Remember: I-squared-R losses go up with the SQUARE of the current. So running 1.2V across a board to your chips loses 400 TIMES as much power as running 48V to the regulator next to them.
Oops. Make that "loses 1600 times as much power". (Multiplied the 10s but forgot to multiply the 4s.)
When a board has several chips running at 10 or more watts apiece you can easily be dealing at currents where the heating of the board consumes more power than the heating of the chips. With a rack of electronics dissipating several KW you can pay for a LOT of tiny switching regulators to avoid more than doubling that.
Give me a call when those fuel cells are ready for deployment, then we can talk about all these wonderful uses. No talk about the carbon footprint of operating fuel cells?
Natural gas (methane) has the highest energy/carbon emission ration of any saturated hydrocarbon (gas or oil) and beats the HELL out of coal. If you're going to use fossil fuels (or renewable fuel from biodegrading vegetable waste, sewage, or cow flops), and a fuel cell is in the efficiency ballpark with a grid plant, why not put the fuel cell in the rack?
The article mixes the use of fuel cells as a power source with efficiency improvements. The only place that makes sense is with the minor savings that may be seen by eliminating DC converters, but you will still need DC regulators which will have some losses.
Fuel cells are not limited by the carnot cycle efficiency limt. They can be FAR MORE EFFICIENT than a heat-engine based power plant.
Modern circuit boards in servers ALREADY HAVE switching regulators near the chips. With the very low operating voltages of modern electronics, the supply currents are SO high that you lose less energy by running 48V on the power planes of the PC board and regulating it down at the load than you do running, say 3.3, 2.5, or 1.8 volts and a correspondingly higher current across several inches of thin copper.
Remember: I-squared-R losses go up with the SQUARE of the current. So running 1.2V across a board to your chips loses 400 TIMES as much power as running 48V to the regulator next to them. It's like high-tension transmission lines in miniature. (They'd go higher except that over 50V gets you out of the easy part of the electrical code and into the region where electrocution becomes a major issue.)
Putting the switching regulator next to the chip also gives you much more stable voltage. When it's already there to save power, this makes good regulation "cheaper than free".
A major oversight of this article is the fact that fuel cells are major heat generators, not something you want in a data center. They would need to be installed in a separated structure, therefore idea that "Rack-level fuel cells would do away with data-centre-wide electricity distribution for servers" is hard to imagine.
They also need and external air supply and to have their exhaust removed rather than dumped into the room air. So you're going to give them their own plumbing. You want to SAVE that heat to keep the oxygen-transport style cells at operating temperature without wasting fuel or power to do it. So you insulate the box and run the ventilation plumbing like a stove pipe - coaxial, with the hot exhaust in the middle and the cool incoming air on the outside. This minimizes the heat loss to the room and acts like a counter-current heat exchanger to preheat the fresh air with the heat from the exhaust, while cooling the exhaust.
It's similar to what I did with my first unix box, back in the '80s or so: The thing put out as much heat as a space heater. So I hooked up a dryer vent hose to the 4" exhaust fan and dumped the hot air outside. Cooling problem solved. (In the winter I dumped it INSIDE to save on heating bills.)
Right now, if I don't have electric and gas, my water heater and furnace become inoperable.
In my remote house my heat and water heater both work fine on only propane. As long as the tank is not empty we're fine (and the tank only needs filling about three times a year).
The water heater is gas with a pilot light and no electric controls (except for the pilot light safety thermocouple, which generates enough power from the flame's heat to control the safety shutdown).
While the regular furnace has electronic controls and blower, I also have a backup: A propane "fireplace" stove in the great room, with a room layout that lets it heat the living area and keep the pipes from freezing.
Again it works with a pilot light, and the thermocouple's few millivolts also provide enough power, controlled by a mechanical-switch thermostat in the middle of the house, to operate the main gas valve as well. Though the stove's room blower will also fail in an outage, convection is more than adequate to circulate the heat in the big-open-space-in-the-middle house design. Kept things nice and comfy when we had a day-long outage in deep winter.
When away the stats are set to 55 for the furnace, 50 for the backup stove. This worked just fine one winter, when the furnace's draft sensor failed and the furnace was dead for weeks until we got there and discovered the issue. That definitely paid for the stove in one event.
If you have a fuel cell that burns methane (i.e. Natural gas) or other fuels the fuel cell will have to reform it into Hydrogen (releasing CO2) before it's used.
And if you're ultimately running from fossil fuels, methane is the least carbon-emitting choice.
Burning the hydrogen atoms to water produces MOST of the power from fossil fuels. Burning the carbon to CO2 produces a little more. But in gas and oil it's mostly there to make the hydrogen easier to handle than H2.
Methane has four hydrogens per carbon (4:1), the best ratio of all hydrocarbons. Ethane: (6: 2 = 3:1), propane: 8:3 = 2.666..."1, and so on. As you transition from gasses to oil you're approaching the large saturated hydrocarbon molecule limit of 2:1.
Then there's coal, where you're JUST burning the carbon. All CO2, much less energy (though still plenty if you burn ENOUGH of it).
Tell me when they come up with a range of affordable, small, light weight, fuel cells that efficiently make a couple hundred to a couple thousand watts by burning odorized propane with ambient air. I want one for my car, one for my travel trailer, and one for each house.
Electrolysis is not cost effective and requires more electrical power than your fuel cell could produce.
By the same argument, power grids "are not cost effective" because they require more electrical power input than they deliver to the cu
Electrolysis may be VERY cost effective. Just think of it as a different sort of energy distribution system, not as a conversion of fuel to electricity, and compare its costs to what it replaces.
(This is similar to the bogus argument against solar panels: "They take more energy to make than they deliver over their lifetime." First that's false - they reach energy breakeven very early in their life. Second, they produce post-carnot-cycle electricity, while most of the energy going into making them is pre-carnot-cycle heat. Third, you need to compare their energy consumption apples-to-apples: How much energy does it take to build and fuel the fraction of the power grid that would deliver the same amount of electricity to the same site - from melting steel to build transformers, to clearing trees and stringing poles, to building power plants and switchgear, to fueling the plants to make the electricity (digging for and transporting coal, drilling for and pumping oil, disposing of radioactive waste,
Lest we forget that all firearms legislation has its roots in oppressing minorities.
Not always minorities. But it's always about government seeking to outlaw resistance.
The NRA, as it happens, was founded by a group of Union officers to fight against southern local governments stealing guns from freedmen shortly after the civil war.
Those so-called "Death Panels" already exist. Why is it alright for Insurance Companies to pick and choose who gets to live, and who must die, while bankrupting even more people in the process? I'd feel much safer if "profit" wasnt a primary variable in whether me or my loved ones get to live or die.
A rhetorical question deserves a rhetorical answer:
Because, with private insurance companies, you can either change policies or insurance companies, to get policy terms (and "death panels") more to your liking. Or you can dump the insurance company altogether and pay the hospitals as many buck as they ask to get as much health care as you both need and can afford.
With Obama care:
- You're forced to buy insurance.
- You're forced to buy a plan the government approved.
- You're forced to buy a plan with some very expensive coverage that you may not need or want.
- And the government picks your "death panel". Don't like that one? Find another country.
On the other hand, the statistics about SAP rollouts would tend to indicate a very high degree of risk inherent in attempting to use that system.
The "other" hand? You're going to take something that's inherently complex and risky even when done professionally by a company with hundreds of developers... and then roll your own? And that's less of a risk in your world?
Risk vs. reward. What have you gained from not wearing seatbelts other than perhaps a few less wrinkles on your clothes
I have learned that most people are utter shit at estimating risk. Especially people who think they're smart and are good at it, but don't actually do the math. We spend trillions to prevent terrorism, but next to nothing to prevent drunk driving. It's because people think that risks they have control over are far less than those they don't, so drunk driving is "Well, I'll be driving, and I'm a good driver, so the risk must be low", and terrorism is "I'll be strapped into the plane and not in control... so it must be much, much worse."
The same kind of thinking applies to rolling your own software, instead of buying it. People are not objective about risk. They flat out suck at it. As for me... what I've learned is to wear my goddamned seat belt, because I read the statistics and know that there's about a 1 in 5 risk of getting into a car accident every year, and the seat belt means a 90% reduction in probable injury -- Without it, I'm just hamburger through the window.
Which is like most companies when they decide to cook their own complex software... they usually wind up paying more, but because they never analyze their own decisions, they, like you, think it's actually less.
You don't get ahead in SPEKTOR by saying "no" to Hank Scorpio. You get fed to the sharks.
You're referencing a character who first appeared on the Simpsons in the 90s... before SAP software as a class even existed. I believe this reference is so damn obscure that only some hipster would recognize it without googling for it. And as for SPEKTOR... Google has nothing. So who knows...
Atleast with homebrew you have a change to ever reach spec and you don't have to spend the same budget again every next year.
In other news, some people believe patching and bug hunting is free and that software never needs modification once installed. There will never be a support cost of any kind.
I disagree. [...] Done right, it works very, very well.
Yes, the same can be said for any risk-taking behavior. "I haven't worn my seat belt for years and nothing bad has ever happened to me."
Many, if not most, IT initiatives with homebrew tech fails. It's nice when it pays off, but almost always it is over budget and under spec. If the CEO got lucky, good for him, but his CIO shouldn't be sitting in the big chair if he didn't at least warn him it could all go horribly wrong.
And how are they not going to do the same for natural gas, or any other form of energy? The one you describe is a regulatory problem, not a technical one.
It may surprise you to learn, but many advancements in technology are due to "regulatory" problems. And it isn't a regulatory problem; energy service providers have a natural monopoly. Before you comment about how that's a regulatory problem, please google what a natural monopoly is. The fact is, you don't want a half-dozen power grids all in the same area competing; AC power doesn't take well to getting out of phase across the grid. And by not taking well, I mean, shit explodes. Violently. You can have many power plants, but really only one grid. It is non-trivial to connect two grids together... it requires a lot of high-power equipment to correct for phase variance, proper isolation, etc. In fact, there are only about 6 major power grids in the United States, and about 10 on the continent. That speaks volumes to the price of having multiple grids.
Where are the calculations that go with a calculated risk?