View Full Version : Gearing from spindle to leadscrew

Lew Hartswick
02-22-2009, 10:48 AM
There was a thread here, or on another board, recently on determining
all the possible threading pitches (some not documented) by looking
at the gears and counting teeth etc. A friend has a Jet 124 OPY
12 x 40 lathe and would like to be able to cut 7-1/2 tpi or close enough
that a nut with that won't bind over a length of about an inch (7 or 8 threads).
I looked at the headstock and counted all the gear teeth but dont
remember how the poster of the thread did all the arithmetic. :-)
I'm only "intimate" with the gear head Clausing Metosas at school
and so am not compentant to figure out even how to adjust the
gearing on the JET. I do see a pair of gears, 120t and 127t on the
same shaft, (english/metric) but don't know how they get changed
SO! does anyone have a book with the "operating" instructions
for the lathe? and/or a source for information that will explain it all?
The QC box has a good range of threads available from 4 to 112
in 5 rows and 8 columns. The 5 rows each succesive row doubles the
previous one.
The entries in the first row are:
4, 4-1/2, 5, 5-1/2, 5-3/4, 6, 6-1/2, 7.
A few other observations:
Two small gears( 21t and 26t) on a lever ? Is this a reversing control?
and if so why are the gears different number of teeth?
Another two gears on a shaft between the "reverse" idlers and the
120/127 of 40t and 50t .
I suspect the lathe hasen't been used for any threading since at
present the 127t gear is engaged.
The way the train is at present is:
Spindle50t - the 21t-26t device - 50t (with the 40t on the shaft) -
127t (with the 120 on the shaft) - 40t on the input of the QC box.
Thanks for any assistance.

tony ennis
02-22-2009, 10:55 AM
Two small gears( 21t and 26t) on a lever ? Is this a reversing control?
and if so why are the gears different number of teeth?

That's probably a reversing tumbler. In one position one gear transmits power, in another position both gears are used to transmit power and reverse the direction of spin. It doesn't matter that the gears have different numbers of teeth when you're transferring power along the gears, not spinning a shaft.
The gears are different sizes on my Craftsman also, so it's probably because the geometry/space constraints make using two gears a better solution.

edit - I thought I would try complete sentences.

tony ennis
02-22-2009, 11:23 AM
If the 50 tooth gear is on the "input" end of a shaft (e.g., doesn't simply transmit power to another gear), you can get a 7 1/2 tpi thread by replacing the it with a 60 tpi gear and putting the lathe's QC tumbers on the "9 tpi" setting.

Basically 7.5 TPI / 9 TPI = .8333, so a speed redux of .833 (which is 5/6, happily) is needed to reduce the 9 TPI setting to 7.5 TPI.

Can you post good pics of the gear train?

And yeeeah. Someone check my math and logic.

reality check - if he's making a nut that isn't too special, it will be cheaper to buy the nut than to buy the new gear. Just sayin'

tony ennis
02-22-2009, 11:46 AM
I should have consulted my spreadsheet more...

Given the constraints about "input side of a shaft", replacing a 40 with a 30 will do the same thing, if the 10 TPI setting is used i the QCGB.

Or if one of the input gears TPI is reduced by 5/8 and the 12 TPI QCGB setting is used.

It looks like, actually, there are a plethora of possibilities. Here are all the easy ones. These all apply to *any* gear in the input side of a shaft. The "#" is just a column separator.

factor # QCGB
1.875 # 4
1.666 # 4.5
1.5 # 5
1.25 # 6
0.9375 # 8
0.8333 # 9
.75 # 10
.0625 # 12
.375 # 20

(There are more but this should do the trick)

So find an input gear in your gear train, count the teeth, and multiply it by the factor in that table. If the result is an integer, the QCGB column shows where to what setting to use.

example: Say you have a 32 TPI input gear. 32 * .9375 (which is 15/16) = 30. Replacing this gear with a 30 tooth gear will carry the day if you put the QCGB on 8.

Finally, you can invert the table and apply it to an output gear, too. That is, a gear on the "leadscrew" side of the power train. But again, it can't be a gear used for transmitting power - it has to be on the output end of a spinning shaft.

In this case, divide the gear's TPI by the number in the 'factor' column above.

For example, say you have a 50 TPI gear on the output. 50 / 1.25 = 40. So by replacing this gear with a 40 TPI gear and putting the QCGB on "6" should accomplish the goal.

Paul Alciatore
02-22-2009, 12:44 PM
Pictures of the external gear train and the threading chart always help in these questions.

The two gears on a lever are probably a tumbler reverse. Look for these features:

1. One of these gears is always in mesh with either the input or output end of this pair.

2. These two gears are always in mesh with each other.

3. The lever allows you to choose which of these gears is in mesh with the other side of the gear train (output or input end of the pair).

If all three of these features are present, it is a tumbler reverse.

As for the tooth count, the gears in a tumbler reverse are idlers. Each idler in a chain will reverse the direction of rotation, but WILL NOT change the speed of the final, output gear. In such a chain, the speed of the output gear is determined by the tooth count of the input gear (spindle) and of the output gear (on QC box). Changes in speed in such a chain are made by changing the first (input) or last (output) gears or by using compound gears (two locked gears on one shaft) inbetween them. Any number of idlers with any tooth count can be inserted with no change in the final speed.

For cutting different threads, without seeing actual pictures, I can say the general procedure is like this:

1. It sounds like you have some kind of compound gear that is involved with English/metric conversion. Since you have English threads on the QC box and want another English thread that is not there, leave this setup alone.

2. Observe the tooth count of the spindle and of the final gear which is mounted on the QC box. The ratio of these two gears is what you are working with. Lets write it as Sp/QC or 50/40.

3. Now, pick a thread on the QC box that looks promising. Tony suggested the 9 TPI setting so lets use that.

4. Since the desired 7.5 TPI is coarser than 9 TPI, the lead screw will need to turn FASTER. To do this, either the 50 tooth will have to be increased, or the 40 tooth will need to be decreased. The ratio of these two threads is 9/7.5 = 1.2. This is the factor we would apply to the find the INCREASE needed for the 50 tooth gear if we elect to do it that way. 50 X 1.2 = 60. So changing the 50 tooth spindle gear to a 60 tooth and leaving the rest of the chain alone the 9 TPI position will cut 7.5 TPI threads.

OR, we can reverse the ratio: 7.5/9 = 5/6 = 0.8333333333... It is best to keep this as 5/6. This is the factor we would apply to find the DECREASE in the tooth count of the 40 tooth gear if we elect to do it that way. 40 X 5/6 = 33 1/3 teeth. Well, 1/3 of a tooth is hard to cut so that won't work.

So, you either have to change the gear at the spindle end or find another position on the QC box to work with. This way, it is trial and error until you find a solution that works.

I like to use Excel spreadsheets to do the above:

First, I create a couple of cells at the top for entering new gears: one for the spindle gear (lets say in cell C3) and another for the QC Input gear (lets say in cell C4). You can enter 50 and 40 in those cells for now.

Then I enter all the threads available on the QC box in the first column below that.

The next (2nd) column will have a formula that looks like this:

The third column will have a formula that looks like this:

These two formulae should be copied down the columns to correspond to all the values in the first column.

The fourth column will give us the converted values for each position of the QC box. Here we use what Excel calls relative references so on the first line of the table we enter an equation that refers to the cell in the first column on that row. So if the first row is row 6, then our formula would look like this:

This formula will convert the value in the first column to whatever new TPI you would get if you use the new gears entered in the cells C3 and C4 above the table. This formula can then be copied down all the rows in the table. Bu using the copy function in Excel, and not adding the dollar signs in the cell references, Excel will adjust all the formulae to refer to the values in the row each copy of the formula is in. Thus, the second row of the table will have the formula: =A7*B7*C7 etc.

Now you can enter whatever gear values you want to try in cells C3 and C4 and instantly see all the possible new threads in the fourth column below. This makes it easy to see what you can do with the gears you already have and also see all the new threads you can cut with new gears.

The same thing would work for metric threads, but metric threads are expressed in lead, not "threads per". So, the ratios would need to be inverted in the above logic.

I suggested cells C3 and C4 to allow room for labels in column B or A. I would do this. Likewise, the columns of the table should be labeled. If someone is using this procedure for another lathe, then the values of 50 and 40 will need to be changed to whatever values are actually used in your gear train. Please note that this includes changing then in all the equations in the second and third columns. This should work for any lathe with a QC box. It is interesting to note that the actual TPI value of the lead screw does not even enter in the above calculations. It is already incorporated in the values of the QC box.

Lew Hartswick
02-22-2009, 09:13 PM
OK I think I have the gist of it but here is a sketch of the layout.
From what I gather the actual input to the QC box is only determined
by the gear on the spindle and the one on the shaft going into the
box. ie. 50 to 40 all the others just "go round and round". So the
object is to make the 40 tooth one biger (slowes the shaft) and use
a higher tpi on the chart, Or smaller and use a lower tpi.
The ratio would be the ratio of the 40 to the new gear.
ie. if the 40 tooth is reduced to 30 then at 10tpi it would be doing 7-1/2.

Is that right? Or do I have it reversed?
If its reversed then maybe change the 40 to 48 and use 9tpi. ????

Now as to the spread sheet work I havent used any at all so need
to get some help on how to use but would like to be able to do
what Paul is talking about.
Thanks again.

02-22-2009, 10:31 PM
According to the manual for the Jet 1024P, it will cut 40 types of inch threads from 4 to 112 and 24 types of metric threads from 0.35 to 4.5.


Steve Steven
02-22-2009, 10:43 PM
I may be wrong here, but I think your lathe may be set to cut metric threads. I think the 127 tooth gear is for metric gears, the 120 for English ones. Try cutting a shallow thread and check to see it cuts what its supposed to using a thread gage or a known screw thread.


02-22-2009, 11:06 PM

The setup as drawn is for inch threads. To do metric threading the 120/127 combo gear is driven on one gear and the output side is driven off the other gear thereby giving you a 127/120 ratio. For example to cut a 1mm thread pitch you would drive the 120 gear from the spindle and drive the change box from the 127 gear then set the change box for a 24 thread per inch. The result will be 127/120=1.058333 then 1.058333 * 24tpi = 25.4tpi or 1mm pitch.

In the above drawing it does not matter which of the 120/127 gear is used as long as both input and output are on the same gear for inch threads. My LeBlond has the same double gear in the gear train and I use the 127 side as an idler because the 120 gear has a couple of broken teeth and the tic tic tic drives me nuts and I don't want to create any more wear on the broken gear than necessary. I only use it if I am cutting metric threads which is fairly rare for me.


02-23-2009, 12:02 AM
The 21 and 26 gears running together are for changing the output ratio of the gear setup, not reversing the gear box. No matter which gear is against the spindle gear the other gear will turn the right direction. All it does is change the output ratio.

rdfeil has it right, the 120/127 gear in his setup is used as an idler to keep the output gear turning the same direction as the input gear. If you want metric you have to use 120 gear and the 127 gear to get the differential ratio.

I was the one that was asking all the questions about trying to get more thead numbers to use.

Paul may have just answered the reason why I can't get my Excel program to work for me. I will try his method. It has to be faster than pencil, paper and calculator.

tony ennis
02-23-2009, 12:36 AM
You sure about the 21 and 26 tooth gears changing the ratio, Carl? They look like standard tumbler gears to me...

Paul Alciatore
02-23-2009, 03:30 AM
You sure about the 21 and 26 tooth gears changing the ratio, Carl? They look like standard tumbler gears to me...

I totally agree. As they are drawn, they look just like a standard tumbler reverse. They are drawn as idlers and as such they can not change the ratio.

I wasn't aware of the additional shaft with a double gear on it (50/40). This may be the opening for an additional range of threads. Is there any way of using this as a compound? Are they locked together in some manner? Like a key or some screws? If so, can the 40/50 be turned over so that the 40 becomes the input gear of the pair and the 50 the output? Then, the 127/100 would be turned over so that the 127 meshes with the 50 which is now on the outside. Of course, the 40 on the QC Box would have have a spacer under it to allow it to continue to mesh with the 127. If this is possible, it will speed up the lead screw by a ratio of 5/4 and all of the threads will be increased in TPI by that factor. So a 6 TPI setting will produce 7.5 TPI.

Steve Steven
02-23-2009, 08:05 AM
Rdfeil, thanks for clearing that up for me, I understand what you wrote.


02-23-2009, 08:32 AM
Well, Lew said there is a lever that rocks the gears back and forth and if they rock back and forth switching the contact of the 21 or 26 on the spindle gear that would change the ratio. Without a photo of the end of the lathe it's hard to really tell what is happening. Assuming that when the gears are moved by the lever they switch contact with the spindle and driven gear.

J Tiers
02-23-2009, 08:43 AM
Well, Lew said there is a lever that rocks the gears back and forth and if they rock back and forth switching the contact of the 21 or 26 on the spindle gear that would change the ratio.


The only way you can change the ratio is to have two gears coupled on one shaft, like the 40 and 50 on that machine. Then one gear turns once, and 40 teeth go by, while the OTHER gear makes 50 teeth go by per turn.....

Otherwise, no matter what the gears are, it's "one tooth in and one tooth out". So no matter how many are in between, if the driver has 20 and the driven has 40, the driver turns twice while the driven turns once.

02-23-2009, 08:52 AM
I disagree, if you have the 21 tooth running on the spindle and the 26 on the driven gear the ratio will be one number. If you switch the gears with a lever and now the 26 gear is running on the spindle and the 21 on the driven gear it will be a different ratio.

21/26=.8076923 and if you reverse the position of the gears with the lever the ration will now be 26/21=1.2380952.

At present we/I are assuming the lever switches the position/contact of the gears on the spindle gear and the driven gear. If in fact it does that the the gear set is a ratio change not reversing.

The other fact that it can't be a reversing gear set is if the spindle is turning clock wise the 21 gear would turn ccw and the 26 cw. Now if the lever rocks the gear set then the spindle turning cw would turn the 26 ccw and the 21 cw. So no matter which gear is in contact with the spindle the third gear will always turn the same direction as the spindle.

02-23-2009, 08:24 PM
If three gears are in a train, only the first and last determine the ratio. The intermediate idler has no affect on the ratio. Even if there are a dozen idlers, they will not affect the final ratio.

J. Randall
02-23-2009, 08:32 PM
Carld, you need to rethink this. I agree with the others, idlers make no difference, only the first and last gears in a train affect the ratio, unless there is a gear compounded in the middle of the train.

02-24-2009, 01:43 AM
:o well, I never claimed to be the brightest nut in the box :D . I went to several sites and this is the method they all gave for determining final ratio of a series of gears. I was not looking at the whole train and I should have. My mistake, sorry 'bout that.

What I am wondering now is why does it have a lever to shift the two gears if the outcome is the same? For all practical purposes the ratio between the two 50T gears is 1:1.

Lew, I sure would like to see a photo of the actual gear train and how the lever works.


Paul Alciatore
02-24-2009, 02:19 AM

Read my description of a tumbler reverse way above in this thread.

Lew Hartswick
02-24-2009, 09:14 AM
Carld, I thnk when the lever swings up it rmoves the first gear and
connects the second one so only one is engaged so that removes
one "reverse" rotation and provides the required reverse of the
output to the QC box.
I can't get a pix since I did the sketch on Sat and won't be able to
get at the lathe again for quite a bit.
I did a little "investigating" yesterday and found on the Clausing/Metosas
at school, a metric thread of 3.375mm which comes out to "so close"
it dosent matter" to 7 1/2 tpi. In an inch (length of the nut) an error
of a little more than .003 inches. I think close enough. So I'm
going to try one and see.
In the long run though I think it would be nice to be able to have
the friends lathe workable for him to do the peices.
Thanks everyone.

02-24-2009, 10:11 AM
Paul, I just reread your post about the two gears. I did a lot of searching and math last night. Using the method to find the final output ratio and using that it's true that no matter what tooth count the two idler gears are the 50T top gear will turn once and the bottom 50T gear will turn .9999999 times. I tried several different tooth counts for the two gears and it always came out .9999999 UNLESS the two idler gears have the same number of teeth. When they have the same number of teeth the ratio it exactly 1:1.

As far as moving the gears to reverse the rotation of the bottom gear it won't happen. If the top 50T is turning CW looking at the gears then the second gear will turn CCW and the third gear CW and the last 50T CCW. If you move the lever and swap the gears (if that's what really happens) then you still have a four gear train. and the results will be the same.

With the two idler gears in constant mesh and just swapping positions there is no reversal of the bottom 50T gear. It will always turn CCW when the spindle gear is turning CW.

Lew, you have to have the two gears in the train to drive the bottom 50T gear the way you drew the gear train. The only way you could have one gear in the train at a time is if they slide back and forth putting one gear at a time between the two 50T gears. Your drawing shows the 21T and 26T in constant mesh engaged between the two 50T gears. Even if the lever put one gear at a time between the two 50T gears it would not reverse the output gear.

tony ennis
02-24-2009, 08:32 PM
With the two idler gears in constant mesh and just swapping positions there is no reversal of the bottom 50T gear. It will always turn CCW when the spindle gear is turning CW.

These tumblers have 3 positions - forward, reverse, and neutral. In one position, the larger gear is jammed into the gap between the spindle and the first real idler in the gear train. The smaller gear spins uselessly. In another position, one gear engages the spindle and the other engages fhr gear train. This gives you the reverse effect. In the 3rd position, the gear train is disconnected from the spindle.

My Craftsman is apart, lemme see if I can get enough of it together for a picture...

tony ennis
02-24-2009, 08:36 PM
See page 45 here (http://books.google.com/books?id=iD5IAAAAIAAJ&pg=PA43&lpg=PA43&dq=lathe+tumbler+gear&source=bl&ots=OBhon5NjPv&sig=rqypGPfEWIYRj3N3VNA9SXc1Lyc&hl=en&ei=u5-kSezVHYzRnQfRv-2fBQ&sa=X&oi=book_result&resnum=3&ct=result#PPA45,M1)

Roy Andrews
02-25-2009, 09:32 AM
while i agree that the tumbler is most likely for reverse it also might be so that the 40/50 tooth gear can be flipped over. one or both of those double gears flip so that you can have a ratio change. maybe the lever makes the adjustment. without a picture or a better drawing it is hard to tell.

Lew Hartswick
02-25-2009, 09:35 AM
See page 45 here (http://books.google.com/books?id=iD5IAAAAIAAJ&pg=PA43&lpg=PA43&dq=lathe+tumbler+gear&source=bl&ots=OBhon5NjPv&sig=rqypGPfEWIYRj3N3VNA9SXc1Lyc&hl=en&ei=u5-kSezVHYzRnQfRv-2fBQ&sa=X&oi=book_result&resnum=3&ct=result#PPA45,M1)
Yep. Tony that's exactly it.
I hope that explains it to the rest.

02-25-2009, 11:00 AM
Yep, that would work for reversing as going from one gear to two gears will reverse the output gear. My Logan uses three gears to do it as shown in the photo below.