Engineer's Flat Files

[Cheshirechappie]

My old medium-sized flat file is a bit careworn, so I felt it was time to renew. As I was out and about today, I dropped into the engineer's merchants on the industrial estate. Unfortunately, on examination, the flat files they had in stock were just hand files with slightly curved edges - no slight bellying to the faces at all. The brand was Draper, by the way - which probably explains it.

Anybody know a sure-fire source of PROPER flat files? Ones with the slight crowning to the faces? (I need flat files, not hand files - I'm knee-deep in those, and can obtain them anywhere without any bother at all.)

 

[Phil Pascoe]

How did I know what you were going to say from the heading??

 

[RogerP]

Well I learn something every day (even after 26,518 of them).

I've just been through a box of several dozen inherited for a toolmaker uncle and I can't find any exhibiting this crowning (despite there being files of all descriptions in the box). Nor can I find any references on the web. Can you point me to one please?

 

[Cheshirechappie]

From 'Workshop Technology' Part1 2nd edition reprinted 1958, by WAJ Chapman.

Page 189 - "Flat File - This file is parallel for about two-thirds of its length and then tapers in width and thickness. It is cut on both faces (double cut) and both edges (single cut)."

Page 191 & 192 - "Use of the File - Cross-Filing. .....But even with the best regulated filing there is always a tendency to rock the file slightly, and the curved surface of a tapered file tends to make a hollow surface, thus counteracting any slight rocking."

From 'Turning and Mechanical Manipulation" Vol II (1847) by Holtzappfel.

Page 868 - "The necessity of the convex form of the file will now be rendered most striking, as where the file absolutely flat on it's face, it would be scarecly possible to reduce with it any small and isolated spot that might become coated with the red chalk from the planometer [surface plate]; but as the file is a little rounded, any precise spot on the work may be acted upon, as the end of the file may be pressed with the left hand on the exact spot to be reduced, whilst the remainder of the file is held just out of contact with the rest of the surface."

 

[Mike.S]

I'm no engineer so was surprised to learn of 'convex' flat files. At first I imagined that the convexity was side-to-side e.g. like an half-round file, and it puzzled me how this could help avoid rocking. But, a bit of searching threw up an educational paper - see page 57 onwards - here. This says:

"The face of most files is slightly convex because they are
made thicker in the middle than on the ends. Because of this
curvature only some of the teeth are cutting at any one time,
which makes them penetrate better. If the face were flat, it
would be difficult to obtain an even surface because of the
tendency to rock a file while filing. "

If this is the convexity being referred to then the curvature is along the length. Is this what CC is seeking, or is there another type of file with (slight) convexity across, as well as along, its length?

 

[RogerP]

.....
If this is the convexity being referred to then the curvature is along the length. Is this what CC is seeking, or is there another type of file with (slight) convexity across, as well as along, its length?

..........PROPER flat files? Ones with the slight crowning to the faces.....

... yes, apparently.

 

[Cheshirechappie]

I'm no engineer so was surprised to learn of 'convex' flat files. At first I imagined that the convexity was side-to-side e.g. like an half-round file, and it puzzled me how this could help avoid rocking. But, a bit of searching threw up an educational paper - see page 57 onwards - here. This says:

"The face of most files is slightly convex because they are
made thicker in the middle than on the ends. Because of this
curvature only some of the teeth are cutting at any one time,
which makes them penetrate better. If the face were flat, it
would be difficult to obtain an even surface because of the
tendency to rock a file while filing. "

If this is the convexity being referred to then the curvature is along the length. Is this what CC is seeking, or is there another type of file with (slight) convexity across, as well as along, its length?

Mike, the flat files I'm looking for have the slight (and it's very slight) convexity in the length. They're flat across the width. In some cases, it's not so much a convex curve, more a very slight change from parallel to tapering in thickness about half-way or a bit more from tang to tip. There are files with curves both ways of the face (properly called 'rifflers' I think) used by pattermakers and toolmakers dressing the dies and moulds used for injection moulding, hot stamping, pressing components out of sheet metal and other such applications, especially components with curves in more than one plane; but they're different beasts entirely from the humble flat file.

I have a couple of 4" precision files made by Grobet (I think) which do have a definite convexity to the length. I bought those from a horological supplier, H.S.Walsh. They're truly superb files to use, but not big enough for any but the smallest engineering work. For really small work, there's usually enough spring in a needle file to give sufficient convexity by bending in the fingers whilst filing, but that dosn't work for larger stuff (it's just about possible with a warding file, but you don't quite have the same control if you're straining to bend the file).

I think I've found one manufacturer - Vallorbe. However, I'm not sure from the description I've seen whether their engineer's flat files are tapered over part of the length, or over all of it, which is no better than flat for the full length. I suppose the answer is to buy one and see for myself.

What is it about modern tool manufacturers? They can't even produce something as basic and well-established as a file without disregarding the little subleties that turn a bit of rough metal into a useful tool. I can understand it with the cheapo imports, but even the big boys like Bahco can't be bothered these days, it would seem.

 

[Mike.S]

CC, thanks for clarifying that.

It was the notion of convexity across the file that had me perplexed but (Rifflers aside), from what's been said, to prepare flat surfaces a slight tapering of thickness along the length can be helpful. I guess this is comparable to trying to 'scoop' the middle of a wood plank edge with a plane in order to keep it flat - rather than the convex edges I usually produce

That's what I like about this forum - always learning something new.

Sorry that I can't actually help with sourcing what you're looking for.

 

[woodfarmer]

Maybe you could do what I used to do. recycle the old worn out file into an engineers scraper. Ideal for removingthe odd high spot.

 

[sdjp]

Probably not ideal, but every square, or threesquare, file I've seen has that transition from parallel to tapering, giving the crowning your looking for? Not ideal, because it'll have a tiny active area in that crowned section, and probably too much of a crown - I suspect your 4" files would give a larger contact patch!

It occurred to me that one of the other major manufacturers of quality files is Dick. I, er, happen to have a wee example of their work, in the form of a 14 inch smooth cut flat file. Which, from measurements, it shrinks from 8.0 to 7.6 mm from handle to tip; and checking with a 'straight edge' [0], it's crowned in the way you describe. Probably a clear signal, the tip of the file shrinks widthwise from 34 to 26 mm, in the top quarter of the file. I say probably, only because I've never twigged on that before (and yet, it's obvious in retrospect).

Checking their website, and I see that they no longer list that sort of file! They have 'hand' files, which clearly have parallel sides. But something tugged at my memory, and I dug out: http://www.ebay.co.uk/itm/Box-of-5-Dick ... 1588476154

Which is both labelled as 'flat', and has that slight taper in width near the tip, to match the one I have. Unfortunatly, unless you want to do a lot of large scale work, a box of 5 might be excessive! Also, 14" files are a beast, and don't really qualify as 'medium' sized, I'm afraid, unless you're used to doing things monster sized. (The same mob also have a 'hand' file from Dick, 14" smooth, which shows the constant width, so I'm inclined to believe their labelling.)

Vallerobe also don't list a flat file either, just a 'hand' file, constant width. Gorobet is a Vallerobe brand; so I'm assuming that they are the same.

Only other major manufacturer I can think of is Nicholson, so I did some Googling, and found: http://www.ffx.co.uk/tools/product/Nich ... File%206In

which says that the Nicholson 'flat' file is tapered in both width and thickness. I can't comment directly, lacking experience of either that site or Nicholson files; but that's the only lead I can shake out of the internet. Still, depending on how keen you are, it might be worth a bit of a stab at a Nicholson.

I'm afraid it looks like the 'flat file' is a dying tool - probably things like surface grinders have replaced it where accuracy counts, and elsewhere it's just jewellers and people who don't need enough accuracy to warrant the specialise tool.

[0] A ruler that must have cost two, if not three whole pence. However, appears good enough to see the crown on the file by rocking on it, and looking for the light.

 

[bugbear]

On woodcentral (I think) someone posted that they used two "normal" (i.e. uncrowned) flat files, wired together at the ends,
with a space (e.g. coin or similar) between to give the "bow".

Not a "true" solution, but a credible work round.

BugBear

 

[lurker]

Check out Proops Bros website
Bill Carter put me onto these folks evidently they bought a pile of British made ones a while back..... I think it was 20 tonnes!!

 

[Cheshirechappie]

Probably not ideal, but every square, or threesquare, file I've seen has that transition from parallel to tapering, giving the crowning your looking for? Not ideal, because it'll have a tiny active area in that crowned section, and probably too much of a crown - I suspect your 4" files would give a larger contact patch!

It occurred to me that one of the other major manufacturers of quality files is Dick. I, er, happen to have a wee example of their work, in the form of a 14 inch smooth cut flat file. Which, from measurements, it shrinks from 8.0 to 7.6 mm from handle to tip; and checking with a 'straight edge' [0], it's crowned in the way you describe. Probably a clear signal, the tip of the file shrinks widthwise from 34 to 26 mm, in the top quarter of the file. I say probably, only because I've never twigged on that before (and yet, it's obvious in retrospect).

Checking their website, and I see that they no longer list that sort of file! They have 'hand' files, which clearly have parallel sides. But something tugged at my memory, and I dug out: http://www.ebay.co.uk/itm/Box-of-5-Dick ... 1588476154

Which is both labelled as 'flat', and has that slight taper in width near the tip, to match the one I have. Unfortunatly, unless you want to do a lot of large scale work, a box of 5 might be excessive! Also, 14" files are a beast, and don't really qualify as 'medium' sized, I'm afraid, unless you're used to doing things monster sized. (The same mob also have a 'hand' file from Dick, 14" smooth, which shows the constant width, so I'm inclined to believe their labelling.)

Vallerobe also don't list a flat file either, just a 'hand' file, constant width. Gorobet is a Vallerobe brand; so I'm assuming that they are the same.

Only other major manufacturer I can think of is Nicholson, so I did some Googling, and found: http://www.ffx.co.uk/tools/product/Nich ... File%206In

which says that the Nicholson 'flat' file is tapered in both width and thickness. I can't comment directly, lacking experience of either that site or Nicholson files; but that's the only lead I can shake out of the internet. Still, depending on how keen you are, it might be worth a bit of a stab at a Nicholson.

I'm afraid it looks like the 'flat file' is a dying tool - probably things like surface grinders have replaced it where accuracy counts, and elsewhere it's just jewellers and people who don't need enough accuracy to warrant the specialise tool.

[0] A ruler that must have cost two, if not three whole pence. However, appears good enough to see the crown on the file by rocking on it, and looking for the light.

Thanks for that! You're right that a 14" file is a bit of a beast; I've used 16" files for reasons other than woodworking, and they're about the largest you're likely to find anywhere these days. In days of old when men were men and 9" angle-grinders weren't invented, they were made in lengths up to 36" according to Holtzappfel (1840s) - they must have been three shredded wheat jobs! I can't imagine many of those being made - pushing a 16" ******* file through mild steel is hard enough work.

That's a really good point about three-square and square files. There are several of them sat in the file-rack waving at me, and it just didn't occur to me. If I get no joy finding 'proper' flat files, I'll resort to using that dodge! Thanks indeed!

 

[Cheshirechappie]

On woodcentral (I think) someone posted that they used two "normal" (i.e. uncrowned) flat files, wired together at the ends,
with a space (e.g. coin or similar) between to give the "bow".

Not a "true" solution, but a credible work round.

BugBear

Thanks BB. That, as you say, is a neat dodge. I'd seen it suggested that a warding file (the thin ones intended for fitting the wards in keys and locks) can be bent into a slow curve with a bit of finger pressure, but when you're getting into 'last thou' territory, you really want as much sensitivity and lightness of touch as you can get, and the muscular effort of bending a file (even a little bit) tends to compromise that.

 

[Cheshirechappie]

Check out Proops Bros website
Bill Carter put me onto these folks evidently they bought a pile of British made ones a while back..... I think it was 20 tonnes!!

I did check the website, but it wasn't clear from their descriptions whether their files are cheap imports (of which they supply a lot) or British-made NOS.

I've ordered a couple of Vallorbe flat files from APTC. MSC Industrial stock them too. I'll report back when they arrive - small parcels are sent first class post, and as the website delivery information advises, "allow five days for delivery".....

 

[bugbear]

Thanks for that! You're right that a 14" file is a bit of a beast; I've used 16" files for reasons other than woodworking, and they're about the largest you're likely to find anywhere these days. In days of old when men were men and 9" angle-grinders weren't invented, they were made in lengths up to 36" according to Holtzappfel (1840s) - they must have been three shredded wheat jobs! I can't imagine many of those being made - pushing a 16" ******* file through mild steel is hard enough work.

In an age when all men were skilled to a high degree in hand tools, Henry Maudsley must have been truly superb to earn this compliment:

It was a pleasure to see him handle a tool of any kind, but he was quite splendid with an 18-inch file.

BugBear

 

[Cheshirechappie]

Ah yes - Henry Maudslay. A much under-rated figure from history. Not only did he make some very significant contributions to engineering, but he trained some real giants - people like Naysmith, Whitworth, Clement and others.

The Vallorbe files arrived today. Sadly, no crowning whatsoever, and if they have any taper in length, it's an even taper for the full length; so they're no better than hand files. Looks like I'll have to resort to the work-arounds mentioned above.

I've not been able to find any source of 'proper' flat files. I can't even find a UK maker of files. Peter Stubs of Warrington packed up making files sometime in the mid 1990s, and Blundell Files and Tools of Prescot seem to have disappeared too. I don't suppose anybody makes file steel now, either. Maybe there's a niche for a small-scale high quality maker a bit like Liogier in France, but where you'd get the kit (and the finance for it) goodness only knows.

 

[sdjp]

The Vallorbe files arrived today. Sadly, no crowning whatsoever, and if they have any taper in length, it's an even taper for the full length; so they're no better than hand files. Looks like I'll have to resort to the work-arounds mentioned above.

Further digging pulled out: http://www.apexhandtools.com/customer_s ... onSect.pdf which described Nicolson's range. Importantly, it includes, as separate entries: flat files, hand files and mill files. That lend confidence to the guess that they might still make a 'proper' flat file.

I don't suppose anybody makes file steel now, either.

I'm given to understand that file are normally a simple 1% C steel - this also makes sense with what I know about steel. That would be, dang as near it 1095 in AISI; CS100 is the same ballpark.

I'm sure that those are not the _perfect_ file steel - but only in the same sense that O1 is not the perfect oil hardening steel for a chisel. That is to say: the difference would be sufficiently small to not be a noticeable problem.

Although, as you point out, the equipment would be a major hurdle. The best rasps are hand stitched, which works in favour of Logier et al. Files rely much more on precision, so that's a lot of infrastructure...

 

[Cheshirechappie]

I've had a look at the Nicholsons, and as far as I can tell from descriptions on several suppliers' websites, they're no different to the Vallorbes. On the strength of that, I've ordered a couple of three-square files to supplement the *******-cut ones I already have. At least I get three working faces with those instead of only two! I thought about the four working faces of square files, but they tend to be a bit narrow.

File steel - yes, I think the most commonly used these days is 1.0%C 1.4%Cr. That's the same (En31, SAE52100) steel used by Stanley for their Sweetheart chisels, and in quantity for ball bearing manufacture. It responds to oil quenching, gives a deeper hard case than straight carbon steels, and the chromium content slightly increases the initial hardness. It's also fairly readily available, given it's use for rolling element bearings. That wasn't always the case, though - my copy of Higgins (Engineering Metallurgy, 5th edition 1983) suggests that a 1.2% - 1.3% straight carbon water-hardening steel used to be commonly used for file manufacture - I'd be surprised if such a steel was commercially available today, though.

The problem with steel grades is that there are thousands of them, so there are quite a few 'very close' near-matches for quite a lot of applications. Also, the advance of metallury over the years have made some of the 'old' steels redundant by producing low alloy steels with better hardening, working, non-distorting or whatever characteristcs than the straight carbon steels. Hence the virtual unobtainability of W1 - there are better steels for just about every application it was ever used for, with the possible exception of the finest of fine edge tools, and the very small quantity likely to be sold for that purpose makes it's production unviable.

 

[Cheshirechappie]

A brief update.

The Nicholson three-square files arrived from Toolbank yesterday. They're stamped 'Made in Columbia'. Quality is, on first impressions, a wee bit indifferent; the tang of one is slightly bent, and they lack 'finish' compared to older UK and USA made files. Not to the point of unusability, but in the little details that set a quality file aside. The toe ends have sharp edges that will need dressing off, the tangs have a slight swell at the tip that will also have to be dressed off before handles can be fitted, and the even colouring of my older files is replaced by rather strange mixture of tones that raise slight doubts about the final heat treatment.

Something that rather came to mind when examining these files was the difficulty in obtaining decent sawfiles documented by FenceFurniture a few months ago in on this forum and others. It does seem as if some engineering hand-tools are going the way woodworking hand-tools went in the 1970s and '80s. Rather sad, that; perhaps there's a niche opening up for someone prepared to start making quality files in smaller quantities. It does seem that the 'big boys' are losing interest in quality.