The document discusses machining difficult features in titanium for the F-22 Raptor aircraft. It provides parameters for machining titanium, including recommended feeds, speeds, coolants and techniques to avoid cutter burns. It also details processes developed for machining thin webs and producing rough surfaces in titanium without requiring expensive backup tooling. Test results showed the new processes allowed thinner web and flange features to be machined within tolerances.

1. an
z
Society of
Manufacturing
Engineers
2000
MROO-110
Abstract to Machining Difficult
Part Features in Titanium for
the F-22 Raptor
author
EDWARD F. ROSSMAN
Manufacturing R&D Engineer
The Boeing Company
Military Aircraft & Missile Systems Group
Seattle, Washington
abstract
The purpose of this report is to present parameters for machining difficult features
in titanium. Begins with a review of efficient machining feeds, speeds, coolants, and
avoidance of cutter burns, but primary focus is on machining thin webs in titanium
without the expense of back-up tooling, and on producing rough finishes in
response to an unusual request. Have spent the past five years developing titanium
processes for fabrication of defense and space parts from titanium plate and cast-
ings. These notes are extracts from my notebook, “Notes on Titanium”. Processes
were developed during the period 1991, to present, 1997, while resolving various
machining problems in support of the F-22.
terms
Titanium
Machining
Thin Webs
Cutter Life
Cutter Burns
Thin Flanges
Ol997, The Boeing Company. All Rights Reserved
Sponsored by the
Fabricating Manufacturing Association of the
Society of Manufacturing Engineers
One SME Drive l PO. Box 930 l Dearborn, MI 48121
Phone (313) 271-l 500
www.sme.org
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

2. SME TECHNICAL PAPERS
This Technical Paper may not be reproduced in whole or in part
in any form without the express written permission of the Society
of Manufacturing Engineers. By publishing this paper, SME
neither endorses any product, service or information discussed
herein, nor offers any technical advice. SME specifically
disclaims any warranty of reliability or safety of any of the
information contained herein.
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

3. Machining Difficult Part Features in Titanium
for the F-22 Raptor
By
Edward F. Rossman
BOElNG
Manufacturing Research & Development
Preface
Background Questions come up daily on the processing and fabrication of titanium.
Current books and reports lack a comprehensive treatise on titanium
processing. My stack of notes on titanium was getting too tall to drag into
meetings, and I often found myseIf away from my desk with the wrong
information in-hand and with a short memory.
So I prepared “<Voteson Titunizm “. This notebook is an attempt to provide
rapid access to titanium processing information. Notes in this report have
evolved from learning to better produce titanium parts. This presentation
“Muchining D~Jfkult Pm Feutures in Titanium ‘*contains excerpts fi-om this
notebook
Technical Background
My work assignment for the past five years has been development of
titanium processing. This work is in support of design and development of
advanced aircraft that make heavy use of titanium.
Tasks have involved heat treatment stress relief alpha case removal,
distortion control, straightening, and development of feeds, speeds, and
cutters for machining.
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

4. Purpose of Report The purpose of this report is to present some parameters for machining
difficult features in titanium
Order of Reporting This report presents fundamental machining parameters for titanium and
describes the processes developed for machining two difficult part
features in titanium. The order of reporting is:
1. Fundamental machining parameters for titanium
2. Machining thin webs in Titanium
3. Producing a rough surface
Acknowledgments are contained in my parent notebook and include many individuals and
companies who have helped.
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

5. Fundamental Machining Parameters for Titanium
Feeds & Speed. As a synopsis, surface feet per minute (SFM) is 60 maximum for cobalt
(8% cobalt) cutters and 120 SFM maximum for carbide insert cutters.
Chip loads are a maximum of 0.008 inches per tooth for both types of
cutters.
Coolants We take a lot of care to use a heavy flood of coolant near the point of
tooling contact. There are many good coolants available, look for:
l Better than average surface finish
l No bacteria problems
l Good cutter life and cutting action
l Environmentally acceptable
Cutter Barns Recent cutter bums in titanium were due to cutter wear that suddenly
accelerated as heat built up on the worn cutter. Bums are best prevented
by:
l Using sharp cutters
. Flood of coolant
l Frequent examination of cutters for wear
l Attention to coolant action to assure liberal flow is a must
l Chips need to be kept away to prevent re-cutting and interference
with coolant
Cutter Life Our goal for cutter life has been a minimum of 45 minutes in all our recent
studies. The cutter life goal is usually met when SFM is 60 maximum for
cobalt cutters and 120 SFM maximum for carbide insert cutters. Chip
loads were a maximum of 0.008 inches per tooth for both types of cutters.
Cutters were considered dull when cutting edge measured 0.004 to 0.008
inches of wear or if the cutter became chipped.
I would recommend that any future cutter life testing be in accordance
with International Standards IS0 3685 (Tool-life testing with single-point
turning tools), IS0 8688-l (Face mill tool life testing, and IS0 8688-2
(End milling tool life testing).
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

6. Machining Thin Webs in Titanium
Background Minimum web thickness (floor of pocket thickness) in machined titanium
parts, such as aircraft spars; was limited to 0.100 inches unless expensive
back-up tooling was fabricated to enable the normal tolerance of +/- 0.010
inches to be held or unless chemical milling was used. The limitation of
0.100 inches minimum often blocked efforts to decrease aircraft weight.
Objective of Study Develop machining methods that would allow thinner webs to be
machined to the required +I- 0.010 inch tolerance without the need for
back-up tooling.
Accomplishments Testing was conducted in two phases:
l A machining process was developed to produce webs that were
0.060 inches thick and within the tolerance of +I- 0.010 inches.
l The process was further refined to produce webs as thin as 0.040
inches thick and within the required tolerances.
The developed process does not require back-up tooling or extended
machining time and the required surface finish of 125 micro inches or
better was met. No oil canning of webs was evident. No tool chatter or
unusual noises were noticed.
New Process The sequence of operations, cutter motion, and cutting parameters used
were as follows:
l Sequence of operations:
1. Rough to within 0.150 -0.300 inches of net thickness.
2. Measure thickness of part at this point if desired.
3. Finish mill to net thickness.
4. Cut off tool tabs (final operation).
l Cutter motion:
1. Cutter ramped at 4 degrees from outside edge of pocket to
centerline.
2. Work from center of pocket to outside walls to produce net
web thickness and fillet radius.
l Cutting parameters.
RPF Cutter SFM: 150
(Tool Life History: 36 minutes)
IPT: 0.004
Test confirmation A test part with five pockets was machined and measured to evaluate
producibility. Web thicknesses varied from 0.039 to 0.046 inches.
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

7. Production Confirmations
We have produced about a dozen production spars similar to the test part
and have met the +/- 0.010 inch tolerance in all cases.
Other Applications This same process helps to minimize wax-pagein other thin sections such
as thin splice plates. Even when the part is resting on a flat plate (which
amounts to back-up tooling), we get good distortion control. Here too,
the tooling tabs are cut off last.
Benefits
Prologue
The new process has allowed designers to reduce weight by
approximately one pound per one hundred square inches of web surface
without the expense of back-up tooling. Applies in those instances where
thinner webs are of sufficient strength.
My thoughts on why this process works are that traditional machining
spring passes do not work well on Titanium. I think thin clean-up passes
tend to push metal away from the cutter in the ‘Z’ axis direction and they
impart heat and surface stresses into the part. I think the reason this
process of leaving a heavy cut for last works is because the pull of the
cutter helix balances the push of the cutter. One of Boeing’s suppliers
said he got best dimensional results on titanium by leaving plenty of
material for the final machining pass. Cutter dullness could certainly
upset the balance of forces. By not doing final spring passes, - you can
actually save time and produce a dimensionally better part.
Cutting from the pocket center first to the outside wall last leaves plenty
of solid/ridged material next to the cutter at all times.
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

8. 6
Background
z 0.12
s
E
.- 0.1
ii
i! 0.08
45
E
+ 0.06
&
f
= 0.04
0.02
0
Minimum flange thickness in machined titanium parts, such as aircraft
spars; was limited to a height to thickness ratio of 2O:l with a minimum of
0.060 inches. To machine thinner flanges required expensive back-up
tooling or chemical milling. These limitations often blocked efforts to
decrease aircraft weight. The chart below shows these design standards.
Minimum Flange Thickness v. Height - Titanium
J
I
5
i
2
3
0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3
Flange Height (inches)
Objective of Study Develop machining methods that would allow thinner flanges to be
machined to the required +/- 0.010 inch tolerance without the need for
back-up tooling.
Testing Tests confirmed the original design limitations. No breakthroughs here.
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

9. 7
New Process Convince engineering to add a note to the drawing allowing machine
mismatches along the sides of the flanges. Flanges as thin as 0.040 inch
thick can then be machined. The sequenceof
operations is:
1) Rough to within 0.5 inches of net thickness.
2) Finish mill the top 0.50 inches of the flange.
3) Step down in ‘Z’ and mill the next 0.50 inch of flange to net
thickness.
4) Continue stepping until finished.
Production Confirmation
We have produced areas on several production parts with 0.040 inch thick
flanges and have met the +/- 0.0 10 inch tolerance in all cases. The cutter
mismatch is then hand blended/sanded as required.
Benefits The new process does allow designers to reduce weight. Tooling costs are
held low because back-up tooling is not required, but the run cost is higher
because of the additional milling and sanding time.
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.

10. How to Produce a Rough Surface Finish
Rough Surface Finish
Sometimes it is necessary to produce a rough surface finish. We recently
needed a rough finish (loo- 175 micro inches) to better allow XRAY
detection of voids caused if the electron weld beam wanders off the joint.
The following worked:
l Cutter body, Carboloy brand, model P220.17-02.00 square shoulder,
2” dia.
l Use (1) insert in this (3) insert cutter body; use false inserts in the
other (2) positions for balance & stiffness.
l Use insert Kennametai #(TPG322KC730) with 0.030’ radius, IS0
#(TPGN-16 03 OSE), Boeing A00 14650.
. 180 RPM.
l 2.9 inches per minute feed. It is critical that the feed rate does not
decrease due to changes in rates while driving comers. If the feed
rate falls outside the range of 2.4 to 3.1 inches per minute, the
generated finish will not meet the required finish.
l Axial depth of cut 0.005”
l Tilt the head 0.1 degree in the direction of travel to keep the heel of
the cutter from rubbing on the cut surface.
Prologue The edges of the initial weld joints fit so well that we could not detect
parts of the joint that were not welded by either XRAY nor by sonic
methods.
Achieving a rough surface finish was not as easy as anticipated. All our
experts remember rejections for rough finish and thought this would be
easy. It proved to be very difficult because all of our efforts for over forty
years have been aimed at improving surface finish.
Lifting the heel of the cutter was a key step because the trailing edge of
the cutter tended to smooth out the roughness created by the front edge of
the cutter.
While we were celebrating our success, the first production part went
through the deburr department without good communications. We ended
up having to re-do the rough finish on the part. The repair used about
0.005” of material.
Copyright (c) 2000 Society of Manufacturing Engineers. All rights reserved.