Page 4 of 6 FirstFirst ... 23456 LastLast
Results 31 to 40 of 60

Thread: Metric thread dimensions :-?

  1. #31
    Join Date
    Nov 2008
    Location
    SF East Bay.
    Posts
    5,777

    Default

    Quote Originally Posted by MattiJ View Post
    And I start with nominal dia stock, ie 8,00 mm for M8 thread, keep cutting until thread crest becomes sharp and then take bit off from thread crest with file or sharp finishing insert until I get small flats on thread tops. Usually gives me tighter fit than store bought bolts.
    That should work IF you manage to sneak up on that sharp crest. What you are doing is cutting a full "height" thread (H) the full depth and then reducing the major diameter till it looks right.

    I don't recommend it because you have a problem with being too loose if you cut deeper than you expected. You also run the risk of the crest being too high as you file it down, causing it to seize against the root of the mating part.

    Why take those risks if you can use a simple formula to get the thread exactly right on the first try every time?

    Dan
    Measure twice. Cut once. Weld. Repeat.
    ( Welding solves many problems.)

  2. #32
    Join Date
    Jan 2003
    Location
    Deep in the Heart of Texas!
    Posts
    7,472

    Default

    Quote Originally Posted by MattiJ View Post
    And I start with nominal dia stock, ie 8,00 mm for M8 thread, keep cutting until thread crest becomes sharp and then take bit off from thread crest with file or sharp finishing insert until I get small flats on thread tops. Usually gives me tighter fit than store bought bolts.
    And that's just it. If you're making things for NASA or some other "lettered organization", precision is paramount. But if you're making hardware for your kids broken wagon, who cares? Who here, in their right mind, would attempt to single point .075mm threads? That's just under .003". Or about 338tpi. Why would such a thread even exist? You're not going to file the crest of those or the thread will be gone!

    You metric guys always brag about the precision then complain about the specs.

  3. #33
    Join Date
    Nov 2017
    Location
    Watford, UK
    Posts
    281

    Default

    I'm hoping to have neatly side-stepped the whole nose radius issue as I managed to find a 10mm shank holder for ER16 inserts. The ER11's are generic profile only and the maximum pitch is 3mm....which I was sure was a massive limitation at the time. Anyway, I picked up a few inserts for 1.0, 1.25, 1.5 and 1.75 which should see me through most of the threads I'm likely to need and I've got an AG60 (and an A60 if I remember correctly) for anything else.

    I'd already come to the conclusion that being able to measure PD was essential if I wanted to make a gauge-perfect thread. I suspect I won't ever need it to be that perfect...but I have real trouble shooting for 'good enough' and if I shoot for perfect but only get 'pretty good', I should be doing ok. I've already picked up some thread wires for when I get that far....just been stumbling over the issue of what OD to start at. Good to know that it isn't that critical if/when I get it wrong....but it's good to have an idea of what I should be starting with and why. Thanks for the link to the chunk of Machinery's Handbook on threads; I'll have a dig through that. I know it's something I ought to get but it's over $100 here so it'll have to wait a while yet.

  4. #34
    Join Date
    Nov 2007
    Location
    Woodinville, WA
    Posts
    8,935

    Default

    If you picked up full profile inserts (what I use unless I'm out of them, then AG60 etc.) then you won't have to worry about the "crest" issue either. Just make your OD "nominal" - like 8mm, and cut until you have the exactly the outer diameter you need as per the specification. No thread gauges.. just a mic, or calipers if you really don't care
    Last edited by lakeside53; 01-13-2018 at 12:21 PM.

  5. #35
    Join Date
    Dec 2015
    Location
    Chilliwack, BC, Canada
    Posts
    3,971

    Default

    Theory is all fine and well. But let's be honest and take a little more pragmatic approach. Who trims down their stock to some crest diameter if single pointing? If I'm making a 6mm thread and I start with 6mm rod I sure as heck do not take the time to cut it down to 5.87mm first. Because single point threading turns up a fair burr I just rely on the use of the file to dress off that burr and call 'er a day as to the crest cuts.

    And 99% of the time we are generally able to test a thread with a nut or some other part. So I tend not to use thread wires most of the time. Other times where I must match something I have certainly used thread wires. Like when I made an exact copy of the threaded spindle nose for the lathe to aid me with testing the fit for back plates and the like. For that I measured the spindle with wires and then cut the sample nose using wires to measure.

    But other than a couple of specific cases like that I'd say that 98% of my single point thread cutting uses the part that the item will go onto as a test piece for fit.


    OK, let's get back to the tapered thread. 6mm is pretty small. And of course as you cut the thread the root gets smaller and even more flexible. So a really good and really sharp tool is needed. And on something that fine very fine cuts. Or even better use of a center to aid with supporting the work.

    You also want to stick to the insert forms. That's fine but did you cut this thread with a basic triangle shaped insert? And if so can you be sure it had enough clearance on front and back to clear the spiral cut which is a thread? For smaller sizes like this perhaps you need to invest in a few specific full form thread cutters. Being shaped to the exact job they may produce less deflection forces.

    The description of the work holding indicates that you also realize there was a problem there too. If there's a very short full size part and threading behind that the very short portion can roll within the collet grip. You really need at least one diameter worth of length for a reasonable supportive grip. And 1.5x the diameter is better and twice or more ideal. Otherwise the metal can fairly easily deform and let the part wiggle around. Or at least recognize that less than 1D is going to give you some trouble and tune the DOC for each pass to some amount that produces less force on the part.

  6. #36
    Join Date
    Sep 2010
    Location
    Wheaton, Illinois
    Posts
    326

    Default

    A lot of talk about a non-problem. All my life when ever I needed an M6 or M8 or M10 etc. thread I would take an 6mm, 8mm, 10mm rod and run a die over it. Sometimes use a little emery paper to take any burr of the top. Never had a problem. I would never turn down a 6mm rod by 0.1 mm. If I am making a M24 thread and single pointing it than I would turn down the OD by 0.1 - 0.2mm. As said before, the thread is measured on the pitch dia. . You can turn the OD way down and if the pitch dia.is not right than you have a bad thread.
    Btw - this goes for imperial and metric. People who are cutting a thread - imperial or metric - and are playing around with four decimals on the OD are just wasting their time. Maybe they have nothing else to do.
    I like the words: "This is Home Shop".
    Last edited by Juergenwt; 01-14-2018 at 03:52 PM.

  7. #37
    Join Date
    Jan 2004
    Location
    Missouri
    Posts
    28,772

    Default

    Quote Originally Posted by danlb View Post
    First, the range of valid pitch diameters directly corresponds to the tolerance classes, which correlate to the Major diameter. So the assertion that there must be one unique correct od is false. It's a range, and always has been. The pitch diameter for a class 3a 5/16-18 thread max 0.2764 min 0.2734 per Machinery's Handbook 27th Edition, page 1736.

    Second, the third paragraph is too difficult to parse due to a lot of missing words and unnecessary punctuation. It seems to show a misunderstanding of where you reference the cut from. It also falsely asserts that there can a a thread with the proper Major and proper crest and pitch that is the correct thread form and too shallow. You are ignoring the fact that the correct thread form is a specific size based on pitch. If it's not that size, it's not the correct form. If the thread is too shallow the crest will be too wide, and therefore not a valid thread form.

    Come to think of it the first paragraph makes no real sense either.

    So that 'proof' is invalid.
    Thank you for repeating my point that there is a range of OD. And that there is a range of PD dependent on the class of thread. WE AGREE ON THAT.

    Re-read what Paul said above. He is perfectly correct.

    Perhaps we should cut to the chase here:

    If you want the thread to fit, and wish to make ONE measurement that best tests the fit, that measurement is the pitch diameter. It is the basis of the measurement over wires.

    VERY few of us have the ability to check the thread form. The best methods for that are either a full form gauge, or an optical comparator. We use an insert, or we grind a cutter, and depend on that for a true form. Some may use full-form cutters, which should be very correct.

    Therefore one has a choice:

    1) Turn to a perfectly correct OD, then advance the cutter to the "depth of thread" as the threading proceeds, depending on the OD and the depth of advancement to arrive at the correct thread. This IS theoretically possible to do.

    OR

    2) Turn to a reasonable OD, then cut the thread, checking the pitch diameter to arrive at the correct threading depth. Clean up the OD.

    I know which of these is the more direct, practical and workable method for me.

    Others may use their own judgement to arrive at their preferred method.
    Last edited by J Tiers; 01-13-2018 at 02:17 PM.
    1601

    Keep eye on ball.
    Hashim Khan

  8. #38
    Join Date
    Dec 2016
    Location
    Helsinki, Finland, Europe
    Posts
    2,231

    Default

    I fully agree with J Tiers that pitch diameter is the most meaningful way to measure the thread if you want good fit.
    Thread wires for me every time I need to turn something with better than general purpose fit. Or trial and error fitting to existing female thread.

  9. #39
    Join Date
    May 2002
    Location
    SE Texas
    Posts
    11,589

    Default

    Dan,

    You say, "I should use full profile inserts which have the exact tip radius for a chosen thread."

    Really? An exact radius for each thread size? So you have how many inserts in stock in your shop? It must be at least several dozen if you are covering English and metric threads. You could probably buy a new lathe or mill with the money you have laying around in inserts. Oh, and who makes these "exact" size inserts for each and every possible thread? If this is true, I again will remind you that this is a Home Shop board and I and many others here can not afford such a collection of different insert sizes.

    I suspect that you have a FEW different sizes of inserts and each size has a tip radius that is within tolerance for a range of thread sizes. If you carefully read the thread specs, you will see that there is a lot of variance that is allowed at the roots and tips of the thread form. This is probably so that the people who make screws by the millions can have long production runs before changing the tooling that wears in use. So, the radius of one size insert may be on the large side for one thread, at a nominal value for another, and on the small side for a third; all of these values being within the range of acceptable root fills for each of those threads. And it would be used for all the threads in that range of "fits". And you would have to use a different cheat factor on each and every one of them except the one thread for which it truly is the exact size (whatever that is). In practice you would probably start with a non-standard, sharp Vee thread form, calculate the penetration from that, and add the tip radius from that.

    Oh, and in reference to my previous comments about the accuracy of the radius on the inserts, since they are made to cover a range of threads, I repeat, WHAT ARE THE TOLERANCES ON THOSE TIP RADII? Please refer us to a PUBLISHED figure.



    Quote Originally Posted by danlb View Post
    I've admitted before that I cheat. I should use full profile inserts which have the exact tip radius for a chosen thread. The radius is part of the spec for those cutters. Note that a radius is optional since the standard calls for a flat at the root.

    But what I do instead is start with the computed Major diameter and then use a sharp V insert and cut it to full depth using the Sandvik multi pass calculations. The root of the thread ends up being a sharp V and the flat crest ends up being the exact height and width called for by the standard. The pitch diameter comes out within spec. It's fully interchangeable with all threads made to the standard.

    If you don't want to do math, use the chart from the online section of Machinery's handbook. ANSI threads are covered starting on page 1732. Metric ISO are covered starting on page 1783

    Dan
    Paul A.

    Make it fit.
    You can't win and there is a penalty for trying!

  10. #40
    Join Date
    May 2002
    Location
    SE Texas
    Posts
    11,589

    Default

    Thank you, J Tiers.

    I am not trying to be anal about this. I have tried to make threads by starting with the OD and plunging in by a calculated amount. And then I wondered why my threads did not fit. To me, that is the point of this discussion and I have experienced it. So I looked at the process further and what I found was that the tip radius of the cutting tool must be the primary factor in this. So I have concluded that if you are going to use that method you must either refine it OR expect the thread to be too large and then use some method to measure or check it as you continue to cut deeper to make it fit properly.

    I do not get anal when cutting a thread for most purposes. If I am making a large or unusual thread, I do a rough calculation of how deep I need to go and them I start checking (gauge or nut) or measuring. If I am making a small, standard thread I cut it about 90% of the way with a single point tool and then run a die over it. That almost always produces a satisfactory thread. If I needed super precision, then I would use either thread wires or a thread micrometer to get a proper measurement of the Pitch Diameter. Also a Go/NoGo gauge can be used in many cases, but they can be expensive and I do not have any. I have a back burner project in mind to make my own micrometer attachment to measure the Pitch Diameter of threads.

    Threads look so simple but there are a lot of details in them. The experience of our original poster shows the need to have a good understanding of these fine points no matter what method you use to cut them. In fact, when using some of the shortcut methods, the need for understanding these fine points is even more necessary.
    Paul A.

    Make it fit.
    You can't win and there is a penalty for trying!

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •