Effective "grit size" of natural stones

Kitchen Knife Forums

Help Support Kitchen Knife Forums:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Check it. For mirror polish. Lol.
PXL_20240304_004125590~2.jpg

PXL_20240304_004242184~2.jpg
 
@ethompson Is there enough consistency in how numeric "fineness" ratings are applied to Japanese stones, assuming reputable merchants, to attempt a grit equivalency on that basis?
 
Japanese Knife Imports has an Oouchi which includes the description: "This regions' natural stones all have about the same fineness of girt to being with (around 6-8k) and the potential finish can go up from there, depending on the stone hardness, the pressure you use during sharpening, water, etc."

Is this mine or designation consistent outside of JKI such that charting "Oouchi" at 6-8k makes sense, or would it have to be "JKI Oouchi" to be meaningful?
 
If I am not just talking to myself now I could use some input regarding this excerpt:

Another vote for a Soft Ark if one is purely looking for a mid-grit natural, though with some caveats...

They're sort of in a 1.5-3K ish sort of range, depending upon how they're lapped, how burnished they are, etc. Like all Arks, the edges off of them are very 'crisp' compared to synthetics, or even some other naturals.

... Jumping from an SG500 to either a soft Ark, or a BBW/Coti on mud is typically a bit much, if you're looking at this sort of progression. 800-1K, sure. 500-600, that's pretty cuspy unless you've got a crazy-fast Coti on full mud.

This is a significantly finer estimate for the Soft Arkansas stone than provided by @stringer @captaincaed and @cotedupy

If you wanted to remove all bevel scratches what kind of synthetic stone would fit in a progression right before a typical Soft Arkansas?
 
Last edited:
If I am not just talking to myself now I could use some input regarding this excerpt:



This is a significantly finer estimate for the Soft Arkansas stone than provided by @stringer @captaincaed and @cotedupy

If you wanted to remove all bevel scratches what kind of synthetic stone would fit in a progression right before a typical Soft Arkansas?

I don't sharpen that way.

So best I can say is it depends.
 
@Mr.Wizard I know we talked a little in PM about this but I'll post up my thoughts here. Maybe @stringer, @Desert Rat, @cotedupy, etc. will chime in with their thoughts on my ramblings.

It's been a while since I went back and played with my Arks and to be honest, I don't tend to overly analyze my stones when I'm using them. I'm more or less just wanting know how they perform on the individual knife and I feel about that performance. So, I apologize if my memory is a bit fuzzy. Also apologize if this was already stated and I missed it.

The biggest issues I see with trying to categorize the grit ratings of something like a soft Ark are 1) some natural variation, 2) pressure can play a fairly big role, 3) the specific knife being sharpened (steel, state of the edge, etc.).

I would often use my soft Arks as a one-and-done stone. Depending on the knife, there were many times that I could go from pretty dull to fairly refined just by changing how I'm using the same stone. I also almost exclusively use pharmaceutical mineral oil on my soft Arks and honing oil on my hards and blacks. If you have a good clean stone, a very light touch of oil, and a knife in a "friendly" steel with a thin grind, you can do basically the work of a SG500. No, it isn't the exact same experience by any stretch, just meaning it is capable of bringing up a dull edge. Then you can add a touch more oil, lighten your pressure and get a pretty nice kitchen-friendly refined edge.

When I used my Arks quite a bit more, I often would toss a Crystolon or India into the starting mix just to save time but I know I sharpened a number of knives just off of softs.
 
Last edited:
The biggest issues I see with trying to categorize the grit ratings of something like a soft Ark are 1) some natural variation, 2) pressure can play a fairly big role, 3) the specific knife being sharpened (steel, state of the edge, etc.).
Indeed, I am trying to include that range as best I can figure and then represent it in a reasonable way.

I would rather not include outliers that behave quite unlike most stones of the type but I do want to capture typical variations. For stones with a particularly large range and/or variation I intend to use two different spans with "range" and "variance" bars, although I haven't decided if I'll reuse same marks or introduce new ones. E.g. for the Coticule I presently have 800 to 14000 as the wide span with 1500 to 8000 as the narrower span. This is intended to represent that a single good quality stone might have a range of 1500 to 8000 (pending review by cotedupy) but that finer and coarser stones exist. It is not practical to try to provide this for every type but I think it should help for Coticule and Cretan, maybe others?
 
@JustAnotherJoe Thank you. I happen to have been aware of that before but I appreciate input like this. I wish they had done experimentation with surface conditioning in that test.
 
About a week or so ago I had typed out a long reply to this thread and some of the questions @Mr.Wizard had asked, and I'd got very nearly to the end, but I didn't finish and post it, and it didn't save. Which was extremely annoying, and I've only just got round to mustering up the mental fortitude to attempt it again.

Buckle in folks, here goes...

1.) The Scandinavian stones you're thinking of including are extremely niche. The only reason anyone's really heard of them is because of G&H, but if you did a poll of all the members here and on Badger & Blade, you'd find no more than a handful of people who own even one of them.

British stones such as Charnleys, Idwals, Tams, Dalmores, WoA, Yellow Lakes &c. are far more common, and are found all over the world. The reason for that is simple historical socio-economics, which in this instance you can read as a euphemism for 'empire building'. If you're going to nab someone else's country - you'll probably want to take a whetstone with you. In fact you might want to take quite a lot of stone with you, because it's extremely good ballast in a ship and more useful at the other end than just filling the hull with sand, which is your other option.

If you were making this chart 1,000 years ago; it'd all be Scandi, and for the same reason. Because back in the day the Vikings got around a bit, and they took a lot of sharp things with them, which in between bouts of pillaging, needed sharpening. 'Norweigan Ragstone' was the No.1 top fave sharpening stone in Europe for a lot of the last 1,500 ish years.

But today; British whetstones quarried in the c.18th, c.19th and early c.20th, are found all over the world. If you want the chart to be most relevant to the largest number of people alive now, then I personally would drop the Scandi stuff and include more British stuff.

(And it goes without saying, but that is a quantitative judgement, not a qualitative one. I'm sure Scandi stones are excellent and deserve wider recognition than they currently have).

---

2.) I don't know anything about scythestones I'm afraid. Again; today that kind of thing is fairly niche, but with quite an interesting history. If taken as a single category, then scythestones have undoubtedly been the most produced type of whetstone in the last several thousand years. In the 1790s; production of a single type of British scythestone (Devonshire Batts) employed many hundreds of people, but by 1830 - no one. Technological advances in agriculture at the start of the industrial revolution effectively wiped the industry out within a generation.

Still though you'd have thought there'd be more old ones around today, but there aren't. Scythestones got used hard and often, but like many low and mid-grit natural stones they weren't desperately efficacious so the wear rate was high, and they also broke easily. In a single harvest one worker could get through dozens of stones.

---

3.) The difference between Rouge du Salm, La Lorriane, and Belgian Blue Whetstone, is at best contentious. And historically the names got applied differently by different mines/quarries.

G&H is an invaluable resource, a work of remarkable dedication and research, without which an awful lot of information would simply have been lost to time. But it was a beginning not an end, and it shouldn't be regarded as the final word on everything. New information comes to light, other old honestones are identified.

And geology doesn't happen in neat boxes either. You don't get some stones that are red and some that are purple and they behave noticeably differently, it's just a spectrum. And in this instance I personally have found no correlation between colour and effect. IME there might be a loose correlation between stones that have the typical 'spots' and those that don't, but in general: The Blue / Red / Purple stone is far less variable than Yellow Coticule...

---

4.) Cotis have been quarried pretty much continuously for at least 2,000 years, and from multiple mines.

I can't emphasize this enough: It's impossible to give a rating for a 'standard' Coticule, because such a thing doesn't exist. It would be like trying to give a rating for a 'standard' Japanese stone, and it would be just as reductive. Coticules can be as coarse as an SG500 or as fine as a Thuringian, as soft as a King Deluxe or as a hard as an Lv.5+++ Jnat. And that's ignoring 'Hybrid Coticule’, a type of Chert similar to Arkansas novaculite, which is found next to Les Latneuses.

Now, I realise all that isn't desperately helpful in this endeavour! One possible solution might be to consider only the veins that are currently being commercially exploited. Whilst I've had quite a number of Cotis and BBWs, if you want some properly authoritative opinion and advice on this - I'd suggest emailing one or both of: Bart Torfs (Coticule.be) and Rob Celis (Ardennes Coticule).

Put a gun to my head though, and yes - I think your suggestion of 1.5k - 8k as a core range for modern Yellow Coticule is as good it could be. In fact when I typed out the previous version of this reply, I think my suggestion was 1k-8k, so basically spot on the same. But as I say - including a rating like that, in that way, would IMO be counter-productice.

---

5.) When @Steampunk puts a Soft Ark at 1.5k - 3k, and a few others (myself included) put them slightly lower - nobody is incorrect. Like all natural stones, Soft Arks vary. And as you've noted - with harder stones like novaculites, surface conditioning plays a very significant factor.

Moreover the definition of a Soft Ark is exceptionally blurry; historically there has been no real, delineated distinction between Soft Arks and Washitas. For instance there’s no mention of ‘Soft Arkansas’ stones in Griswold's 1890 geological survey, they appear to be an invention of Pike's around 1900, in part probably as a way of claiming they were the only company who had access to 'true' Washitas. And the stones that Pike and Norton sold as 'Soft Arkansas' are what most companies today would call 'Hard Arkansas'. While many of the stones sold today as 'Soft Arkansas’ are probably what Griswold would've called 'Calico Washitas'.

The specific gravity of stones that have been sold as Soft Arkansas or Washita can be anywhere from about 2.00 up to about 2.55. That range is huge. The only other type of whetstone I know with a comparable range of possible densities, and which we call by a single name are, you guessed it... Coticules.

---

6.) Regarding Cretan/Turkish Oilstones - changing my mind between 500-7k to 800-8k is really quite a small difference when trying to 'grit rate' natural stones vs JIS. In fact I'm amazed those ranges are so similar. I do have one or two old Turkish that can abrade at around 500, and one or two that would top out finishing at 10k or slightly above. For the purposes here I tried to condense the range down to what someone might expect if buying a modern Cretan stone, which might be something like 1k - 9k or 2k - 10k.

I would no longer stand by what I said a few years back about Turkish being faster than Washitas. They can be, depending on the two particular stones being compared, but in general I would now say that Washitas sit a little below Turkish.

---

7.) Congratulations to anybody who has made it through all of the above! This you will be pleased to hear is my final point. As I've said previously in this thread - I do not think that trying to map the 'grit' of natural stones onto JIS and other systems is a pointless or futile endeavour. I think it can be done, and I think it can be useful, but there is considerable difficulty, and an awful lot of nuance involved...

When trying to rate natural stones in comparison to JIS or ANSI/CAMI or whatever there are actually are two factors that we are considering: The speed of the stone, which might be broadly analogous to the low end of its grit range estimate. And the finishing level of the stone, at the high end of the grit range estimate.

Synthetic whetstones (and other abrasives) don't really have this range or dichotomy to anything like the same degree. Because they are very specifically designed to continue to abrade in the same manner, releasing particles of a uniform size, and constantly exposing fresh abrasive. Until they have no abrasive left, and you replace them.

Natural whetstones on the other hand are not very specifically designed to continue to abrade in the same manner, releasing particles of a uniform size, and constantly exposing fresh abrasive.

And compounding that is the fact that Silicon Dioxide is considerably less hard than Alox, SiC, CBN, or Diamond. So it wears differently; in extreme examples something like a Translucent Arkansas can go from 1k equivalent to 15k, in under a minute.

Natural whetstones get used and experienced in different ways by different people, and they get called different names by different people. And that makes it very difficult and subjective to do something like this. Because we're trying to compare things that are man-made and tightly defined and deliberately unvarying, to things that exist on spectrums and are ineffably plural and variable.

---

I wonder what the rocks make of it all? Probably something along these lines I imagine...

IMG_4015.jpg
 
Last edited:
When trying to rate to rate natural stones in comparison to JIS or Cami or whatever there are actually are two factors that we are considering: The speed of the stone, which might be broadly analogous to the low end of its grit range estimate. And the finishing level of the stone, at the high end of the grit range estimate.
Agreed.
Synthetic whetstones (and other abrasives) don't really have this range and dichotomy to anything like the same degree. Because thy are very specifically designed to continue to abrade in the same manner, releasing particles of a uniform size, and constantly exposing fresh abrasive. Until they have no abrasive left and you replace them. Natural whetstones on the other hand are not very specifically designed to abrade in the same manner, releasing particles of a uniform size, and constantly exposing fresh abrasive.
Eloquently stated.
Natural whetstones get used and experienced in different ways by different people, and they get called different names by different people. Because we're trying to compare things that are man-made and tightly defined and deliberately unvarying, to things that exist on spectrums and are ineffably plural and variable.
Amen.
 
@cotedupy Great thanks for such a detailed reply. I will be re-reading your post in detail and cross-checking my existing notes. I will probably reply multiple times as I find specific points to discuss. Please say when we reach the point where you've given as much of your time as you want to, or if we're already there. But I am eating this stuff up. :)
 
1.) The Scandinavian stones you're thinking of including are extremely niche. The only reason anyone's really heard of them is because of G&H, but if you did a poll of all the members here and on Badger & Blade, you'd find no more than a handful of people who own even one of them.

British stones such as Charnleys, Idwals, Tams, Dalmores, WoA, Yellow Lakes &c. are far more common, and are found all over the world. ...

I need this kind of feedback as I have little exposure to the world of natural whetstones. Nevertheless there are reasons for the initial selection I made that I want to explain.

I can only commit so much space to natural whetstones within the GLGC, the primary limitation in the form of the width of the column. Stones that overlap in grit range stack up, so I have to limit the number of these at any given level. There seem to be a lot more fine naturals than coarse ones, so I have to be quite selective in that area, whereas conversely I struggle to find enough coarse stones to fill out the chart. This lead me to the Orsa and Gotlands stones, then the Loos stone for a more complete family. The Loos stone overlaps others so I might drop it, but at the moment the Orsa and Gotlands are "free" in that there is nothing that could take their place.

I like the idea of having historic items on the chart so long as they do not come at the expense of more pertinent items. For example my charts lets someone reading an old woodworking book get some bearing on what "start with M2 glasspaper" means. Your remark about Norwegian Ragstone more makes me want to include it than the contrary, if it could be represented by a single item and not a whole family.

Given my space issue I asked earlier if it would be reasonable to include "Welsh Slates" verbatim as an item, as I cannot see making room for each of "Yellow Lake Oilstone, Glanrafon, Nantlle Valley" individually. I reiterate. Should I pick a single one if that's all I have room for? Which?

You footnoted Charnley Forest and Llyn Idwal with the disclaimer about the difficulty in quantifying dense novaculite. I included Arkansas stones because they are ubiquitous, but I thought perhaps it was better not to include these other novaculites because of that? (And of course there's still the crowding in fine grits problem.)

I already have Tam O'Shanter, Water of Ayr, and Dalmore Blue on my list.
 
3.) The difference between Rouge du Salm, La Lorriane, and Belgian Blue Whetstone, is at best contentious. And historically the names got applied differently by different mines/quarries.
Alright, so what name do I use and what range do I give it? I believe I recall reading that "Belgian Blue Whetstone" is a more recent concoction. If this is "six of one and half a dozen of the other" I'll probably pick "La Lorraine" as it's the most concise without using abbreviations and I like the vintage label.
 
I already have Tam O'Shanter, Water of Ayr, and Dalmore Blue on my list.
Dalmore stones are good if you can find a soft/soft medium example which is not soaked in oil. I have had hard examples that would not cut until slurried, and very pretty swirled ones that self slurry like a jnat.
 
I need this kind of feedback as I have little exposure to the world of natural whetstones. Nevertheless there are reasons for the initial selection I made that I want to explain.

I can only commit so much space to natural whetstones within the GLGC, the primary limitation in the form of the width of the column. Stones that overlap in grit range stack up, so I have to limit the number of these at any given level. There seem to be a lot more fine naturals than coarse ones, so I have to be quite selective in that area, whereas conversely I struggle to find enough coarse stones to fill out the chart. This lead me to the Orsa and Gotlands stones, then the Loos stone for a more complete family. The Loos stone overlaps others so I might drop it, but at the moment the Orsa and Gotlands are "free" in that there is nothing that could take their place.

I like the idea of having historic items on the chart so long as they do not come at the expense of more pertinent items. For example my charts lets someone reading an old woodworking book get some bearing on what "start with M2 glasspaper" means. Your remark about Norwegian Ragstone more makes me want to include it than the contrary, if it could be represented by a single item and not a whole family.

Given my space issue I asked earlier if it would be reasonable to include "Welsh Slates" verbatim as an item, as I cannot see making room for each of "Yellow Lake Oilstone, Glanrafon, Nantlle Valley" individually. I reiterate. Should I pick a single one if that's all I have room for? Which?

You footnoted Charnley Forest and Llyn Idwal with the disclaimer about the difficulty in quantifying dense novaculite. I included Arkansas stones because they are ubiquitous, but I thought perhaps it was better not to include these other novaculites because of that? (And of course there's still the crowding in fine grits problem.)

I already have Tam O'Shanter, Water of Ayr, and Dalmore Blue on my list.


Ah sorry, I hadn't clocked that some of the more recherche low grit stones were there to fill blank space. Loos is definitely a good one to have cos I think you can still buy them relatively easily(?). And Ragstone certainly has historical interest, Norwegian stones dating between about 500-1500AD are found all over northern Europe.

Other potential options are; Amakusa (around 800-1k), Binsui (1k-2k) and Dalmore Yellow (around 1k).

You're right that the large majority of the natural stones people still use now are finer grits. There are petrological reasons for that, but also simply - the large majority of low and mid grit natural stones just aren't as good as synthetics. Whereas finer natural stones can be at least as good, often better than their man-made equivalents, when it comes to stuff like; single bevel knives, polishing, razor honing &c.

Doesn't help with your layout though unfortunately!

---

If you can, I would try to find space for Idwals and Charnleys, you can do them as one entity though - they are usually very similar, or even lump them in with trans and black arks. Both were produced in very large quantities back in the day, and used widely, and again now popular in the straight razor and woodworking communities. Also interesting historically because Llyn Idwal is the Ur-Novaculite.

The word 'novaculite' isn't a real scientific term, ask a geologist about it and you'll get a blank stare. It was coined as a neologism in late c.18th by a guy called Richard Kirwan in his Mineralogy, from the Latin 'novacula', or razor. Where he identifies a particular type of grey-green rock known for its remarkable honing properties, and it's clear from his description that the stone in question is what we now call ‘Llyn Idwal’.

And as I say - I think you'd also particularly want to include them if you want relevance in terms of stones used historically and still today by woodworkers. As well as being fine finishing, they're also very hard so you can lean into them more without them going out of shape, in comparison to say slates.

One of the reasons so many Idwals and Charnleys are 40-50mms is that if your stone is the same width as your plane blade then it will only dish on one axis - along its length. You also sometimes see them mounted sideways in boxes people have made, ie using a narrower sharpening surface than the main one, which is for the same reason but for sharpening chisels. You see this with Washitas too sometimes. I've won a number of ebay stones on the cheap because they look like they're only 1" wide in the pictures, but if you look at the unusually tall box they're in, you can tell it's actually a wider stone mounted on its side.

---

Yes, I think you can certainly put all Welsh slate together, in fact if you wanted you can probably put all slate from anywhere in the world under one heading. As HB says in G&H; slate honestones are made in almost every country, and it’s a remarkably consistent type of stone wherever it's from. 6-12k at the extremes, with the large majority falling within 8-10k.

The outlier might be Thuringians, which some people call slate, though I personally would call them more like an argillite. They have very little to no fissility.

---

I would call it ‘Belgian Blue Whetstone’, because that's the term that is most widely used, and that's what Ardennes Coticule use. In fact I imagine the term was coined specifically for this sort of purpose - because historically there were a number of different terms being used for effectively identical purple-red rock.

I wouldn't use the term La Lorriane, because it's particularly confusing, as there is also an old type of whetstone sold as 'Lorraine' (that one with the vintage label) that may or may not have been from Belgium. Note that the large majority of the Lorraine is in northern France. There is a small part of 'Belgian Lorraine', but it's not in the Ardennes where cotis and bbw come from, and according to Wikipedia at least - it is 'geologically distinct'.

As far as I understand - that type 'Lorraine' with the pretty vintage label is probably the same, or very closely related to the German 'Goldfisch Wetzstein', and they were probably quarried in NE France, or just over the border in Germany. I do have an example of the Goldfisch Wetzstein and it's quite different from BBW / RdS.
 
Last edited:
4.) Cotis have been quarried pretty much continuously for at least 2,000 years, and from multiple mines.

I can't emphasize this enough: It's impossible to give a rating for a 'standard' Coticule, because such a thing doesn't exist. ...

... I think your suggestion of 1.5k - 8k as a core range for modern Yellow Coticule is as good it could be. ... But as I say - including a rating like that, in that way, would IMO be counter-productive.

How then should I chart this? I don't want to leave it out. As you wrote in post #45 "You don't get all of this range in a single stone" so I don't want to have a single unbroken range, and adding a narrowed range (1.5k - 8k) is one attempt to impart useful information. Do you have a different suggestion? Do I need more than one Coticule item for this to make sense? Also I currently have merely Coticule, do I need to change this to Yellow Coticule? Or perhaps Modern Coticule? I can add information to the README that clarifies this somewhat but I don't want the chart to be misleading in isolation.
 
5.) ... Moreover the definition of a Soft Ark is exceptionally blurry; historically there has been no real, delineated distinction between Soft Arks and Washitas. For instance there’s no mention of ‘Soft Arkansas’ stones in Griswold's 1890 geological survey, they appear to be an invention of Pike's around 1900, in part probably as a way of claiming they were the only company who had access to 'true' Washitas. And the stones that Pike and Norton sold as 'Soft Arkansas' are what most companies today would call 'Hard Arkansas'. While many of the stones sold today as 'Soft Arkansas’ are probably what Griswold would've called 'Calico Washitas'.

The specific gravity of stones that have been sold as Soft Arkansas or Washita can be anywhere from about 2.00 up to about 2.55. That range is huge. The only other type of whetstone I know with a comparable range of possible densities, and which we call by a single name are, you guessed it... Coticules.

Maybe I am going about this wrong. I keep seeing specific gravity come up in the context of Arkansas stones. Is this a more reliable way classify stone performance? As esoteric as it seems should I be charting SG brackets rather than Soft, Hard, Surgical Black, and Hard Translucent? What about Washitas? Does a particular SG range almost always indicate Washita or Washita-like performance?
 
6.) Regarding Cretan/Turkish Oilstones - changing my mind between 500-7k to 800-8k is really quite a small difference when trying to 'grit rate' natural stones vs JIS. In fact I'm amazed those ranges are so similar. I do have one or two old Turkish that can abrade at around 500, and one or two that would top out finishing at 10k or slightly above. For the purposes here I tried to condense the range down to what someone might expect if buying a modern Cretan stone, which might be something like 1k - 9k or 2k - 10k.

I would no longer stand by what I said a few years back about Turkish being faster than Washitas. They can be, depending on the two particular stones being compared, but in general I would now say that Washitas sit a little below Turkish.

I wasn't drawing a distinction between the ranges 500–7k and 800–8k, but rather between them as a group and the newer range of 2k–10k that you gave. I am still a little fuzzy on what to chart. What range can one expect from a single stone?
 
When trying to rate natural stones in comparison to JIS or ANSI/CAMI or whatever there are actually are two factors that we are considering: The speed of the stone, which might be broadly analogous to the low end of its grit range estimate. And the finishing level of the stone, at the high end of the grit range estimate.

Synthetic whetstones (and other abrasives) don't really have this range or dichotomy to anything like the same degree. Because they are very specifically designed to continue to abrade in the same manner, releasing particles of a uniform size, and constantly exposing fresh abrasive. Until they have no abrasive left, and you replace them.

Depending on how my attempts on this are received I may create a new chart dedicated to natural whetstones which would remove a lot of constraints I am presently applying. Within that expanded space I could give two independent ranges for each stone, one for cutting speed and one for finishing capacity. I could use color coding to denote something like polishing ability, and I could use the X-axis for information rather than merely layout, charting stone hardness. We would end up with something of a hybrid between the current format and a bubble chart. You would know better than I if such a chart would be welcome.
 
You can't quantify naturals, they need to be handled and used to understand there behaviour and where they sit in a sharpening lineup which can change drastically if you swap a stone out for another. They are unique individuals with what you could call having a soul.

A simplistic example from my collection is that I have a Norton washita that cuts like a 800 while another at 2-3k and this all depends on what steel I'm using and pressure, the variables are endless. The red striped charnley I use to finish low hrc knives finish to something like 5-6k while another is 8-10k+ super hard only good for razors.
 
Back
Top