AT1 - AT11 taper tolerances?

[J Tiers]

With all due respect (of course) the OP's topic is (was?) to do with a round object - not a flat one - so perhaps the matter of "Blue-ing" should perhaps be addressed accordingly - or are the principles of Blue-ing universal?.

Pretty much universal. The blue material is on a first surface, which is "offered up to" a second surface. Where the space between is less than the blue thickness, the blue marks the second surface, showing the "high spots" between them.

"Topologically" in this case a cylinder or cone is about the same as a flat surface, since the ideal matching surface is at all points touching it, just like flat surfaces, so the principle of marking is the same.

A cylinder would need to have the outer surface split for insertion/removal radially, or the blue won't be any good.... it will get smeared by the insertion process if done lengthwise. But a one-piece conical surface is fine, if you are careful about inserting part B into part A.

Noting the point about the difference being less than the blue thickness, you need a nice thin coating to resolve a small difference. Otherwise you get a big blur and no useful info. More or less as OH suggested would happen.

[oldtiffie]

I agree with you generally JT but with this sort of set up one part is the "reference" and it needs to be as accurate as the user requires. The "adjusted" surface needs to be "adjusted" to accord wit the reference.

In theory, the "adjusted" surface can only approach but not exceed the accuracy of the reference.

In the OP's situation, the surface plate is the primary reference and the "vee" in the v-block is the secondary reference and should at least approach and if possible meet or equal the accuracy of the surface plate.

Note too that the OP is only concerned with the "male" (plug) and not the"socket" (hole) at this stage at least and is/as this is so then "blue-ing" is neither an issue nor required.

[Arthur.Marks]

Thank you to Old Hat and J Tiers for each's explanation of their bluing procedure. I will openly admit that I am not very experienced with it. My level of coverage is evident in the image in the beginning of this thread. I expect it is probably still too thick to pick up "tenths." I put a relatively small dab on my finger and rubbed it along the full MT3 mandrel. Then I inserted the mandrel into the sleeve and twisted it. The blue transferred very evenly between the two. That said, I don't have any idea of the precision it honestly indicates.

oldtiffie, thank you for voicing the obvious! I don't know how I could have missed it, but it makes perfect sense. So if my stack is 3 'tenths' short, indicate with a good 'tenths' reading indicator to correct the error. Sometimes there are too many details to keep track of and one of them goes missing.

Thank you to all the responders here. As always, I am learning a lot!!

[Old Hat]

The funny part is, all you have to do is change what condition you are looking to validate
or qualify and it's all different again.

Case is point, lets say there is no argument on method of application or thickness of film.........
but now, are we looking at proving uniformity, or finding lack of uniformity?

Now, how the two surfaces are joined to get a transfer may become an issue.
Some glide or roll may help achieve one reading, but confuse or invalidate another.

OR
Make possible a reading that technically should not even be attempted.
Example, we are asked for a "blue-off" of a spherical bearing liner....(half)
to it's mating spherical housing... half.
And yet there will be a fit allowance of say .002/.005 on both upper and lowers.

A glide or slight roll. or a deliberately heavier or lighter film may be the only way
to get a print that will help determine a damn thing.
So... to find "uniformity" it's invalid no mater how it's done.
But to find an imbalanced or distorted fit, it will reveal where the offence is
but not determine a value.

===============
the method I outlined earlier, is not my creation.
I learned it from post WW II German immigrant mold-makers.
It came with reasoning that you'll have a hard time ignoring.

There is no other way to to achieve a more uniform deposition.
Further, after further fitting, there is no other way to ensure a repeat
of the same thickness and distribution to subsequent depositions.
Shut-off on molds shot with highly fluid plastics has to be perfect.
And you need it done before the next Holiday on the calender.

One would do well to attempt it before dismissing it.

[oldtiffie]

Why not try the current USA standards authority on such matters - ie "NIST".

https://www.google.com.au/?gws_rd=ssl#q=nist

Morse taper and Jarno taper details are in Machinery's Hand Book 27 p926 "Standard Tapers".

[J Tiers]

I agree with you generally JT but with this sort of set up one part is the "reference" and it needs to be as accurate as the user requires. The "adjusted" surface needs to be "adjusted" to accord wit the reference.

In theory, the "adjusted" surface can only approach but not exceed the accuracy of the reference.

In the OP's situation, the surface plate is the primary reference and the "vee" in the v-block is the secondary reference and should at least approach and if possible meet or equal the accuracy of the surface plate.

Note too that the OP is only concerned with the "male" (plug) and not the"socket" (hole) at this stage at least and is/as this is so then "blue-ing" is neither an issue nor required.

Normally, one is fitting one part to another. One socket is accepted as standard, or an actual taper ring gage for the taper in question is used. The male taper is fitted to that.

If making male tapers, normally one expects to use them in a machine. that being the case, it makes sense to fit them TO that machine. While one might get them perfect according to the standard, and that would be ideal, it may do little good unless one also plans to make the machine's taper perfect per the standard (PPS).

In most cases, when existing commercial tooling fits the machine, one can assume it is "PPS" and it is sufficient to check the newly made tooling against the machine taper. You CAN make the new tooling independently "PPS" by elaborate measuring, but there is little actual need to do that. Of course if you have an appropriate ring gage, you are in business.

If the machine taper is not "PPS", then you have a choice. Either make everything, including the machine, "PPS" or deal with the differences, fixing the tooling to fit the machine.

If making everything "PPS" is the option chosen, then making one taper "PPS" amounts to making a gage for fixing the other. Gages should obviously be "PPS". The "independent" method of checking given is appropriate for that if no actual gage exists.

In either case, a useful means for checking at least one of the tapers will be bluing and checking the "print". Either the male taper against the machine taper, or the male "gage" against the machine taper which you are fixing, or whatever against the gage you have.

[oldtiffie]

Again, the OP is concerned solely with a"plug" (ie "external") taper and NOT a "socket" (ie "internal" taper).

I am surprised that the Op did not "blue" a section of a good surface plate and "roll" the plug on it to see if the pattern transferred to the plug is uniform - to see if the plug is straight throughout its taper.

The plug can be tested for "round" if it is set up between centres or seated in a precision vee-block..

This will be necessary if the taper was finished on a "centre less" grinder which have a well known habit of leaving 3 "lobes" which will show as "round" when the job is not or may not be "round" and cannot be detected with a micrometer.

These tests need to be done to satisfy the minimum level of accuracy for the plug to be used as a reference.

http://en.wikipedia.org/wiki/Centerless_grinding

https://www.google.com.au/?gws_rd=ssl#q=centre+less+grinder

https://www.google.com.au/search?q=centre+less+grinder&biw=1536&bih=732&tbm=isch&tbo=u&source=univ&sa=X&ei=l-epVOLDO4z48QXAx4HwCQ&ved=0CCkQsAQ

[J Tiers]

Again, the OP is concerned solely with a"plug" (ie "external") taper and NOT a "socket" (ie "internal" taper).

And it will never fit into anything? Hmmmmmmmm.........Ok, if'n you sez so......

Don't suppose he set up a centerless grinder for that part, but mebbe, who knows?

[Arthur.Marks]

Don't suppose he set up a centerless grinder for that part, but mebbe, who knows?

Um... no.


The course this thread has taken is bizarre. May I suggest everyone come back to earth for a moment? Small, residential basement. Not commercial. Not attempting to officially NIST certify anything. Only measuring to serve my own curiosity / pride. Part is a mandrel to turn a Jarno 13 to 3MT socket adapter between centers on the lathe.

Keep Calm & Carry On

[Old Hat]

I'm a ROCK of calmness! Try me....

[J Tiers]

If it is a mandrel, then something is a bit odd here.... Why are all the pictures at the beginning of this thread pics of what would seem to obviously be plug gages? Evidently because that is how you are checking tapers....

Anyhow, it seems that far from being bizarre and off center, you DO IN FACT want both a qualified male taper and a qualified female taper.... So comments about bluing the parts etc are relevant. At the very least as to the effect of blue thickness on sensitivity to errors...

You DO want a way to discover if your adapter is "PPS".

Just as a bit of clearing up here..... is this adapter for some far-away use on some machine you do not have? Or is it for direct use on your lathe or other machine?

The question comes up because the accuracy of the taper for use on a particular machine may actually be a secondary consideration.... The total accuracy is composed of the concentricity of the final taper with the machine spindle axis, and the accuracy of the final taper itself.

Everything in-between can be horribly "off", so long as it is consistently "off" and can be replaced in the exact position later. Better of course if it is "PPS" and concentric... but it's the final taper that really counts.

Consequently, depending on usage, you might do better to turn/grind everything on the machine that is the target... same idea as grinding a taper socket with spindle in-place on the machine.

As for "small residential basement" shop, it's clearly a few notches above the average, and there are commercial shops with less nice equipment than your shop seems to have....... and less concern about using it accurately, too. So you can forgive a bit of "upgrading" of our thoughts on what you say....

[Old Hat]

I'm finding way more H S M members are PPS compliant than I would have ever guessed when I joined up.

Possessing a Propensity for Silliness.

Just pick up a hunk of metal and make somethin Cool Hey?

It's fun and rewarding.

[oldtiffie]

Every time (or just about anyway) "precision tapers" are mentioned with regard to a lathe - whether using the top slide or a taper turning attachment and in the absence of any mention of checking the lathe carriage and bed as well as the top slide - and the taper turning attachment for straightness, I can only assume (presume?) that either all those parameters were checked and are adequately accurate or else they were ignored.

I have a very good universal grinder (cylindrical, tool and surface) with fixed centres that would do this job nicely - if I ever had a need to.

http://www.machineryhouse.com.au/G198

This sort of job is either bordering on or within "metrology" limits and needs to be addressed as such or in the absence of such equipment, as well as can be with what the operator has or has access to.

The accuracy (straightness) of top slides is rarely addressed - but it should be checked before taper-turning.

[J Tiers]

I'm finding way more H S M members are PPS compliant than I would have ever guessed when I joined up.

Possessing a Propensity for Silliness.

Just pick up a hunk of metal and make somethin Cool Hey?

It's fun and rewarding.

Looks like "OH" is a "negative free radical".........................

I'll leave it at that.... I have two pieces of assembly tooling to finish in the shop here.....