PDA

View Full Version : Sneaking up to final dimesions good ... Bad ... Or ugly.



Westline
10-21-2011, 08:25 PM
Seems I have a gap in my machining education.
It have been brought to my attention sneaking up to final dimension is a bad idea.
My question is why.
I want to plug this gap but I need to understand why...

Westline

rythmnbls
10-21-2011, 08:40 PM
It would depend on the tooling, e.g. carbide tooling needs a certain DOC to work properly, where as a tool like this http://www.gadgetbuilder.com/VerticalShearBit.html will happily take tenths off.


Steve.

lane
10-21-2011, 09:13 PM
I have been sneaking up on size`s all my life. I hasn't let me down yet.better to be big and take another cut than be too small and scrap the part. 2-3 thousand will file and sand off easy if you have to ,but putting 2-3 thousand back on is a bitch.

914Wilhelm
10-21-2011, 09:31 PM
It would depend on the tooling, e.g. carbide tooling needs a certain DOC to work properly, where as a tool like this http://www.gadgetbuilder.com/VerticalShearBit.html will happily take tenths off.
Steve.

I'm having a hard time picturing where the cutting is taking place on this bit. Does it cut on the end or the side? The angle on the side seems to tilt the opposite way I would expect for a lathe turning counterclockwise (conventional) when facing the spindle.

danlb
10-21-2011, 09:40 PM
What I heard recently made sense. I'll try to paraphrase it.

When you sneak up on it, you are taking a different DOC each time, so the deflection changes as well as the pressure of the tool. This makes it difficult to predict the exact cut that you really get on that last few .00X inches.

If, instead, you bring it within something like .014, then take .007 off twice you are more likely to hit it on the nose.

I'll be testing that later :)

Dan

rythmnbls
10-21-2011, 10:22 PM
I'm having a hard time picturing where the cutting is taking place on this bit. Does it cut on the end or the side? The angle on the side seems to tilt the opposite way I would expect for a lathe turning counterclockwise (conventional) when facing the spindle.


There's a link on the Gadget builder site to a thread on the chaski forum, there you will find more photos and more descriptions on how its ground and used.

Its main advantage is the ability to take very fine cuts and produce an excellent surface finish, perfect for sneaking up :)

Steve.

Chris S.
10-22-2011, 01:29 AM
I'm having a hard time picturing where the cutting is taking place on this bit. Does it cut on the end or the side? The angle on the side seems to tilt the opposite way I would expect for a lathe turning counterclockwise (conventional) when facing the spindle.

So did I until I read the description of cutting action. If you've ever used a wood turners Skew Chisel, to do a peeling cut on a wood lathe, it cuts just like that. I saved all the links because it's sooo unique.

Iraiam
10-22-2011, 03:06 AM
Seems I have a gap in my machining education.
It have been brought to my attention sneaking up to final dimension is a bad idea.
My question is why.
I want to plug this gap but I need to understand why...

Westline


I have better luck "sneaking" up on final dimensions, taking a larger cut as a final pass has burned me before, leaving the piece a little off, still tolerable but wishing it was better.

If I take 2 or 3 smaller cuts to get the final size, I find the results to be more consistent and predictable. This is what I would call sneaking up on it.

I came to this SOP by a desire to produce the most accurate part that I can on the equipment I have available to me.

I guess I'm with you In not understanding why this would be a bad idea. It's also likely that everyone will have a different definition of "sneaking up".

darryl
10-22-2011, 05:17 AM
It would seem that considerable pressure against a cutting tool would 'load' the lathe mechanism against various plays and thus make it more consistent to predict what a workpiece diameter may come to with a known and calculated depth of cut. With some measurements, then some tens of thou dialled in, cut taken, measurement made again, then a repeat of this, you should be able to predict fairly accurately what the resulting workpiece diameter would be after the second cut. If you take a spring pass after this, however, the value in this procedure would be wasted. If your second cut is largely different than the first cut, that also wastes the value in the procedure. If the second cut is modified, but still close to the depth of the first cut, then you should be able to predict fairly accurately the final diameter of the workpiece. In other words, you can modify your second cut by a small amount to suit hitting the diameter right on.

When choosing the doc for the final two passes, you take into account the capabilities of the particular lathe as well as the material and the cutter characteristics. You do this by trial and error, and you get to know what works well in your situation. I use this method, but my choice for the final diameter usually is to leave it about 2 thou large. Then I sneak up using half-thou doc, measure, repeat, etc. If I count it up, I'm probably measuring four or five times after I've brought the workpiece down to within about 50 thou of the desired diameter.

Since nobody is going to turf me out of my home shop for taking too long to make a part, I can continue to do it this way. Sometimes I'll file to final size, but I don't want to have to remove more than about a thou with the file. For the most part, I consider the file to be a way to remove high spots, which will leave the workpiece with more area at the final diameter, rather than a lot of high spots at final diameter with lots of lows in between. In my case, I can more consistently reduce a workpiece to less than wanted diameter with a file than by taking more spring passes or very shallow cuts. The secret to consistent material removal in very small amounts is sharp cutters, optimized relief angles, and play control in the various slides, etc. With larger depth of cut, you don't have to worry about play control as much, but of course you don't want to get caught finding out you removed way too much material.

I'm still not overly confident that my second significant cut will leave me right on the money. I'd rather end up a thou under by removing material in small amounts than three thou under by taking heavier cuts. Most of the time I'm in the 'sneak up' camp.

beanbag
10-22-2011, 05:25 AM
My carbide tools are very sharp and I can sneak up on dimensions just fine.

jugs
10-22-2011, 05:43 AM
Sneaking up to final dimesions good ... Bad ... Or ugly. ?

Like most things - It depends

How much slack in the machine,

Is it 2 ton of solid cast iron (DSG) on a concrete foundation, or a myford on a wooden bench,

Tooling- type, mounting, shape ect.

How good is your coolant system.

etc ect..........


I have a very worn BP (Y axis= 0.012" slack @ ends, 0.173"@ mid( going to fix it one day)), but a DRO allows me to consistently work to 0.001" as long as I take light cuts.


Temperature plays a vital roll on final dimensions, no point in taking the final cut to size if its going to shrink 0.005" when it cools. If your coolant system is crap, go do something else for hr then take last cut.

In the end, YOU do what ever it takes to acheve what YOU want, with the gear/skill YOU have available.

bob ward
10-22-2011, 06:26 AM
Westline, there are fortunate souls who can take a 1/4" finishing cut and finish up on size .0001" with a mirror finish every time. I'm not one of them, I'm a sneaker upper for close tolerance one off work. Repetition stuff is a little different.

There was an interesting discussion on the manual lathe work tolerances on practical machinist recently.
http://www.practicalmachinist.com/vb/general/reasonable-tolerances-manual-machines-232798/

rohart
10-22-2011, 06:57 AM
As long as once you are in sneak mode you plan on taking at least two more cuts, that's fine. Oh, and the two or three cuts should all be about the same depth. This is so the spring of the tool and the work are the same each time.

So, you must estimate what the spring will be. Suppose you think the spring at 6 thou DOC is 2 thou, and you need 18 thou to come off. Cut 8 thou three times, and you're spot on.

If you forget the spring, and you go for 6 thou, and measure, and the spring on 6 thou was 1.5 thou, you'll be down to 13.5 thou. You're in a mess. Up your next cut to 8 thou, you get 6, you're at 5.5 thou.

So estimating the spring is vital.

What's also important is to go to your finishing tool at the same time as you go to your finishing depth. I found this annoying when I was pushing the limits of HSS tooling on hardish material. If the tool lost it's edge on one cut, I would be going by guess work.

The only alternative is to sneak up so slowly, with a real sharp tool, so the spring is neglible. That takes time.

This is all one of the reasons carbide is so convenient. You can mess with DOC, get the spring dealt with, and go for it, without much risk that you'll lose the edge of the tool in the meantime.

Evan
10-22-2011, 07:14 AM
The reasons that sneaking up to dimensions is considered poor practice are several.

It demonstrates that the machine can not be depended on to take a cut to a predictable dimension.

OR

It demonstrates that the operator doesn't either know or trust the machine the machine well enough to cut directly to a tolerance in the shortest possible time.

Time is money in the machining business.

That said, home shop machinists are mostly not in business.

There are still other reasons that it may be a bad plan. In many materials, with stainless steels being a notable example, trying to take a very fine cut can wind up making it nearly impossible to take that last couple of thou off the work predictably. This especially applies to hardenable materials like SS and tool steels. While it is entirely possible to lap off a couple of thou that becomes far more difficult if the attempted final cut produces an irregular depth of cut due to work hardening.

In materials such as brass and aluminum this is not an issue and if time allows it is a perfectly acceptable way to work. But, even when time is not an issue one should know the equipment well enough to be able to work to the final dimension reliably with a DOC (Depth Of Cut) of at least ~ .005" (0.1 mm) in order to deal with difficult materials.

A major reason is that multiple low DOC passes will raise the temperature of the work and change the dimensions.

This is most apparent when dealing with hard turning of already hard materials using high cost tooling. Low DOC passes will heat both the work and the tooling excessively instead of transferring the considerable heat into the removed material.

http://ixian.ca/pics9/cermetfun3.jpg

jugs
10-22-2011, 07:32 AM
As Evan said


Time is money in the machining business.

That said, home shop machinists are mostly not in business.


Cracking picture Evan http://smileys.on-my-web.com/repository/Respect/respect-001.gif

loose nut
10-22-2011, 12:04 PM
What material??????

Free machining OK but 4140 not so much. It needs a bit of depth to get a good finish, unless you crank the RPM's to maximum and run a carbide so the chips are smoking.

Stu Miller
10-24-2011, 09:41 PM
Well, you guys got me enthused enough about the shearing cutter to try it. As reported by others, the finish generated on a piece of scrapbinium was great. That is enough to justify the tool right there, but I went on to try reducing the diameter by .002 inch. Worked so great that I tried reducing diameter by .001. Perfect! If you haven't tried it, do so.

Stu

RussZHC
10-24-2011, 11:10 PM
Stu:

If you are interested in a bit more detailed look at shear tool(s), as I was, someone here linked to this previously:

http://www.chaski.org/homemachinist/viewtopic.php?t=84313

saltmine
10-25-2011, 12:58 AM
Know your machine, and know the limitations of it.

I've got a little HF lathe, and I can turn out acceptable work just by paying real close attention to what I'm doing....Same goes for my little mill.
I seldom cut anything too small, like I used to.

Mcgyver
10-25-2011, 01:10 AM
The reasons that sneaking up to dimensions is considered poor practice are several.

It demonstrates that the machine can not be depended on to take a cut to a predictable dimension.

OR

It demonstrates that the operator doesn't either know or trust the machine the machine well enough to cut directly to a tolerance in the shortest possible time.

Time is money in the machining business.

That said, home shop machinists are mostly not in business.


agreed, there is no one size fits all.....the do-not-sneak-up crowd are right if you're machining to larger tolerances and need to get the job out the door. This environment won't tolerate putting a fine edge on a hss bit, changing tool bits, slewing over the cross slide and making a bunch passes sub 1 thou to gauge how much is being removed per graduation of the compound....but that is hardly the same it cant or shouldn't be done!

anyone who tells you can't take tiny <.001 cuts is also wrong - they just aren't considering the use of a sharp toolbit. Someones who's scraped knows its easy to take a <.0001 depth of cut with a very fine edge on the tool. Not that one turns with a .0001" DOC, however there isn't a physic limit imposed by tooling or cutting action that says you can't take very fine cuts.....provided a suitable edge is on the tool bit

machining parts to fit roller element bearings is probably the most exacting work appearing with any frequency, at least in my shop. Super sharp hss bit, compound at 5.75 degrees (as best you can eyeball it) so each .001 thou of compound movement moves the tool bit in .0001" and use a few tiny cuts noting exactly how much is removed with for and advance of so many markings on the compound. don't move the cross feed between passes. Finish can be mirror if you slow the lathe right down and apply soluble coolant. Its not impossible to work to .0001", nor does it have to be a monarch or hardinge, nor does it have to be in a temperature contolled room (for the smaller dimension we usually deal with). Typically the critial dimension only needs to be held over a short distance so even a worn lathe can take a good crack at it

JoeCB
10-25-2011, 01:45 AM
All good points to consider... the most difficulty is encountered when boring a relatively deep hole. The spring in the boring bar must be considered, and this of course varies with the DOC. "Sneeking - up" will produce unpredictable results unless one keeps track of the bar deflection relative to the DOC.
There was published in the Home Shop Machinest , July/August 2002 an article by Mr. R.G. Sparber that discussed this topic and provided a handy Excel spread sheet for precision boring based on taking three finishing cuts. The differences between the cross slide infeed and resulting actual DOC on the initial two cuts are used in the spread sheet calculations to render a final infeed for the last cut that will give the DOC actually required to produce the desired finished bore size. I find this spread sheet very handy when doing this type of work.

Joe B

DATo
10-25-2011, 04:53 AM
I have always used the 'sneak up' method and always will. If you have doubts regarding the accuracy of your last cut stop the machine and take a measurement of the first 1/16 of the diameter of your last cut to find out where you are with regard to the finish dimension and if necessary adjust accordingly.

Another thing I'll do sometimes is to take a piece of scrap of the same material and take short length cuts till I can reliably determine the finish setting and then put the real part in. This serves two purposes: one, to set the zero and second, to allow the tool to wear itself in.

Other thoughts concerning the reliability of final cuts:

When machining an o-ring groove on the face of a flange I cut the gland on a piece of scrap and recored my internal and external tool positions for the wall settings of the gland. Then I just repeat to these same numbers on the actual part(s).

Another trick (not rocket science - a lot of machinists do this) is to turn a quickie 'go-no go' plug gage for small diameter bores. Transferring a "feel" measurement from the gage to the finish bore is tantamount to taking an OD micrometer reading of an internal bore. Example: On a finish .400 dia. bore X .500 deep I'll put the gage to work testing the first .030 - .050 of cut, record the amount of material removed on the last cut (say .002), then rough the rest of the bore leaving .002 for my finish cut.

A.K. Boomer
10-25-2011, 09:09 AM
Sneaking up is tricky - there are so many variables that can get you - depending on the material there can be surface work hardening - there is also a type of skin effect that deflects - think of it as kinda a surface tension to get over --- even with the sharpest tooling you can touch up to it and there will be nothing - then you go just 2/10ths more and "oops" that part is 4/10ths undersize - (or oversize if ID)

one of the tricks I use has nothing to do with machining at all - if I need an extra tenth or two I just grab a piece of scotch brite, I know - I know - machining taboo - but like I said - it's not because it has nothing to do with machining so it's all good:)