South Bend 13" Metric gears

[GordonL]

I have been through the the posts on the metric transposing gears for the 13" double tumbler lathe with 6TPI leadscrew and I am not really finding the answer. The factory set used a 120:127 set but this is huge and almost impossible to obtain. There is someone selling a set of 3D printed gears on eBay but it does not appear to be 120:127 and he states that it is not exact so I assume that it is some other set. The 37:47 or 63:80 work on the other lathes with 8TPI leadscrew. Any idea what the gears that he is selling may be? I have been trying to figure this out until I am seeing double or triple and have not come close to an answer.

SOUTH BEND 13 Metal Lathe, METRIC Thread Cutting Gear Set! | eBay

SOUTH BEND 13 LATHE Double Tumbler 6TPI. You will NOT be. SOUTH BEND LATHE 13" (Using Double Tumbler QCGB). Maybe better than your lathe and slightly worn lead screw can produce! See below Chart/Table.

www.ebay.com

 

[GenePoole]

The pictures in the listing show different gears. One shows a 63/80, but the one that is attached to a CL145 South Bend (and shown in the video) are different, but don't have the tooth count embossed. He claims it works with the stock SB chart and the chart shown in his description jibes with my genuine SB chart.

The seller is proud enough of his product to guarantee it so what do you have to lose? I guess it depends on how accurately you need to make your threads. I would (and did) opt for the genuine SB set. I believe Ted (@SBLatheman ) still has some of these and to me they'd be worth the extra cost.

 

[GordonL]

My biggest problem is that I very seldom need to cut metric threads. If I can just 3D print a couple of gears I may do that but it is not worth $200 to me. At this point it has kind of become a challenge to see if I can figure it out. I suppose that if it was that simple others would have already done it.

 

[GenePoole]

If you have a 3D printer give it a try. I was able to make a few gears for oddball threads I needed to do, but never tried a full set. Does your CAD have an involute gear workbench? Makes it pretty easy.

 

[GordonL]

I have a printer and have FreeCad so that part is covered. The problem is figuring out which gears that I need. After some more thinking I think that if I duplicate the gear train from the 9 or 10 it should work. The 9 and 10 have a 8 TPI leadscrew but that should not matter. If the QC gear box is set for 16 TPI the carriage should advance 1/16" for every revolution of the headstock so duplicating the gear train should be the same. I think.

 

[GordonL]

I have a printer and have FreeCad so that part is covered. The problem is figuring out which gears that I need. After some more thinking I think that if I duplicate the gear train from the 9 or 10 it should work. The 9 and 10 have a 8 TPI leadscrew but that should not matter. If the QC gear box is set for 16 TPI the carriage should advance 1/16" for every revolution of the headstock so duplicating the gear train should be the same. I think.

Not true. The internal ratio in the gear box is different so headstock to screw gear has to be different. Back to square one. What I have to try to determine is the internal ratio of both the 9" & 10" and the 13" gearbox.

 

[GordonL]

If you have a 3D printer give it a try. I was able to make a few gears for oddball threads I needed to do, but never tried a full set. Does your CAD have an involute gear workbench? Makes it pretty easy.

It looks like you have the 127:120 set. I was hoping to get it to work with one of the other setups, 47:37 or 80:63. So far I have not found the right combination. Lots of folks have done it with the 9" and 10" but the 13" seems to be in another category with the 6 TPI leadscrew. I found another place where the guy had figured out a close setting by just changing the stud gear. He got all of the common threads very close with just the gearbox setting and the stud gear change.

Also your lathe looks like mine looked when I got it. Mine came out of a school and apparently every time some student got out of line the instructor had him put another coat of red paint on the lathe,

 

[GenePoole]

I've never seen another red one before. This one was red because it came from a company that makes fire hydrants, Kupferle Foundary (https://www.hydrants.com). My mentor was a 35+ year journeyman that went to work there as shop supervisor after the shop we worked for shut down. He got me that lathe before they scrapped it. He died shortly after so I keep it Fire Hydrant Red in his memory.

 

[GenePoole]

I've not looked into it as I have the genuine article, but the main thing is to get a 2.54:1.00 ratio in the gear train. 127 is the smallest to get the exact ratio since it is a prime factor of 254, but 80:63 gives 1.2698 which should do the job. The ratio difference should be 1.2:1 when calculating pitch correction. I'd have to do a bit more math than I want to right now, but I would think that if you wanted to say make a 1.0mm pitch thread with an 80/63 gear, you could use the SB chart setting for 1.2mm pitch ( 48T stud, C, and 3 ).

 

[johnoder]

If you think less about the QC and more about the transposing it can be revealing

Say the lathe is set to cut an 8 pitch thread. That is .125"
The transposing set just changes this to a 2.5mm pitch thread - because ,125" divided by 1.27 = .098425197" or 2.5mm pitch

The various transposing sets (like 80 and 63) just make this more or less accurate - 127 and 100 render it perfect

Monarch did the end gearing this way - and I'll guess that was a 4 pitch screw

 

[kitno455]

The real trick is to use a 34 tooth stud gear instead of the 24 or 48. That makes your 6 tpi screw act like a 4.235 tpi screw, which happens to be 5.9972 mm per turn. That's close enough to 6mm to make good fasteners. That will let you cut 11 common metric pitches with less than 0.05% error (1 part per 2000). A smattering of other pitches can be made with a few other gears, with slightly worse error. The usual warning about leaving the half-nuts engaged applies with all of these, BTW.

Gearbox, MM, %Err
144 0.25 0.046
80 0.45 0.046
72 0.50 0.046
48 0.75 0.046
40 0.90 0.046
36 1.00 0.046
24 1.50 0.046
18 2.00 0.046
12 3.00 0.046
9 4.00 0.046
8 4.50 0.046

Also, you can pick up some of the other common screw threads if you add a couple more gears to the mix:

If you replace the stud gear with a 43, the leadscrew will act like it is 7.6mm, which does not sound useful. But with the ratios in the gearbox, you can get the following:

Gearbox, MM, %Err
26 1.75 0.019
13 3.50 0.019

And, if you put the 24 tooth stud gear back, and instead replace the 64 tooth screw gear with a 50 tooth, you get the following:

Gearbox, MM, %Err
26 1.25 0.037
13 2.50 0.037

There are some other, finer common threads, but I doubt you will be cutting them on an SB 13. If you have some other pitch you need, let me know.

allan

 

[GordonL]

Thank you to all who have posted solutions but when I plug those gears into my spreadsheet they do not come out right. I am not sure where things go wrong but apparently there is some difference in either the internal gearing in the QC gearbox or the gear train in front of the gearbox. I have not been able to find the value of the gear train on the 9" or 10" lathe. The 13" has stud gear of 24 tooth, idler gear of 80 tooth and shaft gear of 64 tooth. If the 9 or 10 has the same gear train the result should be the same if either gear box is set to the same TPI. The internal gear box ratio should be set to compensate for the 6TPI or 8TPI leadscrew. The internal gearing should just be a 6 to 8 ratio. Am I missing something?

If I look at the SB chart for the 10" and the 13" the settings are all the same except that the 10" shows a 56 tooth shaft gear and the 13" shows a 64 tooth shaft gear. The input ratio is not the same.

 

[GenePoole]

The real trick is to use a 34 tooth stud gear instead of the 24 or 48. That makes your 6 tpi screw act like a 4.235 tpi screw, which happens to be 5.9972 mm per turn. That's close enough to 6mm to make good fasteners. That will let you cut 11 common metric pitches with less than 0.05% error (1 part per 2000). A smattering of other pitches can be made with a few other gears, with slightly worse error. The usual warning about leaving the half-nuts engaged applies with all of these, BTW.

Gearbox, MM, %Err
144 0.25 0.046
80 0.45 0.046
72 0.50 0.046
48 0.75 0.046
40 0.90 0.046
36 1.00 0.046
24 1.50 0.046
18 2.00 0.046
12 3.00 0.046
9 4.00 0.046
8 4.50 0.046

Also, you can pick up some of the other common screw threads if you add a couple more gears to the mix:

If you replace the stud gear with a 43, the leadscrew will act like it is 7.6mm, which does not sound useful. But with the ratios in the gearbox, you can get the following:

Gearbox, MM, %Err
26 1.75 0.019
13 3.50 0.019

And, if you put the 24 tooth stud gear back, and instead replace the 64 tooth screw gear with a 50 tooth, you get the following:

Gearbox, MM, %Err
26 1.25 0.037
13 2.50 0.037

There are some other, finer common threads, but I doubt you will be cutting them on an SB 13. If you have some other pitch you need, let me know.

allan

This is pretty amazing! I made a spreadsheet (to check my assertion above about the 127/120 v. 127/100 and a 1.2 conversion factor) and plugged in a 34 tooth stud gear and this works rather well. I'm surprised it isn't seen in searches on this topic. A single gear change to a stock SB13 (CL145) and you get some really close metric threads.

 

[GordonL]

This is pretty amazing! I made a spreadsheet (to check my assertion above about the 127/120 v. 127/100 and a 1.2 conversion factor) and plugged in a 34 tooth stud gear and this works rather well. I'm surprised it isn't seen in searches on this topic. A single gear change to a stock SB13 (CL145) and you get some really close metric threads.

That is what I have been trying to do without much success. I am not a spreadsheet expert by anyone's standard. You saved it as a ODS file which opens in my 2007 Excel but none of the formulas work. Any chance of getting it in XLS format?

Thank you: Gordon

Just discovered that the ODS format is OpenOffice and it is not compatible with Microsoft Excel. I downloaded OpenOffice and I can open it with that. Now I just have to figure out hoe to work with it.

 

[GenePoole]

That is what I have been trying to do without much success. I am not a spreadsheet expert by anyone's standard. You saved it as a ODS file which opens in my 2007 Excel but none of the formulas work. Any chance of getting it in XLS format?

Thank you: Gordon

Just discovered that the ODS format is OpenOffice and it is not compatible with Microsoft Excel. I downloaded OpenOffice and I can open it with that. Now I just have to figure out hoe to work with it.

I just threw this together in a few minutes last night. I didn't get fancy with graphics or anything.

The four (or five) fields in row five are the only ones you need to mess with. The standard CL145 has a 24 tooth stud gear, and 80 tooth idler and a 64 tooth QCGB drive gear. I have two for the idler so you can enter transposing gear tooth counts such as 127 and 120 (use 80 in both fields for a standard setup). That's it! The rest is just a depiction of the QCGB settings roughly just like the plate on your lathe. The bottom depiction is in mm. If you put 127/120 in the idler fields and use the stud gears from the standard chart, the lower depiction of the QCGB will show the same values as the metric pitch chart. Below that is just the plunger ratios for the QCGB and should be hidden as they are just refernce values used to build the chart depictions.

I could have put in the "FEEDS IN THOUSANDTHS" as depicted on the original plate as well, but I'm not sure what the ratio is of the worm gear to drive gear in the apron is (seems to work out to about 2.97)

 

[GordonL]

Thanks. That works just fine. It took me a while to figure out that the A1, A2 etc corresponded to the gearbox settings. The Imperial settings mean nothing in the real world once you enter a variable in line 5.

 

[GenePoole]

I've been studying up on this problem and I see that Grizzly Industrial sells replacement gears for their machines at reasonable prices. Looking at the G0824 lathe parts list, there's not much info but I'd guess that these are mod1.5 metric gears, but that's close enough to 16 DP to fit under the cover. You'd need all the gears and have to make some adapter bushings, but way more economical than a genuine SB set. Might be worth looking into.

127T
https://www.grizzly.com/products/P08240207

All G0824 parts:

Parts for 14" x 40" Gunsmith Lathe with 2" Spindle Bore at Grizzly.com

Grizzly Industrial, Inc. is a national retail and internet company providing a wide variety of high-quality woodworking and metalworking machinery, power tools, hand tools and accessories. By selling directly to end users we provide the best quality products at the best price to professionals...

www.grizzly.com

 

[GordonL]

Thanks to everyone for the help. I ended up 3D printing three gears 34T, 43T and 50T. That gets me the most common metric threads and that is close enough for my use. I am not building space ships.

 

[sparkelectronics]

Thanks to everyone for the help. I ended up 3D printing three gears 34T, 43T and 50T. That gets me the most common metric threads and that is close enough for my use. I am not building space ships.

Would you be willing to share your STL files?

 

[GordonL]

No problem except that this forum does not allow attaching STL files. If you send me your email address in a private message I can email them to you. I can also send you the STEP files if you want to modify them and also the chart that I made showing the gear locations for the different metric pitches.