ask me about technology
I have a lot of opinions about technology—what’s feasible, what isn’t, what the best ways to do things are, and so on. Sometimes I write them down on my blog, but usually I don’t. I also know some people who are good at designing technology.
If you want my personal opinion on some technology or technological problem, I guess you can ask me here.
One of your posts on energy suggests that technology like https://terraformindustries.wordpress.com/2023/06/26/the-terraformer-mark-one/ should not be possible at its price point (you wrote that even with free energy the equipment should be too expensive). Do you think the made a genuine advance to make it work?
If so would that basically mean we can plan for using a lot of methane in the future?
comment
The point of Terraform Industries was to trade electrolysis efficiency for lower capital cost, and find a use for lots of hydrogen. I don’t think they’ll be able to do the first part well enough, and their use for hydrogen is suboptimal. They were funded as a bet on excess solar power maybe becoming basically free during the day in the future. That is not currently the case.
We can, however, plan on using a lot of methane in the near future.
If we believe in the god of the straight lines and solar trends in terms of cost and deployment continue, isn’t that a reasonable bet in timeframes of 1-2 decades?
So that basically means that requirements for gas heating to be compatible with running on hydrogen are a bad policy because at the point where we could provide synthetic gas we can have methane.
Big ships seem to run reasonably well on natural gas so we don’t need to do anything to switch them except phasing out non-natural gas run ships.
Should we plan to have different ways to power airplanes or is it easier to count on making the kerosine in 2-3 decades synthetically?
What’s your overall view of how the energy transition will go (assuming the AI transition doesn’t overtake event)? What do you think about claims like this one that we don’t have enough mining or mineral reserves to support the transition plans that people are depending on?
I expect continued heavy use of:
US natural gas from fracking
Chinese coal
LNG in Japan
The current trajectory is towards a mix of natural gas + solar + wind, with natural gas balancing unreliable renewables. Better flow batteries might make solar + storage more competitive, but without natural gas, Europe would need some seasonal electricity storage, which is much too expensive.
There are many different minerals; you can’t lump them all together. Short-term, lithium availability is a big problem for batteries, but in terms of reserves I don’t think that’s the main issue. Nickel is also up a lot, which is part of why people are using LiFePO4 more despite the lower specific energy. (Some people have proposed collecting seabed nodules, but there’s a reason they’re called “manganese nodules”—they’re mostly manganese, which is much cheaper. I don’t expect that to be economically competitive with even low-grade nickel ores.)
More copper mining output should be coming up, but I still expect prices to be a lot higher than they used to be, and you need copper for electric motors. The high-concentration ores of copper and nickel are what’s limited; there’s more supply at lower concentrations, but that can be a lot more expensive.
You may find this post interesting.
When I posted this, I actually thought I’d get stuff like
“Do you think [startup]’s technology is practical?”
“What do you think the best approach to [problem] is?”
“Do you know of any notable promising inventions regarding [subfield]?”
because that’s what I’ve gotten by DM elsewhere, but apparently a public post on LessWrong is different.
What do you think about synthetic biology as a manufacturing technology?
post
You have a post about small nanobots being unlikely, but do you have similar opinions about macroscopic nanoassemblers? Non-microscopic ones could have a vacuum and lower temperatures inside, etc.
What can you do with macroscopic nanoassemblers? Usually for a nanostructure to have an effect on human scales, you need a lot of it. If the assemblers are big and expensive, you wont get a lot of it.
Different things have different optimal scales. In practice, lithography machines + molds + 3d printers + etc are better ways to make fine detail than controlling lots of tiny robotic arms. Big robotic arms are more cost-effective than small ones.
Why did Microsoft have so much success with MS-DOS back in the 80s? I feel like they didn’t actually have that much leverage over IBM. Wouldn’t it have been cheaper for IBM to pay some software developers to build their own OS instead of paying all of the royalties to Microsoft for MS-DOS? Enough to eventually make Bill Gates the richest person on earth.
https://www.cnbc.com/2020/08/05/how-bill-gates-mother-influenced-the-success-of-microsoft.html
Interesting. I don’t think that explains everything though. It explains the beginning, but as MS-DOS had more and more success, it became more and more expensive for IBM to pay royalties to Microsoft. I feel like at some point it would become cheaper for IBM to invest in their own OS instead of continuing to pay royalties to Microsoft.
MS used the IBM deal to sell their OS to a bunch of other computer makers. At that point, IBM switching would have meant an incompatible OS.
Meaning software was written that targets MS-DOS so if a computer didn’t have MS-DOS the software wouldn’t work, and so there was a large demand for MS-DOS? Can’t the software vendor just ship a compiled version of their software? Or their software bundled with an assembler for whatever OS?
Thoughts on this paper about an improved electrochemical process to fix nitrogen? (Originally came across this in a youtube video.)
It’s...interesting work, I guess? I’d like to see a replication of that, actually, since their reported Faradaic efficiency from N2 is much higher than previous papers.
If you’re asking about whether it’s economically competitive with the Haber process (which IMO should be named after Le Chatelier instead) the answer is definitely not.
Electrochemical processes are usually more expensive than non-electrical ones, because you need to have electric current across a large surface area, which is harder than catalyst with large surface area. Consider the cost per mol of NaOH, the main thing made by electrolysis now, and then consider that the linked paper is using fancier stuff than salt water / nafion / etc. While the thermodynamically required voltage is lower, this is a harder reaction than water electrolysis because N2 is hard to react, and I’d expect it to be more expensive than hydrogen from water.
Even if the cells were free, I don’t think their electrolyte is stable for long enough to make that competitive.
Ammonia is cheap to make. If you look at historical prices you can see it can be made for $300/ton with current methods; there have just been price spikes due to supply/demand issues.
Thank you for the answer! One of the things I was curious about was indeed economic competitiveness, so it’s good to know that it won’t be any time soon.
Is there some underlying interest you have, that’s more general?
All knowledge is worth having. But in particular, I’m interested in possible replacements for the Haber-Bosch-LeChatelier process. Yes ammonia is cheap, but we also use a heck of a lot of it. Can you imagine any new process ever bringing down the price further?
Also, from the electro-chemistry side, for this particular paper, what actually does the breaking of the triple bond? Electrolysis of water makes sense to me because I know that water molecules fall apart and come back together all the time in liquid water. So if I think about electrolysis as just kind of nudging the ions to each side of the container and then donating/stealing electrons to/from them once they reach the electrodes, it makes sense to me (this may be oversimplified/wrong). Diatomic nitrogen doesn’t spontaneously fall apart as far as I know? For electrochemical processes in general, is there a large voltage drop at the surfaces of the electrodes?
No need to respond to all or any of these. But you did ask. :)
Water molecules can dissociate into [H3O+] and [OH-] - sort of, they’re complexes with more water with partially-covalent hydrogen bonding. But you can’t just go directly from those to H2 and O2 - for good efficiency you need fancy catalysts and it’s a multistep process starting with OH or H bonding to a metal surface.
Nitrogen conversion generally starts with the end or side of N2 bonding to a metal ion. (Usually the end, IIRC.) The triple bond has electron density far enough away from the nuclei for that to happen. It’s more diffuse and bigger electron density than the orbital of N in NH3 that can get protonated, so it interacts better with positive metal ions bigger than H but smaller than molecules.
Do you have any views on the most promising avenues for human intelligence enhancement through biology? I’d be most interested in approaches that would give us (humanity) better odds in worlds where AI takes off in the next 1–15 years.
The most promising approach is using genetic material from smart people to make children, but there are the obvious social issues with that. In general, I suspect that human intelligence is limited partly by increased value drift leading at some point to people having fewer kids.
The 2nd-most promising approach is genetic screening of embryos and embryo selection, but currently that’s only understood well enough for smallish improvements, and if it was common enough I suspect it could lead to problems from reduced genetic diversity.
I was not impressed by the first linked post. Many relevant genes only affect early brain development. Genetic engineering of a fertilized egg cell is much easier, and significantly improving intelligence that way is still not a thing considered practical to do today. The author’s understanding of biology is poor by my standards.
I’m also not optimistic about Neuralink:
Implants for input are worse than vision and displays.
Neuron development for high-bandwidth output requires direct feedback to neurons, which for muscle control comes from proprioceptors. That feedback is in the form of complex patterns of chemicals being released, which a brain implant wouldn’t be able to do even if those patterns were fully understood, which they aren’t.
On a timescale of 1-15 years, if you want smarter people, the best you can hope for is probably better education. I think that AI training has some insights for designing education, and I suppose understanding AI architectures and operation has made me a little bit smarter.
What’s your opinion on load shifting as an alternative to electrical energy storage. (EG:phase change heating/cooling storage for HVAC). I am currently confused why this hasn’t taken off given time of use pricing for electricity (and peak demand charges) offer big incentives. My current best guess is added complexity is a big problem leading to use only in large building HVAC(eg:this sort of thing)
Both in building integrated PCMs(phase change materials) (EG:PCM bags above/integrated in building drop ceilings) and PCMs integrated in the HVAC system (EG:ice storage air conditioning) seem like very good options. Heck, refrigeration unit capacity is still measured in tons (IE:tons ice/day) in some parts of the world which is very suggestive.
Another potential complication for HVAC integrated PCMs is needing a large thermal gradient to use the stored cooling/heating (EG:ice at 0°C to cool buildings to 20°C).
People don’t want to schedule their washing machines / showers / etc around electricity prices. That’s not worth it unless your country is failing.
Using electric car batteries could make some sense, but for many chemistries, battery wear from cycling is worth more than the storage.
Hot water storage for large buildings could make economic sense with variable electricity prices. CenTrio Plant No. 2 makes ice to use for district cooling. Phase change heat storage doesn’t seem economical for houses but it’s not crazy.
With respect to articulated robot progress
Strain wave gearing scales to small dog robot size reasonably(EG:boston dynamics spot) thanks to square cube law but can’t manage human sized robots without pretty horrible tradeoffs(IE:ASIMO and the new Tesla robots walk slowly and have very much sub-human agility).
You might want to update that post to mention improvements in … “digital hydraulics” is one search term I think but essentially hydraulic actuators fed from a single high pressure fluid rail using throttling valves.
Modeling, Identification and Joint Impedance Control of the Atlas Arms US10808736B2:Rotary hydraulic valve My guess is current state of the art (ATLAS) Boston dynamics actuators are rotary vane type actuators with individual pressurization of fluid compartments. Control would use a rotary valve actuated by a small small electric motor. Multiple fluid compartments allow for multiple levels of static force depending on which are pressurized so efficiency is less abysmal. Very similar to hydraulic power steering but with multiple force “steps”.
Rotary actuators are preferred over linear hydraulic cylinders because there’s no fluid volume change during movement so no need for a large low pressure reservoir sized to handle worst case joint extension/retraction volume changes.
Roller screws have high friction?
This seems incorrect to me. The rolling of the individual planet screws means the contact between the planet and (ring/sun) screws is rolling contact. Not perfect rolling but slip depends on the contact patch size and average slip should be zero across a given contact patch. A four point contact ball bearing would be analogous. if the contact patches were infinitesimally small there would be no friction since surface velocities at the contact points would match exactly. Increasing the contact patch size means there’s a linear slip gradient across the contact patch with zero slip somewhere in the middle. Not perfect but much much better than a plain bearing.
For roller screws the ring/planet contact patch operates this way with zero friction for a zero sized contact patch. The sun/planet contact patch will have some slip due to axial velocity mismatch at the contact patch since the sun screw does move axially relative to the planets. Still the most of the friction in a simple leadscrew is eliminated since the circumferential velocitiy at the sun/planet contact point is matched. What’s left is more analogous to the friction in strain wave gearing.
What? No. You can make larger strain wave gears, they’re just expensive & sometimes not made in the right size & often less efficient than planetary + cycloidal gears.
That’s older technology.
No. There’s a reason excavators use cylinders instead of rotary vane actuators.
No. Without sliding, screws do not produce translational movement.
Not in the sense of you can’t make them bigger but square cube means greater torque density is required for larger robots. Hydraulic motors and cylinders have pretty absurd specific force/torque values.
Yes you can use servomotors+fixed displacement pumps or a single prime mover + ganged variable displacement pumps but this has downsides. Abysmal efficiency of the a naive (single force step actuator+throttling) can be improved by using ≥2 actuating cavities and increasing actuator force in increments (see:US10808736B2:Rotary hydraulic valve)
The other advantage is plumbing, You can run a single set of high/low pressure lines throughout the robot. Current construction machinery using a single rail system are worst of both worlds since they use a central valve block (two hoses per cylinder) and have abysmal efficiency. Rotary hydraulic couplings make things worse still.
Consider a saner world where equipment was built with solenoid valves integrated into cylinders. Switching to ganged variable displacement pumps then has a much higher cost since each joint now requires running 2 additional lines.
Agreed in that a hydraulic cylinder is the best structural shape to use for an actuator. My guess is when building limbs, integration concerns trumped this. (Bearings+Rotary vane actuator+control valve+valve motor) can be a single very dense package. That and not needing a big reservoir to handle volume change meant the extra steel/titanium was worth it.
This is true, the sun and planet screws have relative axial motion at their point of contact, Circumferential velocities are matched though so friction is much less than in a leadscrew. Consider two leadscrews with the same pitch (axial distance traveled per turn). One screw has twice the diameter of the first. The larger screw will have a similar normal force and so similar friction, but sliding at the threads will be roughly twice that of the smaller screw. Put another way, fine pitch screws have lower efficiency.
For a leadscrew, the motion vectors for a screw/nut contact patch are mismatched axially (the screw moves axially as it turns) and circumferentially (the screw thread surface slides circumferentially past the nut thread surface). In a roller screw only the axial motion component is mismatched and the circumferential components are more or less completely matched. The size of the contact patches is not zero of course but they are small enough that circumferential/radial relative motion across the patch is quite small (similar to the ball bearing case).
Consider what would happen if you locked the planet screws in place. it still works as a screw although the effective pitch might change a bit but now the contact between sun and planet screw involves a lot more sliding.
No. Strain wave gears are lighter than using hydraulics.
No. Screw translational movement is proportional to the amount of sliding.
Note:I’m taking the outside view here and assuming Boston dynamics went with hydraulics out of necessity.
I’d imagine the problem isn’t just the gearing but the gearing + a servomotor for each joint. Hydraulics still retain an advantage so long as the integrated hydraulic joint is lighter than an equivalent electric one.
Maybe in the longer term absurd reduction ratios can fix this to cut motor mass? Still, there’s plenty of room to scale hydraulics to higher pressures.
The small electric dog sized robots can jump. The human sized robots and exoskeletons (EG:sarcos Guardian XO) aren’t doing that. improved motor power density could help there but I suspect the benefits of having all power from a single pump available to distribute to joint motors at need is substantial.
Also, there’s no power cost to static force. Atlas can stand in place all day (assuming it’s passively stable and not disturbed) an equivalent robot with electric motor powered joints pays for every Nm of torque when static.
jumping: https://www.youtube.com/watch?v=tF4DML7FIWk
Sorry, I should have clarified I meant robots with per joint electric motors + reduction gearing. almost all of Atlas’ joints aside from a few near the wrists are hydraulic which I suspect is key to agility at human scale.
Inside the lab: How does Atlas work?(T=120s)
Here’s the knee joint springing a leak. Note the two jets of fluid. Strong suspicion this indicates small fluid reservoir size.
I’m getting the impression you didn’t read what I wrote.
Hopefully it helps to get back to the source material Articulated Robot Progress
I apologize if I’m missing anything.
I would argue that the current Atlas robot looks a lot more like the earlier hardiman robots than it does a modern factory robot arm. The hydraulic actuators are more sophisticated (efficient) and the control system actually works but that’s it.
Contrast the six axis arm which has a servomotor+gearing per axis. Aside from using a BLDC motor to drive the pump, and small ones for the control valves, Atlas is almost purely hydraulic. If the Hardiman Engineers were around today Atlas seems like a logical successor.
Perhaps you think Atlas is using one motor per joint (It would be hard to fit 24 in the torso) or ganged variable displacement pumps in which case there would be more similarities. IMO there aren’t enough hydraulic lines for that. Still of the 28 joints in atlas only 4 are what you’d find in a conventional robot arm (the ones closest to the wrist)
Predictively Adjustable Hydraulic Pressure Rails
The patents coming out of BDI suggest they’re not doing that and this is closer to Hardiman than it is a modern factory robot arm.
Again, strain wave gearing (as an approach, including electric motors with high specific power) is lighter than using hydraulics, overall. The same is true for planetary roller screws. This is true regardless of scale. Hydraulics are used for other reasons than maximum performance physically achievable.
Boston Dynamics decreased the weight of their hydraulics system by 3d printing hydraulic channels in the skeleton. That’s expensive, and planetary roller screws are still better if done properly.
Take an existing screw design, double the diameter without changing the pitch. The threads now slide about twice as far (linear distance around the screw) per turn for the same amount of travel. The efficiency is now around half it’s previous value.
https://www.pbclinear.com/Blog/2018/February/What-is-Lead-Screw-Efficiency-in-Linear-Motion
There was a neat DIY linear drive system I saw many years back where an oversized nut was placed inside a ball bearing so it was free to rotate. The nut had the same thread pitch as the driving screw. The screw was held off center so the screw and nut threads were in rolling contact. Each turn of the screw caused <1 turn of the nut resulting in some axial movement.
Consider the same thing but with a nut of pitch zero (IE:machined v grooves instead of threads). This has the same effect as a conventional lead screw nut but the contact is mostly rolling. If the “nut” is then fixed in place you get sliding contact with much more friction.
Perhaps we don’t disagree at all.
a roller screws advantage is having the efficiency of a multi-start optimal lead-screw but with much higher reduction.
A lead-screw with an optimal pitch and a high helix angle (EG: multi-start lead-screw with helix angles in the 30°-45° range) will have just as high an efficiency as a good roller screw (EG:80-90%). The downside is much lower reduction ratio of turns/distance.
We might be talking past each other since I interpreted “a planetary roller screw also must have as much sliding as a lead-screw” to mean an equivalent lead-screw with the same pitch.
No. Did you read my post?
Just to clarify, my above suggestion that roller screws and optimal low reduction lead-screws are the equivalent (lubrication concerns aside) is correct or incorrect?
Are you saying a roller screw with high reduction gets its efficiency from better lubrication only and would otherwise be equivalent to a lead screw with the same effective pitch/turn? If that’s the case I’d disagree. And this was my reason for raising that point initially.
Yes. OK, we disagree about that. Glad we could get to the point.
What is your opinion on general robotics (driven by an early form of AGI) and then robotics self manufacturing?
I have personally thought that using a technique that is a straightforward expansion of LLM training, but you train on human manipulation from all available video, would give you a foundation model. Then you would add rl fine tuning and millions of simulated years of RL training to the foundation model to develop excellent and general robot performance.
Do you think this isn’t that straightforward or near term?
I ask because if robots can go from their narrow and limited roles now to performing most tasks if given information like (an example or a prompt or a goal schematic) it would change everything.
It would trivialize energy transitions for one. And deep mining and ocean mining and so on. (Because you task some robots with building and assembling others, some with building and deploying energy collection, some with mining, and are now rate limited not really by money directly but by materials, regulations, energy, or mining rights)
Your opinions must rest on an assumption that this problem cannot be solved anytime soon. How confident are you in this belief? What are the obstacles you believe are limiting in developing such a robotics technique?
Past efforts failed but didn’t have sufficient compute, and a multimodal technique as described would need all the compute for gpt-4 plus all the compute for image processing, as well as much more training and quality checks during inference. So essentially the technique has never once been attempted at scale.