A podcast for mechanical engineers beginning their studies in Design.
(upbeat music)
Welcome back, this is episode three
of Control Alt Eng.
So we do have a name this week.
I'm Kathy Petkov, I'm joined here
today by Mike Crocco.
And we're gonna walk you through
some of our next steps
in our design journeys.
Welcome everybody.
Yeah, you'll notice we have
a name finally.
We deliberated around that
and invited everybody
to watch a nameless podcast,
which just has silent titles
at the beginning.
But anyway, we'll get this
together, I'm sure,
by about week five or so of semester,
we'll know what's going on.
As you say, if it takes us that long,
I'm really, really worried.
Now, last week, Michael, we were
talking about the fact
that when we talked to our students
about offers,
and then we talked to them about CAD,
we're actually bookending
the design process.
Do you wanna talk a little bit more
about that bookending
and why we do that, just to refresh
everyone's memory?
Yeah, I mean, I guess I think we
looked at offers analysis,
which is a method of getting started,
and because that's where everybody's
gonna start.
Do you remember what it's called,
what it stands for?
No, you always ask me that,
or you always get me to answer that.
It doesn't matter.
Objectives, functions, factors, effects,
requirements, specifications.
Nice.
So this is a great way to set
up a project, right?
To understand what success looks
like eventually,
to get some ideas out and be
able to wash them out
of the system early, or identify things,
ideas that have good opportunities,
or potential benefit, things like that.
And then, yeah, eventually, you're
going to have to put it
down in generally CAD.
I mean, I don't know many designers
that don't work in CAD,
although there are a few.
I mean, it wasn't that long ago,
I still drawing beams on paper and
vellum, but it's rare.
It is rare these days.
And, Yurri, this is, I think,
where engineering
will really start to move and test
and do all those sorts of things,
seeing where CAD goes in the next
sort of five to 10 years.
And I certainly learned tech drawing
on a drafting board.
And I really, I personally
enjoyed it, so.
So we've got, yeah, like you
say, book ending, okay,
a place to get started,
but I wouldn't wanna relegate
offers to being,
the offers method to being something
that you just used at the beginning,
either, right?
Obviously, it's got uses beyond that
and becomes important for
judging the success
or the relative success or other aspects
of the design as you go along.
I guess I would take the opportunity
to talk about how
you may have to change what you've
initially set down
and offers through the design process.
And this is the norm.
And I guess this is kind of where
you're going with it?
Yeah, this is where I'm going with this.
I just sort of wanted a little recap
because what we get our students
to do post offers
is then we start to get them
to think about
the sorts of solutions that might
fit within the areas
that they've identified through
that offers process.
And there's some really specific
reasons why we do that.
Number one is trying to get more
solutions to a problem
so that you cover more of
that design space.
Because I think that the offers,
what it does is it kind of
gives you barriers
and sort of the edge cases
that you sort of are working
up to and around.
But I think it's really important
that people understand
that in engineering we don't go,
this is the only solution
because that's not where we go, is it?
Yeah, I mean, I think there's
a number of things
that can do that too, right?
You can have a process that limits,
let's call it creativity, it is.
It absolutely is.
You can have a process, you
can have a mindset
that limits creativity, right?
And which is very, very common.
People, we often talk about
people who get,
you know, wedded to an idea or
because it's their idea
that they feel emotionally tied to it
or whatever the case may be.
And part of being an engineer is,
I just said wedded, it's
divorcing yourself
from that kind of mindset
to understand that,
yeah, ideas are ideas and
they're important,
but so is expansion of ideas
and then eventual contraction
based on some engineering
judgment, right?
Yeah, I also think that if
you're not going down
that idea and that path of,
this is my only idea,
this is my only port of contact.
And so you've got multiple solutions,
then you can kind of break those down
and you can test those little
bits and pieces
that make that solution up too.
Which means we don't then
just think about,
you know, where do you start design?
And this is all related, I promise.
I find that a lot of students, when
they do start problems,
they start with some random
particular place
within the design and they go, right,
I'm designing from there
because this is the one thing
that I kind of know.
Whereas the reason why we get them
to give multiple solutions
and multiple ideas throughout
those solutions
is because then where that
random place is,
they can do a little bit more
exploration of.
So it's not just, oh, I've got a bracket
and that bracket's got two particular
holes in it.
They're not the holes that anyone
else would use,
but I'm gonna use those
as the whole basis
of my dire design.
Yeah, that flexibility I think
is really important.
I mean, you know, I guess
I think back to,
as I often do on this show, you know,
experiences I've had in industry
and things, you know,
they seem to get bedded down
and things look like they're gonna
work out a certain way
and then you often go back and, you know,
end up redesigning the whole thing.
But that's because you get to a point
where you recognize that, okay,
I thought that this was flexible,
it's less flexible than I thought,
so what else is flexible in the system
and what do we need to rework?
And so I guess what I would
wanna stress here
is that, you know, the offers approach
is a good method of judging, you
know, meeting criteria,
but in that middle bit, you know,
you're going to find yourself
having to combine ideas.
So that's something, you know, ideas
are not siloed, right?
And they have to play off of each other,
they need to communicate with each other
and they need to influence each other,
meaning the ideas need to
influence each other
in your own thought process.
Yeah, and I think this is also,
it's one of those flex things.
This is not, like lots of people
talk about creativity
being a skill set, but in actual
fact, it's a muscle flex.
It's how often do you do it?
So you're not necessarily going
to be great at it.
It goes back to that 10,000 hours
to become a master
of something that we were
discussing last week
is it's really important that
you give yourself time
and space to do these and exploration
of the smaller parts of a design
is often a way of doing that too.
But also I wanted to say at
least to a better,
quite often can lead to a better
or a more robust design as well
because you've explored all
the different options.
I do find a lot of students
do, like you said,
get wedded to one particular idea
and then they follow it until it's
well and truly dead,
beyond dead, trying to still
make it work.
Yeah, and sometimes it's a zombie, right?
And yeah.
That's kind of the nice part
about being a student
is I think in the first episode
we talked about failing
and these sorts of things and
learning from failure.
And ultimately, you're on
a time constraint.
You don't have all the time as a student
to explore every single option
and refine everything to full optimism,
optimal, I guess I should say.
So yeah, you go through that process
as much as you can,
you diverge as much as possible
and then you ultimately have
to pick something
because hey, you haven't built up
your engineering judgment yet.
That's a fledgling sort of skill
and at a certain point you have to jump
and pour your energy into
one particular thing
and then understand that it might
not work as you intended
and be ready to pivot,
but ultimately you do have to kind of
put your finger in the air
and that's why we get every single year
students doing a completely
different design
for every single group almost.
Clearly if they followed some
appropriate process,
theoretically they'd all get
to the same solution,
the same perfect solution.
But we all know that's not
the way it works
in the timeframes that we have.
And I kind of like that we
have such a variation
in different ideas and designs.
I don't ever wanna squash
that out of people.
I think that's really, really powerful
and really important to sort of
follow your own path a little bit
because like we said, there
is a better solution,
but there's not one true overwhelming
solution
that rules them all per se.
It's funny, as you're saying
that I'm thinking
about my experience designing
at Honda first
and then Toyota, right?
And I feel that there's a system in place
both those companies and most likely
many of the big automakers and
a lot of other companies
that make other things that there
is kind of one solution
to the problem in theory within
that company's DNA, right?
So realistically at Honda
when we were there,
we did find that we eventually got
to that final solution
that suited the Honda DNA.
And Toyota was gonna do it different
no matter what,
different than that, but it was
gonna be the Toyota way.
And then I guess I think
of like, yeah, BMW,
those cars end up as BMWs.
And that's very clear and there's
a whole lot of documents
behind that process and approaches
that dictate that final outcome.
Do you think it also then leads into,
especially in the automotive industry,
it becomes branding?
Yes, there is that, but again,
I'm speaking to my experience and it's,
yes, there is a branding aspect of it,
but it's surprisingly ensconced
or captured in the engineering process,
in the engineering design process.
And it is certain technical
decisions get made
based on prior knowledge and approaches
and I guess philosophy on
certain aspects.
And that drives a design to a
particular direction.
So in other words, what you're saying
is that our students have a tendency
to be very different variations
of their own thing?
Yeah, I mean, students are gonna
have their own DNA,
literally, and they're gonna
have a different,
let's say value system behind
all the engine,
like the stuff that they deal
with in classroom
or in the laboratory, yeah,
they've got their,
the kind of process they follow,
but then there are unwritten
or unarticulated influences
behind that, right?
And what all I'm saying is really
is that they are,
once you get to a large company size,
all those things are articulated,
perhaps,
but in the timeframes that we have,
we're not able to document them all.
Yeah, or fully extract them either.
Yeah, sure.
So one of the things I wanna
talk to you about
is that in that process,
we could often see students unwilling
to compromise,
but I found in my engineering experience
that a lot of what I do and a
lot of what we've done
has been around compromise,
whether that's compromising
with high management,
compromising with other engineers
and other departments.
For example, everything that I worked on
when I worked at BAE Systems
was bright orange
because it needed to be,
because you needed to be able to find it.
It wasn't an OEM, it wasn't
an original part
to the air vehicle.
So I have this aversion to bright orange,
but a lot of these things
were compromised.
It was what is best fit for the area
and how do we sort of compromise
on our design
based on the problem that we sort
of come up against.
Now you and I have been exploring an idea
that might help students with
that, haven't we?
Which are you referring to?
Triz.
Ah, Triz, you wanna get into Triz?
Well, just a smidgey,
I reckon just a little overlay.
If I could comment on compromise,
I think that Triz is a system
to orchestrate that,
but engineering is an exercise
in compromise, right?
Always.
Because the way I always used
to relate it is,
when I worked at Honda,
crash performance was really,
really important
and the safety of the occupant
was extremely important.
Well, if you take that to an extreme,
everybody would be riding around
in a tank, right?
Because then that protects the
occupant to some degree,
to the utmost degree.
You have to compromise on
that a little bit
because you also have targets
of dynamic performance
and fuel economy and these
sorts of things.
Yeah, my tank's not very fuel economy.
Exactly.
Like fuel efficient.
It has a tendency to tear up the road.
I live on a dirt road and it ends
up being really bad.
You can't transport it anywhere.
So that's all, the whole point is,
all of these constraints generally are,
at least in some way, competing
with each other.
And so your job is to, A,
find out what the appropriate
compromise is
and that's why it's so important
to have specifications
and weightings on those specifications,
all the rest of it.
And the other thing you're doing is,
in design is you're using
science to inform
a best estimate of what's going
to happen, right?
Absolutely.
Of course things could fail
unexpectedly, right?
True failure points of anything
are a bell curve
and you're determining statistically
what's unlikely to fail,
but everything could fail
at any given moment.
And so you're making a compromise
between,
100% certainty and a level of certainty
of performance that you're happy with.
Yeah, and I mean, and you can't
be 100% certain
of everything all the time.
It just doesn't happen.
Again, except if you use
the analogous tank,
you could, but then it's unrealistic
because then you're forgetting about cost
and that is a driving factor too.
And these sorts of things.
And comfort and being able to fit more
than two passengers inside.
Yeah.
Yeah.
So, Triz.
I wanna see a tank bus.
That would be really cool.
Oh, they exist.
Oh, awesome.
(laughing)
Cool.
It might not be.
Anyway, so Triz, what do you
wanna talk about?
Triz out today.
So where does it, so it's a Russian.
What does Triz mean?
What does Triz mean?
I'm not sure what Triz means.
Oh, come on.
I'd have to, I'd.
We'll do that thing where you
insert the audio here
and it's this thing.
And anyway.
No, we won't do this.
What we might do is we might see
if we can post process
and put it up on the screen
so that everyone knows what Triz is.
Because it comes out of a, it's
a Russian philosophy.
As I understand it.
As I understand it.
Yeah.
And so what it is is a gigantic matrix.
And what it does is it tells you what
you are compromising on
when you change something.
It's something that I wanna introduce
to my students
this year, just as an idea.
So they've got somewhere else to go
because say for example, shafts.
If you have a shaft that isn't performing
the way you want it to perform,
what can you do?
But what do you compromise when
you do change that?
What happens if you can't
change your length?
Are there other things that you can do?
Are there other considerations
that you might need to take
into consideration?
And Triz will actually show
you some of those.
Yeah, I mean, I think it's
a great exercise
to do a little bit later on to look at
if we need to achieve this particular aim
that we're not achieving,
what are some other ways to go about it?
It's things like, I think
some of the things
that stood out to me on the list of Triz,
on that matrix is I'm trying to solve
this particular problem.
Oh, have you considered composites?
Because composite materials have
a different behavior
for everybody's reference to
composite materials
or basically when you have a mixture
of two materials in one.
So it's not just fiberglass,
it's not just carbon fiber, it's
a whole range of...
It's a simplified way of looking at it,
but those have different attributes
than do homogenous materials, right?
And so that might yield a particular
direction,
make it easier to go in a particular
direction.
And so, yeah, it's just a great
repository, I guess,
of here's the approaches that
have been taken
to solve other problems in the past,
to basically give you that
kind of breakthrough.
You've taken this to the limit,
you've taken your current design
direction to the limit,
you can't meet target
or you're not meeting target adequately
or something like that.
And here's another idea about a pivot
that you can make essentially
in terms of material,
in terms of manufacturing and all
those sorts of things.
Yeah, I also think it's interesting
because we've talked a couple of...
In the last two episodes,
we've talked a little bit about that idea
of intuitive engineering.
And this is kind of one of those
tools that I found,
or that we found that can
sometimes help you
with that sort of step between when
you're not there yet,
because we don't know everything.
You know, I'm yet to meet an engineer
who knows absolutely everything,
who doesn't ever reach out
or try and find more from someone
else or another resource.
Yeah, in a sense, it's maybe
like a documentation
or a distillation of somebody's intuition
they built up over a long period of time.
I don't think it's...
I'm reluctant to say it's comprehensive,
but then when you look at it,
it's pretty darn near comprehensive.
It's pretty good.
It's kind of like having three or four
really, really knowledgeable colleagues
in the room with you.
So how would you suggest that a student,
okay, we've just kind of talked
very positively
about this tool.
How would you suggest they might use it
in their project or in the classroom?
I think it's...
I think for, especially for
sort of second years,
which is where I'm sort of teaching
at the moment,
I think the big thing for them is,
okay, so if we've got this scenario,
as in we've got this particular component
and we want to change this,
what are the things that might
be happening in that?
I don't want them to do a full
analysis or anything.
I just want them to start to be aware
that every decision that you
make as a designer
has an impact on something else.
And that goes back to that criticality
of where do you start and what's critical
within your design,
what's not necessarily critical
within your design.
It's really hard to know,
especially like we said, it's
such a short timeframe.
I know it feels long when
you're a student,
12, 13 weeks feels like an eternity.
But from the outset,
that's such a short time space
that you don't always have
the opportunity
of doing that full in-depth exploration.
You don't understand what your
compromise might be
or elongate a shaft,
or if you take away a mass point,
or if you take away a support
within that particular system.
Whereas this gives you just some ideas.
So some of the things that the students
might be designing,
I'd get them to have a look at
and then think about some of the things
that might be compromised
if they're changing things
within that design.
I guess I think a lot of my
time spent in design
was really focused on making sure
parts aren't going to fail, right?
Calculating stress and strain
and working out the strength
equation, et cetera,
load paths and managing all that.
And that was really for me in my
career before education,
that was where I spent my time.
And it's interesting, right?
Because that's my mindset around design
and that's what the bulk of the
time was spent doing.
Are students, well, your students in 2001
are spending almost no time on that?
Would that be safe to say?
Given the nature of the project?
In MMA 2001, they don't have the time.
They possibly don't quite yet have
quite enough technical knowledge
behind them.
However, when they get to
my fourth year unit,
that's pretty much what
we get them to do.
We get them to look at the load paths.
We get them to look at the failure modes
of the particular parts involved.
And we get them to look at
it in three ways.
Number one, we get them to do
the hand calculations.
Then we get them to simulate it.
And then we get them to build and test it
to see whether or not they can get
any of those failures to happen.
So here we are back again,
like forecasting,
for students who are doing that kind of,
I won't say rudimentary, but
simpler projects
where things either break or they don't,
and you may understand or may
not understand why
or whatever it is, things tend
to get overbuilt
and some things just squeak
through, luckily,
and others break.
Design involves more than
just that, right?
It's not just make a shape
that does a job,
it's make a shape that fills
the strict requirement
of performance, but also hits
all those other targets
of rigidity and strength and honestly,
fatigue, right?
And something that we don't get much time
to focus on in undergrad in
terms of the design
and build projects, at least at all.
But all of these things are requirements
and that's why design takes so long
and that's why we have to have
really strict procedures
around locking designs down
and things like that
because as we've been talking,
that iteration process,
well, at some point you have to say,
okay, this is set in stone so
that these other people
can get to work and finalize their design
because we can't just, if you
change one thing,
it influences everything else.
And so at some point you have
to say, it's done now.
Maybe that's a good thing to talk about
with regard to MMA 2001 is,
at a certain point,
you have to say, because we got to build
by a certain date, at a certain point,
you've got to say, this is
the design of that
and other things are gonna have
to respond to it.
Yes.
But we have to stop on that one
and we have to focus on the other things.
And finding that balance is probably
one of the big lessons,
I guess, in that process.
I think, and I think that's really hard.
So every now and then you need
someone external to say,
right, this is the line in the sand.
This is where you need to be by
this particular week.
But going back just for half a
second into that fatigue
and your life cycle and all
that sort of stuff,
I was talking to a colleague that
teaches into MMA 2001
with me and their comment was that,
and this is for every single student
who's going to be taking this
unit to take heed of,
the number of L-shaped brackets
that they have
that don't have any kind of
support for them.
And that fail- As in no gussets.
No gussets.
Find out what a gusset is.
It's really important.
Think about the kind of stress
that that takes
and what that actually does
to that bracket
and how it stops that deformation.
It also stops cyclic loading.
It does a whole range of things
and it's not very hard.
It's funny, I think, there's a
lot of products out there
that do our engineers a disservice
because there are so many just
purely L-shaped brackets
with no gussets out there that do fail
and they're terrible, have terrible
dimensional control
and things like that.
And we see them all the time.
It's funny you raised that.
In my interview at Honda, I was
asked to design a bracket,
an L-shaped bracket for a particular
application
and they were looking for,
are they thinking about rigidity?
Are they thinking about longevity
and fatigue
and things like that?
And how would you form the gusset
and what gusset is appropriately
sized in a given situation?
We had about 15 minutes to develop
that and give dimensions
and you had to do your calcs and
all that kind of thing
and you had to recognize,
you weren't told make a gusseted bracket.
No.
It's here's a thing that you need
to hang on this thing,
design the bracket and they wanna see.
And so, yeah, it's interesting
that you bring that up.
Even more interestingly,
my second year design lecturer when
I was at an undergrad,
John Brown, one of the things
that he actually gave us
was he said, design a bracket,
sorry, design a bracket and walked away.
And then one of the things
that he came back
was that there's a whole range of
questions none of you asked.
So I think brackets are
used as an example
in and around a whole range of places
because we see a lot of them.
We see good ones and bad ones
even in our everyday.
So I think it's kind of interesting
because it's one of those things
that I've always thought about
with students is,
Yurri, should I get them to
design a bracket?
Maybe.
Well, it's so funny because I
spent a lot of my time
designing brackets of different sorts.
I mean, it's really common
because you have to link different
things together
and you do that gently with brackets.
And it's not something to be trifled with
because guess what fails in
any given design?
It's probably gonna be the
bracket, right?
So think about your brackets, everybody,
as you're going and things that
are corollary to brackets.
Yeah, it also doesn't mean
that in MMA 2001,
it's appropriate to make sure
that you already have
completely over engineered
brackets everywhere.
You probably will.
And that's okay too.
The other task I was given in
my interview that day
was I was given a, what's it called?
A dog bone clip.
Is it dog bone?
Dog bone clip?
No, one of those clips,
those paper clips that have
the fold back levers.
Oh, a bulldog clip.
Bulldog clip, dog bone, bulldog.
Bulldog.
Bulldog clip.
And I was asked to draw it in three
view plus isometric.
And so I hope we get to talk about
drawing here soon.
And hopefully that's a future episode
because that was the other thing
that was demanding on me
that day is you have 10 minutes
to draw this to scale.
Blank sheet, no ruler, nothing
like that, just a pencil.
Mechanical pencil?
Yeah.
I just wanted to talk about the
fact that sometimes
when we design something,
we've got this phenomenon in engineering
and it's a really, it happens,
there are two places that I can
think of specifically
that happens quite often.
First one is bearings.
So you've been in automotive for
a long time in the past.
Have you ever designed a bearing
and had a bearing manufactured?
From the ground up.
From the ground up.
Not me personally, no.
Not you personally, no.
And I shudder at the thought
of having to do that
because the specialization involved
in bearing development is quite high.
So many things can go wrong.
But have you ever designed,
whoever done the design analysis
for a bearing
for something?
Yeah, if you're gonna quiz me on it,
I don't know how well I'll do.
No, that's okay.
So one of the things I wanted
to just sort of talk
to students is that there
are a few things.
So like I said, bearings is
one of those things
and often gears are another.
But what you would do is, and
correct me if I'm wrong,
but you start with your
base calculations,
you work out where you wanna go,
you actually do all those calculations
and you get to the end, you go,
this is what my bearing should look like.
And then you go, where's the catalog?
Please, where's the catalog?
Yeah, absolutely.
You don't do anything more
because this is so specialized.
That's a whole career in itself.
There's what, five companies
roughly in the world
that do bearings at any real scale.
And there's kind of two that make
probably 80% of those
even out of the bulk.
You cannot do better than they can do.
No, so don't even try.
So much of it is material science
and they know what they're doing.
Bearings is a great example.
There's all kinds of things out there
where you are gonna pick from
a parts catalog,
which I guess is what you're
getting at, right?
You're not gonna design
the electric motor
that goes into your robot or
whatever, unlikely.
I mean, it's conceivable,
but it's unlikely.
And there's all kinds of other
things out there
that are in that same category.
But when we get to that point
where we pick out catalog from,
oh, the internet these days.
Right, not the big paper catalog,
the SKF catalog?
No, not anymore.
Although I think I have one still.
But what we would do then is that
we go to the catalog,
we find something that is
fit for purpose,
but then we do those calculations
in reverse
to make sure that we've specced
it properly, don't we?
Yeah, sure.
I mean, a corollary would be electrical
design, right?
You're gonna put resistors in a circuit
or something like that.
Well, you're not gonna get
a 1973 ohm resistor.
You're gonna get a 1.9 kilo ohm
or something like that.
And then you're gonna have to check,
okay, it's close, but is it,
do I need to change anything else?
Yeah.
And so again, this is where part of
that compromise comes is,
it also depends on what you can get.
So I just sort of wanted to
put that out there
in people's minds as well.
That's a good thing to think of, yeah.
I mean, that's part of the
design process.
It might feel like cheating
or something like that
to certain people, I'm not sure.
But this is the way the world
works, right?
Yeah, you leave experts to
do expert things
because I don't think I'm going to,
in any way, shape or form,
design a resistor from the ground
up any time soon either.
I've been shocked at the number of times
I've had to design bolts that
were non-standard, okay?
But it was, you had to go through
a pretty exhaustive process
to get approval to design a unique bolt.
And it often was barely worth it,
but you knew, if you were doing it,
you knew you had to do it.
Exactly.
But it's, yeah.
Bolts that are 127 millimeters long.
Oh, okay, yeah.
Because 130 couldn't fit and
120 wasn't long enough.
Yeah. Yeah.
And that is also another
set of calculations
that every engineer will, at some
point, learn how to do.
You don't have to memorize it.
You can always find out how to do it
through your own textbooks
and through other information
you can find.
But it is one of those things is
specking the right kind
of bolt for the right area is another
one of those things
that your engineers do quite often.
Absolutely.
I think that kind of comes to time.
I guess so. Thereabouts.
I just wanted to say, thank
you for joining us.
Today we are in the generator,
which is the old engineering halls.
It's a beautiful, beautiful
space here on campus,
just across from the Allen
Finkel building.
What's the generator for?
The generator is for our
entrepreneurship.
Yeah, it's kind of an, it's
almost an incubator.
Yeah, it is an incubator.
It's very close to an incubator at least.
Absolutely.
And if you don't know what an incubator
is, look it up.
That's where startups get their start.
Startups are small businesses
that don't yet have their own offices
and they need a space.
Yeah, and so this is also something else
that we have here on campus.
I highly recommend you do
a bit of a search
and find out about the generator
at Monash Uni.
We're right across the way
from the maker space,
which is another space that
maybe we'll be using
in the future if we can get
enough quiet in there
to for 30 minutes or so to record a cast.
Might try and see if we can do that.
Okay.
Thanks everyone, have a great week.
See you later everybody,
see you next time.
See you again.
(upbeat music)