Revoscan 5.2.4.872 concerns

Support & Feedback
1

I’ve had a pop 3 scanner for a whopping 8 days and have several question/concerns. I purchased this device to assist me with reverse engineering and I will admit that I have no previous experience with this technology.

I’ve tried scanning ~28 different objects with and without scanning spray in many cases. I’ve read through the forum to answer my question’s and I still have a few. I’ll refer to scans of the sample bust provided in the kit.

Post processing

  1. Fusion - when adjusting the point distance setting in the standard method the slider allows a setting down to 0.05mm, the advanced method bottoms out at 0.2mm. this seems backwards to me but that’s not my concern. What I’m concerned about is that when fusion with the standard method my final mesh, regardless of what point distance I select, has the sandpaper appearance described in much of the forum content. While the advanced method is much cleaner. I’m not yet qualified to judge the quality of either meshes at this point. But I would like you to evaluate these results and advise me if these results are typical. See images attached.

    AdvancedFusionMethod
    AdvancedFusionMethod1920×1032 252 KB

    standardFusionMethod
    standardFusionMethod1920×1032 108 KB

  2. When opening previous scans, a one-click edit function panel is placed in the upper-right-top corner, when executing this feature on a previously saved scan it always increases the point distance to a higher value than I used during the initial scan. I’m interested in why this function is highlighted in this manner when opening a old scan, and why the increased point distance seems typical. Is it that common for someone to reprocess an old scan? I suppose it depends on what type of objects the user typically scans. But it concerns me that both the grainy appearance of meshes when using any standard method point fusing, and the way the one-click edit function is highlighted when opening an older file, that maybe the the software, and maybe the scanner have limitations that are not obvious in your product descriptions.

In my attempts to learn the software and proper scanning techniques, I’ve tried many different approaches, found in the forum, to get consistent results. I just don’t have confidence in this device in it’s current state.
In the forum, one possible cause of the sandpaper appearance is scanner calibration. When attempting to run the calibration software, the program would sometime (but not always) perform a calibration check, which in my case would always fail, and then when attempting the calibration process, it would end by displaying an error which read “calibration failed, repeat”, or “a data transfer error”. It took more than 30 attempts to resolve that issue, which I believe was a USB power issue when my very expensive laptop was operating on its battery. With the laptops power supply connected, I successfully recalibrated the scanner last night, and the images I’ve attached were post calibration. I saw not difference in my scan quality from before the calibration, but I now have another concern, in that throughout the forum everyone is comparing accuracy check values against the 0.05mm standard, with this version of the software the calibration check does not report an accuracy value. This also make me question why that value is no longer reported.

I again admit that I am inexperience with this technology, and I really want to make this device work for me in my reverse engineering projects, but I’m struggling with these potential quality issues. Can you help me with this.

2

I would always use Advance fusion as it uses various algorithms to smooth the point cloud while retaining detail. Standard fusion is more of a what-you-see-is-what-you-get fusion, errors and all. This results in the sandpaper effect. It is useful however if you intend to do post processing outside of Revoscan. I don’t use it at all for any of my scans.

The one-click edit thing is a software idiosyncracy. Revopoint are trying to make the software as easy to use as it can be for beginners and so they highlight the functions that will assist with this. One-click-edit I would say is analagous to doing auto-levels on a photo in photoshop. It will get you some of the way there but is not a replacement for doing it manually.

Your advnaced fusion scan of the bust looks OK by the way. You are missing a section at the back but I’m guessing you missed it when rotating the object.

Fusion at very low point distances isn’t always necessary. It’s a good idea to experiment with fusing at different levels and seeing the effect on the scan. You will find that sometimes 0.3 or even 0.4mm is good enough for what you want, especially for reverse engineering where it’s all about primitives and planes.

1 Like
3

Hi @lcemoore

As @Rilot stated already switch to Advanced mode if you are lacking if skills to edit raw scan Data outside Revo Scan , Standard fusion is to preserve all the data including noises and overlapped points .

Go for Advanced mode , don’t push the fusion slider always to maximum , not all objects require that density or resolution especially objects that you want to reverse engineering.

For example a mug will be fine fused at 0.97 mm
It will not made it less accurate .
Fusing have nothing to do with accuracy , it is a resolution , and if there are not enough original details you adding additional data that you don’t need resulting in sand paper effect .

Usually 50% of the scanned data will be removed from the final scan as it is not needed .
The accuracy is captured while scanning , you can’t improve accuracy by sliding the Fuse slider to higher level since it is just resolution .

You will be very happy with Advanced mode fusing , as it is the best choice for many cases , and the advanced algorithms will do the work for you, so you don’t have to.

Infrared Structured 3D scanners do produce some noises level due to micro reflection and scattering, prepare your surfaces well to avoid it .

4

Thanks to both of you for your quick responses. I’m generally OK with your answers. I do understand the difference between accuracy, precision, and resolution. To honest in my application where the final product will be 3d printed, my printers are definitely not capable of reproducing a mesh to make details that fine possible. But I had other reasons to ask these questions, mostly to qualm my concerns that i didn’t get a compromised unit. Practice does make perfect, as long as my device remains stable.

Let me explain.

When I first unpackaged the scanner and attempted to mount the power pack to the scanners threaded connector, the threaded insert on the top of the power pack fell off. So within the first 15 minutes, there was an quality control issue. Revopoint was quick to respond and a replacement power pack was on the way within 3 days. Then to learn about the device, I followed the procedures demonstrated in the training videos from Revopoint, but I got significantly different results, including a fairly grainy surface contour. I have controlled lighting sources and all my scans have been with the turntable and tripod. To investigateI tried to find a user guide for the Revoscan software, but if it exists, I can’t find it. In fact I couldn’t find a user guide for the Pop 3 either, there is a quick start guide but it’s fairly sparse on specifics.

I watched every tutorial I could find about using the scanner and the Revoscan software, but most of that was for a previous versions of the software. Many of your forum responses related grainy surfaces to calibration issues, so I pursued that path, only to be frustrated for 3 days and many attempts. Even though I finally got through the calibration, I’m still not sure about the the results because the software no longer reports any metrics for the calibration. I read through all the log files and the calibration logs have many entries that state left camera, right camera and/or RGB camera errors, but those errors could have been because I wasn’t doing it right. I just don’t know.

I guess what I’m looking for is some assurance that the scanner is working properly. I agree that advanced fusion scans look pretty good, but the other challenges I’ve had makes me unsure.

5

Hi @lcemoore

I explained to you already , I checked your scan results and that has nothing to do with your device , it is normal to have sand paper results when using standard fusing mode, and there is nothing wrong with your device , all Structured light 3D scanner suffer from a sand paper effect becouse the scanner use so many frames while scanning overlapping everything . Normally you need just 12 frames to build a model , yet structured 3d scanner needs at least 350 frames to track the scanning object , if not cleaned or used in Advanced mode you have to do the cleaning yourself before meshing the fused point cloud .
And that is all secret about … If you had calibration issue , your model would looks like having chicken pox , not sand paper what is result from overlapped point and too high fusing settings ( resolution )

Another method is to simplify the mesh after meshing to 50% , it will keep the model smoother , and remove the sandy surface by reducing the resolution but keeping the details intact .

that’s. all I can help you here , the other issues you have are related to the product itself so not much we can help you out here , you can always contact customer@revoscan3d.com for any issues or feedback or concern .

And please don’t get crazy with the forum posts, most of them was for older software and devices and not much of them in regards of POP3 .

you don’t have to worry about it , the calibration adjusting digitally the positions of the sensors with each other as there are no motors to physically adjusting the small deviations ( errors) of the calibration failed , you would be not able to update the new calibration to your device .
and trust me you will not get rid of the sand paper effect as that has nothing to do with calibration of your device , not what you showed us in the screenshot .

I have all Revopoint’s scanners and more , and all of them showing the same results , MINI is the only one that produces better surface becouse it uses Blue laser light in place of Infrared , infrared sadly cant produce better results at that resolution or accuracy . For that reason you have Advanced fusing mode that removes all the artifacts causes by Infrared light scanning in a one simple click .

Pay attention to the automatic settings , it will tell you what level of details the scanned object should be fused or meshed with according to the level of details . Don’t push everything to maximum as that is not how 3D scanning works . Sometimes less is more

6

i’ve this issue also, advanced mode always smooth but less detail, and basic mode with more detail but sandpaper like surface, it’s confuse to choose

7

That’s only an illusion due to the software shader , no less details trust me … objects with sandy surfaces appearing sharper , visual illusion .
Go to my MINI showcase and see my coins scanned and fused in advanced mode … with 10 Microns visible details yet fused at 80 microns pitch point distance .

1 Like
8

thanks, just went to the postof the ring/ coin you scan in the mini showcase page, did you spray for ring/coin for scanning :o so detail, the screw also

9

No I used my brand 3D scanning liquid at 3 microns , it is not released yet . 3D spray is too heavy for that small details , most details are around 10 Microns on a coin , now you spray 15 microns layer and you missing half of the details .

What you can do at home is use Zinc Oxide with alcohol, it will give you also very thin layer , better than 3D spray and cheapest .

MINI have 20 microns accuracy , still can capture around 7-10 microns at the distance of 9-10 cm … for that reason you don’t want to kill the details with 3D spray .
For the screw I used just Attblime 3D spray and slightly frosted it , I scanned it on live Revopoint show last year .

But the last coin scan is biggest success so far with MINI .

1 Like
10

:o 3microns, more detail than mini, mini2 ? :smiley: thanks for the tip with zine oxide :slight_smile:

1 Like
11

When set proper MINI can capture at higher accuracy than 0.02mm as my scans conformed it . But not bellow 7-10 microns 0.007 to 0.010mm at 9-10cm distance