OK, so now I'm wondering. How much does it cost to buy 1 Kg of plastic? 
And how much does it cost to tool up and turn that plastic into 
something useful? (E.g., the casing for some sort of electronic device.)

Also... I have a keyboard sat next to me. It seems very heavy. Is the 
*entire* thing plastic? Or is there a metal chasis inside?

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Orchid XP v8 wrote:
> Also... I have a keyboard sat next to me. It seems very heavy. Is the
> *entire* thing plastic? Or is there a metal chasis inside?

Let's find out!

*starts putting the thing apart*

Post a reply to this message

> OK, so now I'm wondering. How much does it cost to buy 1 Kg of plastic?

The cost of the plastic itself is almost zero compared to the other costs 
involved (like the tooling cost, cost of running the machine, paying workers 
to keep an eye on the machine, shipping the product etc).

> And how much does it cost to tool up and turn that plastic into something 
> useful? (E.g., the casing for some sort of electronic device.)

A typical hard-tool for a simple plastic part (eg the front cover of your CD 
drive or half of a simple keyboard shell) will cost of the order of 10-40k 
pounds.  That tool will last for perhaps 200k uses until the metal is worn 
away enough that any further parts produced are considered not good enough. 
So a supplier would give you a price of like 10p per part if you were 
ordering large enough quantities.

The exact cost of the tooling usually depends on how many details there are 
in the design, someone has to design and make the tool, so the more details 
there are the more expensive.

> Also... I have a keyboard sat next to me. It seems very heavy. Is the 
> *entire* thing plastic? Or is there a metal chasis inside?

Not sure, maybe in the old days if you had a big heavy keyboard it would 
have one, but I guess with recent cheaper designs they realised that 
actually there is no reason to put a metal chassis in there (it's not like a 
keyboard needs to be that strong in normal use!).

Tools to make metal parts are usually more expensive, because they are more 
complex.  They often need to bend and cut the sheet metal in several steps 
to quite accurate tolerances.

Post a reply to this message

scott wrote:
>> OK, so now I'm wondering. How much does it cost to buy 1 Kg of plastic?
> 
> The cost of the plastic itself is almost zero compared to the other 
> costs involved (like the tooling cost, cost of running the machine, 
> paying workers to keep an eye on the machine, shipping the product etc).

So the plastic itself is pretty cheap?

This leaves me wondering why so many pastical items seem to be 
specifically shaped to be hollow so they use less plastic. I mean, if 
plastic is so cheap, why bother with more elaborate designs to try to 
conserve it?

Is this stuff priced in pounds of Kg or per tonne?

> A typical hard-tool for a simple plastic part (eg the front cover of 
> your CD drive or half of a simple keyboard shell) will cost of the order 
> of 10-40k pounds.

o_O

Ouch...

Now, is that just the mould, or the entire machine? (I would have 
expected the mould itself to be a replacable component, while the rest 
of the machine isn't...)

> That tool will last for perhaps 200k uses until the 
> metal is worn away enough that any further parts produced are considered 
> not good enough.

Really? I hadn't realised they wear so fast.

> So a supplier would give you a price of like 10p per 
> part if you were ordering large enough quantities.

By "large enough" we're presumably talking about tens of millions of units?

> The exact cost of the tooling usually depends on how many details there 
> are in the design, someone has to design and make the tool, so the more 
> details there are the more expensive.

Yeah, figures. So a plastic cup is going to be pretty cheap, but a 
complicated jig for holding a structure together isn't.

>> Also... I have a keyboard sat next to me. It seems very heavy. Is the 
>> *entire* thing plastic? Or is there a metal chasis inside?
> 
> Not sure, maybe in the old days if you had a big heavy keyboard it would 
> have one, but I guess with recent cheaper designs they realised that 
> actually there is no reason to put a metal chassis in there (it's not 
> like a keyboard needs to be that strong in normal use!).

Perhaps I should clarify: I'm talking about a "keyboard" of the musical 
kind. Somebody might attempt to put it on a stand that only supports the 
ends of the unit, and it's over a meter wide. (Even so, all it contains 
inside is a circuit board. No batteries or power supply or anything 
heavy, just the key action.)

> Tools to make metal parts are usually more expensive, because they are 
> more complex.  They often need to bend and cut the sheet metal in 
> several steps to quite accurate tolerances.

Have you ever noticed that cheap PC cases always have jagid edges on 
their insides? Yet if you buy a £800 server case, all the edges are 
beautifully smooth? (And the steel seems about 4x thicker, BTW.)

I have no idea what steel costs per Kg, but presumably a lot more than 
plastic! Also, presumably thick steel takes more effort to bend and cut...

Post a reply to this message

> This leaves me wondering why so many pastical items seem to be 
> specifically shaped to be hollow so they use less plastic. I mean, if 
> plastic is so cheap, why bother with more elaborate designs to try to 
> conserve it?

I suspect the more elaborate designs are to save weight or allow space for 
other components.  Of course the plastic isn't completely free, so if there 
is nothing else to lose you may as well try to use as little as possible.

> Is this stuff priced in pounds of Kg or per tonne?

Definitely talking of the order of pounds per ton.

> Now, is that just the mould, or the entire machine? (I would have expected 
> the mould itself to be a replacable component, while the rest of the 
> machine isn't...)

That's just the "tool", the bit that actually defines the shape and where 
the plastic enters and flows, it fixes into a much bigger machine that 
actually moves the two tool sides together and supplies the plastic under 
great pressure.

> Really? I hadn't realised they wear so fast.

Yep, and that's the "hard" tools that are designed to last as long as 
possible.  During prototyping you often use "soft" tools that only last for 
a thousand shots or so (but are cheaper to make).

> By "large enough" we're presumably talking about tens of millions of 
> units?

Not necessarily, just large enough that you use up most of the lifetime of 
the tool, otherwise things get very expensive (and you would use a cheaper 
method of production than making hard tools).

> Yeah, figures. So a plastic cup is going to be pretty cheap, but a 
> complicated jig for holding a structure together isn't.

You got it, additionally if there are lots of sections that need to have 
accurate tolerances (eg because they attach to some other part) then it gets 
expensive.

> Perhaps I should clarify: I'm talking about a "keyboard" of the musical 
> kind. Somebody might attempt to put it on a stand that only supports the 
> ends of the unit, and it's over a meter wide. (Even so, all it contains 
> inside is a circuit board. No batteries or power supply or anything heavy, 
> just the key action.)

Ah ok, well yeh maybe they do, or they just use thicker plastic?

> Have you ever noticed that cheap PC cases always have jagid edges on their 
> insides? Yet if you buy a £800 server case, all the edges are beautifully 
> smooth? (And the steel seems about 4x thicker, BTW.)

Yes, the tooling for nice smooth edges requires more steps (metal tools 
usually have several steps because it's impossible to bend and cut 
everything correctly in just one stamp) = more cost.

> I have no idea what steel costs per Kg, but presumably a lot more than 
> plastic!

I don't think it's vastly different, again it's usually the tooling and 
manufacturing costs that far outweigh the raw material costs.

> Also, presumably thick steel takes more effort to bend and cut...

Have you seen the size and general beefyness of the machines and tools that 
actually do the bending and cutting :-)

http://www.pdf-inc.com/img/press3.jpg

And an example of a tool:

http://img1.tradeget.com/sophiaxiafei/FR5WPIWW1progressive_die_zy1.jpg

Post a reply to this message

scott wrote:
>> This leaves me wondering why so many pastical items seem to be 
>> specifically shaped to be hollow so they use less plastic. I mean, if 
>> plastic is so cheap, why bother with more elaborate designs to try to 
>> conserve it?
> 
> I suspect the more elaborate designs are to save weight or allow space 
> for other components.  Of course the plastic isn't completely free, so 
> if there is nothing else to lose you may as well try to use as little as 
> possible.

I just notice that plastic objects always seem to be made of *the* 
thinnest plastic possible without breaking, that's all.

>> Is this stuff priced in pounds of Kg or per tonne?
> 
> Definitely talking of the order of pounds per ton.

Wow, as cheap as that? o_O

Does it vary significantly depending on which type of plastic you want?

(I'm still trying to figure out why the plastic cooking utensils my mum 
bought actually MELT IF YOU GET THEM TOO HOT. I mean seriously, *why* 
would you design a product which can't be used for its intended purpose?)

Do you buy the plastic with dye already added, or do you add that when 
you come to use it?

Does it make any difference what kind of finish you want on the item? 
(Matt, shiny, textured, etc.)

>> Now, is that just the mould, or the entire machine? (I would have 
>> expected the mould itself to be a replacable component, while the rest 
>> of the machine isn't...)
> 
> That's just the "tool", the bit that actually defines the shape and 
> where the plastic enters and flows, it fixes into a much bigger machine 
> that actually moves the two tool sides together and supplies the plastic 
> under great pressure.

Right. So tens of thousands of pounds just for the tool itself? Damn, I 
don't even wanna *know* what the machine to work it costs... o_O

>> Really? I hadn't realised they wear so fast.
> 
> Yep, and that's the "hard" tools that are designed to last as long as 
> possible.  During prototyping you often use "soft" tools that only last 
> for a thousand shots or so (but are cheaper to make).

Mind you, I guess if *you* had molten liquid poured onto you under 
extreme pressure thousands of times a day, you'd wear out pretty fast 
too... ;-)

Presumably for prototyping they tend to use all the "3D printers" we 
keep hearing about? (Do they use soft tools to check that the tool 
design works right?)

>> By "large enough" we're presumably talking about tens of millions of 
>> units?
> 
> Not necessarily, just large enough that you use up most of the lifetime 
> of the tool, otherwise things get very expensive (and you would use a 
> cheaper method of production than making hard tools).

Right. So something like a few tens of thousands at least? (I guess like 
anything, it also depends on how fast you want to have the item...)

>> Yeah, figures. So a plastic cup is going to be pretty cheap, but a 
>> complicated jig for holding a structure together isn't.
> 
> You got it, additionally if there are lots of sections that need to have 
> accurate tolerances (eg because they attach to some other part) then it 
> gets expensive.

What is the tolerance on a typical hinge? (E.g., the hings on a CD 
case.) How about something that's supposed to snap together? And while 
we're on the subject, how about something like a USB plug? (Presumably 
that needs to be a snug fit, but not "too" tight.)

>> Perhaps I should clarify: I'm talking about a "keyboard" of the 
>> musical kind.
> 
> Ah ok, well yeh maybe they do, or they just use thicker plastic?

Actually, thinking about it... It probably *is* metal for EM shielding 
purposes! ;-)

>> Have you ever noticed that cheap PC cases always have jagid edges on 
>> their insides? Yet if you buy a £800 server case, all the edges are 
>> beautifully smooth? (And the steel seems about 4x thicker, BTW.)
> 
> Yes, the tooling for nice smooth edges requires more steps (metal tools 
> usually have several steps because it's impossible to bend and cut 
> everything correctly in just one stamp) = more cost.

Plus of course anything that says "server" on it is instantly 10x the 
price anyway. ;-) But yeah - how the hell *do* you smooth all those 
edges? :-.

>> I have no idea what steel costs per Kg, but presumably a lot more than 
>> plastic!
> 
> I don't think it's vastly different, again it's usually the tooling and 
> manufacturing costs that far outweigh the raw material costs.

Well I don't know - they like to charge you 10x for PC cases that are 
aluminium rather than steel. Is the metal that much more expensive, or 
is it just a luxury premium?

I'll bet *wood* costs more! ;-)

>> Also, presumably thick steel takes more effort to bend and cut...
> 
> Have you seen the size and general beefyness of the machines and tools 
> that actually do the bending and cutting :-)
> 
> http://www.pdf-inc.com/img/press3.jpg

OMGWTFBBQ!! O_O

I wonder what the electricity bill for that is like? :-P

Reminds me of a TV program I saw. Guy wanted to make some new leaf 
springs, so he goes to the metal shop. They have a hydraulic press, and 
we watch it casually snip through 23mm spring steel in a way not unlike 
scissors cut through tissue paper. (!) "Mind your fingers" the operator 
says. o_O (Fortunately, the machine rolls the prescribed length of sheet 
past the cutters and automatically cuts it, so your body parts don't 
have to go anywhere near the vicious thing!)

> And an example of a tool:
> 
> http://img1.tradeget.com/sophiaxiafei/FR5WPIWW1progressive_die_zy1.jpg

What the *hell* is it a tool _for_ though?

Post a reply to this message

> I just notice that plastic objects always seem to be made of *the* 
> thinnest plastic possible without breaking, that's all.

Well yeh, someone will have designed the thickness to withstand certain 
forces, there's absolutely no point in making something thicker than it 
really needs to be.  To you and me, saving 0.1p per plastic part seems 
insignificant, but if you are selling 2 million products per month, and each 
one contains 10 plastic parts, that adds up to quite a lot in the end.

> Does it vary significantly depending on which type of plastic you want?

Yes, some plastics are much stronger than others, have different thermal 
properties (both at high and low temperatures), you can even add tiny bits 
of metal or carbon to make them stronger, there are all sorts of 
possiblities.

> Do you buy the plastic with dye already added, or do you add that when you 
> come to use it?

Dunno, we just use plain white, I guess the parts supplier orders the 
plastic from the plastics supplier in the colour you want.

> Does it make any difference what kind of finish you want on the item? 
> (Matt, shiny, textured, etc.)

Yes, but the surface finish is determined by the mould, not the plastic.  If 
you put a rough surface on the metal mould you will get a rough plastic 
finish, put an ultra smooth finish and you get a shiny plastic part.

> Right. So tens of thousands of pounds just for the tool itself? Damn, I 
> don't even wanna *know* what the machine to work it costs... o_O

But then that machine probably lasts for many years.

> Mind you, I guess if *you* had molten liquid poured onto you under extreme 
> pressure thousands of times a day, you'd wear out pretty fast too... ;-)

Exactly :-)  AIUI it's the sharp edges in the mould which start to wear 
first, and depending on how important those sharp edges are in your design I 
guess it can be the difference between you only being able to get 100k parts 
or 1000k parts from the tool.

You know in the days of LPs, you could pay extra to get the high quality 
cuts that were guaranteed to be in the first batch from the set of tools. 
The ones that were made near the end of the life of the tool were lower 
quality.

> Presumably for prototyping they tend to use all the "3D printers" we keep 
> hearing about? (Do they use soft tools to check that the tool design works 
> right?)

3D printers are useful for seeing what things look like, but if you want to 
make more than a few pieces, or actually test them for strength and 
tolerances etc then they're quite useless.  Usually if you are designing 
something with volumes in the millions then you are going to want to make at 
least a few thousand parts to test before committing to mass production.

> Right. So something like a few tens of thousands at least? (I guess like 
> anything, it also depends on how fast you want to have the item...)

Yeh, if you want say 20k parts, then they can probably design and make you 
some tools for that purpose, but don't expect the per-part cost to be as 
cheap as if you went and said you wanted 200k parts per month for the next 4 
years.

> What is the tolerance on a typical hinge? (E.g., the hings on a CD case.)

+/- 0.1 mm is pretty easy to achieve in a single shot plastic injection, 
that sounds about right for a CD hinge to work (if the tolerance was much 
bigger then either the two halves would keep falling apart with very little 
force, of it would be too stiff to rotate easily).

> How about something that's supposed to snap together? And while we're on 
> the subject, how about something like a USB plug? (Presumably that needs 
> to be a snug fit, but not "too" tight.)

A quick google reveals that chapter 6 of the USB spec specifies that :-) 
The outline dimensions of the plug are +/-0.1mm, but some of the pin 
spacings and inner dimensions are down to +/- 0.05

There are tables that specify tolerances of holes and things that fit in 
holes, and what they should be for "very tight fits", "lose fits" etc.

> Plus of course anything that says "server" on it is instantly 10x the 
> price anyway. ;-) But yeah - how the hell *do* you smooth all those edges? 
> :-.

Clever tool design.

>> http://img1.tradeget.com/sophiaxiafei/FR5WPIWW1progressive_die_zy1.jpg
>
> What the *hell* is it a tool _for_ though?

In that case it is just two lumps of metal that are pressed together very 
hard with a thin sheet of metal between them.  When they are pulled apart 
again the sheet of metal has magically been cut and bent in clever ways to 
the shape that the designer desired :-)

Tools can be quite complex, incorporating cutting edges, springs and levers 
etc to do all sorts of fancy folds and things as the two halves are brought 
together.

Post a reply to this message

scott wrote:
>> I just notice that plastic objects always seem to be made of *the* 
>> thinnest plastic possible without breaking, that's all.
> 
> Well yeh, someone will have designed the thickness to withstand certain 
> forces, there's absolutely no point in making something thicker than it 
> really needs to be.  To you and me, saving 0.1p per plastic part seems 
> insignificant, but if you are selling 2 million products per month, and 
> each one contains 10 plastic parts, that adds up to quite a lot in the end.

Plenty of plastic items seem to not really be thick enough to withstand 
even "reasonable" levels of force. Still, I guess the designers 
considered that reasonable for the price.

And yes, I remember going to Maplin and ordering a big of 7400s. And I 
distinctly remember they weren't 21p each, they were something likt 
20.153p each. I remember thinking that was really weird - until I 
realised that if I buy 1,000 of them, I just saved £8.47. o_O

>> Does it vary significantly depending on which type of plastic you want?
> 
> Yes, some plastics are much stronger than others, have different thermal 
> properties (both at high and low temperatures), you can even add tiny 
> bits of metal or carbon to make them stronger, there are all sorts of 
> possiblities.

One would expect that - logically - the more sturdy plastics would be 
more expensive. But is that actually true? Or is it more driven by how 
difficult the pastic is to manufacture?

>> Do you buy the plastic with dye already added, or do you add that when 
>> you come to use it?
> 
> Dunno, we just use plain white, I guess the parts supplier orders the 
> plastic from the plastics supplier in the colour you want.

I suppose it depends what you're making. Some pastics are painted rather 
than having dye inside them. I suppose if you make lots of different 
coloured items, you add the dye yourself, otherwise you just order the 
colour you want...

>> Does it make any difference what kind of finish you want on the item? 
>> (Matt, shiny, textured, etc.)
> 
> Yes, but the surface finish is determined by the mould, not the 
> plastic.  If you put a rough surface on the metal mould you will get a 
> rough plastic finish, put an ultra smooth finish and you get a shiny 
> plastic part.

I'm presuming a mould with a fake-leather texture to it has gotta be 
pretty expensive to make...

>> Right. So tens of thousands of pounds just for the tool itself? Damn, 
>> I don't even wanna *know* what the machine to work it costs... o_O
> 
> But then that machine probably lasts for many years.

Heh. Not a tool for the curios amature. ;-)

>> Mind you, I guess if *you* had molten liquid poured onto you under 
>> extreme pressure thousands of times a day, you'd wear out pretty fast 
>> too... ;-)
> 
> Exactly :-)  AIUI it's the sharp edges in the mould which start to wear 
> first, and depending on how important those sharp edges are in your 
> design I guess it can be the difference between you only being able to 
> get 100k parts or 1000k parts from the tool.

Yeah, that figures.

I wonder - if you want a really smooth object, how long does that last? ;-)

[Of course, it's fashionable for electronic items to have sexy curves on 
the outside, but usually inside they're all sharp edges for bolting 
circuit boards to, "fins" for extra strength, etc.]

> 3D printers are useful for seeing what things look like, but if you want 
> to make more than a few pieces, or actually test them for strength and 
> tolerances etc then they're quite useless.  Usually if you are designing 
> something with volumes in the millions then you are going to want to 
> make at least a few thousand parts to test before committing to mass 
> production.

Yeah, I guess if you want to check the bits fit together, a 3D model is 
fine, but if you want to see how strong it is, you need to use the real 
material that the final product will use.

Reading Wikipedia, it appears that "most" plastic items are 
thermoplastics rather than thermosetting. Is that true? Does that mean 
there's a possibility of reusing your prototypes once you're done with them?

>> Right. So something like a few tens of thousands at least? (I guess 
>> like anything, it also depends on how fast you want to have the item...)
> 
> Yeh, if you want say 20k parts, then they can probably design and make 
> you some tools for that purpose, but don't expect the per-part cost to 
> be as cheap as if you went and said you wanted 200k parts per month for 
> the next 4 years.

It's all economies of scale. Evidently. ;-)

>> What is the tolerance on a typical hinge? (E.g., the hings on a CD case.)
> 
> +/- 0.1 mm is pretty easy to achieve in a single shot plastic injection, 
> that sounds about right for a CD hinge to work.

That sounds really damned tiny! o_O

OTOH, if it's easy to achieve, then why not?

(I would have thought the metal in the mould would expand and contract 
too much due to the heat, but clearly the designers know more about this 
kind of thing than I do...)

> A quick google reveals that chapter 6 of the USB spec specifies that :-) 
> The outline dimensions of the plug are +/-0.1mm, but some of the pin 
> spacings and inner dimensions are down to +/- 0.05

Wow. OK, that's pretty small...

Suddenly I'm wondering what the tolerances are for those *tiny* little 
screws they have on things like spectacles. ;-) Man, the thread on those 
is tiny!

> There are tables that specify tolerances of holes and things that fit in 
> holes, and what they should be for "very tight fits", "lose fits" etc.

You might even know this one... What is the total travel on the buttons 
of a mobile phone? I mean, you can "feel" them click, but visibly they 
hardly seem to move at all! (But then, I hypothesize that the inside of 
a mobile phone is a very cramped place.)

>> Plus of course anything that says "server" on it is instantly 10x the 
>> price anyway. ;-) But yeah - how the hell *do* you smooth all those 
>> edges? :-.
> 
> Clever tool design.
> 
> Tools can be quite complex, incorporating cutting edges, springs and 
> levers etc to do all sorts of fancy folds and things as the two halves 
> are brought together.

Damn. So after some guy designs what the final thing is even meant to 
look like, some other dude has to figure out what seperate bits to make 
and how to slot them together, and then yet another guy has to figure 
out how the **** to build something to make stuff that shape! o_O

You have to design a whole product to construct your product!

Heh. Damn... I wonder how they cut the mould to exactly the right shape? 
Now that crap must have some *crazy* tolerances on it...

Post a reply to this message

> I'm presuming a mould with a fake-leather texture to it has gotta be 
> pretty expensive to make...

You mean like the one you have on the dashboard in your car?  The texture on 
the plastics on your car dashboard is both to give it a "quality" feel but 
also to avoid specular reflections from sunlight etc.

> Yeah, I guess if you want to check the bits fit together, a 3D model is 
> fine, but if you want to see how strong it is, you need to use the real 
> material that the final product will use.

Or if you want to check it will last OK continually in temperatures of -30 
degrees or +40 degrees (other countries outside of England have much more 
extreme temperatures, and things must still work ok).  Or if you want to 
check it still works after you drop it on the floor etc, that sort of thing.

> Reading Wikipedia, it appears that "most" plastic items are thermoplastics 
> rather than thermosetting. Is that true? Does that mean there's a 
> possibility of reusing your prototypes once you're done with them?

I don't think so, the cost of reprocessing the parts into something the 
plastics maker could use would be too high.

>> +/- 0.1 mm is pretty easy to achieve in a single shot plastic injection, 
>> that sounds about right for a CD hinge to work.
>
> That sounds really damned tiny! o_O

But think about it, if the hinge on one side was just 1mm off, it's not 
going to work at all.

>> A quick google reveals that chapter 6 of the USB spec specifies that :-) 
>> The outline dimensions of the plug are +/-0.1mm, but some of the pin 
>> spacings and inner dimensions are down to +/- 0.05
>
> Wow. OK, that's pretty small...

But again, think about what your USB plug would feel like if it was just 
0.5mm too wide, it wouldn't feel like it should and the plug would come off 
too easily.  That's why a smaller tolerance is necessary for some parts.

> Suddenly I'm wondering what the tolerances are for those *tiny* little 
> screws they have on things like spectacles. ;-) Man, the thread on those 
> is tiny!

Yeh, imagine the tolerance on the thread pitch of those :-)  Something 
insane like 0.02 +/- 0.005 mm !

> You might even know this one... What is the total travel on the buttons of 
> a mobile phone? I mean, you can "feel" them click, but visibly they hardly 
> seem to move at all! (But then, I hypothesize that the inside of a mobile 
> phone is a very cramped place.)

Don't know, but usually there are just tiny rubber "poppers" underneath, a 
bit like those toys you used to have as a kid that were like a thick 
hemisphere shell of rubber, you turned them inside out and set them down, 
then after a few seconds they jumped up in the air.  The click you feel on 
the keys is when the rubber "snaps" in and out of shape.

> Damn. So after some guy designs what the final thing is even meant to look 
> like, some other dude has to figure out what seperate bits to make and how 
> to slot them together, and then yet another guy has to figure out how the 
> **** to build something to make stuff that shape! o_O

Exactly - and you begin to realise how expensive, how many people, and how 
much time it takes to design something apparently simple like a mobile 
phone.  Then imagine designing something like a car, or a plane! It's just 
not possible to imagine how much work goes into such a product.

> Heh. Damn... I wonder how they cut the mould to exactly the right shape?

You can get milling machines that are controlled by very accurate stepper 
motors, a computer sends a list of commands to the machine and it moves the 
cutting bit around.

Post a reply to this message

scott wrote:
>> I'm presuming a mould with a fake-leather texture to it has gotta be 
>> pretty expensive to make...
> 
> You mean like the one you have on the dashboard in your car?

Yeah, like that.

Or in fact, just now I was looking at a cheap knife. The blade is a lump 
of metal that's just about sharp enough to cut bread, and the handle is 
a blob of black plastic that looks like it's injection-moulded. (It has 
a fetling line on it.) But then I noticed, the plastic has a rough 
texture to it. (To enhance grip?) How do they do that?

I wonder if they use something like the acid on "frosted" glass...

>> Yeah, I guess if you want to see how strong it is, you need to use the 
>> real material that the final product will use.
> 
> Or if you 
> want to check it still works after you drop it on the floor etc, that 
> sort of thing.

Yeah, well, that's a function of far more than the plastic shell. Gotta 
assemble the whole thing for that! ;-)

>> Reading Wikipedia, it appears that "most" plastic items are 
>> thermoplastics rather than thermosetting. Is that true? Does that mean 
>> there's a possibility of reusing your prototypes once you're done with 
>> them?
> 
> I don't think so, the cost of reprocessing the parts into something the 
> plastics maker could use would be too high.

Apparently by dad spent some time working in a plast where they do 
"vacuum forming", and each time you do that, you have to cut a bunch of 
wasted plastic off the edges. He says they sent it back to the suppliers 
and they got a discount. (But then, how hard can reprocessing a flat 
sheet of plastic be?)

>> Suddenly I'm wondering what the tolerances are for those *tiny* little 
>> screws they have on things like spectacles. ;-) Man, the thread on 
>> those is tiny!
> 
> Yeh, imagine the tolerance on the thread pitch of those :-)  Something 
> insane like 0.02 +/- 0.005 mm !

Hee... must be expensive! ;-)

>> You might even know this one... What is the total travel on the 
>> buttons of a mobile phone?
> 
> Don't know, but usually there are just tiny rubber "poppers" underneath, 
> a bit like those toys you used to have as a kid that were like a thick 
> hemisphere shell of rubber, you turned them inside out and set them 
> down, then after a few seconds they jumped up in the air.  The click you 
> feel on the keys is when the rubber "snaps" in and out of shape.

Right.

And that brings us back to my question about components which are 
supposed to "snap together". I guess it depends on how deformable the 
plastic is? (And that surely must depend on shape...)

>> Damn. So after some guy designs what the final thing is even meant to 
>> look like, some other dude has to figure out what seperate bits to 
>> make and how to slot them together, and then yet another guy has to 
>> figure out how the **** to build something to make stuff that shape! o_O
> 
> Exactly - and you begin to realise how expensive, how many people, and 
> how much time it takes to design something apparently simple like a 
> mobile phone.  Then imagine designing something like a car, or a plane! 
> It's just not possible to imagine how much work goes into such a product.

I wouldn't have called a mobile phone "simple", but yeah. I was thinking 
more something like a pen. (They usually have that click-action that 
allows you to retract the tip. And a lid that needs to fit properly. And 
the tip is usually metal, not plastic. And you need to fill it with ink 
somehow...)

It kinda makes you wonder how anything ever gets made in the first place!

>> Heh. Damn... I wonder how they cut the mould to exactly the right shape?
> 
> You can get milling machines that are controlled by very accurate 
> stepper motors, a computer sends a list of commands to the machine and 
> it moves the cutting bit around.

Heh, and I bet that's even *more* damed expensive again... not to 
mention the quality hardened steel you need for it to mill...

I wonder... do companies buy all this stuff themselves? Or just 
subcontract it out?

I also wonder... Today we have machines that make components of machines 
that are used to make machines for making machines. Some of this stuff 
is made to absurd tolerances - stuff no human could ever make by hand. 
So... how the **** did we make it in the first place?!

Dude, HOW DO YOU BOOTSTRAP CIVILISATION?! O_O

Post a reply to this message

Now, see, this is what I love about off-topic... There's so many people 
here who know interesting stuff! :-D

Post a reply to this message

> I wonder if they use something like the acid on "frosted" glass...

There are both mechanical methods and chemical etching methods to make the 
patterns.  Basically you are just making a pattern on the surface of the 
metal in the mould.  For one of our plastic products (which is an optical 
component so is transparent plastic) they make "fine adjustments" to the 
surface texture by some dude manually "roughening up" the surface in certain 
positions :-)

> Yeah, well, that's a function of far more than the plastic shell. Gotta 
> assemble the whole thing for that! ;-)

Usually you don't do any testing on individual parts, only on the assembled 
product.

> Apparently by dad spent some time working in a plast where they do "vacuum 
> forming", and each time you do that, you have to cut a bunch of wasted 
> plastic off the edges. He says they sent it back to the suppliers and they 
> got a discount. (But then, how hard can reprocessing a flat sheet of 
> plastic be?)

Well yeh, if from every single shot you have some waste then it makes sense 
to work out how to reuse that.  But working out how to reuse 5000 samples is 
a bit pointless if you are planning to make 2 million every month later.

> And that brings us back to my question about components which are supposed 
> to "snap together". I guess it depends on how deformable the plastic is? 
> (And that surely must depend on shape...)

Yep, or technically "stiffness", which is how much something deforms for a 
given force (units DISTANCE/FORCE).  You can of course design the shape so 
that things deform easily for clips, or make springs to hold things in place 
etc.

> It kinda makes you wonder how anything ever gets made in the first place!

Yeh I often wonder that too...

> Heh, and I bet that's even *more* damed expensive again... not to mention 
> the quality hardened steel you need for it to mill...

And then the actual milling bit, which presumably must be harder than the 
hardened steel :-)

> I wonder... do companies buy all this stuff themselves? Or just 
> subcontract it out?

Well you get quite long supply chains sometimes, like take for example car 
company 1.  They will have supplier 2 that makes some big chunk of the 
dashboard.  Supplier 2 will have a supplier 3 that provides one part (eg the 
radio).  Supplier 3 will then use supplier 4 to make the plastic part to go 
on the front of the radio.  Supplier 4 will own the injection moulding 
machine, but could well use a supplier 5 to actually manufacturer the tools 
to go in the injection molding machine.  And of course there will be another 
supplier 5 that supplies the raw plastic to supplier 4.

Companies like Nokia and Ford don't actually make much, they just design 
things and get other people to make them, then assemble bits together :-)  I 
once saw a Nokia factory, each production line was about 50 metres long, 
they simply screwed some components together, did some electrical and 
optical testing, clipped the plastic parts on and packaged it into several 
layers of cardboard boxes.  Very low tech and very simple.

> I also wonder... Today we have machines that make components of machines 
> that are used to make machines for making machines. Some of this stuff is 
> made to absurd tolerances - stuff no human could ever make by hand. So... 
> how the **** did we make it in the first place?!

Well just because you only have equipment to design and make stuff to 
tolerance X, it doesn't mean that you can't make a machine that can make 
stuff to a tolerance lower than X.  eg you could just build a motor and then 
gear it down a lot so that the movements are very small and accurate, or use 
optical lens to shrink your design, or use a lever so that a larger 
tolerance on one end corresponds to a smaller tolerance at the other end.

Post a reply to this message

scott wrote:
>> I wonder if they use something like the acid on "frosted" glass...
> 
> There are both mechanical methods and chemical etching methods to make 
> the patterns.  Basically you are just making a pattern on the surface of 
> the metal in the mould.  For one of our plastic products (which is an 
> optical component so is transparent plastic) they make "fine 
> adjustments" to the surface texture by some dude manually "roughening 
> up" the surface in certain positions :-)

Yay for manual roughening up dude! :-D

>> Yeah, well, that's a function of far more than the plastic shell. 
>> Gotta assemble the whole thing for that! ;-)
> 
> Usually you don't do any testing on individual parts, only on the 
> assembled product.

Not even to test that they actually fit together properly? Or is that 
more or less guaranteed before the parts leave the design stage?

(OTOH, I guess that assembling the final item *is* checking that the 
parts fit! :-D )

> Well yeh, if from every single shot you have some waste then it makes 
> sense to work out how to reuse that.  But working out how to reuse 5000 
> samples is a bit pointless if you are planning to make 2 million every 
> month later.

OK, fair enough.

>> And that brings us back to my question about components which are 
>> supposed to "snap together". I guess it depends on how deformable the 
>> plastic is? (And that surely must depend on shape...)
> 
> Yep, or technically "stiffness", which is how much something deforms for 
> a given force (units DISTANCE/FORCE).  You can of course design the 
> shape so that things deform easily for clips, or make springs to hold 
> things in place etc.

IIRC, if you bend plastics they return to their original shape, until 
you bend them past a certain point, at which they deform permanently. 
(They also change colour for no adaquately-explored reason...)

>> Heh, and I bet that's even *more* damed expensive again... not to 
>> mention the quality hardened steel you need for it to mill...
> 
> And then the actual milling bit, which presumably must be harder than 
> the hardened steel :-)

HOW THE HELL DO THEY CUT DIAMONDS?!

>> I wonder... do companies buy all this stuff themselves? Or just 
>> subcontract it out?
> 
> Well you get quite long supply chains sometimes, like take for example 
> car company 1.  They will have supplier 2 that makes some big chunk of 
> the dashboard.  Supplier 2 will have a supplier 3 that provides one part 
> (eg the radio).  Supplier 3 will then use supplier 4 to make the plastic 
> part to go on the front of the radio.  Supplier 4 will own the injection 
> moulding machine, but could well use a supplier 5 to actually 
> manufacturer the tools to go in the injection molding machine.  And of 
> course there will be another supplier 5 that supplies the raw plastic to 
> supplier 4.
> 
> Companies like Nokia and Ford don't actually make much, they just design 
> things and get other people to make them, then assemble bits together 
> :-)  I once saw a Nokia factory, each production line was about 50 
> metres long, they simply screwed some components together, did some 
> electrical and optical testing, clipped the plastic parts on and 
> packaged it into several layers of cardboard boxes.  Very low tech and 
> very simple.

Heh. That explains why if you drive round any small industrial estate, 
you'll find some nobody company that's never been heard of called 
"Lanstead Way Plastics Ltd" or something. They all mould plastics! :-D

>> I also wonder... Today we have machines that make components of 
>> machines that are used to make machines for making machines. Some of 
>> this stuff is made to absurd tolerances - stuff no human could ever 
>> make by hand. So... how the **** did we make it in the first place?!
> 
> Well just because you only have equipment to design and make stuff to 
> tolerance X, it doesn't mean that you can't make a machine that can make 
> stuff to a tolerance lower than X.  eg you could just build a motor and 
> then gear it down a lot so that the movements are very small and 
> accurate, or use optical lens to shrink your design, or use a lever so 
> that a larger tolerance on one end corresponds to a smaller tolerance at 
> the other end.

Yeah, but without precisely machined gears, wouldn't there be too much 
play between the teeth for the position to hold accurately?

Post a reply to this message

Invisible wrote:

>> A typical hard-tool for a simple plastic part (eg the front cover of 
>> your CD drive or half of a simple keyboard shell) will cost of the 
>> order of 10-40k pounds.
> 
> o_O
> 
> Ouch...

Pff.. That's nothing :) The jigging system I worked on cost about US$1M 
for a decently sized system. ~US$100K per 4 axis table. Our saws ran 
about US$100K-250K

I'm guessing Injection molding machines don't have computer controls. 
I'd imagine the engineering costs are pretty high, though ... getting 
all of the pathways right so the part fills the mold properly, and cures 
properly.

> Perhaps I should clarify: I'm talking about a "keyboard" of the musical 
> kind. Somebody might attempt to put it on a stand that only supports the 
> ends of the unit, and it's over a meter wide. (Even so, all it contains 
> inside is a circuit board. No batteries or power supply or anything 
> heavy, just the key action.)
> 

My el-cheapo Casio has a metal chassis. Ask me how I know ... I had to 
pull it apart to fix the MIDI out port ... stupid thing came un-soldiered.


-- 
~Mike

Post a reply to this message

>>> A typical hard-tool for a simple plastic part (eg the front cover of 
>>> your CD drive or half of a simple keyboard shell) will cost of the 
>>> order of 10-40k pounds.
>>
>> o_O
>>
>> Ouch...
> 
> Pff.. That's nothing :) The jigging system I worked on cost about US$1M 
> for a decently sized system. ~US$100K per 4 axis table. Our saws ran 
> about US$100K-250K

Jig system?

What are you trying to cut? Kryptonite??

> I'm guessing Injection molding machines don't have computer controls. 
> I'd imagine the engineering costs are pretty high, though ... getting 
> all of the pathways right so the part fills the mold properly, and cures 
> properly.

More to the point, apparently IM machines are rated on the amount of 
pressure they can clamp the mould down with - measured in tonnes. (!) 
Apparently the cheap ones manage 3-4 tonnes, and the best reach 6,000 
tonnes or more. O_O

> My el-cheapo Casio has a metal chassis. Ask me how I know ... I had to 
> pull it apart to fix the MIDI out port ... stupid thing came un-soldiered.

You as well?

(Actually, you should be glad it even *has* a MIDI port in the first 
place! Most cheap keyboards don't.)

We had to take my old ghettoblaster apart after the line in became 
loose. After we soldered it back down and reassembled it, we found one 
small metal part left over. We still don't know what that was!

Ah, that was a *quality* piece of equipment though. The box boasts 200W 
of power, but the actual speakers inside had figures of fractional Watts 
crudely printed on them. (2x speakers at 0.5W, 2x woofers at 0.25W.) And 
the manual described the line-in sockets as having "> 14% THD". :-D

Post a reply to this message

Mmm, tasty...

http://en.wikipedia.org/wiki/Injection_molding
http://en.wikipedia.org/wiki/Electrical_discharge_machining
http://en.wikipedia.org/wiki/Electro_chemical_machining

Post a reply to this message

Invisible wrote:

> 
> Jig system?
> 
> What are you trying to cut? Kryptonite??
> 

Jigging -- A means of holding something together while you work on it. 
Tack on another 10K for the rolling press... btw. (You have to press the 
plates into the truss.)

The saws cut ordinary wood, but they're robotic. Lots of engineering and 
programming time goes into them.


> 
> More to the point, apparently IM machines are rated on the amount of 
> pressure they can clamp the mould down with - measured in tonnes. (!) 
> Apparently the cheap ones manage 3-4 tonnes, and the best reach 6,000 
> tonnes or more. O_O
> 

If you can find it in your country, check out the show "How Its Made" 
Here it appears on the Discovery Science network. They detail many 
industrial processes. It blows my mind how these huge presses seem to 
stamp out such things as screwdrivers with nearly no effort.

The U.S. mint apparently uses 84 tons of pressure to press a coin. All 
applied in an instant! Hundreds of times a minute.


> 
> You as well?
> 
> (Actually, you should be glad it even *has* a MIDI port in the first 
> place! Most cheap keyboards don't.)
> 

Well, I had aspirations of creating music on the computer a long time 
ago, and decided a MIDI keyboard was the appropriate input device. That 
didn't work out... :)



-- 
~Mike

Post a reply to this message

"Invisible" <voi### [at] devnull> wrote in message 
news:494655da$1@news.povray.org...

> It kinda makes you wonder how anything ever gets made in the first place!

    Yes, you need a lathe to make a lathe.

     ~Steve~

Post a reply to this message

Invisible wrote:
> Mmm, tasty...
> 
> http://en.wikipedia.org/wiki/Injection_molding
> http://en.wikipedia.org/wiki/Electrical_discharge_machining
> http://en.wikipedia.org/wiki/Electro_chemical_machining

Also neat:

http://en.wikipedia.org/wiki/Ultrasonic_welding

-- 
~Mike

Post a reply to this message

>> It kinda makes you wonder how anything ever gets made in the first place!
> 
>     Yes, you need a lathe to make a lathe.

Hee... My uncle has a 2 tonne lathe.

It's big.


But then, this same uncle also has a 3-foot spannar. And a laser cutting 
tool.

Post a reply to this message

"Invisible" <voi### [at] devnull> wrote in message 
news:49466678$1@news.povray.org...

> HOW THE HELL DO THEY CUT DIAMONDS?!

  If you mean in the first instance when say, a large rough diamond needs to 
be made into several smaller faceted diamonds, well, they 'cleave' the 
rough. Basically, a hammer and chisel are used to smack a rough in half and 
then quarters, etc.

   But if you meant cutting the actual facets, well, with diamond of course. 
;)

   Actually, it's diamond paste. They start with a rough paste and work the 
facets down to a smooth finish with a fine paste and then they use even 
finer diamond paste (or spray) to actually polish the facets.

   ~Steve~

Post a reply to this message

Invisible wrote:
> 
> Hee... My uncle has a 2 tonne lathe.
> 
> It's big.
> 

Where I used to work had a lathe 12 feet long... They used it to machine 
the rollers for the presses. :)

-- 
~Mike

Post a reply to this message

Invisible wrote:
> You have to design a whole product to construct your product!

I always thought it was really complex, until I realized it's exactly what 
you do with software. You go through and solve the problems as they come up, 
and with some experience you find you rarely if ever have to go back to the 
beginning because something is wrong down the line.

-- 
   Darren New, San Diego CA, USA (PST)
   The NFL should go international. I'd pay to
   see the Detroit Lions vs the Roman Catholics.

Post a reply to this message

scott wrote:
> Exactly - and you begin to realise how expensive, how many people, and 
> how much time it takes to design something apparently simple like a 
> mobile phone.  Then imagine designing something like a car, or a plane! 

I've read that mobile phones are actually more complicated than commercial 
jumbo jet airplanes. I don't know how that's measured, mind, but I'd be 
willing to believe it. One rarely has to manage changing speed-of-light 
calculations when driving a car or a plane.

-- 
   Darren New, San Diego CA, USA (PST)
   The NFL should go international. I'd pay to
   see the Detroit Lions vs the Roman Catholics.

Post a reply to this message

> I've read that mobile phones are actually more complicated than commercial 
> jumbo jet airplanes. I don't know how that's measured, mind, but I'd be 
> willing to believe it. One rarely has to manage changing speed-of-light 
> calculations when driving a car or a plane.

Don't believe it!

Just think of all the navigation and control systems in a plane (both 
hardware and software), plus the fact that there needs to be 4-way 
redundancy and everything must be designed to work for 30 years rather than 
3 years.  Surely all that has got to be more complex than a mobile phone?

And that's before you've even considered how complex a modern jet engine is, 
or the retractable landing gear, the control surfaces, the control systems 
to keep the plane stable, the auto-pilot etc.

I guess there's a reason why you can design a new phone from drawing board 
to mass production in 18 months, but a plane takes nearer 10 years.

Post a reply to this message

scott wrote:
> Just think of all the navigation and control systems in a plane (both 
> hardware and software),

Not difficult to top. I mean, heck, we used to do *that* without any 
computers at all.

> plus the fact that there needs to be 4-way redundancy 

As opposed to supporting several versions of CDMA and GSM, bluetooth, GPS, 
HDTV, surround sound, and 3D graphics acceleration, all on one chip, running 
on a tiny battery for a week at a time, talking to assorted carriers all 
with their own versions of the standard implemented? :-)

I think it's certainly close.

> and everything must be designed to work for 30 years rather 
> than 3 years.  

I'll grant you that's a restriction you don't see on mobile phones.

> And that's before you've even considered how complex a modern jet engine 
> is, or the retractable landing gear, the control surfaces, the control 
> systems to keep the plane stable, the auto-pilot etc.

Again, all that was done before people even invented computers.

> I guess there's a reason why you can design a new phone from drawing 
> board to mass production in 18 months, but a plane takes nearer 10 years.

People don't design phones from scratch. Certainly the first analog cell 
phones were under development for some 70+ years, and digital cell phone 
technology was under development for 40+ years IIRC.

(Granted, I don't think people design jet planes from scratch, either, but 
that just means it's a bad comparison.)

-- 
   Darren New, San Diego CA, USA (PST)
   The NFL should go international. I'd pay to
   see the Detroit Lions vs the Roman Catholics.

Post a reply to this message

>> Just think of all the navigation and control systems in a plane (both 
>> hardware and software),
>
> Not difficult to top. I mean, heck, we used to do *that* without any 
> computers at all.

But it's all done *with* computers today.

> As opposed to supporting several versions of CDMA and GSM, bluetooth, GPS, 
> HDTV, surround sound, and 3D graphics acceleration, all on one chip, 
> running

And that's just for the passenger entertainment system on-board :-)

>> And that's before you've even considered how complex a modern jet engine 
>> is, or the retractable landing gear, the control surfaces, the control 
>> systems to keep the plane stable, the auto-pilot etc.
>
> Again, all that was done before people even invented computers.

And they were noisy and inefficient, today they are far more complex.  A bit 
like comparing a car engine from today with one from 50 years ago. 
Completely different and orders of magnitudes more complex today, but in 
return you get a much quieter, cleaner and efficient engine.

>> I guess there's a reason why you can design a new phone from drawing 
>> board to mass production in 18 months, but a plane takes nearer 10 years.
>
> People don't design phones from scratch.

Nor planes, but the typical timescale from concept to MP is a pretty good 
indicator of how complex the product is.

Post a reply to this message

scott wrote:
>>> Just think of all the navigation and control systems in a plane (both 
>>> hardware and software),
>>
>> Not difficult to top. I mean, heck, we used to do *that* without any 
>> computers at all.
> 
> But it's all done *with* computers today.
> 
>> As opposed to supporting several versions of CDMA and GSM, bluetooth, 
>> GPS, HDTV, surround sound, and 3D graphics acceleration, all on one 
>> chip, running
> 
> And that's just for the passenger entertainment system on-board :-)
> 
>>> And that's before you've even considered how complex a modern jet 
>>> engine is, or the retractable landing gear, the control surfaces, the 
>>> control systems to keep the plane stable, the auto-pilot etc.
>>
>> Again, all that was done before people even invented computers.
> 
> And they were noisy and inefficient, today they are far more complex.  A 
> bit like comparing a car engine from today with one from 50 years ago. 
> Completely different and orders of magnitudes more complex today, but in 
> return you get a much quieter, cleaner and efficient engine.
> 
>>> I guess there's a reason why you can design a new phone from drawing 
>>> board to mass production in 18 months, but a plane takes nearer 10 
>>> years.
>>
>> People don't design phones from scratch.
> 
> Nor planes, but the typical timescale from concept to MP is a pretty 
> good indicator of how complex the product is.

The main difference I'm noticing is that... mobile phones aren't safety 
critical.

I don't even wanna *know* what kind of tests a plane has to pass...

Post a reply to this message