25 May, 2013


A Rant Against the Evolution of Things Until They Become Less Useful

Not to sound too much like Andy Rooney or anything, but, "do you remember when ATMs were called cash machines?" That was back in the days like maybe mid-70's when ATMs were widely introduced and began their evolution into being useful things. You could go to a cash machine at any time of the day or night and, assuming you had any money in your account and that the machine was actually working, come away with money in your pocket. At first, I viewed ATMs with a jaundiced eye since they seemed poised to let me down when I was most counting on them working. Or at least that was the situation in the ground in Atlanta, Georgia.

Over time the reliability, expressed as the likelihood of a given ATM being capable of dispensing cash on demand, and thus not pissing you off royally, improved markedly. This was probably a result of the confluence of ATM technology improvements and successively more robust networks linking the machines to the central banks.

As the man who had jumped from the roof of a very tall skyscraper was heard to exclaim as he passed by each floor, "So far, so good." ATMs had become unquestionably useful. And that's a good thing.

But more recently I've noted that ATMs have evolved to perform tasks beyond mere dispensing of cash. This, paradoxically enough, is not a good thing. Yesterday I pulled into the ATM lane. By dint of poor planning, I was in a slight hurry as I frequently am. But all systems looked to be go as there was but a lone car in front of me. What I learned was that this particular motorist was using the ATM to renegotiate a jumbo home mortgage that was seriously underwater, fund a micro-loan facility under the auspices of the World Bank, or perhaps commune with Ben Bernanke on QE3. Okay, maybe I'm exaggerating somewhat. In reality my antagonist seemed to be depositing multiple checks (probably birthday money from his grandma, etc.) and some cash in small, wrinkled bills. The money launderer in me recoils at the thought of ever depositing cash; why not hang onto it, throw it on your bed and roll around in it, or use large denomination bills to light cigars?

Anyway, this transaction was taking a long time and I started to conclude that all this high finance could have been more promptly transacted inside the bank with a human teller rather than outside with his car, my car, and now a third car that had pulled in behind me, idling and needlessly despoiling our air. The bile began to rise to my throat. Just at that moment, the car pulled away from the ATM. So now it was my turn and I was determined to conduct my business with all the alacrity and economy of motion of an Indy-500 pitstop. This would render me a hero to the guy behind me. I hummed my ready card into the machine, banged my PIN into the keypad, and was greeted with the message: "THIS MACHINE CANNOT DISPENSE CASH AT THIS TIME."

The failure of the machine accounted for at most twenty percent of my blood pressure rise. Stuff happens, right? But having to wait behind someone conducting a protracted transaction should be avoidable. It never would happen to begin with if ATMs had not evolved beyond their useful purpose as cash dispensers.

My modest proposal is that banks install cash-dispensing-only machines to augment the all-purpose ATMs that I will promise never to use. I realize that this may be problematic in terms of real estate occupied by the ATMs and any drive-up teller lanes but I don't care. Figure it out and get it done.

13 May, 2013

What technologies will near-term new cars feature?

Looks like $80K and up Tesla's notwithstanding, the internal combustion engine will stick around for some time, especially when one considers hybrids.

Pure electric autos have to get a credible 300 mile range (like a real car) and crack the recharge time problem. I can pump 300 miles worth of oomph into my GTO (which has a hearty thirst) in about three minutes, versus half a day for moving electrons around. I'm not saying this won't happen, but it will take some time and investment. Investment in technology for quick-charge batteries and in infrastructure for more ubiquitous charging stations. One option would be an industry-standard battery pack that could be robotically swapped out in minutes. Speaking of industry standards, a "universal" charger and charge control interface is mandatory as an enabler for widespread acceptance of electric cars. Possibly more so than price (within reasonable limits of course).

I wonder about the long-term viability of hybrids since they seem to be neither fish nor fowl and bring along a lot of complexity and cost. Seems to me that a small, light weight car powered by a modern turbo-diesel could eclipse the effective MPG of a hybrid and do so more cheaply in terms of initial cost (and long-term cost of ownership if the battery pack needs to be replaced).

Having gotten that diatribe out of the way, here is what I am seeing in the crystal ball polished through some work (reasonably priced and unreasonable brilliant!!!) I've done for clients in the automotive industry. Look forward (or not) to:

  • Smaller displacement engines of four (or even three) cylinders. These will eschew natural aspiration for supercharging (tuned for performance) or turbocharging (tuned for efficiency).
  • Direct (high pressure) fuel injection
  • More multi-fuel options and greater tolerance for ethanol blends
  • Variable valve timing and a longer term move to electromechanical valve actuation 
  • Start/stop engine technology to reduce emissions and increase MPG in city driving cycles. Having driven a succession of crappy cars in my time, I find an engine that dies at a stop sign unnerving, but I imagine we'll get over it.
  • A proliferation of engine cycles such as the Atkinson cycle. No gas turbines; no Stirling engines yet.
  • Spark plugs as you know and love them will remain with us for some time. There's no big pressure to supplant them with laser initiation (although we could), an they last a long time without trouble. Coil-on-plug will continue to be the standard, and if performance/efficiency demands, we may see two plugs per cylinder in some applications.
  • 100,000 mile "hood welded shut" expectation. We're pretty close to that already. I see increased use by OEMs of full synthetic oil and really long oil change intervals controlled by on-condition sensors (or at least an ECM calculation).
  • Liquid cooling will be retained, but a non-water based fluid will become more dominant.
  • A motor-alternator unit will replace the conventional starter motor and alternator for weight savings and in response to the stop/start feature.
  • Automatic transmissions in their various forms will continue to squeeze the conventional manual gearbox out as cars continue their evolution from something you drive to something you ride in. Number of gear ratios will stop at around eight (I hope) and CVTs will be used beyond that. Bottom line is that we'll be seeing engines optimized around a tight RPM band for economy and emissions.
  • Some high-end drivetrains will incorporate mechanical or pneumatic energy storage in support of regenerative braking.
  • Pre-start oiling systems may allow your rotating and reciprocating engine parts to last forever if the economics work out. Ditto proven cool down post-shutdown controls.
  • Accessories like A/C, cooling fans, and power steering will be electrically driven for max MPG.
  • Driver "aids" like lane departure warning, back-up obstacle warning, parking assist, radar augmented cruise control, and collision avoidance will increase their market penetration as they come down in cost.
  • Increased use of a "car area network" to eliminate cost, weight, installation labor, and failure-susceptibility of conventional wiring harnesses

Thoughts on air-assisted versus ground launch for LEO missions

Most of what I read on this topic is focused on energy considerations -- i.e. how much delta-v can an air launch save versus a "simpler" ground-based multi-stage system. And when you look at it this way, the answer -- somewhat surprisingly and disappointing to my ever hopeful mind -- is not as much as you might imagine or wish for. You will find a good treatment of the situation here:


The bottom line is, from a purely energetics standpoint, air launch can be viable for relatively small mass payloads (assuming they are within the capabilities of the launcher/carrier aircraft), but it's not anything to write home about. You can compare the perfromance and cost of Orbital Science's Pegasus (air-launched) versus its Taurus (conventional multi-stage ground launch) to get an idea of what I mean.

But I'm thinking that energetics is not the sole or even the primary consideration to be addressed in trades of air versus ground first stages. I see these other factors as being significant drivers to choosing an integrated solution:

  • Options for reuse, especially of high value components
  • Reasonable and survivable (I'm talking of hardware here since I'm not focusing on a human-rated system at this point) abort modes
  • Opportunity for gradual checkout of all vehicle systems prior to commit to flight
  • Enhanced dispatch reliability and launch phasing with respect to an orbital target
  • Minimized time and consumables dedicated to orbital target rendezvous
  • Use of existing hardware to drive down NRE
  • Options for "non-conventional" approaches to system architecture and CONOPS. Things like:
    1. Tow, rather than carry of the flight vehicle
    2. "Leave behind" gear (very massive tires, wheels, brakes, steering actuators, struts, structure)
    3. Transfer of oxidizer and/or fuel from the launcher aircraft while in flight
    4. On-board oxygen generation within the launcher aircraft
    5. Towed UAV that returns to takeoff runway autonomously after releasing the flight vehicle