11.5 TPI

[WobblyHand]

A good question.

It would almost certainly depend on the fit and the length of the female part.

I imagine it wouldn't be an issue on a 1a/1b nut bolt assembly.

But would likely bind on a 3a/3b assembly, especially if the female part is long.

Even a 1a/1b would bind if long enough. Really wondering on a 2a/2b how many threads it would take. Are we talking 3? 10? 15? If 100 threads, I'm not going to worry about it.

 

[Ken226]

Even a 1a/1b would bind if long enough. Really wondering on a 2a/2b how many threads it would take. Are we talking 3? 10? 15? If 100 threads, I'm not going to worry about it.

A few years ago I 3d printed a batch of gears for my lathe, to allow for alot of weird non-standard thread pitches. Like that m16 x 19tpi Whitworth thread on that old locomotive whistle I posted in the Alibre CAD subforum.

In doing so, I created a spreadsheet that calculated all of the ratios in my QCGB and realized that many of the metric pitches are close approximations anyway, even using my OEM change gears.

Most of us have been cutting threads that are close approximations for years without knowing anyway.

Without searching my old removable drives for that spreadsheet, I'd wager that some of the pitches we cut with the oem gears are off by at least the same as his 11.56tpi setup.

 

[WobblyHand]

I know at least on my mini lathe LMS published the pitch error. I don't recall if Grizzly did the same for my 10x22.

I made my own ELS for my 10x 22. I can hit most of the threads exactly, so the accumulated error is zero. Well maybe not a pi mm thread, but practically all other threads have zero accumulated error. (I'm off by 0.0002% for that thread.) But it is still interesting to think about pitch error and what it means in a practical sense. If I make parts for my Grizzly using my mini lathe, I do need to know under what conditions I need to be concerned. It's a math problem that confuses me more than it should.

 

[Ken226]

Edited to correct an error:

The difference between 11.5 tpi and 11.56tpi is an additional .06 threads per inch. .06 times the pitch (.06x.0869) is .005" difference per inch of threads.

 

[Just for fun]

I look forward to hearing if my proposed gear settings generates the calculated 11.5175 TPI value, although measuring anything more accurate than 11.5 maybe be very difficult.

Dave L.

Dave,

I did not come up with 11.5175 TPI with your gearing options. Here is what I had. 30-127 and 120-32 Norton box at B7, looks more like 15 TPI.

 

[Ken226]

Dave,

I did not come up with 11.5175 TPI with your gearing options. Here is what I had. 30-127 and 120-32 Norton box at B7, looks more like 15 TPI.

View attachment 421258

View attachment 421259

Doing the math on that gear combo, 30/127 * 120/32 * .125 * .6153848 = .068141 pitch, hence (14.676)


That combination shows B7 should produce 14.676 TPI, which is so close to 15 that the 15 pitch gauge matches up.

 

[Ken226]

@davidpbest

I opened up your spreadsheet from a few posts back to see if i could figure out why it isn't giving the pitches that Just for funi s seeing when he actually cuts the threads.

It appears that the formulas in the TPI row are the issue. As they are now, they multiply the gear ratio times the thread TPI count. To get an accurate pitch, the gear ratio needs to be multiplied with the actual pitch, then take 1 divided by the gear ratio*pitch product, to convert the pitch back to TPI.

here is an example. For 40/127 * 120/46, in B3 the current formula shown is F15 * R3, which gives 7.806 TPI, but the OP is seeing an actual 11.56 tpi using that setup. The formula for that cell should be 1/ (f15*R4). The same typo appears to exist in all of the TPI cells.

When I edit that cell to show =1/(F15*R4) it gives the correct 11.562 TPI.



The metric formulas are currently correct and shouldn't need any adjustment.

I went through and pasted the formula adjustments into all of the TPI cells. I'm using Openoffice, but tried to save it using the current excel format. Hopefully it still works.

@Just for fun
Just to check my work, does the data in this adjusted spreadsheet now match what you've seen so far in test cutting threads?








Edited to add: Just wanted to point out, according to the adjusted spreadsheet, 35/127 * 120/40 in B3 gets you alot closer to a true 11.5tpi (11.49) tpi) than the other setting.

 

[Just for fun]

Yes Ken, That looks right to me.

Tim

 

[davidpbest]

I opened up your spreadsheet from a few posts back to see if i could figure out why it isn't giving the pitches that Just for funi s seeing when he actually cuts the threads.

It appears that the formulas in the TPI row are the issue. As they are now, they multiply the gear ratio times the thread TPI count. To get an accurate pitch, the gear ratio needs to be multiplied with the actual pitch, then take 1 divided by the gear ratio*pitch product, to convert the pitch back to TPI.

here is an example. For 40/127 * 120/46, in B3 the current formula shown is F15 * R3, which gives 7.806 TPI, but the OP is seeing an actual 11.56 tpi using that setup. The formula for that cell should be 1/ (f15*R4). The same typo appears to exist in all of the TPI cells.

When I edit that cell to show =1/(F15*R4) it gives the correct 11.562 TPI.

@Ken226 @mksj

Ken, thanks for the "new math". LOL As I said before, my source data was provided by Mark Jacobs - I didn't check the formulas. I have updated the files on my system so if this comes up again, I can provide the corrected data. Lesson learned again: metric rules !!!

@Just for fun Tim, sorry to have misled you with the 46-tooth suggestion. Classic case of the blind leading the blind. Mea culpa!

 

[Ken226]

@Ken226 @mksj

Lesson learned again: metric rules !!!

I know the feeling.



BTW:

Not sure if this is something that you guys would be interested in or even care about, but since I had the spreadsheet already open I added a couple things that I have in the version I made for my own lathe.

In the first tab, I added dropdown selections. You can select the gear from those listed (i got the tooth counts from the "master data" tab.

When you select the gears, in the appropriate positions, It will automatically display the pitches for the selected gear combo in both the Imperial and Metric tables, for each respective QCGB lever arrangement.






Also, I added the formulas using OpenOffice Calc, so I'm not certain they'll still work when opened using excel.