I'm waiting for F1 to reintroduce gas turbines

I think a gas turbine electric drive would be great for F1. I know it was tried before, but I think those cars were just ahead of their time, we have the technology now to exploit their potential.

As one might know, gas turbines are not very efficient. That's one disadvantage. And it's exacerbated when the engine is running off peak load as the massive compressor takes a lot of energy to run even on idle.

Another is the response lag when throttling a turbine engine.

There is a big advantage to gas turbines though: they have incredible power to weight ratio.

So optimally, you want to be running the turbine at its peak load all the time. An electric drive allows you to store energy when you don't need it, and delivers massive torque. It's a great combination.

Of course, F1 wouldn't be so eager to change their rules. But it's cool to think about.

Everything leaks

It's just a fact of life that if you have kilometers of piping and hundreds of gaskets, something is going to leak. We expect things to leak.

Refrigerant loops, just like your refrigerator at home, are supposed to be closed loops. It just circulates in a compression cycle forever. If you look at the refrigerant circuit at a natural gas liquefaction facility (huge pipelines, some of the largest in an LNG train), there's basically no reason to drain anything in operation. But you can expect refrigerant make-up to be supplied every week on the order of tens of tons. Obviously small in comparison to the system inventory, but quite a lot considering it's all being lost through leakage.

Steam circuits (used for heating) are also closed loops. The condensed steam that goes through any heat exchanger against hydrocarbons are sent through an elaborate system for cleaning before being boiled again. The default assumption is that there are constantly leaking exchangers, otherwise there'd be no need to spend millions of dollars on a continuous purification system when you could just install a hydrocarbon detector that trips the plant.

Cooling water circuits don't need this kind of cleaning. The water drains out pitch black after months of circulation.

So I guess what I'm saying is, if any PR promises that some petro pipeline stretching across a continent isn't going to leak, don't believe it. It will leak, the question is just how much.

Something gonna blow up..

I found out recently that common practice with onshore facilities is that if an equipment is provided with a pressure safety valve, then a high pressure interlock¹ is not needed. This sounds like such a bad idea.

On it's face it sounds okay.

Except experienced process designers don't do this work, these kind of tasks get passed down the chain.

So I sure hope my classmates did a good job with their PSV sizing calculations back in their second year work terms...


1) A high pressure interlock would be a controller that acts automatically to isolate the vessel from the source of high pressure, e.g. closing inlet shutdown valves

Kind of safety critical

I think there was a study by DNV that was like, 25% of all PSVs are improperly sized.

How could this be?

Oh wait, the PSV calculations were probably all done by interns!

Well no shit then.

A forthright exchange of ideas

"..so we knew we had an operating constraint of 400kW to work with"

"Wait, isn't 400kW really small for a cogeneration unit? How big is the main plant? Like 2MW?"

"I don't understand what you mean"

"How big is the main power plant?"

"We're not working with a power plant, we're working with a food production plant"

"Ohhh! That makes total sense then; I just assumed.."

"Nitta Gelatin! They make gelatin"

"I always thought that was a strange name for a utilities company"

"Dammit, I knew they should've let me do the introduction"

Can't these liberal art guys consult an engineer before publishing this stuff?

From the Economist:

"Morris Technologies has invested heavily in 3D printing equipment and will be printing bits for a new range of jet engines. Morris Technologies uses a number of 3D printing machines, all of which work by using a digital description of an object to build it in physical form, layer by layer. Among the 3D printing technologies used by Morris Technologies is laser sintering. This involves spreading a thin layer of metallic powder onto a build platform and then fusing the material with a laser beam. The process is repeated until an object emerges."

"One of the attractions of printing parts is that it saves material. Instead of machining components from solid billets of metal, in which much of it may be cut away, only the material that is needed to shape the part is used. Printed parts can also be made lighter than forged parts, which promises fuel savings. 

Many manufacturers already use 3D printing to make prototypes of parts, because it is cheaper and more flexible than tooling up to produce just one or two items. But the technology is now good enough for it to be used to make production items too."

No, dammit!

That's not why the aviation industry uses sintering!

Think about the conditions in a jet engine: extreme heat, extreme stress. Any material that can withstand those types of conditions is going to be prohibitively expensive to machine. Sintered alloys are actually much weaker than cast alloys, which means the recipe is going to need to be all the more complex to get that strength back.

Aside from needing to melt substances that are explicitly designed to have an extremely high melting point, the problem with casting is that as the material cools, it shrinks from the mold, so it becomes impossible to cast parts with precision. That means the only alternative to sintering would be taking a big block of superhard alloy, and then trying to grind it down to a blisk or whatever you're making.

Forging produces even stronger materials (through stress hardening), but the logistical impossibility of forging a turbine blade should be obvious to anyone.

Sintering is used in mass production when there's no other choice.

If you're trying to make tank armour on the other hand, where precision is not so important, but strength and cost are..yeah, good luck selling the industry on your 3-D printers.

Don't worry, I'm an engineer!

"Does it cost more pollution to create high octane gas?"

No, there's only a small amount of leeway in how a refinery fractionates its crude. Gasoline itself is a blend of different compounds.

One desirable characteristic of gasoline is how much it can be compressed before it auto-ignites (you don't want this to happen because it'll throw your engine timing off) and a specific band of compounds (known as the B-T-X fraction: benzene, toluene, xylenes) increases this property.

The reason high-octane fuels are more expensive is because there just isn't enough of this stuff to go around. You get your crude and when you separate it into components there is x amount of stuff you can make gasoline with and y amount of B-T-X additives where y is less than x. Supply and demand economics.

"Does it release more pollution when used in a car?"

This is a loaded question.

At high compression ratios, the fuel burns more efficiently so you get more complete combustion, more power for the fuel. So in the strictest sense, no. Let's say you have a VQ35, all else being equal you will have better combustion (so you get a higher CO2 to CO ratio with C being stoichiometrically limiting) and more importantly you get more power per unit of fuel so you need to burn less of it to do the same thing.

There are a number of caveats. First you need an engine designed for high compression, otherwise low or high octane will ignite all the same and you're just wasting your money.

Secondly, high compression engines are usually high power, high performance pieces of machinery. So yes, you will get to 300hp more efficiently with a high octane fuel versus a low octane fuel, but if you are getting to places with a 300hp engine instead of a 100hp engine, you're not really saving any fuel are you?

The corollary is that engines have very different sets of performance envelopes so what's most fuel efficient depends on your driving habits. If you do a lot of hard acceleration, a Nissan Sentra won't have enough torque to push it on the low revs, so your car is going to have to gear down and rev harder. On the other hand, a straight-6 BMW is going to be able to generate the same amount of power without breaching 3000 on the tachometer. An engine operates more efficiently in the 2000 - 3000 rpm band than, say, 5500. So in that case, yeah, you might actually save more fuel in that 300hp BMW than the 100hp Nissan. Jeremy Clarkson might be an ape but he got that right, it's more about how you drive than what you drive.

(I've simplified some things but the point is generally valid)

Thunk

At one time, Japanese autos were much better than what the North American makers were offering. They were more reliable, performed better and came at a great price. Ford introduced the assembly line, but Toyota brought Lean, TPS and 5S to the manufacturing world.

Also, their doors made a great thunk when you closed them.

Of course now, every car door makes that sound. Auto manufacturing has come a long way since then and quality has increased across the board from everyone. But back then American makers were struggling to replicate that sound, the sound of a perfect fusion of parts. Something that comes off a line that measures defects in PPMs. The intangible feeling of goodness.

I just want to say that I get a satisfying thunk every time I close my laptop lid; time to catch up, everyone else