Interpolating vs. Boring?

I know this one is always a fun argument.

Does anyone have a link to a proper study that compares boring holes vs interpolating vs reaming for hole quality?

We have an argument in our shop that since the CNC is supposed to hold .0002, that it can interpolate holes that actually end up being within .0002.

But on the other hand almost every ball bar test result looks something like this:
(Yes, we know the scale is very small)



So, is there a real study out there that analyzes the holes produced by the various methods that can help end this debate?

Boring has to be better: there are fewer variables to it. With interpolation, you're using up most of the allowed tolerance (on a high precision hole) on machine condition/servo tuning accuracy, whereas a bored hole is basically dependent only on the spindle bearing roundness (as is the case for the interpolated tool). Anything you want good, you bore it.

boring will give you a much more round or "cylindrical" hole with the expectation that you have decent spindle bearings.

really depends on how you and/or the customer are going to measure the part to interpret if its good or not.

waste of time looking for a case study...too many variables in the cutting processes and machine tool accuracy. example- if your interpolating on a Haas vs interpolating on a mitsui seiki vs interpolating a .625dia hole with a .5" cutter @ 100ipm vs .125 dia cutter at 20ipm.

Roundness is only part of the picture.

With interpolation the geometric form of the hole is dependant on:

• The ability of the machine to interpolate a circle - affects roundness
• Deflection, runout and coaxility of the tool - affects straightness
• Perpendicularity of the spindle - affects cylindricity

When boring the variables are limited to:

• Circularity of the spindle rotation - affects roundness
• Parallism of spindle axis to Z axis - affects roundness

Under normal circumstances the errors affecting boring are likely to be orders of magnitude less than those affecting interpolating.

One source of error when boring is using a heavily imbalanced boring head on a machine that is less rigid in one axis than the other - an oval hole can result.

Hopefully not far off topic... but I feel it applies.

I did some testing at the last facility where I worked using a Heidenhain KGM: ( KGM Grid Encoder | HEIDENHAIN )

This is in a whole different league than a ball-bar setup. The setup I was testing with the KGM was a cycle-time critical production part on a Robodrill, and we dumped the 2d geometry of the (profile) tool path directly into the computer controlling the KGM. The output (graphical) would show the deviation from "expected" path with logarithmic "enhancement" for easy visualization (actual values supplied as well). We ran test after test, incrementally adjusting feed rates up as we went. Then we went to work on tweaking some settings on the control (Fanuc). Then experimented with different mass on the table (first quick test was just to bolt on an additional Kurt vise.

The results were really amazing, and we gathered a ton of data in probably an hour or two (and learned a *tremendous* amount).

That short period of time was a great demonstration for the engineers that attended as to how feed rates, controller settings, and table mass effect the outcome of a part. The most surprising thing to me was "seeing" how long it takes the machine to "settle" back down to a straight line tool path after making an outside corner at a decent feed rate. There would sometimes be 5 or 6 decreasing sine wave deviations until the path was actually fully straight again. Second most surprising thing was seeing how much the actual path would "short cut" a corner when the feed rate started to get fairly high.

As related to the topic in this post, we quickly set up at the end interpolating some small and large circles. Feed rates and table mass were very, VERY significant influences on how round the actual cutter path was.

Since the KGM is a 2 axis device, Heidenhain brought an angle plate as well so we were able to do some quick checks on X/Z and Y/Z as well. AWESOME tool!

And fwiw: ... the KGM we were using (the larger model) was about $35k back then (around 4 years ago).

PM

RedJeepGuy said:

I know this one is always a fun argument.

Does anyone have a link to a proper study that compares boring holes vs interpolating vs reaming for hole quality?


So, is there a real study out there that analyzes the holes produced by the various methods that can help end this debate?

Click to expand...

If the only question is about "hole quality" there is no argument or conversation, doing tests for something like that would be redundant to an already known answer.

Now, if you add to the question Efficiency, Productivity, Throughput, Time, Cost and Ease of use, then you are really asking for the over the top dispute.

R

precisionmetal said:

The most surprising thing to me was "seeing" how long it takes the machine to "settle" back down to a straight line tool path after making an outside corner at a decent feed rate. There would sometimes be 5 or 6 decreasing sine wave deviations until the path was actually fully straight again. Second most surprising thing was seeing how much the actual path would "short cut" a corner when the feed rate started to get fairly high.

Click to expand...

Shouldn't really be a surprise tho, if you think about how servos work. With no error signal, they don't go anywhere. But to get an error signal you need an error

From the practical side, if you ever tuned up an axis then you know if you jack up the gain too high then the motor starts buzzing. But to get perfection, you'd want the gain infinitely high. Servos just don't work 'perfectly'. Back to "no error signal then no motion" but you have to have an error to get an error signal.

There is more than just the spindle bearings involved in a bored hole tho. Slides don't stay stationary all by themselves. The servos have to be able to hold the slide rigidly to avoid getting any servo-induced errors into the hole. A manual SIP is probably more accurate in an absolute sense

We do aircraft parts. 70% 7075, 20% hard stainless, 10% Ti. I have programed interpolated holes in just about every machine in our shop. New Okuma 5x round within .0002, old Haas mini with new ball screws within .0005, old Matsurra 4x best we could do is .002". If we are doing a bearing hole, we either bore or interpolate undersize and diamond hone it.