How to prepare an object for scanning

[Bongobat]

Hi Pierre,

I think that the solution would reside in the possibility of making the stereotypes of texture after the totality of the scans, but that supposes to precisely find all the positions in which summers have made the scans.
Does somebody consider another possibility?

Yes, you can do this. You simply need to use a stepper controlled turntable. You would need to make sure you do not move the object in relation to its placement on the turntable when you clean off the soluion though. You can also scan all the angles in the dark and then turn on the lights and take all the texture shots. I have experiment a little with that.

Greg

[WalterMo]

Hi,
Or what do you think to do it vice versa as Greg suggested?
Put the object on a turntable and do the texture shots. Then spray it, e.g. with chalk while it is still on the turntable and then scan it.
OK, this shouldn't be done in the living room and an old bedsheet should protect the environment.

Walter

[hal]

Hello to everybody,

I've done some tests with the withing spray from David Shop and my own mix of tempera+water+soap.

For little object (or when you need to scan closely your details, is better the tempera mix, because the coating is very thin and smooth. This save all the little details and gets a good refraction of the laser line.
But is hard to apply it with an brush. Better is an airbrush. I know that an airbrush and an air compressor isn't so easy to manage as a spray bottle.
A trick is to use an airbrush, put a first layer of color, to dry it with an hairdryer and continue with the second layer of color, and so on. This avoid the creation of little drops of color on the surface. Remeber to be generous with soap: with few soap you can have detachments during the fast drying with hairdryer.

The spray bottle is handy and easy to use, but (for my opinion) not the best for little object or small details. The problem is the little roughness that the clark spray create on surface. This spray is washable, but need more care to be removed, especially into rough surfaces (it's a little persistent). If you can't rub hardly the surface, try with a cleaning gel.

Hope that this can help someone,
Mattia

[Normand Blais]

Great Ironman suit Hal , Maybe this stuff could work to prep the parts? http://marketplace.dentalproductsreport ... sp?id=2793

[hal]

Hi Normand,

the spray that you suggest seems very professional, but not washable, right?
I'm searching the best solution for coating, save details and save the scanned objects.
Many others acrylic spray that I've found on market do a good work of coating, but unfortunately not washable on platic surfaces.
However thanks for the advice, I will keep in mind.

Best greetings,
Mattia

[ArchaeoPhoto]

Hello,

there are some posts on coating shiny objects, for example glass or metal.
Let's say there's a need to scan a polished metal surface.
The first step obviously is to coat it. How to achieve this is described in various threads in this forum.

Now the question is (as Santo has also asked): If a coated object is scanned, how to get a texture shot?
One doesn't want to have a texture created from the coated material but from its natural appearance.
But if I have to pick the object up, clean it from the coating, put it back, it's not in the same place as it was before.

What would be the workflow to get good quality textures fused with a scan of a coated object (consisting of n individual scans)?

In my case, I would scan prehistoric (cleaned and polished) metal finds. It will be difficult to convince any art conservator or "Restaurator" to spray or apply some fluid on archaeological finds. That's one thing, but if one tells them "Ok now I have to wash it off" they will be shocked (and likely throw you out )
Right Mattia? I will try the Cyclododecan Spray.
So, what's the approach to this. There are some suggestions but do normal and coated scans allign without problem?
I think Walter's idea is the best so far
Hope there's more help.

Thorsten

[hal]

Hi Thorsten,

sure, the Walter's idea is the best, if you want to take texture with the 3D scanner. Ok, cloud of points + texture from same instrument is the best choise if you aim to obtain a very precise survey of the real object.
But sometimes, in Art field, we need only a "rapresentation" of an object (as documentation and/or base for graphic tables on which mapping pathology and infos about the conservation), and an accurate survey of every shape/color informations are "excessive".

The only important thing is: your final purpose.

If you need a 3D model for a museum's website (as example) or a 3D tool for some presentation, the 3D mesh and a texture taken by a PhotoCamera are enough (workflow: 3D scan - retopology - UV mapping - Texture-match from photocamera shoots). In this case you can coat your object (if possible).
If you need an accurate 3D model and you must measure in its all little details ar use the scan as a scientific tool, you need a system that acquire the cloud of points and texture in same time, with the max. definition possible. In this case coating isn't possible. The only way is to use a 3D scanner that can do easily the job, or, with David, using a more powerfull laser light.

Now, if you have a camera that work at 1024x768 or less, the texure from scanner has a low definition. In this case is the best choise to acquire texture from scanner? The answer is "yes" only if you aren't able or don't want to do the redrawing of the mesh topology and create an UV map (with other 3D software).

I well know the problem of coating with pieces of art, ancient or modern. Obviously the coating is impossible.
But if you use the Cyclododecan (Liquid or Spray) you can jump this problem, because Cyclododecan is a temporary fixative used in art Conservation (especially on archaeological finds).
This product must be tested, first of all, on a limited and "not so visible" area of the object. If Cyclododecan don't change the color/apparence of the surface (if is neutral), you can use it all over the object's surface.
An other little inconvenience could be the Cyclododecan surface: it isn't smooth but have a little roughness (like with very little hair). This could generate a non-regular and non-smooth surface on the final mesh (sparke effect with a laser light).

So, my conclusions is: take in mind your final goal and be sure if 3D points + texure taken with the scanner is the best choise.
Remeber one last important thing: the texture taken from scanner lock its details with the points cloud details. If you have a lot of 3D points, you have a lot of color informations, but if you acquire a small number of 3D points, your texture will be poor.

Best,
Mattia

[no respect]

Hello everyone,

Just wanted to add my two cents. I would try Partall PVA http://compositeenvisions.com/gelcoats-and-mold-releases-126/ 1 mix it down with water 2 spray on (air brush is fine) 3 wash off with water. it leaves a very thin coat preserving all detail or little detail depending on how many coats you decide to spray. Oh, and it dries fast.

Eric

[WalterMo]

Another nice spray to prepare surfaces is "OPN Entwickler Weiss". Normally it is used to detect cracks on metallic items. It is produced here:
http://www.opn-chemie.com/index.php

The matt white coating will be visible some minutes after spraying.
It is easily washable with water. The problem is to get it. Up to now I haven't found a dealer who likes to sell less than a parcel with 12 bottles à 500ml. .

Walter

Edit: In the meantime you can get this spray in the DAVID Shop:
https://ssl.david-vision-systems.de/sho ... e248cf972b

[vw_bus]

In the UK the white powder spray for dye penetrant crack testing is not difficult to get. I have some which I bought for testing welds for porosity or cracks, so it will be the first thing I try when I finish collecting parts for my planeless scanner. The dye penetration crack testing system uses three products, all in aerosol cans - a cleaner, a dye, and a developer. The developer is the white 'chalk like' powder.
'Rocal' is one of the more common brands (Rocol 63135 Flawfinder Developer Spray ), but there are other cheaper ones too.

http://www.rocol.com/corp/index.php?opt ... 85&lang=en

This page has links their distributors:

http://www.rocol.com/corp/index.php?opt ... 14&lang=en

Welding suppliers also often stock it. The can I have is called 9D1B, made by a company called Ardrox, and was a lot cheaper than Rocol (about £10, I think):

http://www.aerospace.chemetall.com/inde ... abc&id=172

It came from an online welding supplies company, but I can't remember who. I hope that helps.

[hal]

Hello,

I want to share my latest scan of an object, completely coated with Cyclododecane Spray. It is a mobile phone (old model, but still in use - first and after the scan). Time ago I've read about some problems to coat delicated object, so, I've tried this auto-removable spray over a cell-phone, in the external and inner parts (on electronic components, too). After that the coating is totally vanished, I've put the battery in the phone and all work perfectly, as ever
Here you can see the scan result. The rendering were done with KeyShot 2 real time renderer, with the trial mode.

In the left side of the phone, over and near the two arrows (small up/down triangular buttons, visible in the second image), you can see some problems: the surface is dirty and rough. This is my mistake: I've sprayed too much Cyclododecane over this area, and the thin layer of the coating became a really big mess. I've leave this problem to alert you. Don't do my same error. Try to use Cyclododecane with parsimony and softness.

To be continued...

[hal]

... to be continued ...

Here you can see the time progression of the Cyclododecane Spray sublimation (vanish by itself).

Keep in mind one important thing: to sublimate correctly and quickly, you need to put your object when the air flow (not close), because the sublimation process need air and good temperatures. In the summer the process happen quickly than in the winter.
If you leave the coated object placed over a support/plane (like a table) the part that are covered and where the air can't flow freely, need more time to react. Same long time are required for deeper parts, furrows and undercut areas.

But in not more than 3 days, all are vanished.

Hope this helps,
Mattia

p.s.: soon other tests. Cyclododecane for coating an ancient World War handgun (metal and wood) and for a tissue/cloth.

[gozos]

Here a few links for white coating which can be use for scanning. I have not test them but they looks good to me and washable...

http://www.plumbersurplus.com/Prod/Krylon-5894-Marking-Chalk-White-%286-Pack%29/163807/Cat/1436?SSAID=198698
http://www.amazon.com/dp/B00455BQB8/ref=asc_df_B00455BQB81576481?tag=thefind0173179-20&creative=395261&creativeASIN=B00455BQB8&linkCode=asn
http://www.skygeek.com/ats-skd-s2.html

And cheaper:
http://be01.rs-online.com/web/search/searchBrowseAction.html?method=searchProducts&searchTerm=495-4995&x=0&y=0

[mipu]

hi, don´t know if someone posted this already: i use this for coating: http://www.conrad.de/ce/de/product/812877/RISSPRUeFMITTEL-500-ML-CRC-CRICK-130/SHOP_AREA_17607&promotionareaSearchDetail=005


called rissprüfmittel CRC Crick 130 (just in case the link does not work)

[Francois]

3 more product references that can be used. And they are cheap ...
- SKINCRIC R 764/S1 from ABAX Industries ( 6,60 € with VAT )
- KF Crick 130 ( 10,35 € with VAT )
- ARDROX NQ1 (replacing the 9D1B mentionned above - 10,35 € with VAT )