Wikidata talk:WikiProject Materials/Properties
How to list and qualify the significant characteristics of a material ?
[edit]Materials have a lot of different properties that can have a high impact on the final product and there is a lot of materials available. Many industrial fails are due to bad material selections. I.e. :
- Selecting classical stainless steels like 316L for under-sea mechanisms is only OK if the part is attached to a metallic structure providing electrons. Most of mechanical parts are not in ideal conditions : water stagnation aeras, differential aeration, random passivation weaknesses, weak polishing, sulfure inclusions, porosities in casted parts, stress, fatigue... Then, materials with higher corrosion resistance or additional protection are needed. Not knowing this rule has lead to catastrophic fails in the development of green offshore energies...
- Commercials often try to sell high cost and trendy materials such as Polyamid 12. Did they tell you that it's mechanical characteristics are divided by 2 at 80°C ? Other Polyamids resist way better to high temperatures. Did you know that a black part under the sun can reach 80°C ?
- Many of sport goods are made in Polyamid 6/66. But beware of water absorption. It has a high impact on dimensional stability.
- ...
It is important that engineers have a quick view on materials pros and cons before refining the search entering in data values details.
As an example, UNS S32205 / EN 1.4462 Duplex stainless steel (Q7388539), the most common duplex stainless steel, is generally used because :
- It has a better corrosion resistance than most of other stainless steels at the same level of price.
- It has a better tensile strengh than most of other stainless steels at the same level of price.
- It has a lower price than most of other stainless steels that are usable under the sea.
However it is often not used because :
- It's maximum service temperature is lower than most of other highly corrosion resistant stainless steels.
- Most of standard elements are not easily available in duplex.
- It was difficult to source in the past and, then, is not very well known.
All these assertions are easy to verify comparing numeric data of main Stainless Steels. However, having it written in such a short summary is very helpfull to let the engineer quickly know that he is, or not, searching in the right place. -- Thibdx (talk) 21:09, 30 July 2018 (UTC)
Before : defining the scope
[edit]Let's suppose that the significant property is the corrosion resistance of our duplex stainless steel.
It is needed to define a reference set items. Duplex's corrosion resistance is an upgrade comparing economic stainless steels but not considering Inconels.
- of (P642) works with some grammatical forms
- relative to (P2210)
- comparing → class of items
- among → class of items
- scope → class of items
- considering → class of items
Create dedicated properties ?
[edit]Wikidata:Property proposal/key strength/Wikidata:Property proposal/Weaknessthere is a risk of seeing these properties miss-used. i.e., for people.- beneficial (material) property / prejudicial (material) property
- optimized value / degraded value
- often used for its / often not used due to its
- often selected for its / often not selected due to its
- has high / has low
- high value / low value
- is high comparing / is low comparing (used as qualifier)
- commercial argument : factual and easy to source. What would be the opposite property ?
Use existing properties ?
[edit]Ex : UNS S32205 / EN 1.4462 Duplex stainless steel (Q7388539)
- has characteristic (P1552) → corrosion resistance (Q55830125)
- Qualifier : criterion used (P1013) → engineering optimization (Q5377895) (could also be average (Q202785), degradation, downgrade, upgrade...)
- Qualifier : of (P642) → stainless steel (Q172587)
- Support Seems to be the most compliant with existing system. We have to ensure that the wording is easy to understand in good english. --Thibdx (talk) 21:09, 30 July 2018 (UTC)
Comments on ASSUMPTION ABOUT ANY DESIGN ENGINEER WHO IS NOT WORKING WITH A MATERIALS ENGINEER
[edit]- Comment A design engineer has many resources available to them, but they are "designing" and not concerned with "materials science" which is best left to a "materials scientist or engineer". This Proposal assumes many wrongful things such as saying "is a pain", when most of the world of engineers know that "the pain" is simply reduced by working with others that are experts in their field. You do not have to know everything as a design engineer, although it is helpful to have some "materials science" in your background, no one is expected to have encyclopedic knowledge and so often you should consult with a "Materials Engineer" or "Materials Scientist", or "Metallurgist" or "Research scientist" or "Aerospace Materials Engineer" while designing for an environment or against a set of working conditions. Thadguidry (talk) 21:31, 30 July 2018 (UTC)
- You are right that it is a bit caricatural. I tried to make it understandable by people who are not in the mechanical engineering field not entering in the details. Most design enginner are working in team, including a material engineer. At least a subcontracor's one. Actually some businesses are so complex that even working with a material engineer is not ensuring a right material selection. Each example above is a real story from big firms where the error has been made by design + material engineers who were not expert enough. However the debate there is not about the right engineering process but Wikidata structure. I remove some of the storytelling to avoid such considerations in the discussion. -- Thibdx (talk) 21:47, 30 July 2018 (UTC)
- In the first example, design engineers from the green energy startups went in contact with shipyards materials engineers. What they didn't know is that the use case on big steel boats is very different because the structure is providing electrons, avoiding corrosion. So that the "experts" material engineer were actually not expert enough for them specific usecase. Millions of dollars were lost...
- In an other company, having access to several internal material engineers, the material manufacturers' ones who are expert of this type of material, the process engineer... Selecting a material for a very innovative product is not a straightforward process. Data is not easy to find about some specific and new situations. The best is to find a material enginner with a very very narrow and deep expertise. For example someone who dedicated his entire life on researchs on the corrosion of stainless steels is an invaluable value. And even with best guy, you have to deal with communication within the team. Having this type of data summary on materials datasheets clearly helps to the whole team to raise them level of information and think about the right things in early steps. -- Thibdx (talk) 23:09, 30 July 2018 (UTC)
- You are right that it is a bit caricatural. I tried to make it understandable by people who are not in the mechanical engineering field not entering in the details. Most design enginner are working in team, including a material engineer. At least a subcontracor's one. Actually some businesses are so complex that even working with a material engineer is not ensuring a right material selection. Each example above is a real story from big firms where the error has been made by design + material engineers who were not expert enough. However the debate there is not about the right engineering process but Wikidata structure. I remove some of the storytelling to avoid such considerations in the discussion. -- Thibdx (talk) 21:47, 30 July 2018 (UTC)
How to list material designations according to different standards ?
[edit]Closed : Proposal 2 win.
Most materials, like SAE 316L stainless steel (Q28453931), have several names depending on the country standard used. Most of standards are in line with others, even if there is some details that can be different.
Here are some of equivalent :
- ASTM 316L
- UNS S31603
- EN number 1.4404
- EN name X5CrNiMo17-12-2
- ...
Despite these different standards, 316L is named 316L quite everywhere in the world. Well... except when used for screws. It is then refered as A4 grade.
A contrario, Martensitic Stainless Steel EN X46Cr13 is named :
- 420C on English forums according to AISI
- Z40C13 at Thiers, according to old AFNOR
- X46Cr13 for most tech guys in Europe...
AISI is not always the referent norm. Aluminium 6082 is named with the new NFA norm.
Feel free to comment or add proposals here under.
Proposal 1 : a page per name linked by a property "is equivalent"
[edit]Each normalised name of the material have it's own page ; and has a proprety "is equivalent" to list the other pages refering to the same material in different norms.
Ex :
- 316L (Qxxx) is equivalent to 1.4404 (Qxxx)
- 1.4404 (Qxxx) is equivalent to 316L (Qxxx)
Discussion :
- Oppose In my opinion most of us will not be willing to create a page for each normalised name of the material on Wikidata + Wikipedias. This can be used if norms have noticeable differences. --Thibdx (talk) 20:01, 20 July 2018 (UTC)
- Oppose, for me, it's not even a question of will (it could be automated) but a question of what would be in this items. I see nothing specific to a name like « 316L » that requires an item. Cheers, VIGNERON (talk) 09:51, 21 July 2018 (UTC)
Proposal 2 : name (P2561) with qualifier named by (P3938)
[edit]Ex :
- name (P2561) : 316L named by (P3938) AISI
- name (P2561) : 1.4404 named by (P3938) EN
Discussion :
- Support Seems the most straightforward and complying with intuitive editing. --Thibdx (talk) 21:38, 20 July 2018 (UTC)
- Support simple and consistent with other items already using this structure. Cdlt, VIGNERON (talk) 09:53, 21 July 2018 (UTC)
- Support This seems straightforward to me. - PKM (talk) 19:13, 21 July 2018 (UTC)
- Support Djhé (talk) 12:25, 23 July 2018 (UTC)
- Comment In other fields, I tend to use the qualifier "criterion used" as the name might not originally be created by the organization and/or standard being followed.
--- Jura 21:25, 30 July 2018 (UTC)
Proposal 3 : complies with (P5009) with qualifier name (P2561)
[edit]Ex :
- complies with (P5009) AISI name (P2561) 316L
- complies with (P5009) EN name (P2561) 1.4404
Discussion :
- Support Seems to be quite logical, with existing propreties in my opinion. I'm wondering about usability with infoboxes since I'm not familiar with it ? --Thibdx (talk) 20:02, 20 July 2018 (UTC)
- Comment why not but isn't this redundant with the proposal 2? Do we want or need both? VIGNERON (talk) 09:53, 21 July 2018 (UTC)
- Of course, not both. The idea is to decide which one is the best. -- Thibdx (talk) 11:10, 21 July 2018 (UTC)
- Comment I'm not knowledgeable about the field. If a standard defines the characteristics of the material, I'd use P5009 to refer to this. This would probably be preferable to a statement that merely defines its name.
--- Jura 21:40, 30 July 2018 (UTC)
Proposal 4 : one property per norm
[edit]- UNS designation
- ASTM designation
- Numerical EN designation
- Symbolic EN designation
- ... (~15 properties)
Ex :
- AISI designation (Pxxx) : 316L
- EN designation (Pxxx) : 1.4404
Discussion :
- Comment I don't see why we would want a specific property for each standard (and there is way more than 15 standard). Cheers, VIGNERON (talk) 09:54, 21 July 2018 (UTC)
Proposal 5 : create a new Standard property with a mandatory according to qualifier
[edit]Ex :
- Standard : 316L according to AISI
- Standard : 1.4404 according to EN
Discussion :
- Support Seems the most compliant with uses in the material science field. --Thibdx (talk) 21:38, 20 July 2018 (UTC)
- Support again: why not but isn't this redundant with the proposal 2? Cdlt, VIGNERON (talk) 09:55, 21 July 2018 (UTC)