Screw Thread and Relation to Linear Adjustment

[Droogs]

Could someone please point me in the direction of a reference source for the following information?
I am currently unable to do any major or prolonged woodworking and will soon be bed bound for a while. So my usually, rather limited and fairly unused grey matter has somehow sparked into life and has started chucking up all sort of ideas for things to occupy it while it can't partake in its favourite hobby of making hamster bedding. Included in this are several ideas for the design of some tools for marquetry and inlaying, which I think it has wanted to be available but aren't, and has now pushed to my front lobes and got me thinking. A couple of the designs involve the need to make very accurate adjustments. So I'm thinking, well use some sort of machine screw doodad. What I need is for some sort of chart that will give me the OD(size) for the screw before threading, along with the relationship to what different threads (TPI) on that diameter screw will mean with regards to lateral movement per full revolution of the screw, ie

5mmOD x 26(tpi) x 1 rev = 1/100mm lateral movement on its axis
3mmOD x 32(tpi) x 1 rev = 1/1000mm lateral movement etc

Hope that explains it, look forward

rgds
Droogs

 

[CHJ]

If you are going to use metric threads then you need to use the pitch not Threads per inch (mixed standards)

A ISO 6mm coarse has a pitch of 1mm for example so advances 1mm per rotation.

http://www.gewinde-normen.de/en/iso-coarse-thread.html

To get the pitch of imperial threads divide 1" by the TPI and this will give you the pitch in inches.

See Sewing machine threadsfor instance, but be aware that some thread forms may be difficult for the average DIY workshop to achieve.

 

[bugbear]

See Sewing machine threadsfor instance,

Wow. That's the largest array of thread information I've seen online. Duly bookmarked, and thank you.

BugBear

 

[flh801978]

Droogs

You can make any size thread with any pitch really within limits

so you could have a 6mm dia bolt with 1mm or 0.75mm or 0.5mm
going larger say a 12mm bolt could be from 0.5 to 2.5mm without looking strange

Though remember with a 0.5mm pitch you are moving 0.5mm with each turn
so no where near the 1/100's or 1/1000's of a mm each turn your example used

micrometers for example use 0.5mm pitch for the screw or 40tpi for imperial mics

Ian

 

[Droogs]

Many thanks guys, lots of info in the responses. I'm not intending to make the screws but buy in some that meet the needed specs as it were. So having looked at the info, if i wanted a srcew that moves 1/10 of 1mm per turn for a metric machine screw regardless of OD of the screw, I'd need a pitch of 0.1? But as there doesn't seem to be any that fine on the list, would it be a case of say getting a 0.5 pitch and having a marker every 72 deg for 1/10 adjustment and that 2 revs would equal 1mm travel?

 

[flh801978]

You will be lucky to get a thread of finer than 0.5mm anywhere the finest thread i have seen is 80tpi on microscope parts which is about 0.3 mm pitch

yes put a big round dial on the end and divide it up to give the resolution you need

Ian

 

[CHJ]

The best method of reading small increments mechanically is to have a graduated Barrel Vernier as per a standard micrometre.

Electronically of course you can use a digital reader as per Digital Electronic Calipers.

 

[bugbear]

Many thanks guys, lots of info in the responses. I'm not intending to make the screws but buy in some that meet the needed specs as it were. So having looked at the info, if i wanted a srcew that moves 1/10 of 1mm per turn for a metric machine screw regardless of OD of the screw, I'd need a pitch of 0.1? But as there doesn't seem to be any that fine on the list, would it be a case of say getting a 0.5 pitch and having a marker every 72 deg for 1/10 adjustment and that 2 revs would equal 1mm travel?

Yes; here's a lathe handle/dial showing exactly that sort of thing.

Rather more than a mark every 72 degrees, of course.

BugBear

 

[Sporky McGuffin]

I'm not sure the threads on machine screws will actually be that accurate, and you'll have backlash to contend with too - think how much you can "wiggle" a typical nut on a bit of threaded rod. Might be worth looking into leadscrews and ballscrews instead.

 

[MusicMan]

I used to do this sort of thing professionally for a scientific instrument company.

First of all this calculation is wrong

5mmOD x 26(tpi) x 1 rev = 1/100mm lateral movement on its axis - it is actually 0.98 mm/rev
3mmOD x 32(tpi) x 1 rev = 1/1000mm lateral movement etc - it is actually 0.79 mm/rev

As others have pointed out, the OD is irrelevant (apart from what pitches are available in that size). Much best to work directly in metric though.

Yes you can divide the revolution by a divided dial, which is how a micrometer works. In fact you can buy micrometer heads without the opposing bracket for just this sort of job, and that will probably be by far the best thing to do. The division is good to 0.001 mm at least - just depends how accurately you can control the rotation. A stepper motor at 200 steps per rev is a neat way to do it.

No, you can't get a thread of 0.1 mm pitch in practice (yes it could just about be made but...). But you can get threads of 0.5 and 0.4 mm pitch. If you couple these, rotate one one way and the other the other way, this gives an effective pitch of 0.1 mm. It's called a differential thread and has sub 0.001 mm resolution. I doubt if you will actually need this for your application, but you can find such differential micrometer heads at (for example) http://www.thorlabs.de/navigation.cfm?guide_id=81.

We look forward to the results of your bedridden creations! Best wishes with the health issues.

Keith

 

[Dee J]

Just about any screw drive arrangement will have some backlash ( except for some expensive ball screws) - so it's best to have some sort of arrangement to bias the lash in one direction - or make sure critical movements are made in one direction only. eg, on the topslide of my old Drummond lathe I always advance the tool to its required location, if I need to back it further out I wind back past the required location and then advance back in.... Just like a Bailey plane blade adjuster really.

 

[Droogs]

Many thanks again guys, I have no real knowledge of this hence the examples were just to express the concept of what I'm thinking about. I'm not sure how accurate the screws have to be. What they are for is I want to make my own version of a string thicknesser but need it to be accurate,repeatable and easily adjustable as the stringing, which will be in various woods, will be part of a veneering layup that will include lapped semi-precious stones and metal work done to tight tolerances. Also they are to go into a stringing compass with micro adjustment. Think of the LN inlay stuff but a lot better in terms of accuracy.

@ Musicman - many thanks for pointing out the micrometer heads, this i will look into in more depth

@ sporky/DeeJ - never heard of leadscrews or ballscrews so will look at these as well

 

[MusicMan]

The usual way to avoid backlash in lightly-loaded screw drives is to spring load a double nut so that the nut defining the motion is pushed against one side of the screw thread at all times. The split nut in a lathe leadscrew does this in effect. An easy way to achieve this in a home-built system is literally to split (saw) a nut (perpendicular to the thread axis) about three-quarters through, bend the two parts together a little bit and then force the screw in. Obviously friction increases but in a lightly-loaded system this won't be significant.

You can get the micrometer heads in all sorts of sizes and I expect this will be the way to go.

In general, you need a precision about ten times better than the thing you are trying to control. So if you want to control your strings to 0.1 mm, your drive system should be settable and repeatable to 0.01 mm. Not hard with micrometer heads.

I suspect leadscrews and ballscrews will be too large for this application but I don't have a clear vision of the machine you are designing. IF you post your design thoughts here, I, and I am sure many others, will be happy to provide feedback.

Keith

 

[Magicniner]

Wow. That's the largest array of thread information I've seen online. Duly bookmarked, and thank you.

BugBear

You will like the Maryland Metrics Thread Identification Chart, you'll need to google it though as my account doesn't have the required permission to post useful links :-(
Regards,
- Nick

 

[monkeybiter]

http://mdmetric.com/tech/tict.htm

 

[woodfarmer]

Why not gear drive the thread ?