Talk:Mechanical filter/GA1

GA Review

Article (edit | visual edit | history) · Article talk (edit | history) · Watch

Reviewer: Materialscientist (talk) 10:32, 25 December 2009 (UTC) This is a relatively high-quality article to which I have only few comments:[reply]

Extended content

I made a number of minor style changes, please check. Many are optional.
In the first part (e.g. Fig. 2) can the resonators solid, hollow or either (sure, solid are practical, but hollow ones may bring extra possibilities)?
Can we wikilink "finite element approximations" ?

Minor style changes. Mostly ok but I reverted a few: "DC current", while apparently a repetition is, as explained in direct current, acceptable terminology, and even necessary in some instances for clarity; in two places (concerning Kennelly and Maxfield & Harrison) the edit lost the sense of historical perspective (M&H were definitive at the time - not necessarily today). The change of "IF" to "intermediate frequency" at the end of the article was reverted, the abbreviation "IF" is used earlier in the paragraph and the term is given in full and wikilinked at the beginning of the article. "intermediate frequency" (without capitals) is ambiguous and could be taken to mean a frequency band whereas it actually means a specific application and not a specific frequency. Couple of others amended which were ungrammatical. SpinningSpark 18:33, 25 December 2009 (UTC)[reply]
All Ok (well, I dislike DC current, but see it too often to get aggravated). Materialscientist (talk) 03:41, 26 December 2009 (UTC)[reply]

Yes, I know. It's like EMF, we all know it isn't really a force, but that's the terminology that has become established. Remember, Wikipedia is not here to put the world to rights, just report it as it is. SpinningSpark 11:29, 26 December 2009 (UTC)[reply]

Hollow resonators. Cavity resonators and hollow transmission line resonators are much used in loudspeaker cabinet design and other audio applications, so obviously they are possible. However, I have no source giving such a design for a mechanical-electronic filter nor have I seen one personally. That's not to say such a thing does not exist. SpinningSpark 17:11, 25 December 2009 (UTC)[reply]
This is a part which could be expanded, but I won't insist on this.

Don't see how if there are no sources. SpinningSpark 11:29, 26 December 2009 (UTC)[reply]

Finite element approximation. I guess we need an article on that, but there is Analogue filter#Approximation which discusses it in outline. Or we could redlink it to Finite element approximation (electrical networks) or Filter synthesis approximation. SpinningSpark 17:58, 25 December 2009 (UTC)[reply]
If not described in a WP article yet, suggest making a note briefly explaining the term. One concern is that "finite element modeling" is very popular in engineering; it has WP article(s), but is irrelevant here. Materialscientist (talk) 03:41, 26 December 2009 (UTC)[reply]

Finite element model is not totally irrelevant, at least on a conceptual level, but would not make a suitable link. I will add something to the article. SpinningSpark 11:29, 26 December 2009 (UTC)[reply]

Unresolved issues

"as nickel wires attached at the nodes" - why nickel (uncommon material for wires)?
Don't have a full answer to that, none of my sources directly address the issue. The wires do not have to be nickel but often are. In some designs the magnetostrictive property of nickel is made use of by making the ends of the wires part of the transducer, thus making a small simplification to the design. This is already mentioned in the article, as is the fact that there are better materials than nickel for the transducers. I have one source that states the wires are the same material as the resonators (which would make them nickel-iron in that particular case) but whether that was necessary, convenient, or just a coincidence, it does not say. I note that nickel has a very high Youngs modulus (the main property of interest in terms of equivalent circuit) compared to the common materials for electrical wires, but it is not especially higher than steel so I would have thought that steel wire would be perfectly ok, but as I say, I have no source that discusses it so would not want to write anything in the article at present. SpinningSpark 17:06, 25 December 2009 (UTC)[reply]
[1] [2] [3] [4] contain some bits of an answer (quick to read even for me :) but I by no means pretend this are the only sources - actually found within minutes on GoogleBooks). What I would like to have in the article is a brief and clear explanation on why these materials (particularly Ni and its alloys) are chosen: magnetostriction, thermal and maybe other reasons. Materialscientist (talk) 03:40, 27 December 2009 (UTC)[reply]
Ok, after trawling through a heap of old patents (example) I agree the article may be giving undue emphasis to nickel, so I propose a bit of a rewrite. Most of the books you found are already referenced in the article and what they have to say about nickel is already incorporated. But what has now become clear is that;
Pure nickel is used mostly in older designs

Nickel's major purpose in these designs is its magnetostrictive property

Nickel is only essential for magnetostrictive purposes in the input and output rod. Something else (steel, nickel-iron) can be used for the coupling wires

Nickel-iron alloys are used in the resonators because there is a particular nickel alloy that maximises Q. This would also apply where the coupling rods are expected to resonate

You mentioned thermal properties, I know nickel has a low temperature coefficient of resistance but I don't think that is a consideration here. This is going to be more than a quick edit and may involve some work on the images. I'm not going to be able to work on it right away, can we put the review on hold for a while? SpinningSpark 20:32, 27 December 2009 (UTC)[reply]

So.. does the article pass or fail? The review seems to have petered out back in December. Wizardman _{Operation Big Bear} 19:45, 5 February 2010 (UTC)[reply]

It fails merely because the author disappeared on WP. I hope he returns, addresses the comment above, after which the article should pass. Materialscientist (talk) 07:03, 6 February 2010 (UTC)[reply]

Ok, sorry for the delay, I found it necessary to go on an extended wikibreak. I am back now and I think I have addressed the above concerns. SpinningSpark 18:02, 5 April 2010 (UTC)[reply]

Removal of nickel in two instances made it not the major wire material here, and the present explanation of it is adequate. That said, I would appreciate a brief note on why steel and nickel-iron. My wild guess is steel is used for superior mechanical properties, and nickel-iron is some compromise for magnetostriction, price, mechanical properties, etc., but I would prefer to have a referenced note. Materialscientist (talk) 01:46, 6 April 2010 (UTC)[reply]

I have now added a section on resonators to explore in more detail the materials used. I have also removed mention of particular metals, such as nickel, in the transducers image to make it more generic. By the way, I have now found at least one design that uses nickel wires couplings throughout (not just for the input/output transducers) but probably best not to confuse things now with this oddity. SpinningSpark 19:12, 8 April 2010 (UTC)[reply]

The last (rather valuable) addition dispersed my remaining doubts. The article is not ideal (meaning extra work would be needed for FAC), but it deserves the GA status. Materialscientist (talk) 09:31, 9 April 2010 (UTC)[reply]