Directions in space?

In article , John Schilling
wrote:

On Sat, 21 Jun 2008 21:49:46 -0600, "David M. Palmer"
wrote:

In article , dlzc\
wrote:

Wonder why we don't use vacuum tubes for high power circuits in
space? We have vacuum, so we don't need the glass envelope...

They actually do use 'traveling wave tubes' (Wiki or Google for
details) for some transmitters on spacecraft. But they don't send them
up without envelopes.

One of the reasons for this is that spacecraft systems require a lot of
testing. And it would be helpful to test them without having to put
the whole system into a huge expensive vacuum chamber.

Actually, spacecraft are in fact tested by putting them in huge
expensive vacuum chambers.

Yes, but that part of the testing is very expensive. It is mostly used
to ensure that the system can work in a vacuum. Since most of the
problems are likely to be overheating/overcooling related, they include
huge banks of lights to simulate the sun and liquid-nitrogen-cooled
walls to simulate the radiant flux environment. This is called a
thermal vacuum chamber.

Cost per hour to run this testing is considered high even by spacecraft
people.

It is also expensive.

There is a lot more testing that a spacecraft requires. Running it for
hundreds of hours to work out the bugs. Shaking it at 15 gees to
simulate launch. Ensuring that all the modes work. Making sure that
when the transmitter turns on it doesn't blow out the sensors.

An anechoic RF chamber is not cheap. Making the same chamber a thermal
vacuum chamber would not be cheap.


Doing all that in a thermal vac chamber would cost more than a glass
envelope around the tubes.

--
David M. Palmer (formerly @clark.net, @ematic.com)

In article ,
"Mike Dworetsky" writes:
So if there is still a Sun in the story, the loss of the Moon can't be a
natural process.

If I'm remembering -- vaguely -- the right story, there had been a
lot of "solar system engineering" during the time the space ship had
been away.

Moving planet orbits turns out not to be as difficult as one might
expect, at least on time scales of tens or hundreds of thousands of
years. I believe there was a _Nature_ paper on the subject several
years ago.

--
Steve Willner Phone 617-495-7123
Cambridge, MA 02138 USA
(Please email your reply if you want to be sure I see it; include a
valid Reply-To address to receive an acknowledgement. Commercial
email may be sent to your ISP.)

On Sun, 22 Jun 2008 14:38:38 -0500, Michael Ash wrote:

I'm sure I'll get flamed for this, but my impresion of tubes used in
guitar amplifiers is that they are essentially psychological at this
point. There's nothing preventing you from building a semiconductor
circuit with identical response, and it would be cheaper and use less
power. But they would fail in the psychological department, because people
*think* that tubes sound better. This, in my mind, doesn't constitute a
decent reason to use tubes.

So, I'm guessing you really don't want a sound card for your computer that
uses vacuum tubes? (I think AOpen made a motherboard that had a tube
socket on it...)




--
chuk

In rec.arts.sf.science Chuk Goodin wrote:
On Sun, 22 Jun 2008 14:38:38 -0500, Michael Ash wrote:

I'm sure I'll get flamed for this, but my impresion of tubes used in
guitar amplifiers is that they are essentially psychological at this
point. There's nothing preventing you from building a semiconductor
circuit with identical response, and it would be cheaper and use less
power. But they would fail in the psychological department, because people
*think* that tubes sound better. This, in my mind, doesn't constitute a
decent reason to use tubes.

So, I'm guessing you really don't want a sound card for your computer that
uses vacuum tubes? (I think AOpen made a motherboard that had a tube
socket on it...)

My computer already puts out an uncomfortable amount of heat in the
summertime without having an element inside it which exists for the sole
purpose of being hot. It might be nice in the winter, but I have more
efficient ways to heat my house.

For sound quality, I'm pretty sure that you could program a digital sound
card to perfectly mimick the tube's output for a fraction of the cost and
power usage.

--
Mike Ash
Radio Free Earth
Broadcasting from our climate-controlled studios deep inside the Moon

In article ,
Michael Ash wrote:

For sound quality, I'm pretty sure that you could program a digital sound
card to perfectly mimick the tube's output

Not perfectly - but probably good enough for the human ear to be unable to
hear any difference. Looking at the waveform on an oscilloscope could
still show some differences though.

for a fraction of the cost

Not likely - hiring a programmer is not cheap. Buying a vacuum tube could
be much cheaper...

and power usage.

Depends on the tube. A miniature vacuum tube doesn't need more power than
millions of transistors...... :-)




--
----------------------------------------------------------------
Paul Schlyter, Grev Turegatan 40, SE-114 38 Stockholm, SWEDEN
e-mail: pausch at stjarnhimlen dot se
WWW: http://stjarnhimlen.se/

In rec.arts.sf.science Paul Schlyter wrote:
In article ,
Michael Ash wrote:

For sound quality, I'm pretty sure that you could program a digital sound
card to perfectly mimick the tube's output

Not perfectly - but probably good enough for the human ear to be unable to
hear any difference. Looking at the waveform on an oscilloscope could
still show some differences though.

You choose how close you want to get, then you choose your components to
match it. A 24-bit DAC will get you to within one part in 16 million of
where you want to be, and they're quite common these days.

for a fraction of the cost

Not likely - hiring a programmer is not cheap. Buying a vacuum tube could
be much cheaper...

Sure, if you want to build *one*. But generally you produce these things
for sale, at which point the cost of the programmer is divided across all
of the units you produce, whereas you pay for the tube for every one.

And of course for the digital solution there's no reason to be building
hardware at all. Use a commodity sound card and just write a little driver
that preprocesses the audio being output.

and power usage.

Depends on the tube. A miniature vacuum tube doesn't need more power than
millions of transistors...... :-)

I very much doubt that. "Millions of transistors" is not a lot these days.
You're talking perhaps one watt. Can you get a miniature vacuum tube that
only requires one watt?

If you write a preprocessor than your power consumption is effectively
zero. You'll increase CPU utilization fractionally and thus CPU power
consumption, but the amount of work such preprocessing requires compared
to the power of a modern CPU is miniscule.

--
Mike Ash
Radio Free Earth
Broadcasting from our climate-controlled studios deep inside the Moon

Dear Michael Ash:

"Michael Ash" wrote in message
...
In rec.arts.sf.science Paul Schlyter wrote:
....
and power usage.

Depends on the tube. A miniature vacuum tube
doesn't need more power than millions of
transistors...... :-)

I very much doubt that. "Millions of transistors" is
not a lot these days. You're talking perhaps one
watt.

His comparison was not correct. We were (at one point) talking
about a power amplifier. "One watt" is less than comes out of PC
speakers, and all solid state does this handsomely. We are
usually talking tens to hundreds of amps, at frequencies of 40 to
20 kHz (audio), or up into the MHz for radio.

Can you get a miniature vacuum tube that
only requires one watt?

Yes. There are always the cold cathode tubes ... ;)

David A. Smith

In article ,
Michael Ash wrote:
In rec.arts.sf.science Paul Schlyter wrote:
In article ,
Michael Ash wrote:

For sound quality, I'm pretty sure that you could program a digital sound
card to perfectly mimick the tube's output

Not perfectly - but probably good enough for the human ear to be unable to
hear any difference. Looking at the waveform on an oscilloscope could
still show some differences though.

You choose how close you want to get, then you choose your components to
match it. A 24-bit DAC will get you to within one part in 16 million of
where you want to be, and they're quite common these days.

Some things:

1. Even if the bit resolution corresponds to one part in 16 million, do you
really believe the D/A converter is accurate to within 0.000006 % ????

2. Even if the D/A converter also has this accuracy, it's still not perfect!
There will be deviations up to one part in 16 million, and a perfect simulation
requires zero deviation. Right?

3. And when we're at it to this level of accuracy - exactly WHAT should we
emulate digitally? Individual tubes of the same tube type will surely differ
from one another more than 0.000006 % -- don't you think so? Of course, if
we aim at "good enough" rather than "perfect", the accuracy requirements will
be much lower than this.


for a fraction of the cost

Not likely - hiring a programmer is not cheap. Buying a vacuum tube could
be much cheaper...

Sure, if you want to build *one*. But generally you produce these things
for sale, at which point the cost of the programmer is divided across all
of the units you produce, whereas you pay for the tube for every one.

And of course for the digital solution there's no reason to be building
hardware at all. Use a commodity sound card and just write a little driver
that preprocesses the audio being output.

OK - unless I want to start my own software company which implements a
vacuum tube simulator intended to run on standard sound chips - which
software of this kind can I buy today?

and power usage.

Depends on the tube. A miniature vacuum tube doesn't need more power than
millions of transistors...... :-)

I very much doubt that. "Millions of transistors" is not a lot these days.
You're talking perhaps one watt. Can you get a miniature vacuum tube that
only requires one watt?

One watt is perhaps a bit hard, but I can get down to two watts. This little
cutie:

http://www.r-type.org/exhib/aac0073.htm

requires approx. one watt for the heater (it utilizes direct heating) and
another watt for the rest of the tube.

If you write a preprocessor than your power consumption is effectively
zero. You'll increase CPU utilization fractionally and thus CPU power
consumption, but the amount of work such preprocessing requires compared
to the power of a modern CPU is miniscule.

--
Mike Ash
Radio Free Earth
Broadcasting from our climate-controlled studios deep inside the Moon


--
----------------------------------------------------------------
Paul Schlyter, Grev Turegatan 40, SE-114 38 Stockholm, SWEDEN
e-mail: pausch at stjarnhimlen dot se
WWW: http://stjarnhimlen.se/

In article ,
N:dlzc D:aol T:com \(dlzc\) wrote:

"One watt" is less than comes out of PC speakers,

One watt is less that comes out of all but the very largest speakers !!!!

Spekares are horribly inefficient -- a typical hi-fi speaker will have an
efficiency of only about one percent. Which means if you pump some 100 Watts
of electricity into the speaker, only about one watt will get out of it
as sound - the remaining 99 watts are lost as heat. Otoh, one watt of
acoustic power is VERY LOUD unless the room is very large.

and all solid state does this handsomely.

Solid state outputs electricity, not sound.... ;-)




--
----------------------------------------------------------------
Paul Schlyter, Grev Turegatan 40, SE-114 38 Stockholm, SWEDEN
e-mail: pausch at stjarnhimlen dot se
WWW: http://stjarnhimlen.se/

Dear Paul Schlyter:

On Jun 25, 8:15*am, (Paul Schlyter) wrote:
In article ,
N:dlzcD:aol T:com \(dlzc\) wrote:

"One watt" is less than comes out of PC speakers,

One watt is less that comes out of all but the very
largest speakers !!!!

[Speakers] are horribly inefficient -- a typical hi-fi
speaker will have an efficiency of only about one
percent. *Which means if you pump some 100
Watts of electricity into the speaker, only about
one watt will get out of it as sound - the
remaining 99 watts are lost as heat.

This is untrue, in this context. What I believe (see blow) gets out
of the speaker is better than 70% conversion of electrical energy into
directed, usable wave motion. What is absorbed by any particular
human's ears is much less than 1% of what is used to drive the
speaker... and eventually, yes all that energy that is not absorbed as
"signal" or as work, becomes heat.

Now here is a paper that makes glancing reference to "Rossing, p.
413", that says cone speakers are 0.5 to 5% efficient...
http://web.media.mit.edu/~meyers/mcg...udspeakers.pdf
... page 4

I wonder if one could couple a speaker to an "inverse speaker" (IS)
(not a microphone), and see how much power can be generated at the IS
output leads. I know they do refrigeration acoustically...

*Otoh, one watt of acoustic power is VERY
LOUD unless the room is very large.

I am unconvinced, so I'll have to do more research...

and all solid state does this handsomely.

Solid state outputs electricity, not sound.... ;-)

Silly.

I recall electrolytic capacitors making quite a bit of noise with
electricity as input, and even at my age I can still hear flyback
transformers in TVs screaming... I used not to be able to walk into a
"Broadway" store, because the security system (I guess) gave me real
headaches.

David A. Smith