7 ms·

The Mechanical Transmission of Power – Jerker Line Systems

129 points by youngerdryas 14y ago

ck2 14y ago
Hmm, the very first "smart grid". Well not so smart in itself but very cleverly used. First steam-powered cable-car systems were also at the end of the 1800s.
js2 14y ago
Wonder how many digits and limbs were lost to such systems.
- ars 14y ago
  Probably not many. The lines don't move very fast - about 1 cycle per second.
  - gus_massa 14y ago
    But the tension is (probably) very high, if a digit gets trapped it'd (probably) get crushed.
jchrisa 14y ago
This company builds tech for using municipal water supplies as a power distribution system http://www.lucidenergy.com/ http://www.lucidenergy.com/
gus_massa 14y ago
Is this really more a efficient way to transmit energy than electrical currents? Anyone has a link to a economic analysis?
- jstanley 14y ago
  These were invented in the 1500s. I imagine in those days this was far more pratical than electricity :)
  - gus_massa 14y ago
    I was curious because the article says: > 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.
    - schiffern 14y ago
      >It looks like the modern version of this technology can be useful today. Low-Tech Magazine has, surprise!, a pro- low-tech bias.
- wlesieutre 14y ago
  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'.
  - wiml 14y ago
    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.
    - wlesieutre 14y ago
      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.
rwinn 14y ago
Video: https://www.youtube.com/watch?v=ALPkDoX9Wz0 https://www.youtube.com/watch?v=ALPkDoX9Wz0
mjb 14y ago
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.
- bigiain 14y ago
  "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.)
bane 14y ago
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. [1] - they looked something like this https://www.youtube.com/watch?feature=endscreen&NR=1&v=e_7r4FHt0zI https://www.youtube.com/watch?feature=endscreen&NR=1&#38...
nickpinkston 14y ago
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. http://www.drakewell.org/ http://www.drakewell.org/
pontifier 14y ago
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.
- sliverstorm 14y ago
  Wait, but then you have to regulate speed to each joint. Is there an easy way to do this I haven't thought of?
  - bigiain 14y ago
    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/ 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.

129 points by youngerdryas 14y ago

ck2 14y ago

Hmm, the very first "smart grid". Well not so smart in itself but very cleverly used. First steam-powered cable-car systems were also at the end of the 1800s.

js2 14y ago

Wonder how many digits and limbs were lost to such systems.

ars 14y ago
Probably not many. The lines don't move very fast - about 1 cycle per second.
- gus_massa 14y ago
  But the tension is (probably) very high, if a digit gets trapped it'd (probably) get crushed.

jchrisa 14y ago

This company builds tech for using municipal water supplies as a power distribution system http://www.lucidenergy.com/ http://www.lucidenergy.com/

gus_massa 14y ago

Is this really more a efficient way to transmit energy than electrical currents? Anyone has a link to a economic analysis?

jstanley 14y ago
These were invented in the 1500s. I imagine in those days this was far more pratical than electricity :)
- gus_massa 14y ago
  I was curious because the article says: > 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.
  - schiffern 14y ago
    >It looks like the modern version of this technology can be useful today. Low-Tech Magazine has, surprise!, a pro- low-tech bias.
wlesieutre 14y ago
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'.
- wiml 14y ago
  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.
  - wlesieutre 14y ago
    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.

rwinn 14y ago

Video: https://www.youtube.com/watch?v=ALPkDoX9Wz0 https://www.youtube.com/watch?v=ALPkDoX9Wz0

mjb 14y ago

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.

bigiain 14y ago
"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.)

bane 14y ago

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. [1] - they looked something like this https://www.youtube.com/watch?feature=endscreen&NR=1&v=e_7r4FHt0zI https://www.youtube.com/watch?feature=endscreen&NR=1&#38...

nickpinkston 14y ago

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. http://www.drakewell.org/ http://www.drakewell.org/

pontifier 14y ago

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.

sliverstorm 14y ago
Wait, but then you have to regulate speed to each joint. Is there an easy way to do this I haven't thought of?
- bigiain 14y ago
  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/ 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.