Tech Tuesday 06/11/19 – Back in WA Edition

Oscar Gordon

A Navy Turbine Tech who learned to spin wrenches on old cars, Oscar has since been trained as an Engineer & Software Developer & now writes tools for other engineers. When not in his shop or at work, he can be found spending time with his family, gardening, hiking, kayaking, gaming, or whatever strikes his fancy & fits in the budget.

Related Post Roulette

36 Responses

  1. Michael Cain says:

    TT03: Every time new tech for remote methane sensing is deployed, it turns out that some industry or another is leaking methane at an order of magnitude higher rate than previously assumed.Report

    • Oscar Gordon in reply to Michael Cain says:

      Is that an indictment of how we officially asses methane discharge, or of how remote sensing tech tends to overblow the leakage?Report

      • Philip H in reply to Oscar Gordon says:

        More the former then the latter, as methane discharge in regulated industries is significantly self-reported.Report

      • What Philip said, at least in the oil and gas industry. The EPA’s official methodology is bottom-up, and makes assumptions about best practices, equipment failure rates, etc. A few years back Colorado toughened its standards and the companies were required to do a lot more measurements. What they found was more leaks, and bigger leaks, than anticipated. All of that lost gas was money flying away — the industry has improved its practices considerably. Even though drilling and production activity is up quite a lot since the new regs were put into effect, the number of reported leaks and their size are down significantly.

        My son-in-law supervises the fabrication shop for a company that builds custom manifolds and such for oil and gas producers. The company’s forte has always been quality — better welding, zero corner-cutting on testing, etc. Starting a couple of years after the new regulations were in place, their business has been growing quite a bit.Report

        • George Turner in reply to Michael Cain says:

          I double checked the article’s numbers, and sure enough, they fail the metric system.

          Given their finding that 0.34% of the natural gas used by fertilizer plants is leaking, they correctly extrapolated to an estimate of 28 gigagrams of leakage, which is 28,000 tonnes. That would mean the fertilizer plants consuming 8.2 million metric tonnes of natural gas, whereas based on CO2 emissions (including re-use), I’d calculated a usage of 8.6 million metric tons, with the difference probably due to using a slightly different year’s data.

          Where they screwed up was with the EPA emissions figures. The EPA and everyone else said that industry leaks 8,000,000 tonnes of methane, which was updated by a surprising finding that the actual leakage rate was 60% higher than assumed, at 13 million tonnes.

          Google thought that 8 million tonnes, which is eight billion kilograms, is a gigagram, but it’s actually a teragram. They then used the erroneous eight gigagrams figure to find that the fertilizer plants emit 3.5 times more than everything else in the US, when the actual number would be 0.35% as much as all the other emission sources.

          Now they’ll get laughed at by everyone in industry for not knowing how to shift units between tera and giga.Report

  2. George Turner says:

    TT03, the story about Google researchers finding that fertilizer plants are emitting three times more methane than the rest of the US, is probably a reflection of garbage data. Natural gas isn’t pure methane, and I suspect some other component of it that isn’t used in the Haber process might be going through the high-temperature, high-pressure process and producing a compound that pegs the needle on Google’s remote sensors, even at trace amounts.

    The US fertilizer industry only consumes 8.6 million metric tons of methane, based on how much CO2 they produce. The hydrogen is stripped to combine with 9.6 million tons of nitrogen, resulting in 31 million tons of CO2, of which 24% is recycled to make urea and other products. The reaction produces the same waste heat as burning 4 million tons of the methane input, and 51% of that waste heat is recovered.

    The EPA’s estimate of methane leaks from just the production side is 2.3%, or 13 million tonnes. Just using that figure would mean that the fertilizer industry is emitting 4.5 times more methane then they’re buying from the gas companies.

    So if Google’s data was accurate, the oil companies should just run pipes from the fertilizer plants to their other customers and quit drilling holes in the ground.

    Fertilizer production is a very expensive undertaking in a highly competitive world market, and the chemical engineers optimize it every way they can because they have to buy all the natural gas and other inputs, unlike an oil company that can just vent the gas because it’s an unwanted byproduct of drilling holes in the ground.Report

    • George Turner in reply to George Turner says:

      The article says:

      In addition, this figure far exceeds the EPA’s estimate that all industrial processes in the United States produce only 8 gigagrams of methane emissions per year.

      Actually, just from leakage on the production side, the number is 13 million metric tons, which is 13,000,000 tonnes, 13,000,000,000 kg, or 13,000,000,000,000 grams. The older EPA number would have been about 8 instead of 13. But that’s a teragram, not a gigagram.

      What they’ve probably done is assume that a gigagram is a billion kilograms, which is off by a factor of a thousand.Report

    • DensityDuck in reply to George Turner says:

      A former co-worker left to start a company doing that kind of survey from the air, and it took three PhD’s and several experts in optical sensors to develop the kind of equipment needed and turn the measurements into actual data. This isn’t something you can do with a gas detector that you bought at Lowe’s duct-taped to the roof of your car.

      That said, the company was hired by industrial gas providers and producers, and they weren’t allowed to republish the data on leakage afterwards. He did let on that what they was a lot higher than anyone was really willing to talk about.Report

      • George Turner in reply to DensityDuck says:

        I pulled up the original paper and their science and technique section looked really thorough. I suspect that when they reached their erroneous conclusion caused by screwing up the metric system on the EPA’s emission figures, they went over their methods really carefully because they thought they had a truly shocking result and wanted to be absolutely sure their data would hold up. Unfortunately, they apparently never double checked their conversion of the EPA figure from tonnes to grams. Peer review didn’t catch it either.

        I left a short little comment on the paper pointing out the mistake.Report

  3. George Turner says:

    TT04: So Michelin has managed to use a complex shape made out of advanced polymers to replicate what everyone else can do with compressed air, which is virtually free, and which results in a tire whose innards don’t get filled up with gravel when you zip down a rural road.

    If we had been using the Michelin tires all along, and some garage mechanic in Alabama came up with a way to make a tire using just compressed air, we’d have hailed him as the genius of the age.

    Yeah, I’m probably in a bit of a skeptical mood after TT03.Report

    • Oscar Gordon in reply to George Turner says:

      Modern tires are an exercise in complex composite materials, and when you get a hole in them, you run a very good chance you will have to throw them out, and maybe the rim too.Report

      • George Turner in reply to Oscar Gordon says:

        Well, given how those tires flex, gravel that lands in the open area, where it will likely be held by centrifugal forces until the car stops, is probably going to chew them up. They might want to consider a fabric sidewall made out of something like Kevlar, just to keep out debris, and especially mud.Report

        • Oscar Gordon in reply to George Turner says:

          Google images of airless tires mounted on construction vehicles, see how they deal with it.

          Alternatively, we can assume that there will be different variations on the theme, so if you frequent gravel roads, you get a different kind of wheel.Report

  4. Philip H says:

    TT03 – the Google Car is a driven vehicle that does all the Street View imaging. SO its not a hazard to navigation. Yet. The Methane sensor is just a small add-on that puts additional data into the cars memory bank. Interestingly, NOAA has similar under reporting of methane emissions from fracking out west.Report

  5. DensityDuck says:

    [TT10] oh hey there extreme-aft CG, how are you today? still flipping the aircraft over? cool, cool, catch ya later!

    I mean, it’s at least an interesting approach to solving the problem of “elliptical cross-sections are not as efficient as round tubes”, but I that looks like the aerodynamic center is going to be *way* too close to the center of gravity. The thing’ll flip like a coin! I kinda feel like the solution would be to fill in the opening of that V-split, but that adds weight, and eventually your design isn’t any better than the way we’ve been building airplanes since forever. Ye canna change the laws’o physics, cap’n!Report

    • Oscar Gordon in reply to DensityDuck says:

      That’s where I was going with my comment. Can you make it work? Yeah, sure. Is it worth the effort? Boeing kept looking at it, and kept saying ‘no’.

      Maybe now it is worth the effort, if part of the effort is on someone else’s budget. Perhaps TU Delft got themselves a fancy new computing cluster and is itching for something they can use it for.Report

  6. DensityDuck says:

    [TT19] the shocking secret here is that we’ve had the capability to do this since the 1970s, we just didn’t, because it was cheaper to hire migrant workers and pay them cash. It wasn’t just a California and Texas thing, it was all over; retirees or Okie pickers driving from state to state, and up north it was Canadian families who’d move across the border for a season and pick blueberries and apples. As the illegality of hiring migrant pickers is being increasingly enforced, the cost of work increases, and it becomes worth it to buy machines to do the job.

    I mean, that’s the thing; as with coal mining, ag-labor jobs are gone and aren’t going to come back. If we make the cheap labor illegal people will just mechanise, and then you’ll need skilled labor to operate and maintain the machinery.Report

  7. DensityDuck says:

    [TT07] I’d say it’s probably the number of thin membranes undergoing buckling as the load hits them. Buckling and crippling cause a great deal of plastic (that is, permanent) deformation in a material, and plastic deformation is very dissipative of energy (this is why cars have crumple zones.) Foam, by its nature, has plenty of thin membranes, which means it’s very dissipative of energy.

    You could probably get a similar effect by using honeycomb, but the honeycomb is directional–if you don’t line up the axis of the honeycomb holes with the bullet, then it’s just punching through the sides of the foil walls and that hardly does anything. The foam’s advantage is that the bubbles result in “walls” oriented in every direction.Report

    • Oscar Gordon in reply to DensityDuck says:

      This. Sorry, I was knocking this out in a hurry.Report

    • George Turner in reply to DensityDuck says:

      Due to what you mention, the area or volume undergoing deformation is probably much larger, a bit like a foam mattress, instead of the thin cone of deformation around the projectile’s path that you get with homogeneous armor.

      And each bubble in the direct path might be acting like a saw tooth to shred the surface of the projectile, as opposed to a solid cylinder (with a cone in front) applying uniform pressure that helps maintain the projectile’s shape. A foam has places for the shavings to go.

      If that is part of what’s going on, it would be interesting to see how foams do against shaped charge projectiles, where they might be better at turning the uniform stream of molten copper into more of a spray pattern you’d get from your faucet’s aerator.Report

      • JoeSal in reply to George Turner says:

        I have concerns about the rates of ablation of that foam. It appears to be ejecting considerable volume per impact.Report

        • Oscar Gordon in reply to JoeSal says:

          That’s not the foam ablating, it’s a ceramic cover plate.Report

          • JoeSal in reply to Oscar Gordon says:

            Ah, so that brings up the next question on how it performs after the ceramic is ablated?Report

            • Oscar Gordon in reply to JoeSal says:

              Good question. One thing to note is that the researchers admit that they were using what was basically COTS facing and backing material, and the bonding agents employed were not ideal for mating that material to the foam, so a lot of what you see may be the result of ‘cobbled together’ armor, rather than something that had been designed with purpose.

              Which means, once they begin optimizing for the foam, it could be scary good.Report

  8. Pinky says:

    The sun doesn’t assault your skin in WA, but for four months it doesn’t register in your brain as “daylight”.Report

  9. George Turner says:

    Here’s another piece of tech/defense news.

    Defense News: Supersonic speeds a problem for F-35 stealth coating

    The stealth coating isn’t compatible with the heat at the back end of the plane.

    For example, an F-35C can only fly at Mach 1.3 in afterburner for 50 cumulative seconds, meaning that a pilot cannot clock 50 seconds at that speed, slow down for a couple seconds and then speed back up. However, the time requirements reset after the pilot operates at military power — an engine power setting that allows for less speed and thrust than afterburner — for a duration of three minutes.

    The F-35B can fly for 80 cumulative seconds at Mach 1.2 or 40 seconds at Mach 1.3 without risking damage.

    But for both the C and B models, flying at Mach 1.3 over the specified time limits poses the risk of inducing structural damage to the aircraft’s horizontal stabilizer.

    And this potential happens at Mach 1.3, which is just peachy.Report

    • DensityDuck in reply to George Turner says:

      So many asterisks in that statement that appear when you read the actual article and actually know something about how this works.

      “The F-35 stealth coating melts when the airplane flies!*”

      *at the maximum speed possible**
      **with full afterburner***
      ***which it ordinarily will never do, even in combat****
      ****and this is the old coating anyway, and we’re about three iterations beyond that*****
      *****and it only happened once******
      ******and the “structural damage” they talked about was superficial scarring on the surface*******
      *******which has to be reported because the terms of the contract demand it, not because anyone thinks it’s actually going to be a problem********
      ********which they know because this will, like, reduce the structural performance margin for FS 2.5 from 3.0 to 2.8, which means you’re still 280% stronger than you need to be for the a load state that’s 2.5 times worse than the absolute highest loads the jet would ever experience in its design life

      So, does it need to be addressed? Sure, these planes will presumably be flying for the next 50 years and we don’t really want them to be degrading just from flying around. But is this a worse problem than, say, compressor stalls or inertial coupling? No, nowhere near, and we fielded operational systems with those problems.Report

      • George Turner in reply to DensityDuck says:

        Yes, a whole lot of caveats in that article.

        That such a low Mach number can be considered the edge of the envelope caught my eye. They went to a whole lot of trouble to a make a jet that’s only somewhat faster than a Harrier?

        In any event, my first guess is radiative heating from the glowing afterburner exhaust and its pretty Mach diamonds. If that is the case, and given that they couldn’t replicate the conditions, I’d ask what might have been different that would make the exhaust brighter than normal in the IR spectrum, such as more soot or water vapor in the exhaust stream, or some type of aerosol or particulate matter in the atmosphere when the incident occurred.Report

        • DensityDuck in reply to George Turner says:

          Or it’s a different surface material, or a different application method. Pretty early on they switched from hand-trimmed and -applied decals to robot-sprayed treatments, with machine-cut decals applied by hand only in areas like door edges where treatments get ripped off through the operation of the system. It could be that the tech putting those treatments on just didn’t work out the air bubbles well enough, and that’s why the treatment bubbled in a way that they couldn’t reproduce in testing.Report

          • George Turner in reply to DensityDuck says:

            That’s probably more likely, especially early in the production run. A lot of low-rate aircraft production is essentially craft work, and that can vary quite a bit as they try to figure out a good process.

            One time I was in a bar outside a Grumman plant and one of the workers told me how they did they carbon fiber layup for the tail cone of some military transport. The cone was made pointy end down, and they couldn’t figure out a good way to lay plies in near the inside tip of it. So they had a big guy hold a small ex-cheerleader upside down by her ankles so she could reach down in there.

            There’s a video of wartime B-24 production that shows how they used midgets inside the wing to bolt the outer section to the inner section.Report