decisions through product autopsy
by Justin Petro
Dissection is a Core 77 feature segment focused on understanding how the synergy
of materials, manufacturing techniques, and design lead to innovative products.
Its goal is to uncover the secrets of a design by conducting an
autopsy on a coveted design object.
This feature is about more than merely appreciating a product's aesthetic, or
even its "cool" factor. It is an understanding of a product’s life,
and through its dissection
how well it lived.
As designers and engineers, we spend much of our time justifying new processes, materials, and manufacturing techniques to our clients, when, in fact, many of us aren’t daily consumers of the products we create. We also rarely have
the chance to review our products post mortem and witness how well they aged. However, if we can learn how products live, and in turn degrade, we can make better choices in our material selections, manufacturing processes, and their
As is the case of any scientific dissection, the ability to learn about our specimen
begins with the quality of the cadaver. The criteria for candidacy is that the
object must be instantly recognizable, sought after, and utilize multiple and
progressive manufacturing techniques. Today’s sacrifice to the design gods is
my beloved Motorola
For those not familiar with our cadaver, it has a long history of design recognition. And, it has no doubt inspired myriad cases of gadget envy. The V.70 has been featured extensively in the design press
including ID magazine and Innovation (Fall 2003)
and it won a Gold IDEA award in 2003. It was brought to market in 2002 as part of Motorola’s resurgence
in the mobile handset market where it broke new ground in both design, use of materials, and form. It was designed by Iulius Lucaci
in Motorola’s Chicago
My V.70 came courtesy of Amazon.com and T-mobile a year and a half ago. It served me diligently in its time. In fact, the electronics even survived being transplanted
into a new housing. As you’ll shortly see though, dissecting
and reassemblingthis phone is not for the faint of heart. It takes nearly 2 hours, if you're good
(and lucky), to complete a full transplant.
::Assessment of exterior treatments, materials, and techniques::
The designers of the V.70 strove to create a phone with a tough exterior shell,
like an exoskeleton, complimented by a more comfortable, soft, and tactile inside.
The exterior metal of this phone is particularly interesting as it involves multiple
production processes to achieve its final look and feel.
This phone was a showcase product for Motorola, and as such, they did not spare any expense in producing a very unique device. The exterior aluminum
shell serves not only as a structural element, but also as a design feature. It is also the most interesting of the phone’s components. The battery cover housing of this phone begins as 0.70mm aluminum (fig.01). The housings are
first stamped to get their overall shape. Secondary operations include the stamping of holes for the battery release door, fasteners, and selection button. Tertiary operations include the folding of retaining guide rails. And finally,
the company brand on the back of the outer housing is stamped, then completed by a mill
Internally, removal instructions are pad-printed
on surface of the aluminum (fig.02).
The final process of milling the back brand mark is perhaps the most unique, and expensive (fig.03). Although it is difficult to tell from the production parts, it would appear the entire process is a secondary stamp, then a machine
treatment; this feature thus appears as a hand craftly adding to the jewel-like quality of the phone. The expense of this operation comes from its precision. The stamp of the mark is 0.2mm deep, while the final machined
mark is a mere 0.07mm in height.
As I alluded to earlier, I have two sets of housings for this phone. One from
Motorola, the original equipment manufacturer (OEM), and the other from and
after market source (AM). AM
are very easy to find these days, and offer
users almost limitless upgrades for their phones.
I have thoroughly compared the two housings, and will continue to note the
differences throughout this dissection.
Though my AM pieces look similar,
under close inspection they are different. I cannot be sure where these parts
come from; however I have severe doubts that they are OEM parts from Motorola.
While they could be 2nd or 3rd generation parts, the more likely assumption
is that they are imitations. The parts themselves are constructed out
of two different gauges of aluminum, 0.70mm for the OEM part, and 0.50mm for
the AM part. I don’t think
this much variance could have come from the same production line, yet it
didn’t have any effect
on the fit of the part.
The other noticeable difference in the housings is the treatment of the brand mark. The OEM housing has mill marks across the entire mark and circle. The AM part merely has machine marks on the mark.
This is a trivial difference, yet one that I cannot say I fully understand. The cost savings would be minimal, if any. My only assumption is that by not duplicating the part in its entirety, the AM manufactures exempt themselves
from any legal infringement. This trend is apparent in many of the other parts as well. There are differences in relatively insignificant items, such as the font used on the keys or the quality of the spray finishes. However,
as you will see, the quality of the AM's manufacturing techniques dramatically affects the longevity of the parts.
The top antenna cover is created in a similar manner; however its material begins as 0.50mm aluminum. This indicates that the top and bottom housing were created independently of each other with two separate tools. Additionally,
the variation in thickness may be to allow better radio reception in the top housing, or may be within the manufacturing tolerances of the part. The antenna housing also has a small rubber
stopper which covers part of the antenna assembly. Again, this is most likely to aid transmission signals, or provide access to the phone. The stopper appears to be a medium durometer elastomer. The antenna housing is fastened to
the body by two Torx
fasteners on either side. This adds yet another unique and jewel-like touch to the design.
Both of the housings appear to be satin finished
via a media blast or brush technique. The texture is very light and no doubt applied to mimic the silver paint of
the top surface. The surface then appears to have a very light clear coat. However, it is not anodized, and as a result, scratches easily.
The choice of aluminum has impacted the longevity and durability of the part. Where other painted elements of the phone show wear, the aluminum remains intact. Although this phone has been dropped numerous times, the only blemishes
to the metal are a few scrapes and scratches. In short, far better resilience then a painted surface would have seen over time.
This part also has some interesting history prior to the V.70. The V.60, Motorola’s top phone prior to the V.70, also had a similar aluminum back housing. However, it had nowhere near
this amount of complexity. Interestingly enough, as the V.60 became more main-stream and Motorola downgraded it to their mid-level entry phone, they experimented with additional
techniques to reduce cost. One such technique was to injection mold the outer housing and then plate the plastic. This represented a large cost savings for Motorola and had little, if any, ill effect on the design.
In fact, these plated parts were absolutely indistinguishable from the original stamped metal parts.
As we move to the top side of the assembled phone we see a great mixture of manufacturing techniques
. The top half of the phone is painted silver. The rotor housing
shows a very apparent difference between the OEM and AM parts (fig.04) The Motorola parts lasted for a little over the year and had little signs of wear. The AM parts however, took merely a couple of months to wear completely to
the plastic. This can likely be attributed to the quality of the paint used and the preparation of the plastic.
||.06-soft touch vs chrome
The top housing has allotment for a small speaker port at its top as well as
a navigation switch near its bottom. The navigation switch is a rather ingenious
pass through to the internal key pad when closed (fig.05) This allows for operation
of the phone when closed, but doesn’t require running additional electronics
to the switches. The speaker has two faux openings and one real speaker port.
There is a thin mesh behind each of the faux ports which prevents you from seeing
all the way through the housing.
While the front of the rotor unit is composed of standard ABS
the back snap fit piece is coated with a soft-touch
spray. Given the thinness of the coating, it appears to be sprayed, vesus overmolded
or doubleshot. This finish really adds to the tactility of the phone and makes it feel more human than a plastic or even metal part.
The soft-touch treatment is repeated on the number pad, another ‘high-touch’ area (fig.06). This is one of the great surprises of using this phone, as it is something you wouldn't expect on first encounter. The number pad is also
pad-printed with numbers, letters, and answer/hang up icons.
most obvious difference between this part and my OEM housing
is the additional Asian language characters. I believe them
to be Kanji,
but that’s a novice's guess. Upon a very close examination
I found other peculiar differences. The actual font used
for the characters is slightly different, as is the treatment
to the brand logo. |
Both of the soft touch pieces are made of translucent plastic. The plastic
is, again, most likely an ABS or PC/ABS blend. The plastic on both sets of
parts exhibits a good amount of flex; therefore its mixture is heavily weighted toward ABS. Another significant reason for its flexibility is for the large amount of snap-fits
that connect the phone's housings together. While the phone’s tolerances are high, the slightly more flexible resin allows for easier assembly and therefore fewer defective parts.
The rotor is adorned with a chrome bezel that encircles the phone’s screen.
The bezel begins as an injection molded plastic part, which is then plated. The
bezel is completed by the addition of the navigation
button assembly. These buttons, though they appear trivial, are actually a rather
complex part as we will see when we disassemble the bezel.
The screen is protected by a piece of Lexan or similiar Polycarbonate resin. This piece is created using IML/IMD
. IML/IMD, commonly referred to as In-Mold Labeling/Decorating,
is a maturing new technique which allows for the integration of a graphic film into a plastic part. This is a great example of when and why to use this technique. Not only does IML provide a window to the screen, it protects the
screen, and provides a space for both product branding and screen matte. The use of IML also ensures the longevity of the screen as PC is very resilient to wear.
While pad printing the matte and logo may have been slightly more cost effective (as the AM part illustrates), it has a higher tendency to wear and degrade.
Though the AM part looks similar the OEM, it is in fact quite different. The piece begins as an injection molded PC part. It is then pad printed with the navy matte on the top of the plastic. A third pad is used to put the logo
on the plastic. While on the surfaceand right out of the boxthe difference seems minimal, after just weeks of use its degredatin is obvious. While the OEM part maintains its gloss black and silver logo, the AM part
quickly looses its sheen and begins to wear. This example of IML dramatically illustrated the power of the technique.
Though the OEM part would have required the graphic film, the trim tool, and
the injection mold tool, it would have been able to produce the part in far less
time than the pad printing technique used for the AM part. The dry time alone
would have been nearly 2-3 times as long to get to a finished part; conversely,
the OEM IML part would have been ready as soon as the plastic cooled. This is
a win both for production time and for part longevity.
Ironically, this pad-printing technique might have been more effective if it
were simply mirrored and applied to the underside of the part.
First level interior
::Disassembly of mechanical fasteners and user removable components::
The exterior disassembly begins by removing the back battery cover. This is accomplished
by depressing the battery switch and forcing the cover over the switch. The next
step is to remove the two Torx .75 metric thread 8mm fasteners (fig.07).
|.07-battery cover removed
Doing so uncovers the underlying plastic skeleton. Though the outside of the phone relies on protection from the aluminum housings, by and large, the real structure of the phone is provided by plastiastic is a
transparent resin, which appears to be identical to the other ABS/PC ABS material used on the touch surfaces of the phone. Its wall thickness varies from feature to feature of 0.75mm to 1.1mm. The internal structure is slightly
textured (Mold tech
: MT1005-1), though is still fairly unrefined as it is the b-surface of the part and would rarely, if ever, be seen by a user.
This illustrates another difference between the OEM and AM part. The OEM part, even though it is a b-surface part, has a great deal more attention paid to its refinement
than the AM part. The plastics also differ in opacity and color, the AM parts offer less clarity, more opacity, and generally feel cheaper. Further, the OEM part is generally stiffer and is probably the result of a higher quality
resin or a custom blend.
The removal of the covers reveals the GSM
card, battery contacts, two additional metal contacts, and part of the antenna
assembly (fig.08). I was unable to draw a conclusion as to the function of the
two metal contacts. My first assumption was that they informed the phone if the
covers were not present (fig.09). Having removed the covers, and was still able
to operate the phone, I rendered that hypothesis false. Further, as the phone
did operate without the covers, my second hypothesis about the contacts being
a ground, or completing a circuit, was also rendered moot.
Upon opening up the phone, one of the first things I noticed is the large amount of dust and dirt in the phone; mostly from my
, no doubt. As there were little-to-no environmental barriers in this phone to keep out dirt and other particles (except for the LCD and speaker mesh), the phone was at the mercy of its manufacturing tolerances. The phone
proved sub-optimal in this area, as I had to clean behind its LCD cover every couple of weeks by disassembling the bezel.
The bottom battery housing is connected to the top keypad housing by 6 snap fits.
These two housings encase the internal electronics.
I continue the first level interior dissection by turning the phone over and removing the front bezel. The front bezel is a removable "style" accessory; users can choose from three different options that ship with the
phone (gold, silver, and black). However, there are other AM options available
. The bezel is removed by pressing down and turning counterclockwise. Once the
bezel is removed, we are left with the aforementioned IML LCD cover, three dome switches, and the metal rotor housing.
The difference in the OEM and AM bezel parts is dramatic. While plastic
plating is becoming a very popular surface treatment for products, it is
clear that not all platers are created equal (fig.11). The OEM bezel is
far superior in shine, lack of pitting, and color consistency.
||.11-OEM vs AM chrome
The bezel is home to three buttons. The construction of these buttons is truly interesting, and frankly, a bit bewildering. The assembly consists of three parts. First, the blue buttons are injection molded (fig.12). They have a
circular boss that protrudes through the second injection molded and plated part (fig.10,11). Finally, that assembly is adhered to a thin acetate-like substrate (fig.11). The dome switches beneath don’t require anything but pressure
to make contact, so it would seem that this process has largely been utilized for aesthetic reasons. The buttons are then captured by pressure between the ring and phone substructure. You will see a similar technique used for the
number pad keys, though implemented slightly differently.
This amount of construction to achieve this part seems like an extravagant procedure. Especially when all that is gained is a blue dot on the keypad. I am unsure of the
reasoning behind creating the buttons in this manner. Yes, they will last a life-time as the blue button is a physical part, but the cost associated with it far out weighs its longevity. If the part’s permanence was an issue,
a pad printed dot, or in-molded and back filled dot, then sealed with a clear coat, would have been less expensive and likely as durable.
Second level interior
::Disassembly of electronic components and non user serviceable components::
The second level of dissection is a realm into which few people venture. The
reason for this much destruction of a device is either that it is completely
broken, or you’re completely crazy for trying to fix it. In my case, it is
a bit of both.
To begin the second level of dissection, I remove the second set of Torx screws (fig.13). I then unfasten the 6 snap fits that attach the battery housing plastic. This releases the back housingand everything else inside it.
With the back housing free, the electronic package is allowed to move freely, and disengage as one piece. The phone now exists in three pieces: the back housing, the electronics package, and the rotor assembly and keypad.
The upper rotor assembly now consists of IML lens cover, neoprene gasket, dome
switches, keypad keys, top plastic keypad housing, silver rotor housing, and
its back cover (fig.14). All of which are joined to the metal substructure.
The back housing, now devoid of much interest, remains home to two metal contacts,
an elastomer selection button, and the battery cover release button assembly
The two contacts connect to two metal surfaces within the phone, but I still can't ascertain their functionality. A volt meter might render and answer to this question.
However, as I’m a designer and not an electrical engineer, I’ll leave this question be.
|.16-bezel and domes
||.17-seal and L brackets
The IML window settles itself within the metal substructure and removes easily
as it’s a press fit part (fig.16). The gasket is lightly adhered, and pulls off
easily, while the chrome key pad is simply captured by pressure between the circuit
board and the top plastic housing. The dome switches on the bezel must be ‘fished’ through
the housing to be removed (fig.16,19).
The chrome key pad uses a similar technique to the navigation buttons of the
chrome bezel (fig.18). However, the implementation of this key pad is slightly
different. While the bezel buttons were adhered to the film substrate, the chrome
buttons appear to be either heat
. I would guess that they
are heat staked to the film substrate, as I have successfully managed to pry
one or two off of the film. While the bezel buttons are two individual pieces,
these are only one.
To disassemble the remainder of the upper rotor assembly requires some manual
effort, a bit of dexterity, and some luck as to not crack any of the plastics.
The metal sub structure captures the silver top housing by grabbing onto the
bottom housing via two metal L-bracket tabs (fig.21). To release the two parts,
the metal substructure must be disengaged from the bottom plastic. Once free,
we can see how the phone communicates its rotor state electronically (fig.20).
||Now that we can see the undersides of these parts, ever wonder what those weird circular markings mean on the c-side of a part? They can tell us a number
of different pieces of information. Unfortunately, every manufacture uses a different set of markings, so it’s hard to say exactly what they mean. However, we can usually figure out some of the following: what day or month
the product was shot on. What revision of the tool the product was shot with. And, if using a multi-cavity tool, which cavity the part came from.
|.19-domes ribbon cable
||.20-rotor contact ring
||.21-steel sub structure
Removing the top housing reveals the metal contact strip which communicates whether the rotor is open or closed. It does this in a relatively simple and elegant way. In the phone’s case, this metal conduit has two very important
jobs. It must first determine open vs. closed, and secondly it must transfer sound to the speaker. To determine state, the metal ring is broken up into three sections, with dead spots separating the active areas. When the phone
is closed, its sensors read a dead area, and when open, vice versa, hence telling what configuration it’s in (fig.22). It’s a very simple solution to what seems to be a complex interaction. While communicating digital voice data
is a bandwidth constrained process, analog sound reproduction is simple, and the phone’s connection points provides ample signal strength to drive the speaker
To completely disassemble the upper housing, the back plastic cover has to be removed. This is accomplished by unseating the snap fits which hold it together. The back cover is a very well made part. Its tolerances render it extremely
difficult to remove, and by doing so, allow it to capture the speaker assembly in place without adhesives ( also textured on the underside (MT11010). This helps add to the depth of the soft touch finish,
reduce the light transfer, and hide the lifter/slider lines
. In fact, from a manufacturing standpoint, the lifter lines are barely noticeable. This is an excellent example
of the ability for texture to hide sink marks, lifter lines, slides, and flow inconsistencies (fig.24).
This disassembly reveals the internal speaker assembly as well as the toggle
switch assembly (fig.22). The toggle assembly consists of a plated plastic
part and an elastomeric sleeve. The sleeve serves to help seat the part, as
well as ensure that the user’s pushes get translated to the underlying keypad.
The elastomer also aids in dampening the switch’s action.
Third level interior
::Disassembly of electronics package::
The final focus of this dissection is the phone’s electronics package. One of
the biggest selling points of this phone was its size, and
the sheer amount of electronics that Motorola has integrated into the package
is quite amazing. While I’m not qualified to deconstruct everything from an electrical
engineering stand point, designers should take note of how Motorola was able
to reduce not only component height, but also connector height. In doing so,
they made the electronics portion an easy assembly for factory technicians/machines.
The electronics package consists of an electroluminescent keypad, the main board, the LCD, and the SIM card readerall of which use low profile, pressure-fit connectors. At its highest point, the entire assembly is a mere 12.0mm.
There are four very intriguing elements to the electronics package which were either unique to this phone, or were revolutionary at the time.
|.25-EL under plastic
The first, and perhaps the most “whiz-bang,” is the integration of electroluminescent
lighting, or EL
for short. While EL has been around for quite sometime, Motorola was one of the early adopters. This application showcases the true usefulness and practical application of EL (figs.25 ,26, 27). Motorola has also gone a step farther
and integrated the EL directly into a very unique dome switch
||.29- underside of domes
What we can see by peeling back the EL layer is a very unique approach to using
dome switches (fig.28). Motorola took a different approach in assembling
this key pad. Instead of assembling an array of dome switches where individual
components communicated via cable to the board (like the bezel dome switches),
they cut out the middle man. The domes are adhered in their array directly to
the board (fig.29). The EL layer is then adhered to the film and electrically
connected to the board. The keyboard contacts are built into a single
PC board. This implementation dramatically reduces the amount of electrical traces,
component height, and cable routing associated with the keyboard design (fig.30).
Although small, it allows for better tactility than an elastomer solution
would have provided, not to mention a substantial height savings. The keyboard
connects to the mainboard via a simple pressure fit connector (fig.31).
The top of the package is also home to both the microphone and communication
connectors between the phone and bezel (fig.31,32). The microphone implementation
reveals an interesting solution to the problem of connecting a microphone to
its required 2 leads. Instead of using two wires, or concurrent chases, the implementation
here calls for two concentric contact rings. These rings allow the microphone
to sit freely on the main board with no worries regarding adhesion or loose solder
joints (fig.33). This implementation requires far less space and ensures that
the microphone can actually spin 360 degrees without loosing contact;
thereby making both component assembly and the phone more reliable. The bezel
contacts which transmit sound to the bezel are interesting as they are simply
spring-loaded posts which protrude on either side of the LCD (fig.35).
|.34-low profile connectors
||.35-switch and contact
Finally, the main board is home to an oscillating motorthe same type you’d find in your home hand sander, only much smaller (fig.36). Motorola was one of the first companies to include vibration as an alterative call notification
technique. Prior to on board motors, manufacturers were accustomed to incorporating oscillating motors into their battery packs
It is clear that the level of both manufacturing and electronic complexity which went into this V.70 is immense. This phone’s complexity from a materials and production point of view is amazing. However, it is equally impressive when considering
its technical and assembly hurdles. This phone served Motorola well as a showcase for design ingenuity and their ability to package complex electronics.
It is certain that the manufacturing decisions behind this phone greatly impacted its appeal and captured consumers’ attention. However, the choice of these production techniques greatly impacts the phone’s longevity, not merely
aesthetics. This phone illustrates also that beyond the technique used, the vendor providing those services is critical in producing both an aesthetic and lasting product.
The differences in the AM and OEM parts is immense. On the surface, and right
out of the box, the two sets of parts are largely indistinguishable. However,
upon minimal use, and closer inspection, we can see that the AM parts are in
no way comparable to what Motorola was able to achieve. I surmise the AM parts
are a victum of circumstance. Their poor quality may be attributed to vendor
choice, the need to cut corners/cost, or the AM’s
lack of attention to details. Plating, plastics, spray finishes, and overall
quality were orders-of-magnitude worse then the OEM parts. While it is important
that designers continually push the boundaries of materials and processes,
especially where they serve both form and function, it is imperative that they
be manufactured under close supervision to ensure that quality is maintained.
Our job as designers is not only to design aesthetically, but also for the elegant degradation of our products. The V.70 illustrates this point perfectly. While not inexpensive to produce, the cost is justified through both its
longevity and looks.
As you encounter your next design problem, think about how material selection, spray coatings, electronics packaging and manufacturing techniques can help develop a unique and compelling experience for your product. Remember it
is not merely aesthetics that sells products, but longevity and user
which gains trust, builds brands, and ulitmately creates loyal, repeat consumers.
The Motorola v.70 consists of 31 major components:
2 p. Stamped and formed aluminum
1 p. Stamped and formed steel
1 p. LCD
3 p. Electronics boards
1 p. EL film
2 p. Environment seals
1 p. Microphone
1 p. Speaker Assembly
4 p. Torx Fasteners
23 p. Plated Plastics
3 p. Elastomer parts
1 p. LI battery
4 p. Plastic housings
18 p. Dome switches
A graduate of Carnegie Mellon University, Justin
lives and works in Austin, Texas. He works at Design
Edge where he oversees Research and Communication. He spends
his days dissecting strategic business problems and then developing
unique user-centric design solutions. Justin has worked with Dell,
Trilogy, Maya Design, OHSO, Bluespan, FrogPad, Siemens, and Merrill
Lynch. He has been published in AIGA Quarterly, awarded patents
in both hardware and software design, and has guest lectured A href="http://www.cmu.edu">Carnegie Mellon University, the
University of Texas, and Savannah
College of Art and Design. If you’d like to be dissected,
he can be reached at firstname.lastname@example.org.
> back to top
> back to core