Author Topic: Japan quake  (Read 21258 times)

ccarlson

  • Guest
Re: Japan quake
« Reply #150 on: March 28, 2011, 01:06:15 PM »
Didn't say it was better, but that age doesn't necessarily mean it's inferior. Analog devices have some advantages. Digital devices have a tendency to lock up with power surges, but analog devices will normally recover quickly.

That said, digital devices offer far more features, so yes, they are definitely a better technology. I doubt either technology would have prevented what happened in Japan.
« Last Edit: March 28, 2011, 03:30:57 PM by ccarlson »

Offline bluesman

  • I must live here
  • **********
  • Posts: 8825
  • Delaware
Re: Japan quake
« Reply #151 on: March 28, 2011, 01:42:15 PM »
Didn't say it was better, but that age doesn't necessarily mean it's inferior. Analog devices have some advantages. Digital devices have a tendency to lock up with power surges, but analog devices will normally recover quickly.

That said, digital device offer far more features, so yes, they are definitely a better technology. I doubt either technology would have prevented what happened in Japan.

If anything we can chalk this up as a real eye opening experience and hopefully we can learn from it.
Ron Price

Offline phillamb168

  • Brewmaster General
  • *******
  • Posts: 2351
  • Lardy, France
    • My Job
Re: Japan quake
« Reply #152 on: April 07, 2011, 03:05:04 PM »
Bof. http://www.jma.go.jp/jp/quake/2/20110407233752391-072332.html Another quake, and another Tsunami.
I'm on twitter: phillamb168
----
morticaixavier for governing committee!

Offline bluefoxicy

  • Brewer
  • ****
  • Posts: 305
Re: Japan quake
« Reply #153 on: April 07, 2011, 06:49:33 PM »
Didn't say it was better, but that age doesn't necessarily mean it's inferior. Analog devices have some advantages. Digital devices have a tendency to lock up with power surges, but analog devices will normally recover quickly.

Analog devices tend to be superior in pretty much all settings except for cost-wise and inherently digital applications.  Computers MUST be digital, they COMPUTE s***.  Your car stereo?  You'd be far better off exchanging that DSP for a real volume and tone stack, complete with pots and capacitors and all; though a 20-bit DAC instead of a 1-bit DAC would improve sound quality dramatically.

The big thing about digital devices is they rely on absolute precision, on programmatically predictable cases, and on stable and consistent internal states.  Digital devices in mechanical systems attempt to emulate analog devices without analog loss:  they measure a temperature or voltage or pressure at a source, encode it, then transmit it; while the analog device produces a voltage anomaly which decays in transmission (longer line, electromagnetic interference, etc affects the precise signal).

Let's take a car for an example, and let's take something simple:  a viscous coupling all-wheel differential system (the general design of a torsen works the same way, but with different mechanical basis).  This involves two axial differentials (front and rear) and a center differential. There's a lot of ways to do this, and depending on topology both analog and digital systems can do axle-first (rear left slips, rear right gets the power) or can do opposite-wheel (i.e. rear left slips, front right gets the power) power transfer.

In an analog system, you have a viscous coupling fluid which resists deformation.  Under light stress it flows, and under heavy stress it solidifies--think silly putty, which flows when under weak force (even under gravity pressure) yet bounces like a set elastic (or in some formulations, shatters like a hard ceramic) under sudden impulse.

When all four wheels are on the ground, the front and rear differentials each have internal connectors churning this fluid at the same rate, so no energy goes into deformation, and no solidification happens; similarly, the front and rear wheels turn at the same rate, and the center differential doesn't experience anything either.  Simple so far, right?

If the rear right wheel slips, though, the increase in power there puts more load on the visco-elastic fluid in the differential, which causes it to react by partially solidifying.  Because it's now somewhat solid, it cranks against the other wheel's drive shaft, transferring the power from the rear right wheel to the rear left.

So far, we're getting 100% power to the rear wheels, and 0% to the front.  What if both rear wheels slip?  Well then the power going to the rear spins them, and as they start to slip, power goes to the center differential (as its input shaft--from the rear input shaft--is no longer sending power directly to ground).  The rear wheels are limited from spin by the resistance encountered by the center differential, which starts to solidify from the force applied, turning the front wheels--power goes up front, and your car is now partially front wheel drive.  Only enough power will reach the rear wheels to drive them without slip; if they slip at all, then their power is not being transferred to the ground (no load on the system), and instead goes into the center diff, which transfers the load to the front wheels, which of course takes the power that is spinning the rear wheels into slipping off the rear wheels.

This system works.  It is 100% reliable as long as nothing outright breaks.  Even when it is unmaintained and starting to fail, it does what it's supposed to do.  When a sudden shock occurs (complete loss of traction to the rear wheels, etc), it quickly reacts because the laws of physics demand a certain mode of operation.

It is also not exactly 100% efficient.  Torsen is more efficient (and somewhat better, as it always transfers power:  25% of the power goes to each wheel by default, rather than being 100% RWD until the rear slips, meaning you can shove about twice as much power against the ground BEFORE any loss from slipping wheels occur).

Even more efficient:  slap slip sensors onto each wheel.  When a wheel is detected to be moving faster than another wheel while it's moving, it's slipping (and when it's detected to be moving slower when brakes are applied, it's slipping and your anti-lock brakes kick in and release the brake on that wheel).  Then the computer manipulates a differential to adjust how much power goes this way or that.

Obviously, the computer is trying to remove the "deformation of fluid" or "heat lost in gear system" part of the equation and instead use a less lossy gear system on a computer control to provide a more efficient differential.  Rather than laws of physics, just a bunch of calculations go into deciding how to shift the power and to what wheel.  This takes, in the end, actually less energy than using a viscous or even a Torsen system; and in fact, a profile can be produced that starts off with a more efficient layout (Torsen starts 25% to all wheels and redirects it when a wheel slips; but putting about 95% to the rear wheels is more efficient at launch) and even selects different layouts depending on vehicle dynamic rather than on "which wheel is slipping right now?"

The downside is, of course, if the digital electronics experience any minor flaw--decaying chips, decaying solder joints, localized static build-up, electromagnetic interference, anything--that changes a bit inside a chip, weird s*** happens.  If the computer detects input that doesn't make sense within its algorithms, weird s*** happens.  The system is not durable.  It can respond faster based on known data rather than natural physical consequences of physical events; it can make predictions and pre-adjustments and such; but it can't handle something going vaguely out of spec.

Analog devices are quite functionally superior.  They also quite often drain a decent amount of extra power out of the system, and use more costly or bulky components.  Don't take that to mean that analog is the answer to everything, though:  some things can [em]only[/em] be done digitally.  Analog computers include the Japanese Soroban--lies, it's digital!  Beads are set and cleared, and if they're dangling in between then you have a nonsense state (unless you assign a separate meaning to a "hanging bead" or whatnot).  Hand-written word is digital:  the letter "a" is an "a," and the vague variations in handwriting style and shape are meaningless (this is OTHER information, which is analog).

Plenty of systems to control nuclear reactors and the like are digital because they are computers, they compute things, and they are not a digital analog to an analog system that does something else.  It's not a simulation of an oil-filled chamber that's heated by the flow of pumped coolant and somehow measures whatever by expanding and flowing up a graduated tube or whatnot; servos are controlled by systems where we say, "Put this here," and we are sure as hell not going into the nuclear reactor core and pulling a lever to do it.  Lots of the automation is logic, not consequence (and thus, on a computer, logic implementing consequence).

If you really want to, maybe you can design systems for that; but are they academic "analog computers" or are they actually systems that are self-stabilizing?  Pebble bed reactors are self-stabilizing:  when they get hot, they become self-limiting, squelching the rate at which fission can even occur.  You take away all coolant and all safety systems and dump the raw fuel into a big, uncontrolled, exposed pile and it will get hot and stay that way; it will not experience runaway fission past its design specifications, and there aren't even any moving parts to control this.

Can you create a system around a BWR that does the exact same thing?  No computers, and when something is quite wrong it's okay because the entire design reacts to failure by shutting off elsewhere not because of fuses and triggers, but because some fluid was expanded or some metal flexed or whatever simply because everything was in proper operation, and when that went out of spec those things moved into a different, safe state by design as dictated by the laws of physics?  It would be a system where the computer controls activate something, but that something relies on pushing i.e. a fluid around that's only in place because the reactor is i.e. getting proper coolant flow, and when that flow fails something deforms from the heat (temporarily, like a metal that flexes and shifts hydraulic fluid around) and drops the coolant rods into the fuel even though the computer says to raise them and go full-on hot?  "Analog safety system" (with a manual override, an actual lever that requires a human to crank it over to supply the torque to undo what it's done)

Offline maxieboy

  • Senior Brewmaster
  • ******
  • Posts: 1155
  • Mid MI
Re: Japan quake
« Reply #154 on: April 07, 2011, 08:19:00 PM »
Didn't say it was better, but that age doesn't necessarily mean it's inferior. Analog devices have some advantages. Digital devices have a tendency to lock up with power surges, but analog devices will normally recover quickly.

Analog devices tend to be superior in pretty much all settings except for cost-wise and inherently digital applications.  Computers MUST be digital, they COMPUTE s***.  Your car stereo?  You'd be far better off exchanging that DSP for a real volume and tone stack, complete with pots and capacitors and all; though a 20-bit DAC instead of a 1-bit DAC would improve sound quality dramatically.

The big thing about digital devices is they rely on absolute precision, on programmatically predictable cases, and on stable and consistent internal states.  Digital devices in mechanical systems attempt to emulate analog devices without analog loss:  they measure a temperature or voltage or pressure at a source, encode it, then transmit it; while the analog device produces a voltage anomaly which decays in transmission (longer line, electromagnetic interference, etc affects the precise signal).

Let's take a car for an example, and let's take something simple:  a viscous coupling all-wheel differential system (the general design of a torsen works the same way, but with different mechanical basis).  This involves two axial differentials (front and rear) and a center differential. There's a lot of ways to do this, and depending on topology both analog and digital systems can do axle-first (rear left slips, rear right gets the power) or can do opposite-wheel (i.e. rear left slips, front right gets the power) power transfer.

In an analog system, you have a viscous coupling fluid which resists deformation.  Under light stress it flows, and under heavy stress it solidifies--think silly putty, which flows when under weak force (even under gravity pressure) yet bounces like a set elastic (or in some formulations, shatters like a hard ceramic) under sudden impulse.

When all four wheels are on the ground, the front and rear differentials each have internal connectors churning this fluid at the same rate, so no energy goes into deformation, and no solidification happens; similarly, the front and rear wheels turn at the same rate, and the center differential doesn't experience anything either.  Simple so far, right?

If the rear right wheel slips, though, the increase in power there puts more load on the visco-elastic fluid in the differential, which causes it to react by partially solidifying.  Because it's now somewhat solid, it cranks against the other wheel's drive shaft, transferring the power from the rear right wheel to the rear left.

So far, we're getting 100% power to the rear wheels, and 0% to the front.  What if both rear wheels slip?  Well then the power going to the rear spins them, and as they start to slip, power goes to the center differential (as its input shaft--from the rear input shaft--is no longer sending power directly to ground).  The rear wheels are limited from spin by the resistance encountered by the center differential, which starts to solidify from the force applied, turning the front wheels--power goes up front, and your car is now partially front wheel drive.  Only enough power will reach the rear wheels to drive them without slip; if they slip at all, then their power is not being transferred to the ground (no load on the system), and instead goes into the center diff, which transfers the load to the front wheels, which of course takes the power that is spinning the rear wheels into slipping off the rear wheels.

This system works.  It is 100% reliable as long as nothing outright breaks.  Even when it is unmaintained and starting to fail, it does what it's supposed to do.  When a sudden shock occurs (complete loss of traction to the rear wheels, etc), it quickly reacts because the laws of physics demand a certain mode of operation.

It is also not exactly 100% efficient.  Torsen is more efficient (and somewhat better, as it always transfers power:  25% of the power goes to each wheel by default, rather than being 100% RWD until the rear slips, meaning you can shove about twice as much power against the ground BEFORE any loss from slipping wheels occur).

Even more efficient:  slap slip sensors onto each wheel.  When a wheel is detected to be moving faster than another wheel while it's moving, it's slipping (and when it's detected to be moving slower when brakes are applied, it's slipping and your anti-lock brakes kick in and release the brake on that wheel).  Then the computer manipulates a differential to adjust how much power goes this way or that.

Obviously, the computer is trying to remove the "deformation of fluid" or "heat lost in gear system" part of the equation and instead use a less lossy gear system on a computer control to provide a more efficient differential.  Rather than laws of physics, just a bunch of calculations go into deciding how to shift the power and to what wheel.  This takes, in the end, actually less energy than using a viscous or even a Torsen system; and in fact, a profile can be produced that starts off with a more efficient layout (Torsen starts 25% to all wheels and redirects it when a wheel slips; but putting about 95% to the rear wheels is more efficient at launch) and even selects different layouts depending on vehicle dynamic rather than on "which wheel is slipping right now?"

The downside is, of course, if the digital electronics experience any minor flaw--decaying chips, decaying solder joints, localized static build-up, electromagnetic interference, anything--that changes a bit inside a chip, weird s*** happens.  If the computer detects input that doesn't make sense within its algorithms, weird s*** happens.  The system is not durable.  It can respond faster based on known data rather than natural physical consequences of physical events; it can make predictions and pre-adjustments and such; but it can't handle something going vaguely out of spec.

Analog devices are quite functionally superior.  They also quite often drain a decent amount of extra power out of the system, and use more costly or bulky components.  Don't take that to mean that analog is the answer to everything, though:  some things can [em]only[/em] be done digitally.  Analog computers include the Japanese Soroban--lies, it's digital!  Beads are set and cleared, and if they're dangling in between then you have a nonsense state (unless you assign a separate meaning to a "hanging bead" or whatnot).  Hand-written word is digital:  the letter "a" is an "a," and the vague variations in handwriting style and shape are meaningless (this is OTHER information, which is analog).

Plenty of systems to control nuclear reactors and the like are digital because they are computers, they compute things, and they are not a digital analog to an analog system that does something else.  It's not a simulation of an oil-filled chamber that's heated by the flow of pumped coolant and somehow measures whatever by expanding and flowing up a graduated tube or whatnot; servos are controlled by systems where we say, "Put this here," and we are sure as hell not going into the nuclear reactor core and pulling a lever to do it.  Lots of the automation is logic, not consequence (and thus, on a computer, logic implementing consequence).

If you really want to, maybe you can design systems for that; but are they academic "analog computers" or are they actually systems that are self-stabilizing?  Pebble bed reactors are self-stabilizing:  when they get hot, they become self-limiting, squelching the rate at which fission can even occur.  You take away all coolant and all safety systems and dump the raw fuel into a big, uncontrolled, exposed pile and it will get hot and stay that way; it will not experience runaway fission past its design specifications, and there aren't even any moving parts to control this.

Can you create a system around a BWR that does the exact same thing?  No computers, and when something is quite wrong it's okay because the entire design reacts to failure by shutting off elsewhere not because of fuses and triggers, but because some fluid was expanded or some metal flexed or whatever simply because everything was in proper operation, and when that went out of spec those things moved into a different, safe state by design as dictated by the laws of physics?  It would be a system where the computer controls activate something, but that something relies on pushing i.e. a fluid around that's only in place because the reactor is i.e. getting proper coolant flow, and when that flow fails something deforms from the heat (temporarily, like a metal that flexes and shifts hydraulic fluid around) and drops the coolant rods into the fuel even though the computer says to raise them and go full-on hot?  "Analog safety system" (with a manual override, an actual lever that requires a human to crank it over to supply the torque to undo what it's done)

Pretty much what I was thinking...
A dog can show you more honest affection with a flick of his tail than a man can gather through a lifetime of handshakes." Gene Hill

[47.7, 310.8] AR

AHA Member

Offline Bret

  • Assistant Brewer
  • ***
  • Posts: 210
Re: Japan quake
« Reply #155 on: April 07, 2011, 08:27:16 PM »
 ;D
Member #: 900000767

Offline gordonstrong

  • Senior Brewmaster
  • ******
  • Posts: 1355
    • BJCP
Re: Japan quake
« Reply #156 on: April 07, 2011, 08:31:04 PM »
"Brevity is the soul of wit." -- Shakespeare

(And here I thought it was coriander)
Gordon Strong • Beavercreek, Ohio • AHA Member since 1997 • Twitter: GordonStrong

Offline maxieboy

  • Senior Brewmaster
  • ******
  • Posts: 1155
  • Mid MI
Re: Japan quake
« Reply #157 on: April 07, 2011, 08:42:45 PM »
"Brevity is the soul of wit." -- Shakespeare

(And here I thought it was coriander)

 ;D
A dog can show you more honest affection with a flick of his tail than a man can gather through a lifetime of handshakes." Gene Hill

[47.7, 310.8] AR

AHA Member

Offline hopfenundmalz

  • Global Moderator
  • I must live here
  • *****
  • Posts: 9819
  • Milford, MI
Re: Japan quake
« Reply #158 on: April 08, 2011, 01:22:16 AM »
Way back when, I learned to solve DEQ's on Philbrick Analog computers.  Those were all analog.  Really, they were.

Not so long ago, a very good engineer was talking about doing somethign in Matlab/Simulink.  I said here is how you set up the equations, that is easy.  He looked at me and said where did you learn this?  I said, "Well, this was how we learned it back in the day.  The equations have not changed much." He had worked with me often, said OK, and went with it.  Then others were asking him how he got the answer so quick.   ;)
Jeff Rankert
AHA Lifetime Member
BJCP National
Ann Arbor Brewers Guild
Home-brewing, not just a hobby, it is a lifestyle!

Offline bluesman

  • I must live here
  • **********
  • Posts: 8825
  • Delaware
Re: Japan quake
« Reply #159 on: April 08, 2011, 01:37:15 AM »
Didn't say it was better, but that age doesn't necessarily mean it's inferior. Analog devices have some advantages. Digital devices have a tendency to lock up with power surges, but analog devices will normally recover quickly.

Analog devices tend to be superior in pretty much all settings except for cost-wise and inherently digital applications.  Computers MUST be digital, they COMPUTE s***.  Your car stereo?  You'd be far better off exchanging that DSP for a real volume and tone stack, complete with pots and capacitors and all; though a 20-bit DAC instead of a 1-bit DAC would improve sound quality dramatically.

The big thing about digital devices is they rely on absolute precision, on programmatically predictable cases, and on stable and consistent internal states.  Digital devices in mechanical systems attempt to emulate analog devices without analog loss:  they measure a temperature or voltage or pressure at a source, encode it, then transmit it; while the analog device produces a voltage anomaly which decays in transmission (longer line, electromagnetic interference, etc affects the precise signal).

Let's take a car for an example, and let's take something simple:  a viscous coupling all-wheel differential system (the general design of a torsen works the same way, but with different mechanical basis).  This involves two axial differentials (front and rear) and a center differential. There's a lot of ways to do this, and depending on topology both analog and digital systems can do axle-first (rear left slips, rear right gets the power) or can do opposite-wheel (i.e. rear left slips, front right gets the power) power transfer.

In an analog system, you have a viscous coupling fluid which resists deformation.  Under light stress it flows, and under heavy stress it solidifies--think silly putty, which flows when under weak force (even under gravity pressure) yet bounces like a set elastic (or in some formulations, shatters like a hard ceramic) under sudden impulse.

When all four wheels are on the ground, the front and rear differentials each have internal connectors churning this fluid at the same rate, so no energy goes into deformation, and no solidification happens; similarly, the front and rear wheels turn at the same rate, and the center differential doesn't experience anything either.  Simple so far, right?

If the rear right wheel slips, though, the increase in power there puts more load on the visco-elastic fluid in the differential, which causes it to react by partially solidifying.  Because it's now somewhat solid, it cranks against the other wheel's drive shaft, transferring the power from the rear right wheel to the rear left.

So far, we're getting 100% power to the rear wheels, and 0% to the front.  What if both rear wheels slip?  Well then the power going to the rear spins them, and as they start to slip, power goes to the center differential (as its input shaft--from the rear input shaft--is no longer sending power directly to ground).  The rear wheels are limited from spin by the resistance encountered by the center differential, which starts to solidify from the force applied, turning the front wheels--power goes up front, and your car is now partially front wheel drive.  Only enough power will reach the rear wheels to drive them without slip; if they slip at all, then their power is not being transferred to the ground (no load on the system), and instead goes into the center diff, which transfers the load to the front wheels, which of course takes the power that is spinning the rear wheels into slipping off the rear wheels.

This system works.  It is 100% reliable as long as nothing outright breaks.  Even when it is unmaintained and starting to fail, it does what it's supposed to do.  When a sudden shock occurs (complete loss of traction to the rear wheels, etc), it quickly reacts because the laws of physics demand a certain mode of operation.

It is also not exactly 100% efficient.  Torsen is more efficient (and somewhat better, as it always transfers power:  25% of the power goes to each wheel by default, rather than being 100% RWD until the rear slips, meaning you can shove about twice as much power against the ground BEFORE any loss from slipping wheels occur).

Even more efficient:  slap slip sensors onto each wheel.  When a wheel is detected to be moving faster than another wheel while it's moving, it's slipping (and when it's detected to be moving slower when brakes are applied, it's slipping and your anti-lock brakes kick in and release the brake on that wheel).  Then the computer manipulates a differential to adjust how much power goes this way or that.

Obviously, the computer is trying to remove the "deformation of fluid" or "heat lost in gear system" part of the equation and instead use a less lossy gear system on a computer control to provide a more efficient differential.  Rather than laws of physics, just a bunch of calculations go into deciding how to shift the power and to what wheel.  This takes, in the end, actually less energy than using a viscous or even a Torsen system; and in fact, a profile can be produced that starts off with a more efficient layout (Torsen starts 25% to all wheels and redirects it when a wheel slips; but putting about 95% to the rear wheels is more efficient at launch) and even selects different layouts depending on vehicle dynamic rather than on "which wheel is slipping right now?"

The downside is, of course, if the digital electronics experience any minor flaw--decaying chips, decaying solder joints, localized static build-up, electromagnetic interference, anything--that changes a bit inside a chip, weird s*** happens.  If the computer detects input that doesn't make sense within its algorithms, weird s*** happens.  The system is not durable.  It can respond faster based on known data rather than natural physical consequences of physical events; it can make predictions and pre-adjustments and such; but it can't handle something going vaguely out of spec.

Analog devices are quite functionally superior.  They also quite often drain a decent amount of extra power out of the system, and use more costly or bulky components.  Don't take that to mean that analog is the answer to everything, though:  some things can [em]only[/em] be done digitally.  Analog computers include the Japanese Soroban--lies, it's digital!  Beads are set and cleared, and if they're dangling in between then you have a nonsense state (unless you assign a separate meaning to a "hanging bead" or whatnot).  Hand-written word is digital:  the letter "a" is an "a," and the vague variations in handwriting style and shape are meaningless (this is OTHER information, which is analog).

Plenty of systems to control nuclear reactors and the like are digital because they are computers, they compute things, and they are not a digital analog to an analog system that does something else.  It's not a simulation of an oil-filled chamber that's heated by the flow of pumped coolant and somehow measures whatever by expanding and flowing up a graduated tube or whatnot; servos are controlled by systems where we say, "Put this here," and we are sure as hell not going into the nuclear reactor core and pulling a lever to do it.  Lots of the automation is logic, not consequence (and thus, on a computer, logic implementing consequence).

If you really want to, maybe you can design systems for that; but are they academic "analog computers" or are they actually systems that are self-stabilizing?  Pebble bed reactors are self-stabilizing:  when they get hot, they become self-limiting, squelching the rate at which fission can even occur.  You take away all coolant and all safety systems and dump the raw fuel into a big, uncontrolled, exposed pile and it will get hot and stay that way; it will not experience runaway fission past its design specifications, and there aren't even any moving parts to control this.

Can you create a system around a BWR that does the exact same thing?  No computers, and when something is quite wrong it's okay because the entire design reacts to failure by shutting off elsewhere not because of fuses and triggers, but because some fluid was expanded or some metal flexed or whatever simply because everything was in proper operation, and when that went out of spec those things moved into a different, safe state by design as dictated by the laws of physics?  It would be a system where the computer controls activate something, but that something relies on pushing i.e. a fluid around that's only in place because the reactor is i.e. getting proper coolant flow, and when that flow fails something deforms from the heat (temporarily, like a metal that flexes and shifts hydraulic fluid around) and drops the coolant rods into the fuel even though the computer says to raise them and go full-on hot?  "Analog safety system" (with a manual override, an actual lever that requires a human to crank it over to supply the torque to undo what it's done)

There's only one thing I can say about all of this....

you get the AHA Forum award for the longest post ever on this forum.  ;)

but we still love you.  ;D
Ron Price

beveragebob

  • Guest
Re: Japan quake
« Reply #160 on: April 08, 2011, 04:08:25 AM »
I was wondering when the movie is coming out.  ;)

Offline punatic

  • Official Poobah of No Life.
  • *
  • Posts: 4583
  • Puna District, Hawaii Island (UTC -10)
Re: Japan quake
« Reply #161 on: April 08, 2011, 05:49:54 AM »
All I know is, at the power plant where I worked, the ease of operation and operational awareness went way up and forced outage rates went down with each control upgrade.  This was most pronounced when we retrofitted from analog - pneumatic controls to digital - electronic controls.  It was very handy and efficient to have the ability to bring up any system control screen at any control console throughout the power plant site.

Now I direct the operations of several well fields, RO water treatment plants, and distribution systems.  I can access the controls to these systems from anywhere I have broadband access for a laptop via SCADA.  Remote operation certainly has its hazards, but SCADA allows me to monitor operations and help my operators operate when they are onsite.  They are there to see if the equipment is actually doing what the controls indicate it is doing. This was not be possible with analog controls and the tone-telemetry systems that we upgraded from.   Unfortunately It also sometimes makes for working vacations.

Theory is fine, but operation-wise I'll take digital controls over analog any time.
There is only one success: to be able to spend your life in your own way.


AHA Life Member #33907

Offline euge

  • I must live here
  • **********
  • Posts: 8022
  • Ego ceruisam ad bibere cervisiam
Re: Japan quake
« Reply #162 on: April 08, 2011, 06:00:22 AM »
Wasn't there another quake and tsunami? Not a peep from the news I follow.
The first principle is that you must not fool yourself, and you are the easiest person to fool. -Richard P. Feynman

Laws are spider-webs, which catch the little flies, but cannot hold the big ones. -Anacharsis

Offline tschmidlin

  • I must live here
  • **********
  • Posts: 8198
  • Redmond, WA
Re: Japan quake
« Reply #163 on: April 08, 2011, 06:04:27 AM »
Wasn't there another quake and tsunami? Not a peep from the news I follow.
A 7.5, no real tsunami.  I think there was a warning briefly and some areas were evacuated, but nothing materialized.
Tom Schmidlin

Offline Hokerer

  • I spend way too much time on the AHA forum
  • ********
  • Posts: 2654
  • Manassas, VA
Re: Japan quake
« Reply #164 on: April 08, 2011, 04:30:46 PM »
Wasn't there another quake and tsunami? Not a peep from the news I follow.
A 7.5, no real tsunami.  I think there was a warning briefly and some areas were evacuated, but nothing materialized.

They're blaming three deaths on it so something must have materialized...

http://www.cnn.com/2011/WORLD/asiapcf/04/08/japan.quake/index.html?hpt=T2
Joe