Boxford aud metric gears

[RRMBKParticipant @rrmbk]

Hello all.

Can anyone tell me please what the standard stud gear ( headstock spindle Driver ) Idler gear and Leadscrew gears should be for an early ( 41/2 inch) AUD underdrive fitted with a factory original equipment  5mm pitch metric leadscrew and gearbox.

I do have the large double gear idler which I presume is to allow imperial threads to be cut with the metric leadscrew but its the original setup I am really interested in to get the correct pitches from the gearbox plate as currently they do not correspond.

Hope this makes sense.

Kind regards all .

[HopperParticipant @hopper]

The guy to talk to is forum member Brian Wood. He literally wrote the book on this stuff. Sure he will be along at some point.

[Clive FosterParticipant @clivefoster55965]

According to the Boxford book “Know Your Lathe” 1988 edition the standard set up for a metric machine with a gearbox is:-

stud gear       20 teeth

idler gear       80 teeth

gearbox gear  45 teeth

For coarse pitches replace the stud gear with the 50 teeth spare gear that is normally parked outside the gearbox gear.

The 45 and 50 teeth gears are of very similar size so it would be easy to inadvertently interchange them on the gearbox drive shaft. Which will give incorrect pitches. A likely error on a used machine and hard to uncover if you don’t have the book.

The book is not clear as to what constitutes a metric coarse pitch, presumably its on the gearbox data plate, but for English threads coarse is said to be from 4 to 7 tpi, so I guess a metric box coarse will be roughly 3 mm to 6 mm pitch. So a matter of little concern to the user who is unlikely to cut such coarse threads. Especially as they are mostly greater than the lead screw pitch and so widely considered unsafe to do on a small machine. Its often recommended that if such things have to be done the lathe be driven via handle on the lead screw end rather than by its motor. This practice is said to greatly reduce the stresses on the drive train.

Clive

[Clive Brown 1Participant @clivebrown1]

Probably a typo, but the Boxford metric leadscrew pitch should be 3mm, not 5mm. The compound gear for cutting imperial threads is 135T / 127T.

[Craig BrownParticipant @craigbrown60096]

Here is a picture of the chart off the gearbox of a later AUD but to my knowledge it was the same all the way through. These later charts were better, giving all the information required. Hope this helps

[Michael GilliganParticipant @michaelgilligan61133]

I hope you don’t mind, Craig

… I’ve just spent a couple of minutes brightening-up that very useful chart

MichaelG.

[RRMBKParticipant @rrmbk]

Thank you all for your most helpful information. The 5 mm was not actually a typo, just my poor eyesight in reading the info plate ! I will try and upload a photo i’ve taken of the older plate for future reference.

You have uncovered the mystery. The leadscrew was fitted with a 56 t gear, hence the problems.

I use the 135/ 127 as the idler just for storage space reasons, but do have the correct 80 t idler.

So that explains the 80t , 50t & 45 t gears, The lathe however came with the following gears so i’m not sure what the relevance of the others is !

56, 52, 44,40, 38.  anyone got any ideas?

[Clive FosterParticipant @clivefoster55965]

The 56, 52, 44, 40 and 38 gears are needed when cutting the full range of officially possible Imperial threads on a Metric Boxford model A.

John Stevenson put up a picture of the relevant chart, which should be on the inside of your change gear cover, in a thread back in 2011.

Hopefully this link works :-

BOXFORD Metric Threads on Imperial Model A

3 rd post down.

Maybe someone who knows how can transfer it, or a more readable copy should one exist on the internet, to this one.

Why it’s not in Know Your Lathe, the obvious place for it, I shall never know. Along with why Boxford copied SouthBend using white text on a black background reduced at not wonderful resolution in the book.

Clive

[peak4Participant @peak4]

My version of Excel is too old, but maybe someone with 365 could have a play with the various tables and screenshots

Insert data from picture Excel for Microsoft 365

https://support.microsoft.com/en-gb/office/insert-data-from-picture-3c1bb58d-2c59-4bc0-b04a-a671a6868fd7

Bill

[Mark EasingwoodParticipant @markeasingwood33578]

I don’t know about Boxford Lathes, but the 38 tooth is usually required for 19 tpi threads.

Mark.

[Howard LewisParticipant @howardlewis46836]

IF it is any help, on my Lathe (Not a Boxford; basically Metric and has a Norton gearbox) a 120/127T gear is used as an Idler. For cutting Metric threads the 120T serves as the Idler.  To cut Imperial threads, the 127T side of the compiund Idler is put into mesh, and the 40T gearbox input gear and spacer exchange positions.

HTH

Howard

[DC31kParticipant @dc31k]

We should note that thus far in the discussion, we have had mention of three different compound gear sizes.

135/127 by Clive Brown 1 (repeated by RRMBK, the original poster).

100/127 in the chart posted by John Stevenson and cited by Clive Foster.

120/127 by Howard Lewis.

The point is that the extra gears that RRMBK has might not make the threads shown in JS’ chart unless he has the same compound gear as shown in the chart.

[Clive Brown 1Participant @clivebrown1]

The 135 / 127 gear is the Boxford item for cutting imperial threads on metric lathes. Which is what the OP has.

The 127 / 100 gear is the Boxford item for cutting metric threads with an imperial lathe, as described in the JS link, and not applicable to the OP.

Both gears were supplied as “extras” by Boxford; ie not part of the standard change gear set that normally came with their lathes.

I’m not aware that Boxford supplied a 127 / 120 gear. They don’t feature in their published gear trains.

PS. I’ve never grasped where the 135 count springs from, but my lathe’s imperial.

[Clive Brown 1]

On Clive Brown 1 Said:

The 127 / 100 gear is the Boxford item for cutting metric threads with an imperial lathe, as described in the JS link, and not applicable to the OP.

So is it just coincidence that the gears he has match the ones in the ‘not applicable’ situation?

Can we say that the reason for the OP having the gears in his possession with the machine he has is so far unexplained?

That is what we are trying to find out here – why he has those gears (and if they are of any use to him).

In principle, as he has a 127t gear, the numbers he has makes sense for cutting imperial pitches on his metric lathe – there is a 7, an 11, a 13 and a 19 in the factors of those gears.

Maybe the take-away from Clive Foster’s post is that there might be a similar chart (something in the same spririt, but a different language) available for his machine. That is what we need to find or construct in order to answer the open question.

[BazyleParticipant @bazyle]

edit – I got a number wrong and posted rubbish so deleted.

[Clive FosterParticipant @clivefoster55965]

Following on from what DC31k says the extra gears needed for cutting imperial threads on a gearbox equipped lathe are specific to the compound conversion gear and the position of the extra gear in the train. Leadscrew pitch doesn’t matter because the conversion is done via specific pitch settings on the gearbox.

So in principle a conversion set specified for any lathe could be used so long as the appropriate gears are used in the appropriate position and the relevant gearbox settings chosen. But you also have to take into account any differences in the initial drive train ratio so it’s not as simple as it sounds.

My metric Smart & Brown 1024 uses a 120 / 127 compound gear with a set of extra gears to go on the gearbox input shaft. So different set of gears. Drive train is 30 to 120 so 4 to 1 ratio in this case.

This link on Thingverse

https://www.thingiverse.com/thing:5515671

refers to a 3D printed set of metric to imperial conversion gears based on a 65 / 76 compound gear. Also charts of the settings but the threads are said to be approximate.

Apologies for getting metric and imperial wrong way round in the link to JS post. The gears the OP has are, of course, the official imperial to metric conversion set.

Clive

[Howard LewisParticipant @howardlewis46836]

The essential thing, for an accurate conversion from Metric to Imperial , or vice versa, is the inclusion of a 127T gear (25.4 x 5 = 127) in the train.

The number of teeth on the other half of the compound Idler is determined by the needs of that particular machine. (EXAMPLE : Mine is intended to cut Metric pitches using 40/120/40 in the drive to gearbox.  For Imperial  pitches the train becomes 40/ 120+127/ 40)

For a Metric machine with changewheels, it needs a 127T in the train for accurate translation from Metric to Imperial pitches.

The Driver and Driven will depend on what pitch is required.

An Imperial machine will need the 127T to translate the Imperial pitch Leadscrew to produce Metric threads

63 being ALMOST half of 127 will translate Metric to Imperail, (or the other way round) but with a slight inaccracy, which will vary according to the pitch being cut.  It might be so small as to be unnoticeable over a short thread, or could be importanr over a long length of thread where the errors begin to add up.

An error of 0.005 mm might not matter, particularly, over 10 pitches, because of clearances, but will be obvious when the total error becomes totals 0.5 mm over 1,000 threads!

There are books which give the percemtage errors for a pitche, alongside the gear trains for that pitch.

Howard

[Howard Lewis]

On Howard Lewis Said:

 

63 being ALMOST half of 127 will translate Metric to Imperail,

But 63T is not  used as half of 127 as an approximation.

The 63T is used with an 80T gear to give a ratio of 1.2698, an error of 0.0125 per cent from the perfect 1.27 ratio of a 100/127 tooth gear set. Plenty good enough for practical use and used by various lathe manufacturers.

Likewise, the combination of 47/37 is used by other lathe manufacturers (Hardinge, Southbend, Logan)  for metric conversion. It gives a ratio of 1.2702 , an error of just 0.0213% from the perfect 1.27. Again, plenty good enough for practical work.

I presume Myford uses the 21T gear because 21 is one third of 63. The gear train would have to then use other gears to multiply by 3 to achieve the final overall ratio. But I don’t have a 21T gear so have never looked into it. I use the standard change imperial change gears and Martin Cleeve’s charts to approximate metric threads within an error of 1 thou in three inches to 1 thou in 8 inches.

[Clive Brown 1]

On Clive Brown 1 Said:

 

PS. I’ve never grasped where the 135 count springs from, but my lathe’s imperial.

127/135 in effect turns a 3mm pitch leadscrew into a 9TPI leadscrew. That’s why in the chart the 9tpi setting has 32T input and output gears and most of the pitches contain a gear that has a tooth count divisible by 9.

[RRMBKParticipant @rrmbk]

Once again thank you all for your advice, assistance and expertise. At least my pitch measurements now match the table on the gearbox.

Interestingly there is no chart on the inside of the change wheel cover and this is most certainly original and still in its original paint, so not been overpainted.

[Pete Rimmer]

On Pete Rimmer Said:

On Clive Brown 1 Said:

 

PS. I’ve never grasped where the 135 count springs from, but my lathe’s imperial.

127/135 in effect turns a 3mm pitch leadscrew into a 9TPI leadscrew. That’s why in the chart the 9tpi setting has 32T input and output gears and most of the pitches contain a gear that has a tooth count divisible by 9.

Brilliant! Thanks for revealing the mystery of the 135T. It’s been bugging me all night. Never would have given a thought to a virtual 9TPI leadscrew!

[Hopper]

On Hopper Said:

On Pete Rimmer Said:

On Clive Brown 1 Said:

 

PS. I’ve never grasped where the 135 count springs from, but my lathe’s imperial.

127/135 in effect turns a 3mm pitch leadscrew into a 9TPI leadscrew. That’s why in the chart the 9tpi setting has 32T input and output gears and most of the pitches contain a gear that has a tooth count divisible by 9.

Brilliant! Thanks for revealing the mystery of the 135T. It’s been bugging me all night. Never would have given a thought to a virtual 9TPI leadscrew!

Yes it doesn’t appear to be intuitive. Until you swap out the 45 gear on the gearbox for the supplied 40. 45/40 = 9/8 and now you have a virtual 8TPI leadscrew giving you a bunch of even-number pitches on the box. Somebody somewhere must have this conversion chart

Did anyone notice that on the Metric gearbox if you start on any row on the right, the numbers go up in 1/8th increments as you move left?

[Clive FosterParticipant @clivefoster55965]

Pending a simpler alternative I guess you could always brute force things using a spreadsheet as ratios in the box can be derived from the settings. As the 127/135 compound gear driving 40 teeth gear turns the 3 mm pitch lead screw into 8 tip , 0.125″ pitch, we know that 3 mm on the box becomes 8 tip.

So as standard 20-80-45 at lever positions 1 & D gives 3 mm pitch then 20-127/135-40 at lever positions 1 and D gives 8 TPI.

Alternatively standard 50-80-45 at lever positions 4 & C also gives 3 mm pitch and so will 50-127/135-40 at lever positions 4 & C.

Either way the internal ratios of the gearbox can be figured out and a suitable spreadsheet constructed to give the actual TPI for an appropriate selection of gears used in the first driver position g on the standard chart with the 127/135 – 40 conversion pair fitted.

I think it certain that the actual gears Boxford used will be a subset of the accessory gears supplied for converting a lathe without a gearbox. According to the accessories list Boxford used 7 gears for the conversion set. Presumably the 127/135 and 6  simple gears. The simple lead screw set has 11 gears so the gearbox doesn’t seem to save many gears although I suspect several of teh gears are for fairly exotic purposes.

Clive

[Clive Brown 1Participant @clivebrown1]

Thanks for replies to my puzzlement over the 135T gear. Due to its use with a 127T gear my thinking was that it should be an exact metric / imperial conversion, rather than a close approximation, as has been suggested. It makes me wonder if Boxford could have used a gear combination giving a comparable result with more economically sized gears, but I’m not going there!

 

[HopperParticipant @hopper]

Now I am confused. So the 135/127 combo turns the 3mm leadscrew into a virtual 8TPI, not 9TPI as posted earlier?

[Author]

Posts