Cutting a shallow taper

I think I see where the discrepancy in the tables comes from, the values given appear to be tangents not sines. For very small angles the difference is negligible, but they diverge significantly once you get over a degree or so, wildly once you approach 90* - at 90* sine is 1.000, tangent is infinite! Jim's method appears to use the tangent, as he's measuring the slope relative to carriage movement (adjacent of the angle) rather than topslide (compound) movement, the hypotenuse, which would give a.sine ratio?

I'm not clear how Jim gets from measuring the (known) angle to a topslide setting for the taper, as most will be locking the carriage in place and using the topslide to cut the taper?

My preferred method (without a taper attachment or for steeper tapers) is to set up a dti on the toolpost and rotate the topslide to an approximate angle, then adjust for no dti deflection against a 'master' taper (if copying) or a ground bar plus gauge block over a fixed distance, 5" if I have enough topslide travel (if cutting a taper from scratch).

If I want 90* included angle, I just engage both feeds... ;)

I did see a nice setup on another forum's shop made tools thread with a table and fence that popped in the chuck/collet to set a sine bar 'on its side' for setting the compound, looked very handy and once I have some spare tuits....

Dave H. (the other one)

Edit: one more thing, when bluing the taper with a collet, put an accurately sized ground bar in the collet to be sure the collet's taper is correct!
 
Last edited by a moderator:
You folks are killing me with the math questions, but let me take a try at it. Tigtorch, the sine of 6.5 degrees is 0.1132 but the number we are looking for is the length of the opposite side. You are assuming that the hypotenuse is 1.000", actually the adjacent side is 1.000" and the hypotenuse is 1.0065" in your example. I used the tangent function to calculate the length of the opposite side and as British steel points out there are some differences depending upon the number of decimal points that you use and how close to zero you are. The objective of the post was to show that for many tasks using inexpensive equipment, in this case imported lathes, reasonable accuracy level could be achieved using this method to set the compound slide angle versus the built in protractor. British Steel is correct on using blue with the collet compress to the native size however in the small home shop that is frequently a challenge. For ER collets, since they are double cut you will be very close with or without compressing them in my humble opinion. R8 collets less so.

If you would like to check my math I suggest this as a good trig calculator to use:

http://www.carbidedepot.com/formulas-trigright.asp

I have not tried to make a ER collet nut, these need to be hardened and cutting the retainer piece is beyond my skill level and the limited equipment that I have. For a few bucks you can buy a good nut and save the aggravation.

British Steel is working at an accuracy level that is beyond my skill level and the capabilities of my equipment. Thanks for the very good feedback.

Jim
 
JimSchroeder said:
I have not tried to make a ER collet nut, these need to be hardened and cutting the retainer piece is beyond my skill level and the limited equipment that I have.
Jim
Click to expand...

Jim,

From what I've seen of your work, I think you're selling yourself a little short. Believe me, if I can make one there's no doubt in my mind that you can make one.

http://www.hobby-machinist.com/attachment.php?attachmentid=38905&d=1360464663&thumb=1

http://www.hobby-machinist.com/attachment.php?attachmentid=38906&d=1360464663&thumb=1

Steve
 
JimSchroeder said:
SteveCMO, you do nice work. How did you cut the retainer lip? Very nicely done.

Jim
Click to expand...

Jim,

Thanks for the kind words.

It seems like it would be harder than it actually is. I basically turned the locking ring, or lip, concentric with the rest of the nut. Then using the four jaw, offset the nut .070" and used the boring bar to remove one side of the ring. This basically created an eccentric ring. The collet can "spring" and lock into the shorter side of the ring. It was a little hard to snap the collets in and out at first but after smoothing the rough edges with a little emory cloth it works great.

Hope that helps. Give it a try!

Steve
 
Align tool post EXACTLY first?

I know this thread is a few weeks old but I want to confirm an observation I made while successfully using this method. Since we are measuring the angle from a tool placed in the tool post (reference the photo in the first post of this thread) doesn't the tool post need to be aligned EXACTLY to the compound travel first? Once the angle is set, the tool post angle is not so critical since it travels in the compound direction regardless, but since the angle it presents to the dial indicator is used to set the compound it seems to me that MUCH care should first be used to set the tool post parallel to the compound travel.

Is this correct or am I missing something? BTW, I carefully used this method to make my ER40 chuck and it worked beautifully, my TIR including the collet was less than 0.0005.
 
Hi tigtorch,

I'm probably not understanding correctly, but......

In the first photo Jim has his indicator touching the compound and is using that for his measurements, not the tool or tool post. As long as the compound is set to the correct angle, and you only use the compound to cut the taper, the angle of the tool doesn't matter - it always moves at the same angle as the compound.

Does that answer your question or did I not understand it?

Steve
 
Steve,

I see that now, thanks! My compound doesn't lend itself to this as easily. I guess my old eyes failed me.
 
tigtorch said:
Steve,

I see that now, thanks! My compound doesn't lend itself to this as easily. I guess my old eyes failed me.
Click to expand...

Yeah, sometimes I wish my old lathe had more of the "billet" type compound and cross slide. It would sure make it easier to mount mag bases and indicate off.
 
You must log in or register to reply here.
Back
Top