What grit/micron should I have mirror polish?

[philippasteur]

[quote quote=“AnthonyYan” post=4346]I like to use the optics standard for mirror finish, which means that a surface is smooth enough to do high quality imaging, such as for a camera, telescope, or microscope.

Once your scratches start going below half of an optical wavelength, you will be at or close to an optically smooth finish (ie: mirror surface). Loosely speaking, light cannot notice features which are much smaller than its wavelength. This is why astronomical telescopes are typically accurate to 1/10th wavelength. More accurate mirrors exist, but that is at the point of diminishing returns in terms of image quality. For astronomy a mirror that is accurate to 1/4th wavelength is considered to be the lowest quality that is still (somewhat) usable.

In practice, being able to read fine-print text in the reflection of your knife bevel is, in my opinion, a good enough definition of mirror-finish. Say, your eye 6 inches away from the knife and the text 6 inches away from the knife. One might also require that the surface have no noticeable “haze”.[/quote]

“Visible light is approximately in the range of 0.4 to 0.7 microns (okay, if you are really picky, the 0.38 to 0.74 microns). If you want, that’s 400 to 700 nanometers, or 4000 to 7000 angstroms.”

Is a quote that is helpful from one of Anthony’s previous posts.
If we go with the half wavelength figure as indicating optical mirror quality we would have to polish to about 0.025 microns to get to this optical quality level. This assumes that the depth of the scratches we get are equal to the grit size, which is not always the case. Of course there is the additional polishing as the grit on the stones fractures and becomes smaller. We may get there with 0.1 micron abrasives as they break down… and with more time.

This site has an interesting chart:
http://www.warleypolishing.co.uk/technical/surface-finish-table-guide.php

They have two grades at the high end. Both claim a surface roughness figure of 0.5 microns. The difference is in the amount of reflectivity. It would seem from this that there is more to polishing than surface finish alone. There is the part that I call brightness. Maybe I can see a reflection enough to see print at 3 micron grit, but it is much brighter and better refined at 0.5 microns.

I like the 6 inches to the subject and 6 inches to the eye definition, because at least it is some kind of definition. Maybe we could refine it and state the font size that represents “fine-print”. Perhaps a number 4 point font? The one thing that is not there, and without sophisticated measuring equipment, is very subjective, is brightness of the image. I am not sure how we cover that one.

One thing for sure though, we never can get an optical quality mirror surface on our edges. The max reflectivity of typical stainless steels is about 65%. This does not come close to the 90%+ (some as high as 99.9% in dielectrically coated flat mirrors as in the case of a couple of the star diagonals that I use)in an optical mirror.

Phil

[anthonyyan]

[quote quote=“PhilipPasteur” post=4352] This site has an interesting chart:
http://www.warleypolishing.co.uk/technical/surface-finish-table-guide.php

They have two grades at the high end. Both claim a surface roughness figure of 0.5 microns. The difference is in the amount of reflectivity. It would seem from this that there is more to polishing than surface finish alone. There is the part that I call brightness. Maybe I can see a reflection enough to see print at 3 micron grit, but it is much brighter and better refined at 0.5 microns.
[/quote]

The science of surfaces is extraordinarily complex and I’m no expert in it. It can involve a huge number of different fiends, such as surface physics, tribology, friction, wear, optics, etc. In the chart from the Warley Polishing website, they’re using R_a as a measure of surface roughness. You can find the formula at the link below. It’s basically an average of absolute height of peaks and valleys.
http://en.wikipedia.org/wiki/Surface_roughness#Amplitude_parameters

Unfortunately, R_a is only a single number, and is only a measure of vertical deviations, so it’s not clear how wide the surface scratches are. Perhaps it is reasonable to assume the width of scratches is about the same size as their depth? If not, then there are many different surface finishes with the same R_a, just as there are many distributions with the same average. This might account for two different surfaces having the same R_a but different levels of reflectivity.

One thing for sure though, we never can get an optical quality mirror surface on our edges. The max reflectivity of typical stainless steels is about 65%. This does not come close to the 90%+ (some as high as 99.9% in dielectrically coated flat mirrors as in the case of a couple of the star diagonals that I use)in an optical mirror.

I think I would say this a little differently. In astronomy, high reflectivity is necessary because the light they are collecting is very dim. So, as you mention, they require reflectivity of 90% and higher.

But not all mirrors are used in astronomy. And so there are companies that manufacture optical steel mirrors (see link below). My understanding is that these are optical grade mirrors, just not designed for astronomy.
http://www.precision-metal-optics.com/stainless-steel-mirrors.htm

It may be my own personal terminology, but I would say glass has a mirror finish, even though, generally, glass reflects only a small fraction of incident light (because most of the light is transmitted). If you put a black screen behind a sheet of glass (to block transmitted light), then you could take a very high quality photo using the reflection alone. It just might be a bit dim (so you could increase the exposure time, etc.). I would say the same for the surface of still water.

I like to think of “mirror finish” as mostly about smoothness and somewhat independent of “reflectivity.” But that’s just my preference, and it is reasonable to me that “mirror finish” could mean “mirror like” and therefore have high reflectivity as well as being optically smooth.

Sincerely,
–Lagrangian

[anthonyyan]

Here is another company that manufactures stainless steel mirrors. Since steel does not shatter, these are considered to be a type of safety mirror.

http://www.mirroredstainlesssolutions.com/

Not as bright as a conventional mirror due to the lower reflectivity of steel, but I think it would be difficult to notice unless you had a steel mirror and a regular mirror side-by-side.

Similarly, many people consider chrome plating to have a mirror finish. Yet chromium has a reflectivity of less than 60% for light which is perpendicular to the surface. In the link below, you can enter the wavelength of light in microns and get a plot of how reflectivity varies with the angle of incidence.

http://refractiveindex.info/?group=METALS&material=Chromium

Sincerely,
–Lagrangian

[philippasteur]

You make valid points. They don’t answer the question of how we define a mirror edge on a knife so that it is understandable to all. My original thought was that it is hard to tell a person such as the OP what grit they need to use to get a mirror edge (which was the original question) if we don’t have a definition for a mirror edge. This came about specifically because I have had people show me knives that they thought had a mirror edge, but didn’t look at all like the edges that I think have a mirror edge. All of our discussion about the minutia has not got us closer to getting a working definition.

Answering this question may be impossible as there seem to be many different types of mirrors with widely divergent properties in common use. Personally I would be quite happy to get my edges to the point where they look like the picture that you posted… Other than the picture, did you find out how they polish the surface. Maybe we can tell people that they have to match that process to get a mirror edge. At least that allows us to give some kind of answer based on a specific outcome.

[davidhamilton2]

There is some interesting and cool science here, but maybe a better definition of mirror like might be a term borrowed from coin collecting (hey, they work with metal, not glass). {You can skip down to the last paragraph if you wish to skip the background}

rare coins absolutely can NOT be polished without destroying it’s value. a coins grade relates to the condition it was in the moment the dies struck it. anything that happened after that is damage, The value of a damaged coin is often orders of magnitude below undamaged coins (though honest wear lowers the value but doesn’t destroy it).

There are many companies (some of them highly respected) that grade and encapsulate rare coins. coins are rated on a scale of 1 (you can tell it is a dime) to 70 (unobtainable perfection) with a few qualifiers sometimes used after the number. Depending on the coin, the difference between say, a MS65 and a MS 66 grade might be 10, or even 100’s of thousands of dollars.

—Read the next two paragraph only if you want to know more about coin grading—

Coins can be graded either Mint State 1-70 (MS1 to MS70) this means the coins were made for commerce, the coins in your pocket probably average MS 45 or so, when you get a brand new new nickle out of a fresh from the mint roll, that is likely a MS 63 or so.

coins can also be Proof’s, they “can” be graded PF 1 to PF 70, though the lower grades are rare. Proofs are usually minted for collectors. they are made in small quantities, they use specially prepared blanks and are stamped much slower and with much more pressure so they have much more detail. If a proof is in a low grade it was probably stolen from a collection and spent, more likely, on the rare time a proof got circulated, it will lose it’s proof diagnostic and become an MS coin

Proofs are usually mirror finished, but not always. but they are pretty much always specimen strikes

-------------!!were AT info for the KNIFE guys now!!

SORRY IT TOOK SO LONG TO GET HERE, BUT THIS IS WHAT THE COIN GUYS SAY

An official grade qualifier from the most respected grading company is DMPL That qualifier can add Thousands to the price of a coin. It stands for Deep Mirror Proof Like.

the coin that earns that DMPL is not a proof, it is proof like, but by definition, news print can be read from 4 inches away under good (but not direct light).

____________________________end of the line

So, I propose knife nuts follow coin collectors. If we can read news print 4 inches away, it is a mirror. discuss as you wish

[philippasteur]

http://oxforddictionaries.com/definition/english/mirror

noun
• a surface, typically of glass coated with a metal amalgam, which reflects a clear image: he checked his appearance in the mirror
• a thing regarded as accurately representing something else: the stage is supposed to be the mirror of life
http://www.thefreedictionary.com/mirror
mir•ror (m r r)
n.

1. A surface capable of reflecting sufficient undiffused light to form an image of an object placed in front of it. Also called looking glass.
2. Something that faithfully reflects or gives a true picture of something else.

http://dictionary.reference.com/browse/mirror

mir•ror

noun
1.
a reflecting surface, originally of polished metal but now usually of glass with a silvery, metallic, or amalgam backing.
2.
such a surface set into a frame, attached to a handle, etc., for use in viewing oneself or as an ornament.
3.
any reflecting surface, as the surface of calm water under certain lighting conditions.
4.
Optics . a surface that is either plane, concave, or convex and that reflects rays of light.
5.
something that gives a minutely faithful representation, image, or idea of something else: Gershwin’s music was a mirror of its time.

1. mirror(noun)
   polished surface that forms images by reflecting light
   unattributed supposedly from Princeton’s WordNet

Ok, If you read the dictionary definitions of what a mirror is above it is easy to see how hard it will be to come to a consensus on what a mirror edge is. The definitions are all pretty vague and subjective, to say the least. I can see that Anthony did a good job in showing me that my personal perception of what a mirror should be is a bit limited. Of course the idea of setting some kind of standard is in itself limiting. The hard part is to get people to agree with and use the standard. We could use the ISO or ANSI standards for surface finish, but then the measurements to show compliance is a problem.

I think that the idea of reading text of a specific size at a specific distance may be as good as we can get people to use. It is a simple concept and easy to perform. Unfortunately it will not show the difference between a 2K grit reflection and one done with submicron abrasives. I am sure that both will pass this test but will look completely different. It may be like the difference between the 65 and 68 rated coin, you would need to submit the blade to some kind of certified expert and let then subjectively rate them. Of course this will not happen.

Anyway, it whas been a good discussion and I learned quite a bit in doing research and reading other posts.

So, who else would be in for definng a minimal mirror edge to be one where newsprint can be read at 4 inches under “good” (whatever that means but not direct light??

Phil

[stevenpinson]

I can only write, I cannot read. :silly:

[anthonyyan]

Here are three possible tests for “mirror finish”:

(1) Reading newsprint/fine-print in the reflection at some reasonable distance (4 inches? 6 inches?), and no visible “haze” on the surface.
(2) Laser-pointer test: http://stellafane.org/tm/atm/polish/polish.html#Polished_Out
(3) Microscope test: No visible scratches under magnification. (100x? 200x?)

Any one of these seem reasonable to me. I think (1) is the most informal and least stringent. But between (2) and (3) I don’t know which is more stringent (and it might depend on magnification and microscope quality, as well as depend somewhat on laser wavelength).

Sincerely,
–Lagrangian

P.S. Apparently,for the laser-test, amateur telescope makers tend to recommend red lasers for the test. For details, see http://stellafane.org/tm/atm/test/quality.html#polish

[philippasteur]

I can only write, I cannot read. :silly:

You can count belly button lint fibers at 4 inches…

[philippasteur]

I think number one and three would be good criteria. I am not sure how number two would be applicable to polished steel.
Here is a quote from the article that is linked:

“But you should see NO reflection from the front surface. Yes, that is correct, a fully polished mirror will have essentially no reflection (that most people can discern) – it appears the polished surface is not there at all! If your mirror in not fully polished, the “rough” surface will scatter some light to the side, which you will see as a front surface reflection”.

So they are looking for scatter off of the polished surface of the mirror contrasted to a good reflection from the unpolished rear of the mirror. I guess we could look at how accuratelly the laser is being reflected from the edge. Rather than looking at the reflected laser (not god for the eyes) we could focus the reflection on a piece of white paper to examine the image.

Anyway, some good thoughts !!

Phil

[quote quote=“AnthonyYan” post=4379]Here are three possible tests for “mirror finish”:

(1) Reading newsprint/fine-print in the reflection at some reasonable distance (4 inches? 6 inches?), and no visible “haze” on the surface.
(2) Laser-pointer test: http://stellafane.org/tm/atm/polish/polish.html#Polished_Out
(3) Microscope test: No visible scratches under magnification. (100x? 200x?)

Any one of these seem reasonable to me. I think (1) is the most informal and least stringent. But between (2) and (3) I don’t know which is more stringent (and it might depend on magnification and microscope quality, as well as depend somewhat on laser wavelength).

Sincerely,
–Lagrangian

P.S. Apparently,for the laser-test, amateur telescope makers tend to recommend red lasers for the test. For details, see http://stellafane.org/tm/atm/test/quality.html#polish[/quote]