That's very cool stuff, and a really clever solution to a highly constrained problem. I'd really love to see one in action - and it looks like I could.
> Why do this instead of powering each device individually? One: you save three electric motors. Two: there is no need to provide batteries or electric outlets at any of the locations. Three: you can balance the system so that one device helps power the other, saving a considerable amount of energy.
Pairing rocking machines like this is a real efficiency improvement, especially with IC and steam engines where you don't really have the option of capturing the return stroke's energy directly. Even with an electrical system, where this is possible, it would add a whole lot of complexity.
> In these cases, you can distribute mechanical energy without conversion losses.
Without conversion losses, maybe. However, you are going to lose a lot of energy to friction (especially if you have a lot of the rocking posts, friction posts, hold-downs and so on). Also, unless the rods connecting the machines are perfectly elastic you're going to lose a bunch of energy to heating the rods themselves.
You can have all the "Doubts on progress and technology" you want, but if you only count one kind of loss (electrical conversion) and not another (mechanical friction and inelastic fraction) you're going to end up with a non-realistic bias in favor of mechanical systems every time.
"You can have all the "Doubts on progress and technology" you want, but if you only count one kind of loss (electrical conversion) and not another (mechanical friction and inelastic fraction) you're going to end up with a non-realistic bias in favor of mechanical systems every time."
I've been wondering for a while whether anyone's trying to re-capture some of the "lost" energy from household electrical systems. I wonder why my fridge, aircon, and stove aren't "pre heating" the cold water going into my hot water heater? I even wonder why all my "fixed electronics" - I'm looking at the TV, Playstation, audio gear, media center PC, home networking gear - have all got heatsinks and fans "disposing" of "waste heat" (which I sometimes then spend more money dealing with by cooling the room down with aircon).
Is there _really_ no good way to capture all that "waste heat" and use it to offset the energy I consume having a hot shower every day?
(I'm guessing I'm not the first to ever have these thoughts, and that the answer right now is "power is so cheap that the payback time for any available technology that could do this is way longer than practical". Which satisfies the pragmatist in me, but still makes the tree-hugger in me slightly melancholy.)
>I wonder why my fridge, aircon, and stove aren't "pre heating" the cold water going into my hot water heater?
I asked the CEO of SunFrost (who make super-efficient fridges for solar houses) whether he had examined the possibility of a water-cooled refrigerator. He said that people didn't want to pay a plumber to install and remove their refrigerator, and that the lifetime savings doesn't really pay for the extra time and hardware.
There is a very simple solution to this if you are in an area where electricity is cheap and the climate is cold. Simply use electrical heaters with a thermostat. Using electronic appliances, including light bulbs, increases the temperature and reduces the time your heaters have to run.
This is also the reason why low-energy light bulbs are less profitable in certain countries.
Your solution is penny wise and pound foolish. Using heat pumps will save you enormous amounts of energy overall, even though your appliances look superficially worst under that analysis.
The economic equivalent would be:
You: "Hey Bob, those are expensive fishing trips. Think of the opportunity cost!"
Bob: "Thanks, you've saved me a bundle. I'll go quit my job immediately!"
(The effect you're citing is real, but the explanation is more basic: at low energy costs, efficient goods have longer payoffs. This happens regardless of what heating method people use.)
Yes, this is very obviously true. But there are very good historical reasons why this is the way things have traditionally worked in Norway.
First, heat pumps were very uncommon until only a few years ago, since no one needs A/C in this cold climate. A heat pump can have a 10-year payoff with our electricity costs, although this has gotten better in recent years. There is no gas infrastucture. And it makes no sense to switch to high-energy appliances untli you're switched to more efficient heating.
Yep. I've mostly go the other problem though. Here in Sydney Australia, it's reasonably rare for us to need to heat the house - especially right now, in the middle of our summer. Even in the middle of a Sydney winter it's uncommon for us to actively heat the house - we've only got one "space heater" in the house, a little 1000W fan heater that mostly gets used to dry rained-on motorcycle jackets/boots/gloves. Very rarey (perhaps 2 or 3 days per year) we run the aircon in reverse-cycle heating mode.
I guess that your comment probably reveals another reason why this idea isn't currently economically feasible - the "richest market" where the pre-conditions for it to work (households with significant amounts of "waste heat generating electronics" running permanently or close to), is surely the USA and the 2nd richest market is probably western Europe. Both regions where for a large part of their population, the climate is cold enough for significant parts of the year for your simple solution to be "good enough". A market the size of Australia (~20million people) probably isn't anything like big enough to get the economy of scale to work.
Now my tree-hugger side is even more melancholy...
I would think using electronic appliances would save heating effort on any closed-loop heating system, electrical or not. Am I missing something? Other than that, great point.
In cold climate waste heat from electronics isn't wasted.
Also I think some old customs were really ecological just because of being used to scarce resources. For example - my grandpa have old ceramic stove for cooking in the kitchen. It has pipes with water inside - additionaly heating water for washing, when you cook. I've always thought it's a great idea.
My grandma also had big plastic bucket in the tube in the bathroom, when you washed your hands during the day you did it over that bucket. When you needed to flush the toilet, you first used the water from that bucket, only if it was empty you used built-in "flusher".
These things weren't really needed anymore, but that generation got used to not waste anything.
My father grew up in Southern Illinois when there was still oil being pumped out of the ground. His father, grandfather and great-grandfather worked on systems that sounded like these. A trip to area let me see an old single piston engine [1], but we never found any of these systems still working.
According to my father, my grandfather would go to the central hub and grab hold of one of the lines for a few moments, then to the next and the next. He could tell by the way the line was moving and felt if there was a problem down field he needed to attend to.
I never quite "got" what he was talking about in my mind, and my father was very young when he saw them in action so I never really had a great explanation of what these "lines" were.
Great post and finally fills in a gap in my father's childhood stories.
To see one of these systems in action, the Drake Well museum in north Western Pennsylvania has the system used to pull the first oil wells. From steam engines to the well pumps - it's very cool place if you're up by Erie. Also, a nitroglycerin blasting demo is pretty cool.
It's transmitting power directly instead of taking it through a generator then back again using a motor, so it's quite possible. A quick search suggests that motors and generators can both be over 90% efficient in the electrical/mechanical conversion, but I would guess that you could do better in a purely mechanical system.
Over short distances we already use mechanical energy transfer all the time (drive shaft in cars, chains in bikes, etc.), this takes the same idea and extends it to a longer distance.
My hunch is that while it would still work in theory, electrical systems would win out because of construction costs and maintenance. Motors and generators are widely available, wire gets laid once and works for decades, and contractors know how to deal with all of it. A jerker line needs more frequent lubrication, has custom parts that wear out, and when they do you can't just say "I need a new NEMA 6 motor" and have it in a day.
EDIT: Also worth noting that the high efficiency numbers for the electrical system are going to be at peak load, if you run it at partial they'll drop. The frictional losses on a jerker line are probably a bigger percent of the total at part load as well, but I don't know whether they'd be more or less significant than a generators'.
It's really easy to overestimate the efficiency of a mechanical system. Each of those swiveling joints, wheels, etc., is going to lose a few percent even if perfectly maintained. Some of the devices like the friction supports will lose more.
Regarding replacement, I suspect that in an oilfield in the back of beyond in the late-19th/early-20th century, many of the replacement parts for these systems would be made on-site with a saw or a farrier's anvil.
It might come down to whether you're using the oscillating motion directly or trying to turn it back into rotation. There's no way that you'd get rotation more efficiently than a motor, but elements like the tripod pendulums seem like they'd have comparably small losses to the generator/motor. Then again, I've never seen any measurements of energy losses on pushing a pendulum back and forth, so as you say, I could be overestimating the efficiency of those too.
> The technology, which still operates in a handful of small oil fields, could also work with renewable energy sources, and shows great potential for efficient small-scale energy use.
It looks like the modern version of this technology can be useful today. My guess is that the energy is more efficiently transmitted by electricity (electric resistance vs. mechanical friction), but it's more expensive to put and maintain a electrical motor in each final node. But I don't have enough information, I'd like to see some real numbers.
I have often thought of using a similar system for power distribution in a robot. a single powerful motor with flywheel, and multiple cables to transmit power to each joint.
If you've got a (mostly) symmetrical action, like bipedal walking, there almost certainly is some way to recapture "returned power" from footfalls to reuse for the next step. It'd be a little different from just balancing reciprocating pumps against each other, since you'd need to "time slip" the recaptured energy - but a flywheel might make that possible without incurring too much loss to make it worthwhile.
For multi-legged robots, a bit of smart geometry can pretty much do this already - see these things: http://www.strandbeest.com/
I suspect using some computer-adjustable versions of the "multipliers" from the article would allow you to build something like a strandbeest which you could steer while only losing a little of the efficiency.
> Why do this instead of powering each device individually? One: you save three electric motors. Two: there is no need to provide batteries or electric outlets at any of the locations. Three: you can balance the system so that one device helps power the other, saving a considerable amount of energy.
Pairing rocking machines like this is a real efficiency improvement, especially with IC and steam engines where you don't really have the option of capturing the return stroke's energy directly. Even with an electrical system, where this is possible, it would add a whole lot of complexity.
> In these cases, you can distribute mechanical energy without conversion losses.
Without conversion losses, maybe. However, you are going to lose a lot of energy to friction (especially if you have a lot of the rocking posts, friction posts, hold-downs and so on). Also, unless the rods connecting the machines are perfectly elastic you're going to lose a bunch of energy to heating the rods themselves.
You can have all the "Doubts on progress and technology" you want, but if you only count one kind of loss (electrical conversion) and not another (mechanical friction and inelastic fraction) you're going to end up with a non-realistic bias in favor of mechanical systems every time.