Multi-pass

[Michael Rosiles]

Iam conducting a study using the stress analysis features in creo simulate but I can't explain the steps the process is taking. I am using the multi-pass adaptive convergence method using maximum von mises stress as my measurement with a targeted convergence of 1%. I set the minimum P-order to 1 and the maximum P-order to 9.

What I can not explain is why this method is using only 7 passes to calculate a solution. The solution is not converging until the 9th order bias functions so it is not stopping because it has converged.

I would expect the method to run a pass for each polynomial order and increase those elements that have not yet converged until the maximum p-order is reached or all elements have converged. The method appears to be skipping a couple of these passes for the 3rd and 5th polynomial order.

I have attached the run status summary file from my analysis. It can be seen that the second pass uses 2nd order functions and the third pass jumps to 4th order functions. the same jump can be seen between the fourth and fifth passes.

I tried setting the maximum p-order to 3 and the results returned a maximum p-order of 2. If I set the maximum p-order to 4 and the results returned a maximum p-order of 4 and included elements with a p-order of 3.

Can anyone can shed some light on why the multi-pass adaptive method is not running a pass for each p-order or why the solution returned (which had not converged) when I set the maximum p-order to 3 did not include 3rd order elements.

It is normal for a MPA to skip orders if it detects that the difference between runs is nowhere close enough for converging with the next order. I suspect the idea being to save you time in running the analysis by skipping orders that won't help much. I wouldn't be surprised if that is the case for the runs that didn't converge and never used the maximum order setting. If nearly every element and edge hasn't converged, there's almost no way on earth that the next run will converge within 1%. I don't know what the criteria is for Simulate to make these kinds of decisions but that is the reason behind only selecting the maximum and minimum orders. Let their algorythm do it's job to optimize the time spent on the analysis.

You might want to do a plot of the element orders in the model so you can see if there is something in the geometry that the automesher needs help with by manually increasing the number of elements in a particular area using AutoGEM controls, adjusting the geometry, tolerance, etc.

Depending on the geometry and discretization of your part, 1st order and perhaps 2nd order runs can be useless and even misleading...particularly for the p-method which uses less elements....unless, of course, that is your aim. The default minimum is 3rd order and I'm usually inclined to leave it there so that the shape functions can accurately depect the geometry.

Finally, are you only converging on the max vm stress measure? That could be misleading and dangerous, espcially if you run into a nasty singularity in another analysis...doesn't appear to be the case this time. Is there a particular reason you didn't want the default options of local displacement, strain energy, and RMS stress? Ultimately, you might want to include more measures for convergence. In the report, is looks like max plane stress values for convergence are high even if the values are low. Displacement in Y also didn't converge to your target of 1% (which might be too strict). Just my $0.02, hope that helps.

Convergence scheme for MPA is quite intricate. Iif there's no convergence for an element edge, order of the displacement field is increased by one. If there's no convergence for an element face, order is increased by two.

I'm not sure why you think drum scanners are more like flatbeds than a film scanner. Sounds like you are trying to justify purchasing a flatbed scanner. Be aware, Flatbed scanners from Epson are not very good, you can do web stuff and only very small prints. Even with those limitations you have to do a lot of aggressive sharping. So, in this regard they are pretty far from a drum scanner. An older used film scanner will produce superior results than any consumer flatbed. I would advise caution in believing bias reviews and marketing material on Epson scanners, I believed and was not happy with the results.

That said, mult-scanning has nothing to do with resolution. It is a way of dealing with noise in shadows, some advocate that it is only good for shadows on slides. The process is to take several samples at the same point, then average them to get the value for that pixel. Read up on noise sources in scanning and it will make some sense. Better scanners do multi sampling, which read the same location multiple times with out moving the sensor.

A drum scanner looks at a single point of the film as it rotates, while a flatbed has a CCD array that covers the entire width of the scan as the optics move down the length of the image area. The dynamic range of a photomultiplier cell (drum scanner) is two or three levels greater than that of a CCD. Furthermore, fluid mounting to the drum eliminates Newton's Rings, and nearly eliminates the effect of scratches and surface defects in the negative.

Multi-pass scanning reduces the random noise while scanning the densest parts of the image by average the results of several scans. Film scanners make consecutive passes at one line before progressing to the next line, whereas flatbed scanners start from the top at each pass, with much greater risk of misalignment between passes.

Although multi-pass scanning takes much longer than single-pass scanning, the file size is the same since only the average values are recorded. DMax is not changed, but the quality is improved somewhat, especially with transparencies. I find no significant improvement when scanning negative film, which has a lower DMax than transparencies.

Why do you say so? I have yet to touch one but what I've seen on the web (Imacon) seems like the surface where you put the film is going to be bended when inserted in the body, no flat at all. True, the film size they can scan may be bigger than just 24x36mm. I think it is useless to use a drum scanner at its highest resolution with a print paper, because print paper does not have that much.

Files don't get bigger for me with multi-pass, it just takes longer and the scanner noise is different. Add image depth? I can't say, but at least in theory it is no. Superior results? YEs, less errors. What I have noticed, is that if you have line separating areas of large contrast, the CA (chromatic aberation) between the two may be important, such as the line is not very fine, but it is doubled by a red and blue thicker line. When multi-passing this line can be reduced in thickness, giving better contrast. But given that the film itself is very good and has a good thin line there. It may be that in the reality that line is pure separation, but then there's film too, and lens.

My impression is that the Nikon/Minolta style "film only" scanners have more trouble with focus than do flatbeds...there have been posts about that with the latest 5000 and 9000 as well as the Minolta. These scanners cannot hold film flat, therefore focus logically has to be more active than on flatbeds.

John, Nikon film scanners focus during a preliminary pass, and not continously throughout the scan. It is true, that with 120 film especially, you have to use a glass holder to maintain critical flatness. 35mm film is stiffer, and bare film is held flat enough by the standard holders. Mounted slides sometimes exhibit excessive curvature, which can be partially overcome by changing the location of the focus point.

It's not that flatbed scanners don't need to focus, rather they have fixed-focus (with a few exceptions). The question is moot, because the resolution of inexpensive flatbed scanners is so poor compared to a Nikon film scanner, that focus is a minor issue.

John, it's not the scanners that have trouble holding film flat, it's the holders or slide mounts. You can use glass mounts or holders and have no significant flatness problems from that point on. But there will be dust issues then (unless ICE is used). Fluid mounting can also be used with many non-drum film scanners and that should solve both problems but leave you with cleaning work.

Focus is a minor problem for dedicated film scanners compared to differaction with a flatbed. Usually its very easy to pick a focus point in the normal work flow. I have a Nikon 9000 and a Minolta 5400, both are easy to set the focus point. Even a slightly soft scan from a film scanner is superior to a Epson scan. While you don't have to think much about focus with a flatbed, its because one it is fixed and two Epson relies on a large depth of field to try to get everything in focus around their focus point. The draw back to this approach is very significant diffraction, which is very noticeable. I'd rather have focus controls and no diffraction.

With a film scanner multi sampling can help in some situations, but not all. With Epsons, it can help and it can hurt the image. I've had more problems with registration of the different passes than I had success in the shadow areas, that's with a Epson 3200. That is the subsequent passes do not line up with each other, so the calculations are not comparing the same pixels from each pass.

Personally, I don't use it much at all. The Nikon 9000 is so good it does not need it, have not used it on the Minolta either. Yes, I did buy a glass film holder for MF. However, when I tried scanning MF on an Epson I just threw a way the results, could not even print at 3x.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

3a8082e126