Talk:Machine taper

Jacobs taper 2 1/2

I changed the description of JT 2 1/2 from "obsolete" to "obscure". I just spent a few hours looking for info on it, and it seems that, as far as anyone can tell, it's only used on cheap Taiwanese drills from the 1970s and 80s, with no spares available. Its dimensions are confusingly close to JT6, but not close enough, and the Jacobs Chuck company doesn't acknowledge it in their literature. At least to me, "obsolete" seemed to imply that if I go looking for vintage chucks, one with a JT 2 1/2 arbor mount might pop up, but I've found no indication that this is the case. It's good that the dimensions are listed here though, judging by forum posts people tend to confuse it with JT2 and JT6 — Preceding unsigned comment added by 91.157.138.44 (talk) 15:56, 22 April 2016 (UTC)

How many Morse tapers???

This article is internally inconsistant in a number of ways. In one place, it says, comes in 8 varieties, from "Morse 0" [...] to "Morse 8" (which would be 9 sizes). In another place, it says, Morse tapers come in seven sizes identified by number -- #0 through #7 (which would be 8 sizes). The table lists seven sizes, #0 through #6. Could somebody who is really familiar with machine tools please fix this? --RoySmith 17:02, 16 August 2005 (UTC)

More information on why Morse tapers of different sizes have different angles and why those angles are seemingly randomly different would be interesting. Why is it not a fixed consistent angle? and why are the bigger sizes 6 and 7 so different? — Preceding unsigned comment added by 24.64.96.165 (talk) 03:20, 11 May 2023 (UTC)

Morse Taper 4 1/2

Just to add more confusion, there's also the Morse "four and a half" taper. My understanding is that this was inserted later as a size small enough to fit inside a standard 2 1/4 inch threaded spindle head, but large enough to accommodate 5C collets. The 4 1/2 is listed e.g. here http://littlemachineshop.com/Reference/Tapers.php. JohnAspinall (talk) 20:10, 2 June 2009 (UTC)

Another thing to add

There should be a discussion of "self-holding" and "self-releasing" tapers. SASGW 18:23, 23 January 2007 (UTC)

________

Do you have any information on this matter? I have try everything at my university and only found 4 page about the subject in the hanbook of mechanical engineering calculations by Tyler G. Hicks —Preceding unsigned comment added by 70.50.252.192 (talk) 21:48, 21 February 2008 (UTC)

Self-holding tapers hold themselves in place both axially and with respect to torque. Once they have been lightly pushed or tapped into place, friction holds them stationary. (Chips or burrs can prevent them from holding properly.) Self-releasing tapers have a larger degree of taper and rely on things holding them stationary, such as a drawbar or a retention knob to pull the tool into the spindle, and tangs or driving lugs to resist the torque. They are the taper of choice for automatic toolchanging with indexable toolholders. When the machine is ready to change the tool it stops pulling on the retention knob, allowing the taper to be released. A blast of air keeps any chips off of the taper surfaces during the toolchange. — ¾-10 02:12, 22 February 2008 (UTC)

Strange angle measurments

"16 degrees 15 minutes 37 seconds" (refrencing the NMTB taper angle)? I've never seen minutes used in measuring angles, outside of 19th century navigation...

Replaced the NMTB taper information with the (correct) decimal equivalent.

Sadly, engineers still use minutes and seconds.Kallog (talk) 07:51, 4 June 2009 (UTC)

Finding the angle for 3.5 per 12: wrong and right

Wrong

To the anon who has twice recently changed the degree measurement of the NMTB tapers (3.500 inches per foot) from 16.2602047... degrees (correct) to 16.59428994... degrees (incorrect): Please note that tan 16.2602047° = 0.2916666 = 3.5/12. You can check this for yourself using Google.com (Google Calculator). Please note that when you type tan(n) into Google, it assumes by default that n is in radians. 16.2602047 degrees = 0.283794109 radians. tan(0.283794109) [radians] = 0.291666666 = 3.5 ÷ 12. Thanks. — ¾-10 02:18, 13 November 2007 (UTC)

Right

Please note that the correct way to calculate the angle is:

The taper is the change in the size of the diameter as you travel down the axis of the taper. So, take 1/2 the change size (1.75") and divide that by 12", arctan of the result is 1/2 the included angle. So 16.5942899... is correct. —Preceding unsigned comment added by SamELLI (talk • contribs) 17:45, 29 April 2008 (UTC)

D'oh! I can't believe I made that obvious mistake. 1,000 apologies. Thank you for explaining the correct calculation. — ¾-10 03:12, 30 April 2008 (UTC)

Advantages/disadvantages of tapers?

Could someone please clarify the advantages or disadvantages of tapers when compared to say, chucks or collets? Perhaps I just didn't get it, but to me it seems like the article states that tapers are outdated, and are being replaced by chucks and collets... —Preceding unsigned comment added by 24.45.0.133 (talk) 17:51, 15 January 2008 (UTC)

Not true at all (i.e., that tapers are outdated, and are being replaced by chucks and collets). Tapers are not going anywhere, because of their ability to provide both (1) repeatably high concentricity with successive tool changes and (2) easy CNC toolchanging with repeatable tool length offset. In fact many CNC toolholders have both a taper (to interface with the spindle) and a collet (to interface with the tool). Neither of those components is going to replace the other. It is hard to explain this if the audience is not familiar with how CNC mills do automatic tool changes. I lack the time to tackle it now, but hopefully will make time to try it in future. This may be something that is very hard to teach via Wikipedia alone. But I can imagine some diagrams that could help. — ¾-10 03:13, 16 January 2008 (UTC)

I should qualify my comments. It is true that, specifically regarding new manual machine tools, such as the average new manual benchtop drill press or new manual vertical mill, you are not going to be doing your tool changes by drifting out one Brown-&-Sharpe-taper-shanked drill bit or milling cutter and tapping in another. There is going to be either a drill chuck or an R8 collet setup mounted semi-permanently in the spindle. (The drill chuck itself will usually be held in the spindle by a self-holding taper). Tool changes will not involve unseating that taper; they'll be done with the chuck or collet. So in that limited respect, people are right when they say that a milling cutter with a taper shank is old-fashioned. However, that drill chuck is still held by a taper, even if you go years without unseating it; and when you change contexts and talk about CNC mills with automatic toolchanging, tapers are still being seated and unseated on a daily basis. (In this case, self-releasing tapers, 3.5/12.) The fundamental reason is that tapers provide indexability—i.e., you can separate a matched pair of tapers and put them back together again, and the geometry of the two parts is back to the same place it was last time. (And in more than one axis—both concentricity and tool length offset). As I said earlier, this is easier to understand than it is to explain. I will try to improve Wikipedia's coverage of these concepts when I get a chance. — ¾-10 15:37, 16 January 2008 (UTC)

R8 collets

The section on R8 collets is somewhat confusing, and imho wrong. A machine with an R8 collet system (usually a manual Bridgeport style vertical mill, aiui) normally has a set of collets of varying sizes. To use a 1/2" shank end mill, you install the 1/2" collet, insert the end mill in the collet, and tighten the draw bar to retain the end mill. To use a different size tool, you use a different size collet. Of course, you can also get tools that have an R8 shank, and are installed directly into the spindle, or an R8 shank toolholder that isn't a collet. (I have a face mill of the former type, and a 1" set screw style end mill holder of the latter for the machines at work.) And then there are taper adapters, for eg holding a drill chuck in an R8 spindle machine. If someone more knowledgeable than I wants to tackle a rewrite, that would be great. Otherwise, I'll get to it fairly soon. (Also, the whole section needs references.) Evand (talk) 15:12, 8 September 2010 (UTC)

Design questions

History? could some one please explain how morse discovered the correct taper?…was it a happy accident or was it born on an engineers drawing?where does stiction begin if that is what it is? 45 degrees plus? — Preceding unsigned comment added by 124.182.139.213 (talk) 09:22, 8 November 2011 (UTC)

I would be more interested in why Morse's tapers all have different angles! I assume that he made master tapers and that these were copied, rather than the tapers made to a deliberate standard and the basic errors in these have been inherited to test the skill of machinists who make their own tapers down the ages... but it would be nice to know for sure. Stub Mandrel (talk) 20:45, 15 August 2014 (UTC)

Agree (would be interesting to know re why different angles). If he had set up a lathe with a slide rest on an angle (either by a taper attachment or custom clamped setup), he could have used the same setup to cut the exact same angle on all of his arbors and simply declared successive axial segments of the self-same cone as No. 1, No. 2, No. 3, and so on. The fact that each of his tapers has almost the same angle but different by 10 or 20 minutes of arc makes me suspect that he was attempting to cut the same angle on different setups, and the resultant tiny error between setups ended up enshrined in the standard (kind of like how the error in measuring the Earth's size ended up enshrined in the platinum 1-meter standard bar). Would be interesting to find out whether that hypothesis is true.
  Regarding the question from 124.182.139.213 in 2011, sorry this reply came about 3 years late, but here are some pieces of the answer ... the distinction between self-holding and self-releasing tapers comes much sooner in the parade of degrees (the arc of the protractor) than 45°. For example, the self-releasing NMTB/Cat taper is 3.5/12, which is only 16.5943° cone angle (8.2971°per side) (pictures available by typing "NMTB taper" into images.google.com). If one of the questions is "how did Morse know what degree of taper would produce self-holding rather than self-releasing", the answer is undoubtedly simple "cut and try". He and others had probably made countless self-holding tapers via trial and error before he got to the point of declaring his standard series. Kind of like how countless screws and nuts had been made (by Maudslay, Clement, Whitworth, and countless others unnamed) before there finally became an official Whitworth thread standard. — ¾-10 17:13, 16 August 2014 (UTC)

The boundary between self-holding and self-releasing tapers is determined by comparing the tangent of the included angle to the coefficient of friction between the male and female surfaces. King of Tea Tree (talk) 00:01, 10 March 2019 (UTC)

All tapers under drawbar load transmit a good bit of torque, although not always all of it

Looking at various sites online, one finds that some people think that because a self-releasing taper under zero drawbar load is freely releasing, the same taper under drawbar load doesn't transmit torque. They imagine that all of the torque is transmitted by the keys engaging the slots. This is not only false but also quite fanciful/imaginary. By this same misapprehension, the only thing holding any bolted joint from twisting would be the bolts pushing against the sides of their bolt holes, and nothing but magical elves keeps the disc of a friction clutch from slipping when the pressure plate is engaged. This is certainly a comical way to understand what is happening. No, the fact is that when you hold any object, with any angle of taper or no taper at all, against another one with hundreds or thousands of pounds of force, it puts up great resistance to being slid away from its current position. This is why you can't push a 1000-pound loaded crate across the garage floor, even though you can push an empty one across it. It's not the inertia of the 1000 pounds that's stopping you. Which is obvious because if it were on a cart, the wheels would let you move it. What's stopping you is the static friction of a pallet undersurface with a half a ton of force pushing it against the floor. I suppose that some people misapprehend that the only thing driving a closed 5C collet is the tiny pin that sloppily engages the keyway of the collet (which exists merely to keep the collet from spinning when it is open). I don't have time to work on this right now, but if this article is going to cite refs that fancy that an NMTB taper under drawbar load doesn't transmit torque, then that sentence needs to be prefaced by the clause, "some people mistakenly think that XYZ [ref]." The reality is that the keys are there to keep it from moving if the taper by itself supplies insufficent force (for example, the taper only resists n ft-lb, but the tool is giving it 1.5n ft-lb). — ¾-10 23:18, 15 January 2015 (UTC)

OK, I see that the article as currently edited now says "high torque". And the refs cited implicitly convey what I was getting at above. It is not that the taper transmits no torque, it is that it can only transmit so much. When I deleted the sentence last night, it just said "torque". Thanks for improving it. To follow up a bit more, I think most people would be surprised in that most drilling and milling never threaten to slip the taper. It is only the heaviest milling that does so. This is why this site is partially correct—they are correct that you don't see wear on the slots because most of the time the taper alone is transmitting the torque. But their implied idea that the keys never save it from slipping under any cutting load is false, at least on some machines. — ¾-10 23:26, 15 January 2015 (UTC)

I agree with your bolt analogy that there is always some resistance to torque. The 1918 article distinguished < 4 degree tapers as not slipping and > 10 degree tapers as slipping with the middle being in doubt. I suspect that assumes some sort of typical load. The statement also disagrees with the patent's claim that 3.5/12 is free releasing. R8s are 8 degrees, but I always had to tap them out of the spindle in my limited machining experience; the R8 also has a key slot, so it does not depend on the taper grip for everything; (maybe the R8s locked up on the key/slot). I've read about CNC mills not releasing their tools, but I forget the underlying cause. The sources also talk about breaking (I presume Morse) tangs under high loads; the tang is apparently needed to supply high torque, and that implies that Morse tapers slip. I think most people see low power machining such as small end mills used on aluminum. I dealt with some guys using a 50 HP CNC vertical machining center, and they could not get the performance they expected from it -- they estimated they got half the horsepower before the spindle stalled. I found that one of the three phases driving the spindle was dead; they got their 50 HP after it was fixed. I would not be surprised if a tool shank always runs up against the key when machining low RPM at 50 HP. I think the keys are replaceable for that reason, but I also don't expect them to be replaced often. Even if the shank slipped often under medium loads, I'm not sure that there would be much damage to the slot or key. All of the high power stuff is out of my experience. Glrx (talk) 01:04, 16 January 2015 (UTC)

Axial location.

"located both concentrically and axially by the taper" States the current lead. Is this still true?

Due to the advent of high speed machining and the high accuracy being demanded today, the axial location is, on newly developed tapers, being provided by a flange rather than expecting the accuracy of the taper to control this. HSK is an example. Spindle temperature controls were introduced, at least in part, to reduce the axial errors in tapers.

Coromant Capto, controls not only the concentricity and axial position but the angularity of the taper. It thus enables quicker tool changes. Sorry if this sounds like an advert but I think it is brilliant. Capto has an ISO standard, worth a mention? AnnaComnemna (talk) 18:10, 1 November 2015 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Machine taper. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20060208104532/http://www.woodturners.org/tech_tips/morse_taper_sizes.htm to http://www.woodturners.org/tech_tips/morse_taper_sizes.htm

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 12:53, 11 January 2018 (UTC)

Length Jacobs taper

What side of a virtual triangle is the length of a Jacobs taper? — Preceding unsigned comment added by 81.207.104.254 (talk) 19:58, 19 July 2018 (UTC)

Morse taper table

Why is the Morse taper table in millimeters?  Morse taper is inch-based and should be defined as such.

216.152.18.131 (talk) 17:17, 3 September 2020 (UTC)

I was puzzled by this, too. It appears to date nearly the creation of this article. In May 2005 DerrickOswald added the Morse taper dimensions to this article, and they were the only actual dimensions in it (the article was stubbed in March of 2005). I think there's a good argument for having both, of course — most people actually using Morse tapers work in imperial measurements, but it's desirable to be able to compare one taper to another. Probably means this needs some magic wiki stuff to allow conversion of tables and a control to select what is displayed. I'm not familiar with what's available for those templates, but I suspect they do exist... jhawkinson (talk) 20:02, 10 May 2023 (UTC)

The word "Tang" ( in the use section)

"The tang is not engineered to withstand twisting forces which..." This is the first use of the word "tang". What tang, please? Maybe someone knows? Longinus876 (talk) 20:49, 28 December 2021 (UTC)

Use of "Morse taper" in medicine & dentistry

Various implants in the medical and dental fields make use of tapers, and they often refer to these in literature as "morse". However, by the definition of the tapers per this page, these would seem to be incorrectly labeled. Many of these implant tapers have vendor-proprietary specifications, with different dimensions, angles, etc; none of which fall into the 8 specific sizes called out in this page.

Medical literature justifies in calling these tapers morse, because the fundamental mechanism used in implements specifically matches the morse design pattern -- ie the angles between the male & female components are mismatched, causing an interference fit. ( ref: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3779551/ among others. ) This is especially interesting as taper interference angles are neither mentioned on this page, nor in in any machining documentation I've come across for morse tapers. (The taper angles for Morse Machining Tapers are consistent between the male and female specifications.)

Which leaves me to question:

• Are medical implant tapers truly morse?
• If not, when and why did they start using this terminology, and should they have a separate page, or a separate section in this page?
• Should inclusion of details of the taper interference angle be documented on this page or a medical-specific implant page? The implications are useful and relevant and should be documented.

Superatrain (talk) 03:45, 5 April 2022 (UTC)

I read the article referenced as having mismatched male and female components makes no sense to me. I don't think the article says that but the pictures seem to imply it. I don't think the article is accurate on that point.

The two parts of the taper mating mechanism have to match exactly in order for the surfaces to mate along the entire length of the taper. In high horse power situations like boat shafts joining to propellers machinists use an abrasive paste on the taper to lap the taper and make the surfaces mate perfectly. The abrasive paste is applied to the taper and both parts are pressed together and turned against each other in order to make a custom and exact fit. Reference: https://www.youtube.com/watch?v=a5EbjgfklaI

To prevent the tapers from coming apart a fastener could be used to pull the male side into the female side - this is done on power hand drill chucks for example or in the referenced video above boat propeller shafts. Why don't the medical manufacturers do this? The article needs some expertise from a design person in that field. 24.64.96.165 (talk) 03:02, 12 May 2023 (UTC)

CAT 40, CAT 50 vs NMTB

Cat 40 and Cat 50 tooling is the industry standard for CNC mills - at least in North America. Close reading of the article section NMTB is required to decipher CAT40/50 is a type of NMTB. If a reader searches for CAT40 or CAT 50 nothing is found! Can this be fixed? No one says NMTB taper tooling in practice. Finally the section on NMTB does not give measurements of the various types and how they are the same or different. Pictures of the various kinds would help. These would be good improvements. 24.64.96.165 (talk) 03:03, 11 May 2023 (UTC)