overcome this by using the building in math in the edit field. Let's say that you have a 4 meter object that you want to be sim meters. Then type in 6/4 and there you have the scale factor.

Don't forget to convert the relevant to meshes if it's not a mesh. Otherwise mesh smoothing doesn't work.

I agree with Jeff. I never have any problems importing obj files from the source I use. Absolutely rock solid.

This is my version. Sail_roof.mp4

Yes you need a solid table, that's for sure. Just shifted from a Canon TM200 to a HP. Way better and reliable. I hated the Canon from day one over the three years I have had it. Ridiculously fragile, over engineered, very expensive print head, and the worst drama queen ever, though printed well when being in a good mood and a lot of persuasion

Yes good illustration what I meant. 3D printing requires a special mindset designing for 3D printing, where you have to extract a way to work and how to split the model into parts, and finding a good orientation for the print. You just need to practice. I typically glue the parts together, using a glue that can handle the slightly non sticky surface of thes plastics. On very thin skins, like the boat hull above having a thickness of just 1.2 mm, I typically add a small flange along the joint, and of course on the inside, looking a bit like a frame or bulkhead. This adds surface area to the glue joint, but also adds a bigger footprint towards the printer bed, which is useful as it is a skinny shape relative to the mass that moves around above, and therefore risks falling over when the bed moves.

There are a few basic rules that are useful to follow, assuming that you are using an extruder style printer with a single extruder so that you can't add supporting material that you remove in a step 2: 1/ Make sure that you have enough thickness. On my fairly basic printer, I have found that 1.2 mm always works. In my boat example, the contractor wanted 5-6 mm on the full scale model (about ten meters long). When I scaled it down to fit my printer having a build height of 250 mm, and printed in two parts giving a model height of about 500 mm, the material thickness fell below 1.2 mm, so I had to produce another model with thicker walls relative to the full scale model. 2/ For single extruder printing, the rule is to avoid angles exceeding 45 degrees. Imagine building something out of dry sand. If the angle is too steep, it will simply collapse. On my boat example, I had to find a setup that avoided thes 45 degrees. Standing on the transom would not work as it leans outwards 12-14 degrees, which would make the model fall over. Splitting it along the centerline and printing it in two pieces lying on the side would not work as the sides would exceed 45 degrees by a wide margin. The solution was to split the model in the middle. Each part could then stand up without any problem, and I could double the size of the model. The transom and two plate was however a problem, so I printed them as separate parts with joints at an angle to to hide it as much as possible. On a house, you can think it a similar way. How can I for example print a roof having a flatter roof angle than 45 degrees? You simply print it as a separate piece standing up on the side, or for more complex shapes, you can ask yourself if it's possible to split it in the same way as I did with the boat hull, so that you manage the angles and can maintain a shape that stands without falling over. Windows and doors can of course be a problem if you want to print it with see through properties, but you can print a wall lying down instead ? 3/ How can I check if a model is a true solid model, and would therefore cause a minimum of problems? The basic criteria simply means that the part or parts all consist of closed shapes. If you still have a problem, you can check if the normals all face outwards. A quick test is to convert a given shape toa mesh and then calculate the volume of it. If you do get a volume, the model is most likely correct. If not check closed shapes and volumes again until it works. 3/ You can of course print a house model as a solid object, having a mesh inside but it does have the disadvantages of just being an exterior and it would also take longer to print. It would therefore probably make sense on very small models representing volumes instead of being models looking like real houses. Below is an example of a solid print, that is, it's filled inside with a mesh structure.

It was a combination of drawings, laser measurements, and scans, where I tried to merge them into one unit. The house was built in 1939, where architects did not have VW to work with, and nothing was done in 3D, so even the drawings had to be cross checked against one another. The house also gradually changed over the years. As you can see there are a lot of factors that had to be merged into one 3D model. I have to say that I have made good use of these scans since my posting. I used it for locating sewage pipes in the house as it is time to reline them in the basement. For some reason we could not find the original drawings, but using combinations of scans from all seven stories, I was able to generate a good picture where everything was. Another example was a leak in a terrace, where a joint turned out to be the problem. I could place a scan of the terrace on top of the plan drawing below and it became obvious what the problem was. I even double-checked the measurement with a Leica laser, and the scan was 14 mm off vs the laser scan over 6.225 meters.