Tech Tuesday 05/14/19 – Back From Vacation 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

10 Responses

  1. TT01:
    One assumes that a third temperature causes the salt to precipitate out as well, so the solvent can be reused.

    This is such a fundamental point that its not appearing in the article, the press release, or the abstract of the paper seems telling. My assumption is that they can’t easily precipitate the salts out (yet) in a way that allows the solvent to be directly reused. I’m still working on getting a copy of the complete paper.

    It also seems telling that they only apply the technique to hypersaline brines. This suggests that the technique may not be suitable for desalinating, say, normal seawater. I’m also suspicious that they worked with NaCl as the only dissolved salt; I can think of a number of places in the oil and gas industry where this type of technique would be useful, but the big nasties are metal salts other than NaCl.Report

    • Avatar Oscar Gordon in reply to Michael Cain says:

      If you get a copy of the paper, let me know, I’d like to read it.Report

      • Here. I’ll leave it for a couple of days.

        We both misunderstood the chemistry (or at least I did). At the low temperature the solvent “captures” water molecules that are not tied up with the salt’s component ions. The solvent-plus-water can be separated from the further concentrated brine based on density differences. At the high temperature, the solvent-plus-water gives up the captured water, with separation again by density differences. Any salt ions in the heated solvent-plus-water go with the water during separation, not the solvent.

        This probably also addresses my question about non-NaCl salts.Report

    • Avatar Philip H in reply to Michael Cain says:

      Here’s the abstract:

      Hypersaline brines are of growing environmental importance but are technologically under-served by today’s desalination methods. Temperature swing solvent extraction (TSSE) is a radically different desalination technology that is membrane-less and not based on evaporative phase change. TSSE utilizes low-temperature heat and a low-polarity solvent with temperature-dependent water solubility for the selective extraction of water over salt from saline feeds. This study demonstrates TSSE desalination of high-salinity brines simulated by NaCl solutions with three amine solvents: diisopropylamine (DIPA), N-ethylcyclohexylamine (ECHA), and N,N-dimethylcyclohexylamine (DMCHA). We show that TSSE can desalinate brines with salinities as high as ≈234000 ppm total dissolved solids (i.e., 4.0 M NaCl) and achieve salt removals up to 98.4%. Among the solvents, DIPA exhibited the highest water extraction efficiency whereas ECHA and DMCHA produced water with the lowest salt content and solvent residue content, respectively. Lastly, a high water recovery of >50% was demonstrated for TSSE desalination of 1.5 M NaCl brine using DIPA in semibatch experiments with multiple extraction cycles. This study underscores the unique capabilities of TSSE for the desalination of hypersaline brines.

      My office doesn’t have an institutional account so I cant download the PDF.Report

      • Avatar George Turner in reply to Philip H says:

        Last week I looked up the various amines, but it didn’t really help me understand their process. I did run across a reference that said it’s not as efficient as reverse osmosis for the more usual salt levels, except that it can run on waste heat, which is basically free.

        As for recovery, the amines I looked up boil at quite low temperatures (a little above 70C), so evaporation and distillation should be a simple matter.Report

  2. Avatar George Turner says:

    TT04: The aircraft uses pneumatic amplifiers where a puff of air through a slot can detach a much larger stream, which is I’m sure perfectly workable through most of the normal flight regime.

    However, modifying the airflow over a wing does not give an immediate response because the change in circulation builds over time, mathematically similar to charging a capacitor, and takes several chord lengths of travel to complete. That means that active aerodynamic control can’t completely close the loop to stabilize certain aerodynamic instabilities, and thus the ubiquitous swept-back, reflexed-tip flying wing designs.

    I came up with a way to make a dynamically unstable flying wing, straight and using a non-reflexed airfoil, which would be much more efficient than conventional flying wings. To do it, you have to abandon airflow control for locking down the instantaneous change in angle of attack and go straight back to applying a controlling torque with Newton and the rotational inertia of an electric motor.

    So take a model airplane wing and add a high-output DC motor (like you’d use in almost any model vehicle these days), with it’s axis aligned from wingtip to wingtip. For even more moment of inertia, affix the batteries to the motor’s rotor and pin the motor shaft to the airframe. The idea is that when you lay the wing on your kitchen table and rev the motor to maximum torque, the whole wing flips and flops like some toy, because the wing is extremely light and the motors are fairly heavy and extremely powerful. Basically, the motors can flip the lightweight balsa wing anyway they want. That allows direct control over angle of attack.

    So the motor control circuit is fed from a rate gyro. At one level the control loop is just trying to maintain a constant angular rate by sending a signal the motor controllers to accel or decel either in forward or reverse. In flight, there will be some initial deviation from the desired AoA, which if unchecked will result in a dynamic instability that ends in a tumble. But the torque on the motors will counter act this with a constant rotational acceleration, slowly drifting toward either the maximum forward or reverse speed for the motor.

    This is especially true since the wing has a different moment coefficient at different angles of attack, ie. the center of pressure isn’t constant throughout the flight regime, which is why such wings can’t normally be flown. So to keep the control motor’s velocity from maxing out, the whole motor mount is shifted forward and back with a regular servo linkage, just like a hang glider is controlled by weight shifts.

    So the AoA motor is rapidly adjusting to maintain a constant pitch rate, which is set with a control loop based on the commanded angle of attack, which stems from the pitch input of the person flying the plane. The forward-back servo then looks at the pitch motor’s RPM and tries to return it to zero by shifting the weight forward or back, just like a hang glider pilot flying a well-behaved, naturally stable aircraft.

    It should work great until the battery is depleted, when it goes into flutter and crash mode.

    However, the better the control loop, the smaller the deviations it has to counteract, and the less energy it uses per flight hour.

    Anyway, I wish I’d found time to build it, but I never did.Report

    • Avatar Oscar Gordon in reply to George Turner says:

      Got a drawing?Report

      • Avatar George Turner in reply to Oscar Gordon says:

        I had a few sketches years ago, but mechanically it would be really easy to just get a common model airplane kit, such as one of the many balsa 4-channel trainers, to take advantage of the space in the fuselage. Just don’t use the horizontal stabilizer or elevator. That lets you retain the ailerons and rudder and focus purely on the pitch stability mechanism.

        It also gives you some space for the pitch torque motor inside the fuselage, instead of the more inaccessible wing, which is the eventual goal for a flying wing design, but not really a requirement for a proof-of-concept vehicle. The weight shift servo doesn’t have to necessarily move the pitch motor forward and back, as that was just a way to shave some weight by having the mass do double duty. As long as you’ve got a motor controller handling the pitch torque motor, and a servo shifting the weight to try and keep the pitch motor near zero speed, you should be able to establish pitch stability.

        Assuming that was sketched out, you’d move on to how the control loop should act.

        To best emulate a conventional aircraft, you could use an angle of attack sensor and an airspeed sensor as inputs and then figure out how the AoA should be changing, eventually converting that to a position in the pitch axis, in reference to one of the many inertial sensors now common in model drones, and use that to command a new pitch angle, with the final control loop using motor torque to adjust the pitch rate, tuning it to avoid over or under damping.

        Then a motor RPM signal (which is available from a good brushless motor controller) is used to drive the weight shift servo, again tuned to properly damp, although in this case the pitch motor would compensate with rapid torque adjustments, but at the cost of wasted current.

        And of course all the data would need to be recorded so the performance could be tuned, but that’s not a problem with the better flight controllers, either, which can download all their data via USB to a laptop.

        Only then would I start trying to pack everything into a wing, using split ailerons or other devices for yaw control.

        So basically, if I had a sketch, it probably wouldn’t be more useful than the above descriptions, because the main concept is simply trying to use inertial torque from a motor to stabilize a conventional wing’s pitch or angle of attack (depending on what you think would be the better approach to the problem).

        ETA: For the first attempt, you could ignore AoA and other such inputs and just let the pilot eyeball it, since you’re just trying to establish that the aircraft is flyable and controllable in the most comfortable parts of the flight envelope (able to get up, make a few circles, and land).Report

  3. Avatar Oscar Gordon says:

    For anyone interested in what I used to do in the Navy…


  4. Avatar pillsy says:

    Today I learned what unsprung weight is!Report

Leave a Reply

Your email address will not be published. Required fields are marked *