Re: Something good

On 2007-12-15, Rowland McDonnell <real-address-in-sig@xxxxxxxxxxxxxxx> wrote:
Whiskers <catwheezel@xxxxxxxxxxxxx> wrote:

Rowland McDonnell <real-address-in-sig@xxxxxxxxxxxxxxx> wrote:
Whiskers <catwheezel@xxxxxxxxxxxxx> wrote:

Rowland McDonnell <real-address-in-sig@xxxxxxxxxxxxxxx> wrote:
Whiskers <catwheezel@xxxxxxxxxxxxx> wrote:

[...]

Ironically, the car-pound charges less than some less secure
long-term car-parks in central London, so the owner may not have been
quite as daft as appears at first sight.

He is that daft 'cos his car was going to be crushed and he appeared -
from the quotes in the article - to be completely unbothered by the
prospect. One of these horribly rich people who has homes in various
big cities around the world and a flashy fancy expensive car in each of
'em - from the sound of it, anyway.

Ah, well, if a chap can forget that he parked a Lamborghini somewhere in
London, then he has bigger problems than mere money.

[...]

I suspect that the nutating disc engine may not be reliably self-starting,

Could be - I've not looked to see if I can figure it out, but if not -
well, it's not a lot of bother these days to fit an electric motor to
spin the thing for starting, is it?

No, but that adds something else that can go wrong; if you're trying to
keep your spy-plane drone or stealth kill-bot as light and simple as
possible that counts as a drawback even if it isn't a
forget-this-idea-then problem.

However, from my reading of the mode of operation, there doesn't seem to
be any position in the cycle - as there is with a crank-and-piston
arrangement - where you can have a `no torque on the crank' when
applying force to the piston. Seems that the thing'll nutate when you
shove stuff in, no matter what the position of the bits. But I could be
wrong - it's all a bit tricky to figure out from static 2D drawings,
innit?

Yes. I think that the working fluid has to be moving for the thing to
work, and unless the disc is nutating then the fluid won't move so the
engine has to be running before it starts, if you see what I mean. In the
old water-driven version, heaving on the fly-wheel would probably do the
trick (or heaving on a rope wrapped around the rim of the flywheel but not
fixed to it, if you wanted to avoid getting kicked by what would probably
have been a mighty big kick when the thing did start).

and a version using combustion products as the working fluid might be
self-destructing if it fails to start.

One possible solution to that might be to either have a starter motor to
spin it up to speed - after all, one spins a conventional IC engine for
starting - or some kind of pressure release vent that's activitated for
starting to permit only a bit into the power conversion chamber so it
can't be destroyed no matter what. Or maybe a bit of both? I dunno -
even with hand-waving from the outside, it doesn't seem like a problem
that'd stop a serious engineer for very long.

Quite, there will be solutions.

I think that the expansion works in a very dynamic way somehow. Umm.
To do with the intertia of `stuff already inside the chamber' moving
around and preventing or allowing movement here and there at different
times.

I think I'm going to have to sit down with an orange, a cocktail stick,
and a paper collar just to get it straight in my head.

The moving part doesn't change the two volumes within the enclosure, it
just moves them around a bit.

Yeah - that's what it seems like, and frankly it does my head in trying
to figure out how the hell it works, quite. Something like a turbine,
certainly - but it's weird. One gets used to `the standard ways of
doing things' and this is very different, isn't it?

Weird is the word for it. But fluid dynamics /are/ weird; almost as weird
as gyroscopes and electricity and light. Not to mention gravity and
magnetism.

The original mill engine was powered by
water, as a direct substitute for a conventional water-wheel;

Yep - although it was meant as a supplement to conventional water
wheels, the point being that they needed more power than they could get
with water wheels from the water they had. But water wheels mostly just
extract power from the water falling over them (and a bit of the KE of
the flow, as well). What this engine was for was extracting energy from
the PE of high pressure water with a big head behind it, then permitting
the exhaust to be used for the existing water wheels with no loss of
power.

Which I thought was dead ingenious.

Effin' brilliant, it was! :)) And far easier to make than any
'conventional' sort of turbine (which they may not have known or thought
of anyway).

you don't
get much expansion with water as the working fluid, particularly with no
heat-source involved. All you have there is the pressure difference
between the inlet and exhaust, just like a water-turbine or 'hydraulic
ram' <http://virtual.clemson.edu/groups/irrig/Equip/ram.htm>.

Hyrdaulic rams are a bit other, mind. They work on compressing air in -
erm, chamber no. 15 in that picture - using the KE of water flowing
through the thing. If you don't have enough air in the chamber, it'll
hammer itself to bits. For sure you need a pressure difference to get
the air compression, but that comes from suddenly stopping the flow of
the driving water supply - it's KE you're using in that design to get
the pressure increase.

Kinetic energy in the form of the shock-wave reflected from the mouth of
the intake pipe, which is created by the sudden closing of the stop-valve
which is caused by the turbulence of the intake water flowing around said
valve ... it's another of those weird things. There may be shock-waves
and turbulence involved in the operation of the nutating disc engine - in
fact I'm sure the answer lies in that somewhere.

I've got a book on how to make one, bought at the alternative
technology/energy centre in N Wales back in the 1970s, IIRC. Never done
anything with it - turns out that the water flow you need to operate the
design in the book was a fair bit more than you got out of a single tap
at my parents' house.

[snip]

If you were connecting the intake of the ram directly to the tap, then the
problem was a combination of far too much pressure and an effective intake
pipe going all the way to the water-works, which is far too long. The
shock-wave from the closing of the stop-valve would have been absorbed in
the water mains and never reflected back so you wouldn't get much of a
pulse to push water into the pressure reservoir so the pump would never
work properly, if at all. A hydraulic ram is a gentle, subtle, thing, and
the diameter and length of the low-pressure intake pipe are critical to
the operation of the device. You only need a foot or so of 'head' to make
one work. A farm near where I grew up had one fed by a few yards of iron
pipe laid in the bed of a stream running down a steep slope, and fed water
to a cattle trough about 100 feet higher up the hill and a few minutes
walk away. Only a very small proportion of the water that flows into the
thing is actually pumped out at higher pressure; most of it just flows away
when the stop-valve is open.

There are published tables for working out the proportions of the intake
pipe and pressure-reservoir for various 'heads' and flow-rates.

There's a nice Wikipedia page for
Stirling engines <http://en.wikipedia.org/wiki/Stirling_cycle>.

I've long felt that someone should figure out a way to use all that heat
energy that you dump to the `environment' via radiator to get them
things to work.

Then again, I've long felt that way about the heat energy dumped from
normal engines.

Stirling engines were popular as a way to scavange heat from solid-fuel
kitchen ranges to drive spits or ventilators or even small electric
generators.

Hmm! Yes.

A modern variation is to use the Stirling engine as part of a
combined heat and power system, with the waste heat from the engine being
used to provide water and space heating.

Nice!

The quietness of the engine
makes it more acceptable than a diesel close to housing, and the
relatively continous combustion should make for cleaner exhaust too.

Yep. I suppose one nice thing about 'em is that they'll take almost any
heat source you can think of to power 'em.

Even sunlight, if there's enough of it around. There are 'executive toy'
versions that run from the heat of a cup of coffee, or even the palm of a
hand. Or you can find a source of 'cold' instead of 'heat', if that
approach is easier. And of course, a Sterling engine works as a 'heat
pump' if you apply an external drive to it.

Given so many alternatives whose proponents insist are so much better,
it's remarkable that Otto and Diesel engines still dominate.

Those designs are the ones that are highly developed, the ones that
people feel comfortable with, and so on.

Wankel engines, for example, *OUGHT* to be more efficient than piston
engines - but they're not, because they've not had as much development
money spent on them.

There is the little matter of sealing the tips and sides of the rotor;
that will probably always be the biggest problem with that system.

That problem was solved back in the 1970s. The big problems now are
down to getting combustion working efficiently - they've got problems
with `quench' due to the typically long, thin combustion chamber shape.

I'd like to see a multiple-spark ignition and a ceramic (i.e., hot
running to reduce quench) combustion chamber and rotor. I reckon if you
had a pair of spark plugs and fired each one with multiple sparks, you'd
do something useful. Certainly at least work modelling in the
CFD-combustion-modelling package I'm sure must have been developed by
someone.

A compression-ignition version might work better - if they could get the
seals to work at diesel-engine pressures. Or take the cumbustion into a
seperate chamber and use the rotor as a sort of turbine.

Once upon a time, I'd've said `certainly worth building one just to find
out' - but you don't have to do that any more.

The trouble with computer models, is that they can only model what you
already know. Which is why they don't work spectacularly well for weather
forecasting, and shouldn't be relied on at all for predicting anything to
do with 'climate change'.

WHy do you think that the technically *appalling* x86 ISA has lasted so
long? The real reason is that `those who have learnt how to work with
it are opposed to change because they like what they know'. It's all
down to human conservatism and nothing to do with engineering or
economics.

That was probably just as much a matter of luck for Intel as it was for
Microsoft when IBM were panicked into producing some sort of desktop
computer in a hurry;

It wasn't so much of a panic - so I've heard - as a tentative first step
into the PC world with a preliminary design that was meant to be much
slower than its mini computers (hence the inefficient internal
architecture that couldn't be persuaded to work quickly, which is why it
put up such a poor showing against the BBC Micro in speed tests, for
example)> and replaced with a proper PC design if IBM's foray into PCs
looked like it was working out. That's part of why they didn't bother
with any sort of design protection - it just wasn't worth it from IBM's
point of view.

The proper PC design never turned up because the first intentionally
crappy design sold so bloody well - in part because people could clone
it, once a `clean room' IBM-compatible-but-not-written-by-IBM BIOS had
been created.

Also because mere desk-jockeys (such as me) could actually get our hands
on the DOS and BASIC supplied and make them do real-world useful stuff
without having to get the data-processing specialists in to do feasability
studies, systems analysis, program design, de-bugging, etc, etc, all at
great expense and taking far too long for solutions required yesterday.

For some reason, management were much easier to convince to buy ugly great
boxes labeled IBM than they were to buy sleek pocketable 'programmable
calculators' and 'pocket computers' labeled Casio or even, God forbid, JVC,
even if they could do much the same stuff at far less cost and had been for
years - my Casio FX-something or other, circa 1979, ran for months off a
pair of button cells and could run rings around the original IBM PC as
shipped, and was still doing sterling work in '87 when my employers
finally started getting 'personal computers' installed more than one per
floor of the building. I expect the successors to my employers still
have micro-filmed (or perhaps even digitised, by now) copies of print-outs
I produced on my little 2" silver-paper thermal printer.

And then along came Wordstar and Lotus with real programs that could
replace whole departments of clerks and typists with just a few cheap
machines; businesses were not going to let that get away!

once the ball started rolling it became almost
unstoppable. Better technology can't beat market forces; look at Betamax
versus VHS, open-reel versus cassette audio, discs versus cylinders, Mac
versus DOS, ...

Well, what's `better' in those cases? Cassette audio is better than
reel-to-reel from the point of view of user convenience. Imagine trying
to handle 1/4" mag tape in a car.

That's what the '8-track cartridge' was all about ;)) Audio from
cassette tapes is ghastly, even with a lot of clever electronics and
"Dolby". The format was intended for office dictating machines.

VHS was superior to Betamax in one
respect: you could get a whole feature film on a single tape.

Betamax could be made to do that too; I have Betamax tapes of feature
films. (What I don't have, is a working Betamax machine). The real killer
was the tape-hire shops doing a deal to sell subsidised VHS machines from
JVC - and JVC allowing their specification to be 'open' so that you didn't
have to rely on JVC for all the machines, unlike Sony and Philips who kept
their stuff proprietary.

The fact
that VHS did this by accepting lower visual and audio quality didn't
seem to bother people.

Still doesn't; people think that CDs are 'the best ever' and that "MP3s"
(even if they're really AACs or WMAs or something) are 'just as good
really'. Which they probably are, if you're only using in-ear ear-phones
to listen to the Spice Girls while riding the tube.

One reason that MS-DOS computers out-sold Macs is that they had one
technical advantage: they were *HUGELY* cheaper than Macs - expensive
compared to almost all other PCs, but a lot cheaper than Macs which back
in the early days weren't just more expensive to make (and they were,
lots, because of the more RAM and more equipment and faster CPUs and all
that, not to mention the built-in monitor and compulsory mouse and so
on)

Yes.

but also because the idiot from Coke in charge thought Apple'd do
better if it deliberately over-charged for the Mac - they asked a much
higher profit margin than anyone else in the biz thought reasonable.
And of course when they sold 'em in the UK, giving us a price that was
`the same number of pounds that USians paid in dollars' - well, only
rich people could afford them at all as PCs.

Coke have always charged far more for their product than it's worth, so it
gets to be a habit, I suppose. Levi tried to lever-up the price of their
work-clothes in Europe to turn them into 'high status' items, a year or so
ago, but Walmart-Asda and Tesco didn't let them get away with it.

Apple had stopped all that nonsense by the 1990s, but the damage had
been done by then: everyone `knows' that Macs are more expensive. These
days they're usually cheaper than the equivalent spec `dark side' PC
(from a brand name, that is - must compare like-with-like) - but the
`dark side' zealots still think that Macs are over-priced. Oh yeah, and
not only are Macs cheaper than the competition, but apparently they run
Windoze better than the average `dark side' machine. And there's not a
lot of difference betweeen Macs and the `dark side' at all these day.
Not a lot of difference at all.

Why do you think that Macs have suddenly become so much more popular now
they've got x86 CPUs?

Have they?

Yes. I've read the stuff from various people - I'm convinced that the
reason the zealots have switched opinions is the x86-ness of modern
Macs. Just that, puts it in their comfort zone.

It couldn't be to do with PC World starting to sell Apple
laptops?

Doubt it. Shops that sell Macs as well as `dark side' machines
generally - well, the sales people don't like Macs, and they put people
off buying them. I've heard dozens and dozens of tales of sales people
in PC World and Dixons explaining to prospective customers how it is
that Macs are not what they want, they're incompatible or whatever,
Windoze can do everything a Mac can, you don't want a Mac, you really
don't.

Surely no-one takes 'advice' from the staff in PC World? They seldom know
anything more about the stock than how much commission they get for
selling it (and the 'service agreement' or 'insurance' too of course).
There's probably a lower 'margin' on Apple than on, say, e-machines or
Packard Bell.

I've asked a few questions of sales people in such shops - about Macs -
and, well, they either don't know what they're talking about, or they're
deliberately trying to put me off buying Mac.

Try asking 'does this work with Unix?'. Or even 'how much do I save if I
get this without Vista installed?'.

They've been selling Macs in such shops since the 1980s - it's always
been bad for Apple in a way, despite the sales that must be generated.
Macs need to be sold from dedicated Mac outfits.

Or they give advice that's just so *WRONG*. It's why Apple came up with
the whole Apple Store idea.

Computers need to be sold by people who know what they're talking about;
like cars and cameras and shoes and lots of other things. But people who
know what they're talking about expect more than minimum wage and a
temporary contract, and that puts the prices up. So don't go to the cheap
shops unless you think /you/ know what you're talking about. (Not that PC
World is cheap very often).

Or Apple's making it possible for people to get Windows running
in their Apple hardware

It's been possible pretty much ever since Windoze came out - via
emulation, mind.

Ah, but that means having to run Mac ...

[...]

One thing that x86 types don't like is the `other way' of bit ordering
numbers that quite a lot of CPUs use - I'm used to `LSB first'; Intel
uses `MSB' first. And MSB first doesn't feel right to me; I expect x86
users feel the same about LSB first. That's actually a big reason for
Macs now being more popular amongst a certain type of zealot - although
I suspect it's not a *HUGE* group.

A tiny group; who needs to get down to the level of the CPU architecture
these days?

The 'Intel inside' campaign has convinced a lot of people that anything
else just isn't 'as good', although they have no idea why or how that might
be so beyond what it says on the shiny stickers. Technical aspects have
nothing to do with it. It's got so bad that if a machine lacks the Intel
sticker some people hesitate to buy it even if it has the latest Microsoft
stuff pre-installed. (My father for instance, had to be convinced that the
AMD-based PC his local computer expert was trying to sell him might
actually be /better/ than the more expensive branded one in the magazine
that had the Intel sticker on it; he'd never heard of AMD, but the Intel
brand was all over the place). As for considering something that doesn't
even have Microsoft stuff on it, well, you can forget it as far as many
people are concerned. Even if they can't tell the difference when using
the thing. Marketing goes a long long way if you get it 'right'.

their OSs can be run on a wide range of CPUs from mobile phone to
super-computer - that most people use them on Intel or similar CPUs is
merely due to those machines being the most easily acquired,

Yeah, frequently hauled out of skips (or `dumpsters' as they call 'em
over the pond where most of this goes on).

Some afficionados love to rescue old Sun work-stations and Apple servers
and such like and make them run Linux. Others hack into routers and
mobile phones (and iPods, probably) and turn them into 'something else'.

thanks to
Microsoft (whose OSs mostly won't run on anything else).

Back in the early days, MS-DOS was meant to be run on anything at all.
It was supposed to be a platform-independent OS. Thing is, what with
compiled software being compiled software, the fact that you had MS-DOS
software in the box didn't mean it was going to run on the particular
MS-DOS machine you happened to own - unless, of course, it was `IBM
compatible' with an x86 CPU...

That's down to the 'proprietary software' model; you can't distribute the
source code for people to compile or port on their preferred systems if you
want to keep the code 'secret'. So buyers have to get whatever system the
program they want will run on from the binary packages available, or make
do with what they can get for their existing systems, or create their own
thing from scratch.

--
-- ^^^^^^^^^^
-- Whiskers
-- ~~~~~~~~~~
.