This is from memory so take it with a grain of salt.

I remember reading the following tidbits.

Metal fasteners didn't start to show up until around the 15th century but did not become common until the industrial revolution in the 19th. The story I remember is that before machines were created to produce repetitive threads they were hand filed and the nuts were tied with a string to the bolts that they fitted. I.E. they were only used on large structural projects like bridges where their ease of use outweighed their difficulty to manufacture.

Have a look at this book:
One Good Turn - A natural history of the screwdriver and the screw by Witold Rybczynski.
http://www.amazon.com/exec/obidos/search-handle-form/103-2273759-3479857

I haven't read it, but it should answer a few questions. The extracted pages from it on Amazon are from the introduction and index rather than the body of the book, so they're pretty unhelpful in judging its content.

My memory agrees with Hush, but I'm wondering if woodscrews might have been around before then - after all it is a damned sight easier to make an external thread than an internal one.

Does a corkscrew count?

Another field where they might show up earlier is jewellery. I could ask my brother the archaeologist.

looking through the sample index pages of that book it looks like it'll have all the answers - I'm not ordering it - the postage will be twice what the book costs!

Cheers

Greg Locock

A bit of an aside--I remember an interesting discussion on NPR a few years back regarding screws.  The scientist/historian/author (I don't remember which) commented that the screw was the only invention which allowed us to find it in nature.  

The only naturally-occurring screw is the DNA double-helix, which could never have been discovered without first the invention of precision screws. So we imagined the shape, then we found it occurring naturally.

Hopefully not too off-topic.
Brad

I can't think of a visible double helix in nature, but anybody who's seen a creeper twisted around a tree trunk would have more than enough inspiration to make a screw shape. Nature is full of spirals and twisting shapes, so it's not like the idea of a screw developed in a vacuum.

Talking about a double helix occurring in nature, has anyone ever seen two snakes mating? Very close!

Y'All - I'm from Southern Massachusetts

Out of the clear blue sky I stumble on a thread that reminds me of something I haven't thought about in years. I asked this question of the curator of a machine tool museum in the Massachusetts/Rhode Island Blackstone River Valley. This is the birthplace of the Industrial Revolution in the United States. He didn't have a good answer, but I suspect someone out on this list will know. Here goes.

Many of us have probably cut our share of lead screws and know it's not too difficult with a decent lathe fitted with a good lead screw.

The big question for me is "How the heck did they fabricate that first "good" lead screw?

My guess is that it wasn't too difficult to turn a uniform cylinder and to lay out a right triangle on a piece of paper and wrap that aroumd the cylinder. That would give a decent layout, but how was the thread cut? Probably by hand with a file, but would this give enough accuracy? Would this screw being used to cut a second screw "filter" out some of the imperfections and would repeating the process a few times with these theoretically, progressively better screws finally result in an acceptable lead screw?

I'm not losing any sleep over this, but would like to know the answer if anyone out there does.

Thanks in Advance.

I don't know the answer but have always believed that machine tools "evolved" as world-class craftsman hand-made "precision" tools that were used in rudimentary machines to make better and better tools.

In this case, I would suspect that it was ground or filed although I can picture how it might have been cast. I think casting's been around for a long, long time and it would be much easier to cut that first lead screw into clay or something.

There is a rather nasty tool called a thread chaser that can be used to handcut threads on a lathe without a leadscrew. It is basically a flat plate with 5 or so teeth cut into it, and you just push it along so that the 5th tooth engages in the helix cut by the other 4. It would be possible to shape these teeth to get the correct advance for a particular diameter, rather than having to guess.

Cheers

Greg Locock

I guess thgat the thread chase makes sense, but how do you get the needed accuracy into the chase?

JC

Well, you've taken a rotational problem and turned it into a linear one. I guess you do the old trick with three chasers, if they all fit each other then they are good. I didn't say it was easy!

Cheers

Greg Locock

I have cut threads into a steel rod, starting with a hand held hack saw to make an angled groove, and then following the initial groove with a thread chaser of the type GregLocock mentions.  It took over an hour, and the accuracy was merely passable; but the nut went on and stayed on.  (This was for an axle on a traveling lawn sprinkler.  The original design uses gawdawful clinch nuts.  And, yes, I have a die set for cutting threads, but couldn't get the die started for some reason.)

So it would seem that some of the methods mentioned above would work very well.  A practised craftsman could file a very accurate thread chaser and then file an accurate initial groove on a rod.

The question of how machine tools became progressively more accurate, despite having only less accurate machines to make them, is quite interesting and would make another good thread.

Returning to Binary's original question, I have a book called "Ancient Inventions" - which, by the way, I can thoroughly recommend. There is on page 17 of my paperback edition a photograph of a reconstructed gynecologist's instrument, a "speculum", which contains a beautiful example of a long positioning screw thread. The device, of extraordinary precision and workmanship, was excavated from Pompeii, which was buried in 79 AD. Many other instruments found in the "House of the Surgeon" were not seen again until the seventeen hundreds. It does not look to me as though the thread would have been cut with a hand chaser, but I could be wrong - the photograph is, after all, of a reconstruction, which was of course made on a modern lathe.

As a follow up to the post by "Steamboat44" pertaining to the use of one tool to make a better tool.

The addition of back gearing, lead screw, and roller bearings on a lathe was a great leap forward in the machine shop.  In the 20's to make a very precise lead screw a very simple method was used to average out imperfections.  The very best lead screw was made on a lathe and it it turn was installed a lathe and a follower made with leather was used to make another lead screw.  Some of the original ruling engines used this approch to make very fine lines for diffraction gratings.

I have seen some amazing work done on machinery I thought was totally worn out. It was the machinist not the machinery.   

Here is the history of the screw according to

http://inventors.about.com/library/inventors/blscrewdri...

unclesyd:  My post flatly contradicts the last sentence in the opening paragraph of the article which you have cited by Mary Bellis, in which she says metal screws did not appear until the 15th century, and that only wooden screws existed before that time. The device unearthed in Pompeii was made completely of metal and was beautifully machined, but you can't tell from the picture whether the nut was internally threaded, or whether it just used a short pin to engage the helix. The Romans also had metal surveying devices rather like theodolites, with worm and wormwheel mechanisms for setting the inclination in two axes, so if you classify a worm as a type of screw, then that's another example.

Likewise the link provided by unclesyd indicates that screws were invented somewhere between the 5th and 1st centuries BC

TTFN

IRstuff : Yes, that seems at least to be consistent with all the references I have - but I don't quite follow you - are you taking issue with that claim or concurring with it ?

A little expansion on metal screw history:

http://www.hayesbolt.com/news%2C_seminars%2C_events.htm

Perhaps Ms. Bellis' article was mis-edited and should have said "only appeared in large quantities in the 15th century"

TTFN

Here's another interesting link:

http://www.findarticles.com/cf_dls/m1026/5_158/67161980...

TTFN

I could be wrong, but I have a hunch that many metal screws produced in Roman, and perhaps Greek times were not hand filed as your first link claims, but produced on a lathe, probably using a thread chaser, but possibly something more sophisticated. However, it's just a hunch - I can't prove it. But the more I learn about earlier civilisations, the more I think we underestimate them.

Re: how to make things more accurate starting from less accurate:

In "Foundations of Mechanical Accuracy" published by Moore Tools, there's a nice description of making precision lead screws.  The final step is hand lapping.

Laps are devices that average out errors, so if you have a screw with decent overall accuracy but local errors, a lap (skillfully operated) will improve it.