HN Theater @HNTheaterMonth

⬐ jcims

Unfortunately the explanation of the A/C fluid cycle is haphazard and the description of how forces relate to crankshaft position is completely wrong. The crankshaft actually applies maximum force to the piston while approaching top and bottom dead center, when the mechanical advantage is greatest, and the least at mid stroke, where the torque of the motor is most directly opposed by the pressure of the gas against the face of the piston. This is good, however, because it marries well to the available power through an alternating current system.

The design described does seem to have fewer moving parts, which is good. But the advantage of linear vs rotary motion with respect to forces seems sketchy. Field strength in the linear system is likely to vary across the travel and would still need to be synchronized to the alternating current to prevent motor stalls. (Edit 50 of this comment: this effect is actually illustrated directly in the video here: https://youtu.be/2TFiL5BM3ss?t=348 Imagine the field lines directly correlate with force applied, and mentally plot the quantity of field lines over time as you watch the animation. Turns out it looks precisely like the force chart of a rotary compressor.)

The primary advantage of this pump, to me, seems to be reduced weight/complexity and lower friction losses. For it to operate with low starting load it would likely need to essentially vibrate its way to full travel (confirmed at 7:47 in the video), something rotary compressors can’t really do. If you try to restart rotary compressors too quickly after stopping, the high head pressure will stall the motor. This causes it to effectively be a dead short minus winding resistance, pulling much more current than average (known as LRA or locked rotor amps) with no cooling. This will heat the motor quickly until it thermally trips or burns up. This is why you will frequently see restart delays in A/C systems.

For what its worth, scroll compressors have been around forever and share some of these advantages, but aren’t dominant in the industry because of the additional complexity in manufacturing. They have an additional advantage of somewhat continous output pressure, which *significantly* reduces ambient noise and vibration. They also have fewer moving parts and are generally more reliable.

All of these advantages convey to standard air compressors by the way, something I'm a little surprised they didn't mention.

⬐ phnofive

You sound like you have expertise, so could you have a look at his patents? I couldn't make much of them other than they looked familiar in some aspects and impractical/impossible in others:

https://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=...

I assume that this is pretty common in trying to improve linear motors - I am trying to get my hands on a small one to drive an air pump and it's just inscrutable to me.

⬐ jcims

I just have experience rather than expertise, hoping someone with actual expertise shows up to correct me where I'm wrong. :)

I love the idea of low hanging fruit in every dimension of technology, but the reality is that linear pumps have been around for quite some time and aren't dominant in any high (sales) volume application that I'm familiar with. That tells me there are likely some tradeoffs here that are not being explained. Vibration would be one obvious area, but could be handled with a harmonic balancer or an additional pump operating 180 out of phase (unfortunately increasing cost/complexity).

The other potential issue is that the motor needs to be built for the application for the most part. With rotary compressors, you can incorporate pulleys to gear the system up or down in accordance with the cheapest motor you can put to use. With this system the motor is the pump, so it's harder to add some kind of lever to gear it up or down. (This is less applicable to small A/C systems which tend to use hermetically sealed compressor systems, but for larger systems this can come into play.)

⬐ ironmagma

The cost is likely to be less important as time goes on and consumers become more aware (and caring) of the external cost to having inefficient machines.

⬐ jcims

Not to be trite, but cost is always important to the manufacturer. Tesla, for example, has been able to build a vertically integrated company through sheer will of its leadership. However, this is highly unusual and most product companies are going to want to invent as little as possible in order to ship a product. In this case a company wanting to sell an HVAC system based on this tech would have to develop the manufacturing pipeline to build these compressors vs. buying them from Copeland.

Now Copeland may choose to build these, but they are going to have to build the motor vs. buying them from one of hundreds of companies that build AC motors to spec.

Now one of those companies might decide to build these linear motors, but they are going to have to license the IP to do so, putting them at a distinct cost disadvantage vs. standard rotary motors. They are also going to be a sole supplier for some time, something manufacturers are very wary of unless it's a major differentiator. These are likely to cancel out.

⬐ gpm

Since you mention them - I wonder if Tesla might actually be an ideal customer.

As you say, they are vertically integrated, they are well set up to produce the full stack of the technology. They also need compressors (for air conditioning, and at least in their trucks for other systems), that need to be compact and energy efficient. The technologies at play (pressurized fluids, eletric motors and magnets) are all very familiar to them.

⬐ toomuchtodo

They have their own patent for their highly efficient octovalve heat pump in the Model Y.

https://cleantechnica.com/2020/08/03/teslas-octovalve-enable...

> In the Tesla Model Y, the heat pump was a true engineering marvel. Elon admitted on the podcast that he has a trophy rack of impressive tech his teams have come up with in his bedroom at home. The new printed circuit board (PCB) for the Tesla Octovalve was so impressive to Elon that he added it to his bedroom trophy rack.

⬐ gpm

You got me curious. It looks like their innovation here is in fluid routing, and they're just using a standard compressor: https://jalopnik.com/the-tesla-model-ys-octovalve-and-coolin...

It's probably pretty easy to swap out the compressor for a new design, fundamentally all compressors do the same thing.

⬐ Danieru

> linear pumps have been around for quite some time and aren't dominant in any high

Every house in Japan with a small scale septic system has a linear pump for injecting air. This may sound niche but in fact large parts of the tokyo metro area are not on centralized sewer systems. Thus there should be at least 10M+ houses with such motors running continuously in Japan alone.

The pumps run 24/7 without maintenance for well over a decade.

Which also suggests there are bigger reasons not to use such pumps in AC. As any Japanese manufacturer will know of such pumps.

⬐ leviathant

>Which also suggests there are bigger reasons not to use such pumps in AC. As any Japanese manufacturer will know of such pumps.

Yup! This reminds me of a similar pitch from a few years ago. When a Kickstarter went live for a small, quiet window air unit called Noria, people signed up in droves, thinking a startup in Philadelphia had somehow solved a physics problem that Johnson Controls and the companies it's sucked up over the years somehow whiffed on. This new device would chill the whole room, it would be quiet, compact, and most importantly, it wouldn't be heavy. It would easily slot into your window, and store under your bed in the cooler months when you don't need it. They raised nearly $1.5M USD in 2016, and in the last year finally started to deliver window units that were ... big, heavy, and noisy.

The physics of heat exchangers are very well established. It's wise to take any wild claims of higher efficiency with a massive grain of salt - someone missed something in their equations and it's going to come out during manufacturing.

⬐ danans

> If you try to restart rotary compressors too quickly after stopping, the high head pressure will stall the motor. This causes it to effectively be a dead short minus winding resistance, pulling much more current than average (known as LRA or locked rotor amps)

Inverter based heat pumps and ACs that are already available on the market today have low LRAs as a consequence of their ability to operate at continuously variable speeds.

⬐ y04nn

Thank for the explanation, what about screw compressor? Are they too expansive to manufacture and need too much maintenance? How would you compare them to scroll compressor? I know they have industrial use, but I'm not sure why they are not more widespread.

⬐ jcims

Screw compressors are more well suited for high volume applications (like superchargers) than high pressure. This is because the geometry of the screws is difficult to sustain with very high tolerances. Small leaks aren't a problem when you're talking 10-50psi, but the 'high side' pressure of an A/C system is typically over 200psi.

⬐ mech422

My Dad made the screws for screw compressors back in the 70-90s... He primarily worked for joy/lovejoy, but also made screws for ingersall and a few others.

It was amazing to me how small the screws were for say, a shopping mall unit.. and how PRECISE the machining was.

The math behind the profiles was insane to, it was a 'self generating' curve.. You calculated a point on the curve, and that was fed in to generate the next point. Generating a new profile shape was a HUGE process... Wonder what it would be like with today's computers :-P

⬐ jcims

That's awesome! Seeing how wholly dependent modern manufacturing is on CNC makes me respect those early machine operators even more. They were constantly innovating on how to apply physical geometry and minimize incremental errors in order to build parts with tolerances that would still hold up today.

⬐ mech422

My Dad used to do a 'trick' where he would zero out a 6 decimal place readout on the big ground glass comparator..

You could then move the table anywhere you wanted in X/Y, cover the display, and he would bring it back to 0.0000X by eye :-)

He actually went to some institute in Sweden that did ultra-accurate measurements to show them how he did it. I think the 'trick' was, he moved the shadow till it was just past the blueprint, and back it off till a sliver of light shone thru, rather then trying to 'walk' the shadow up to the blueprint..

Neat stuff - but I decided on a tech career when I was mucking out the Holyroyd mill and he turned the cutting pumps on :-P Being drenched in cutting oil convinced me I liked computers better :-P

⬐ jhallenworld

I'm worried about vibration. Maybe they should have two masses operating in opposite directions at once to balance out the vibrations. Maybe they will license my idea..

⬐ flavius29663

wouldn't that effectively double the cost of production? It might be cheaper and easier to invest in some spring based dampening. Anyway, even in their video, the vibration is out of control, moving the pipes around and making it look like it won't last too much in real life applications.

⬐ jcims

You could probably do some kind of damped harmonic balancer with a simple mass on the end of a damped spring tuned for the expected operating range. Still increases cost but not as much.

⬐ jhallenworld

Yup, "force is lowest at TDC" is completely wrong. Mechanical advantage is highest at this point..

A question I have is how much does it vibrate externally? Maybe they should have two moving masses running in opposite directions to balance it.. I should patent this, maybe they will license my idea.

>scroll compressors have been around forever and share some of these advantages, but aren’t dominant in the industry because of the additional complexity in manufacturing

Guess the easiest shape to manufacture to high precision for a close seal? Anything else is not worth considering unless there is some other quality which outweighs this issue / cost. So for example Wankel engines burn oil because they are not using the simplest shape.

⬐ rozap

VW mass produced scroll superchargers 30 years ago for the G60 and they were unreliable but super efficient. Kind of surprising that with 30 years of manufacturing advances they're not more common. Such a cool design.

⬐ ip26

Modern turbos more or less killed off superchargers. Largely thanks to advances in computer fuel injection managing all of the tricky parts of turbos.

(Superchargers are simpler, but cost ~10% engine power to drive)

⬐ _hyn3

Superchargers are far from dead. Most of the highest performance American muscle cars all derive their higher horsepower (650 to 797HP) with large superchargers.

⬐ ants_a

The comparison to a turbo got me thinking if some of the compression energy could be recovered in the expansion stage. Turns out the answer is yes, but the energy recovery is limited to 7% at best: https://www.sciencedirect.com/science/article/pii/S187661021...

⬐ Confiks

Do note that Magtor is currently only claiming to be able to compress air, not air conditioning fluids / refrigerants. They add that "[a]lthough air compression is a first milestone for Magtor, the medium-term goal is to bring Magtor efficiencies to climate control engineering allowing products such as fridges, heat pumps and air conditioning units to become 30+% more efficient" [1].

It's hard to verify more on the design. They offer a lot of explanation and tests, but no independent review is given. Main question on YouTube about scrolling compressors (used in newer fridges) is anwered a bit vaguely [2].

The patent portfolio page is very grandiose [3] but doesn't care to link anything. A quick search doesn't turn up much to do with air or fluid compression, only "Magnet device comprising stators and translators" [4] and similar, which don't mention compression at all.

[1] https://www.magtor.tech/who-we-are/

[2] https://www.youtube.com/watch?v=2TFiL5BM3ss&lc=UgxHEJkjlZ9fZ...

[3] https://www.magtor.tech/intellectual-property/

[4] https://patents.google.com/patent/US10943721B2

⬐ nashashmi

How can something like a rapid compressor generate SO much heat AND transfer it so quickly. Isn’t there some kind of time delay in both generation and transfer?

⬐ robocat

They appear not to be developing the compressors themselves, but instead are trying to license their IP.

From their ‘Who are we’ page[1], their Director of Production “oversees the production of the Magtopressor Demonstrators”, so not really production per se.

The focus on worldwide patents and number of patents in the videos is also noticeable.

[1] https://www.magtor.tech/who-we-are/#our-key-personnel

⬐ poorjohnmacafee

FYI this is an advert. Not just in resemblance to an infomercial but the youtube presenter is reaching out and quoting CTO of the company as responses to youtuber questions in comments.

⬐ fastball

Fairly certain YouTube requires you to disclose paid-for adverts now.

⬐ elil17

Fairly certain humans have been breaking rules almost as long as we’ve been making them.

⬐ fastball

Sure, but it would be very unlikely in this particular case, given that YT is very trigger happy when it comes to violations and YT is the main source of his livelihood.

⬐ filleokus

I regularly watch this channel and I honestly believe his claims of independence (no revenue from any kind of sponsorships or adverts, just from Patreons). I think he simply want to be helpful for his audience and forward questions/responses if he has the chance to do so.

Overall I like his content, he is usually fairly technical and realistic about the prospect of things he cover. He has also has done some deep dives in e.g how to calculate levelized cost of energy to help analyse/compare different new inovations he showcase on the channel.

But sometimes I guess it can be slightly too pop-sciencey, reminding me (both in a good and bad way) of the glossy science magazines of my youth. Extrapolating barley lab bench-viable results to full scale production, with very optimistic timelines etc.

⬐ mda

Even if this is not an advert, he could have done better research and put a more balanced view. Are there prior work similar to this? What are the risks for the companies approach? Will it really provide 30% better efficiency? In current form this indeed looks like an advert, whether intentional or not.

⬐ atoav

IMO the most important thing to make air conditioners more efficient is to design, build and modify buildings in such a way they do the heavy lifting passively. This means also we need to avoid building idiotic things like sun-facing glas fronts etc.

We have to stop pretending we are going to save the day by making AC units (or any other thermal machine) a few percentage points more efficient — because this is peanuts and nowhere near the amount of efficiency needed to grant humanities survival on earth. The best AC unit is the one that you don't need.

⬐ dzhiurgis

It's probably cheaper to put AC + solar panels in each home than rebuild all of them...

⬐ atoav

Depends what costs you factor in down the line. E.g. I remember reading a study about Paris, that looked at the impact widespread AC use would have on the ambient temperature of the city. The optimistic guess was a rise of about 5°.

Where I live (Central Europe) air conditions are the total exception. Adding AC to flats is rarely ever done, thermal insulation much, much more often, even in old buildings. Maybe this is because the winters also get cold so thermal insulation has the extra pro of reducing the heating bill.

⬐ dzhiurgis

AC is somewhat cheap of source of heat in winter too, provided your system is scaled enough.

⬐ KozmoNau7

Exactly! We need more efficiency, and heat pumps/air conditioners are already very efficient at what they do. Of course further efficiency improvements are welcome, but they are small gains.

We need to look at where big gains can be made, and improving the thermal efficiency and air circulation of buildings holds a lot of promise. You don't have to go full passivhaus, even just improved insulation, double/triple-layer glass, sun shades and well-placed foliage to provide shade can net much larger improvements than some tweaks to AC design.

⬐ MauranKilom

Isn't the video inaccurate about the force of current pumps near maximum compression?

The power put into the system by the motor is constant throughout the entire revolution. It follows that, since the piston moves slower near maximum compression, force increases (since power = force * speed). This is also intuitively the case: You need much more force to push back the piston (against a constant motor torque) near its apex because the moment arm of the crank shaft gets smaller and smaller.

Yes, at the maximum compression point the force is not well-defined. In a sense it is infinite (you could not actually turn the rotor by pushing on the piston). But certainly not zero.

I'm struggling to understand what the creator might have meant instead. It's not work or power (which, again, is constant). The rotary motion works, if anything, in favor of the system, as the force gets larger with compression (past the midway point). Is there some quantity I'm missing? Maybe something to do with gas laws?

I'm also skeptical about the illustration with the additional magnets to contain the magnetic field lines. As drawn, that would not work for AC as the magnetic field would be the wrong way around half the time. Maybe they just use extra iron to shape the field - but that is extremely far from being innovative and more like electrical engineering 101.

It could well be that I'm missing something, but overall this video had multiple strange explanations that lower my confidence in the overall accuracy of the reporting...

⬐ tuatoru

I don't get it. This is just a solenoid. Solenoids have high "start-up" (saturation) currents.

Using the inertia of the plate mass at the end of each stroke is clever though.

I'd have thought a rotary electric motor connected to a rotary compressor would be quieter and easier to design the bearings for, and would operate in a more steady-state regime, so be more reliable. You also don't need the valve where the fluid from the two sides meets up, avoiding that efficiency loss.

I think the latest fridges (which are also air conditioners) use rotary compression.

Modern fridges and industrial electric motors have had "soft start" circuits in them for at least a decade. The start-up current thing is a non-issue.

⬐ flavius29663

Existing fridge compressors are ridiculously reliable. They are a close circuit, self lubricating motors. My grandma's fridge from the 70s was still running strong when we threw it out a few years ago.

I didn't quite understood how that new design is more reliable or serviceable? The closed compressors are literally zero maintenance. Also, the vibrations of that thing will tear itself apart quite fast, just look at their own promotional video: the pipes themselves cannot possibly last long with such vibrations: https://youtu.be/2TFiL5BM3ss?t=471

⬐ gpm

This comment by the author on youtube is worth reading:

Attempt at direct linking: https://www.youtube.com/watch?v=2TFiL5BM3ss&lc=UgzdQZ_jgrbde...

--- Context (post by Ed Williams)

Most residential (and many commercial) AC units, refrigerators, heat pump and such use scroll compressors. Yes, they have a startup in rush, but I'd be interested in seeing efficiency of this tech versus current scroll compressors.

--- Post by "Just Have a Think" (the creator)

Response from Magtor CTO : There are many types of compressor in use and the industry somewhat specialises within particular technologies based on the power requirements, type of applications, manufacturing costs, end product costs, etc.

Regarding efficiency of scroll vs reciprocating compressors, not all sources agree but the trend is that scroll compressors are at least slightly more efficient, based on operating conditions…

There is this interesting white paper from Schneider Electric, “The different types of cooling compressors” https://download.schneider-electric.com/files?p_Doc_Ref=SPD_... and I also found this study: “Comparison of hermetic scroll and reciprocating compressors operating under varying refrigerant charge and load” at https://docs.lib.purdue.edu/cgi/viewcontent.cgi?referer=http...

⬐ re

The meat about the new technology starts at about 4:30 into the video, after some background on the history of A/C and how it works.

> Most modern compressors and pumps use rotary motors to shove a fluid from A to B inside a chamber, but rotary motors can lose as much as 30 percent of their output when they're asked to convert their rotary motion into linear motion required by compressors and pumps. That conversion is done by a crankshaft and once a crankshaft is in the system you get some friction losses but more importantly you get diminished efficiency because you're only getting maximum shove force in the middle of the piston stroke. The start and end of the stroke are both dead points where zero force is available to do any useful work. In a compression task you really want the maximum force to be applied at the end of the stroke not the minimum force. Overcoming these inertia forces every time the motor starts up requires a big spike in power demand which can be as much as five times the motor's normal operating consumption.

> The technology that Magtor have been quietly developing for over a decade now is a magnetically driven linear motor that completely eliminates the need for a crankshaft and achieves far higher operating efficiency as a result. Here's how it works. A fixed electromagnet stator sits at the centre of the motor. On either side of that are magnetic plates that are connected to each other via a shaft that runs through the middle of the stator. That connection effectively turns the two magnetic plates into a single moving part. As a voltage is applied across the electromagnet its polarity attracts the magnetic plate at one end and repels the magnetic plate at the other end. But because it's an electromagnet its polarity is changing 50 or 60 times a second as a result of the alternating current flowing through the copper windings. [...]

> You also avoid that costly power spike at start up as well. Because there's negligible mechanical inertia to overcome, as soon as an electrical current is applied the movement caused by magnetic attraction and repulsion can get going instantaneously. And on top of all that, Magtor have developed a very smart additional element that optimizes the motive power provided by the magnetic flux of the system. [...] In this working example a magtopressor is connected to a five-litre tank of air. Magtor's tests have shown that their device, which is smaller and lighter than a traditional reciprocating piston air compressor, can pressurize a cylinder 38% faster while consuming 33% percent less electricity.

(emphasis added)

⬐ handol

> That conversion is done by a crankshaft and once a crankshaft is in the system ... you get diminished efficiency because you're only getting maximum shove force in the middle of the piston stroke.

I don't think that's right. Why would diminished force when the piston is turning around lead to a power loss?

⬐ ars

Isn't that the point of a https://en.wikipedia.org/wiki/Scroll_compressor ?

They run in a continuous way rather than with a piston. And my understanding is that most A/C compressors use them.

⬐ visviva

Thank you, I was never going to watch a video to get the answer to the title question.

⬐ makomk

I'm pretty sure some of the more high-end fridge compressors used by companies like LG already work more or less like this, with marginal efficiency gains but some improvement in noise.

⬐ davelondon

Yes, look from 2014: https://www.youtube.com/watch?v=r2HDJjrQiEQ

⬐ beervirus

And some disadvantages in reliability, if the internet is to be believed.

⬐ None

None

⬐ xxpor

Yep, according to https://en.m.wikipedia.org/wiki/Linear_compressor

⬐ fake-name

How did this get a patent? This is just an linear piston pump [1][2].

You can literally buy fish tank air pumps that use this exact mechanism [3].

The only thing mildly novel here is that they're driving it both directions via magnets & more complex drive electronics. Most of the commercial versions just use a steel piston, and single-direction actuation because they can then run the electromagnet directly off mains AC to get the cycling action.

1: https://www.techmagpm.com/en/linear-piston/ 2: https://www.nitto-kohki.eu/en/operating-principles-en/linear... 3: https://www.jehmco.com/html/central_air_pumps.html

⬐ everybodyknows

This is a variation on a long-established linear-motor compressor design, for sale in these products: https://engelcoolers.com/collections/powered-fridge-freezers

Google "Sawafuji Swing Compressor".

⬐ zeristor

Of course they key thing is to get data from commercially available systems, but a 33% saving could be huge deal.

For air conditioning, fridges, ground source heat pumps.

https://www.magtor.tech/

⬐ cogman10

Cutting the cold crank power requirement to 0 is also pretty nice. It means ACs with this thing would play nicely with solar/battery systems as they wouldn't require big current spikes to start.

⬐ zeristor

Yes, it’s mentioned that that can peak at five times the running load to get it started.

⬐ GekkePrutser

Even more. My simple 100W fridge shows peaks up to 3000W when it switches on. Very short but sometimes my Shelly EM catches them (and messes up my graphs :) ). I can also see the lights dim a bit when that happens.

Even if it works I'm not going to buy a new fridge just for that though. Manufacturing a whole new fridge will be worse for the environment than running this for many years especially on green power.

However when it does go, it would be nice to have.

⬐ timonoko

Apropos: Helsinki has never experienced heatwave longer than 5 days (above 25°C). Last summer it was phucking 45 days.

However the bottom temperature of the near-by Baltic Sea is -4°C. I know this because I was hunting for halibut.

There is already a separate system which refrigerates government offices from the Baltic Sea.

I would suggest is to use the system to cool all homes by using the central heating system. All you need is lower is temperature to 15°C in those pipes.

⬐ ajnin

What would be the impact of all that heat on sea life and fishing activities ? Also, as we tend to learn the hard way, natural resources are finite. How big of a heatsink is the bottom of the Baltic sea ? For how long could you cool all of Helsinki's homes per, say 1˚C temperature increase ?

⬐ timonoko

No worry. All life in the the deeps is all ready gone. This is because of lack of oxygen and some chemicals, like phosphors mostly from from agriculture.

⬐ supperburg

This is unmitigated bullshit and probably sponsored by the vapor ware company that is making this thing. Just add more insulation and some solar panels.

⬐ bin_bash

I’m surprised this wasn’t about Exergyn which is far more interesting tech: https://www.economist.com/science-and-technology/2021/07/21/...

⬐ Abishek_Muthian

I too thought the video was about Exergyn, I was pleasantly surprised when I read the Economist coverage on them. It's not like we come across breakthrough technology which is cheaper, efficient and planet friendly.

It is supposedly 30-40% cheaper than current ACs and being solid-state stack it's definitely more efficient(Not sure how it compares with the calimed efficiency of Magtor).

⬐ londons_explore

For efficient AC and heat pump systems, we need to invent a compressor which is close-to isothermal rather than close-to adiabatic.

And we need to switch to pressure change rather than phase change systems.

That pretty much means the compressor needs to be built into the 'condensor'.

The closest we get to today is two stage compressors - where you compress a bit, cool the gas, then compress some more. But we need to get to 50 stage compressors - where we compress slightly, then cool, and repeat.

Also, we will need to switch out the pressure regulating orifice for a turboexpander to recuperate the energy which would otherwise be lost letting something high pressure become low pressure.

All these add complexity and cost, but ultimately will be needed to make efficient enough systems when energy is more expensive in a carbon-free world.

⬐ toxik

Interesting nerdsnipe. Tell people (incorrectly) how an AC system works, then segue into a full-blown ad piece for some "new" tech.

⬐ _nalply

Additionally air conditioners should transfer heat to tap water, not outside air. Water as a medium to discard heat is more efficient.

There are downsides to this: wasted tap water and hot sewers. Probably certain municipalities should disallow such air conditioner water heaters except to make hot water for showering and bathing.

Also see: https://www.hotspotenergy.com/air-conditioner-water-heaters/...

⬐ baybal2

LG been making linear piston compressors for mass market for more than a decade.

The true breakthrough here will happen with solid state gas compressors.

Solid state compressors already approach, or pass 50% efficiency — something only possible in very large, staged, centrifugal compressors for large AC systems. A solid state compressor, unlike that, will be equally efficient regardless the size.

⬐ saltedonion

So what is the approximate overall efficiency of this system now? Is there still viable room for improvement after this 33% increase?

⬐ adflux

I have a single hose AC (renting right now). did some research on how to improve the efficiency of the unit. I ran into this report by the australian government:

https://www.energyrating.gov.au/sites/default/files/2020-06/...

Adding a second hose to a single hose AC can improve effiency by around 100%... This is because portable air conditioners are tested differently to other air conditioners.

"The DoE found the addition of a second duct that uses outdoor air to cool the condenser would yield an efficiency gain of 103 per cent for a retail price increase of around US$8 (from US$534 to US$542)—a price/efficiency ratio of 1:103 (i.e. a one per cent increase in price provides a 103 per cent increase in efficiency). Because double duct portables are no longer available on the Australian and New Zealand markets, suppliers are unable to communicate their benefits compared with single duct products. The existing energy efficiency regulations on portable air conditioners appear to have had the perverse effect of removing the more energy efficient product (double duct portables) from the market, while encouraging the marketing of single duct portables, by allowing unverified claims to be made about the effectiveness and efficiency of these products. This RIS considers options to resolve this perverse situatio"

"The single duct portable’s performance information is likely to have been obtained from a test that measured the cold air produced. Fixed air conditioners, conversely, are tested to measure their cooling effect on an enclosed space. While a single duct portable may blow cold air onto a person, if its performance was tested in the same way as a fixed air conditioner, it could have a net heating effect on the room it was supposed to cool."

⬐ Gravityloss

One point in the presentation is not correct. This is instead correct: the maximum force occurs at minimum speed at the top and bottom of the stroke. The minimum force and maximum speed of piston occur mid-stroke.

https://youtu.be/2TFiL5BM3ss?t=284

⬐ LeifCarrotson

You have to distinguish between the available force that the constant-torque motor is able to produce and the actual force that is exerted on the working fluid by the piston.

Yes, the crankshaft confers the greatest mechanical advantage at the top and bottom of the stroke; that's an excellent observation. The position vs time plot of a crankshaft follows a sine wave, the velocity (as a derivative of position) follows a cosine, and the acceleration available from a constant input torque follows the derivative of the position, again a sine wave. However, this available force does not meant that the exerted force at the top of the stroke is high, as it needs a reactive force to work against. This force is mostly provided by the fluid pressure, rather than by a velocity-dependent force like friction.

At the top of stroke, the gas is not under significant compression, so it takes minimal force to move the piston. Near the bottom of the stroke, when the gas is maximally compressed, there is maximum force on the piston. When the pressure exceeds the output of the compressor, the air begins to exit through the check valve. Another complicating factor is that there are typically two pistons on the same crankshaft, so as the piston returns the gas remaining in the cylinder is working with the motor. The final and most complicated part of the problem is that the whole point is that the gas being compressed is not an ideal Hooke's law spring where PV = constant, it's undergoing adibatic compression, where P x V x gamma = constant, where gamma is the specific heat of the refrigerant, and the temperature changes through the compression.

You can improve your intuitive understanding of these relationships with an ordinary bike pump and the bike's pedals. When the pump is fully up, there's little resistance, you can feel the force increase when it's almost at the bottom, and you can feel and hear the air going into the tire for a short length of the stroke at the bottom. Likewise, pushing vertically on the pedals (avoid pushing forward or pawing backwards) does the same thing as a crankshaft.

⬐ jedberg

Is this just an ad for a company trying to patent their technology?

⬐ woleium

Heh, title should be "A non-revolutionary solution"

⬐ elil17

Clever design, perhaps. However, it only applies to small, residential AC. Commercial systems use more efficient designs already, such as screw compressors. Also, good luck using this with a turbine - many of the largest compressors directly driven by gas or steam turbines, which are way more efficient than electric motors because they eliminate inefficiencies associated with electric generation and distribution.

⬐ sandreas

Well, the idea behind this solves ONE problem of the whole process... here is an alternative, that rethinks the whole process and already is used in many german projects: https://fahrenheit.cool/en/

⬐ TaylorAlexander

If anyone is curious I built some brushless linear motors in 2011 and I have recently put some relevant code and background docs on GitHub.

https://github.com/tlalexander/open_linear_motor

⬐ smnscu

The fact that the video wasn't from Technology Connections was both surprising and disappointing. Great share though! I wish ACs and heat pumps were more common in the UK, though I understand why they aren't (yet! I expect an AC installing craze in the next couple of years).

⬐ cryptonector

There's an internal combustion engine that works on roughly the opposite as this magtor compressor: a pair of opposing pistons moving the connecting rod linearly to drive a linear electric generator. I can't remember the name of it.

⬐ xondono

I don't understand what is innovative or new about this, and I'm highly dubious of the quoted numbers.

"33% more efficient" is repeated in several company documents and videos, but I can't find the baseline for that measurement.

⬐ iamgopal

Years ago, I did try to built on similar concept but instead using reluctance for further reducing hysteresis losses. Due to lower power density of such system the size of the compressor was increased by about 2 times.

⬐ nobrains

Analogy wise, this seems like AC (alternating current) instead of DC (direct current), but for compression motors.

(Edit: But only until the unit that combines both the sides and makes it back to DC (still using analogy) back again)

⬐ hellgas00

LG uses linear compressors in their fridges, very similar to this design and they have been out for years. They generally fail after only a few years, terrible reliability in comparison to a piston compressor.

⬐ gpapilion

Mine failed within 5 years, and this seems to be a common thread among LG owners.

⬐ swamp_gas

Some ACs are already 33% more efficient.

https://www.pickhvac.com/faq/what-is-the-best-seer/

⬐ fernly

If this[1] is the "Magtor" mentioned, they say nothing at all about this tech on their website, that I can find.

[1] http://magtor.com/

⬐ jf

From another comment in this thread, it looks like the site is: https://www.magtor.tech/

⬐ fernly

It's magtor.tech as shown in the next post.

⬐ phnofive

Replying elsewhere, but here's a USPTO link on the inventor:

https://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=...

⬐ basicplus2

The biggest inefficiencies come from combining heating and cooling in the one unit.

if you buy a airconditiiner that only cools (or only heats) it can be aything from 3 to 5 times more efficient than a reverse cycle type.

⬐ tjmc

Can you please link to datasheets that verify this because I haven't seen these discrepancies in smaller (split) systems.

⬐ thorin

Live in areas that don't need air conditioning :-)

Coming from UK / Europe I'm always surprised by the number of people in the US and of course Asia that are totally dependent on aircon.

⬐ oblio

Yup, let's abandon 10-20% of the globe, and maybe 30% soon, as global warming ramps up.

That's going to be quick, easy and politically viable.

⬐ thorin

Maybe forcing buildings to be created in such a way to minimize aircon and be more energy efficient would be helpful rather than relying on powered devices to fix the problem in place. I could deal with the 1 or 2 weeks a year it might be might to nice to have aircon where I live.

Building in excessively Dubai for example seems crazy to me.

⬐ oblio

I agree with you, but the global tendency is towards bigger and harder to heat/cool houses, spread apart (so again, no shared heating/cooling), whenever living standards improve.

People seem to be living in crowded cities just because they're forced (jobs are nearby, apartments are generally cheaper than apartments because they're generally smaller, etc).

Desert environment should definitely be super high density and possibly largely underground in my opinion. Things don't seem to be moving that way, though.

⬐ thorin

I'd struggle to live somewhere where you have to spend most of your time indoors in and air conditioned environment anyway, but this doesn't seem to bother a lot of people, I guess they are used to it. If we carry on this way maybe colonizing other planets won't seem so bad...

⬐ robbmorganf

I'll give it to them that the solution is an elegant reduction in parts count. It's just a shame that they compare it to a straw man and exaggerate the benefits. "It solves this problem, and this one, and this one" but modern compressors don't suffer most of those. Other sibling comments have good specific qualms, so I won't duplicate them.

That said, I'm glad people play around with new designs. That's how innovation works. I just wish investors would be satisfied with "we're trying a reasonable alternative design to look at benefits" rather than requiring "the old design is a dinosaur, and our design solves every problem under the sun"

⬐ sargun

Why wouldn't they use scroll compressors?

⬐ AussieWog93

Impressive tech, but I wonder how much power we could save if we just started putting doors in living rooms.

⬐ gpm

How would doors in livings rooms save power?

⬐ AussieWog93

Maybe it's an Aussie thing, but most air conditioners I've seen are based in the living room. Most modern living rooms also don't have doors, meaning that these air conditioners are cooling 3-4x the volume of air that the owner actually wants them to.

At best, this wastes power. At worst, this wastes power and means the one room you actually want to cool doesn't get cooled properly.

I've lived in a unit with a door in the same room as a split system and it was amazing. The room would become completely cool in around 5 minutes, then the compressor would shut off and run for maybe 2 minutes every 15 minutes or so to maintain this cool temperature.

⬐ gpm

Canadian (Toronto) here, pretty much every air conditioner I've seen uses the same forced-air vent system as the furnace, and thus has outputs to every room of the house.

Definitely sounds regional, but also we air condition the entire house so we aren't winning the efficiency war here.

⬐ stephen_g

From what I've read and seen in videos about it, that seems to be the case in much of the US as well.

In Australia you can get ducted systems, but at the residential scale it's both more expensive and less efficient than split systems. So houses will often have larger split units (5 - 10kW heat/cool capacity) in the main living areas and then smaller ones (2.5 - 3.5kW capacity) in all the bedrooms. Sometimes you have a multi-head system with one outdoor compressor unit and two or three indoor units.

This is even the case for large apartment buildings (often have a separate AC compressor for each unit), but most office and commercial buildings have central AC.

⬐ thrdbndndn

In my country (and most of East Asian countries I have been, like China, Japan, etc.) residential ACs are mostly in individual rooms (like this: https://moderndiplomacy.eu/wp-content/uploads/2020/08/air-co... ) . You would have one for the living room, one for each bedroom, etc.

Central HVAC exists, but 90%+ homes don't have them.

It's not as convenient and comfortable (one common problem older houses have is that there is no AC in kitchen) but the energy cost difference compared to the norm in the North America is night and day (the houses are significantly smaller here too).

⬐ TeMPOraL

That makes sense if you actually want to only cool one room. Otherwise, you may want to have doors open/close independently of your cooling needs.

I've spent some time trying to turn a portable AC unit into proper AC solution during the recent heatwaves over here in Poland, and the doors were a problem for me, not a solution. For example, I work in a study, and my wife continuously moves back and forth between kitchen and the living room. That's already 3 rooms to cool, 4 if you count the corridor (which shouldn't cause you a thermal shock every time you navigate it). Which doors are best closed or open depends on whether our kid is home. E.g. when she is at home, my study doors are closed, and I get to cook inside.

I'm on a rental now, so I can't duct the apartment the way I'd like (also the reason I bought a portable unit), but on the next apartment, I plan to either have a split system with a terminal in every room, or a duct system for supply air from the AC unit.

I understand that for large homes, some rooms may not see much use during the day - but in apartments, I can't imagine such a thing, unless you're living alone.

⬐ chadcmulligan

There's a few other things you can do in hot places

- plant trees outside windows, deciduous on north facing

- build your home facing north/east and have windows / doors so the wind can flow through

- put eaves on the buildings - there was a trend for a while to not have eaves

⬐ mrfusion

Btw what’s the story with teslas octovalve? Does that add a lot of efficiency to their heat pumps?

⬐ stephen_g

From reading some links from here, I don't think so. Despite the breathless amazement of the writers, it doesn't seem to really be an innovation in heat pumps itself, it's just doing things in a way that's much more efficient than Tesla's previous heating/cooling systems...

⬐ trhway

sound similar to free-piston vs. crank-shaft in case of ICE engines.

⬐ mianos

Pretty much. And the answers to all the technical questions are the same BS paste: "Response from Magtor CTO :"

⬐ ekianjo

Protected by patents = will get no traction whatsoever.

⬐ uCantCauseUCant

The answers is - reservoirs.. earth cools itself at night, we need to pump the heat into the night sky and store the cold in salt reservoirs.

⬐ SevenSigs

if your AC is ~10 years old... just replace it and you will probably get > 33% efficiency gains

⬐ OJFord

I'm going to go out on a limb and assume 'a revolutionary solution' is not going to come from '(youtube.com)'...

⬐ Weryj

I don't understand the dismissal of content because of where it's hosted.

There is a large scientific community all publishing well researched videos. It's a free platform that has a large audience, seems like a natural fit for anyone looking to teach.

https://www.youtube.com/c/TheAIEpiphany

https://www.youtube.com/c/pbsspacetime

https://www.youtube.com/c/SciShow

As some examples

⬐ OJFord

I enjoy a lot of YouTube content, I just reacted to the 'revolutionary' assertion attached to it. I'm not going to watch your links right now at least, but I assume they are not announcing breakthroughs, that they are great content sure, but following earlier press releases, publications, whatever's applicable?

I believe HN is quite efficient, in that I expect to see AC 33% efficiency boost in paper/startup/PR/etc. form. Then maybe the 'how it works' or explainer videos later, sure, but you can't claim 'revolutionary' at that point.

Someone else pointed out it is new, it's just that for some reason the submittor chose the video instead of the company site announcing it which includes the video.

⬐ mhh__

Well, the first YouTube video was published on YouTube and that was pretty influential

⬐ OJFord

Was it? I don't even know what it was. YouTube itself is.. well it's at least arguable that it was revolutionary, but that's the platform, not whatever the first video was, if there even was a meaningful first (wouldn't you seed it with as much as you could from other sources?).

⬐ mappu

The first video is well known - https://en.wikipedia.org/wiki/Me_at_the_zoo

⬐ OJFord

I think I'll stand by 'not revolutionary' then?

⬐ TheCycoONE

The first video was founder Jawed Karim visiting the zoo: https://m.youtube.com/watch?v=jNQXAC9IVRw

It gets pulled out as a factoid periodically.

⬐ jffry

Please take a moment to go out on a limb and read the video description or watch the video.

The video is discussing a new technology from a company which purports to have come up with a more efficient way of compressing refrigerant in cooling systems, https://www.magtor.tech

⬐ OJFord

Fair enough, I stand corrected. I think your link would've been a better submission than just the video specifically.

The submission lacks '[video]' and took me a couple of seconds to decide it wasn't for me, just in terms of presentation style even after you made it more appealing.

⬐ OJFord

Can no longer edit, but the submission has been updated to remove the silly click-baiting ('revolutionary'), and to add '[video]'... I don't know if the flaggers and down-voters missed the earlier title so lack context, or what, but it's sorted now - I take no issue with the current title 'How can we make air conditioners more efficient? [video]', that's not what my comments above are referring to.

⬐ kova12

Jeezus, what a bunch of wankers. Just stop screwing around and get nuclear energy prioritized already. Solve energy crisis and global warming all at the same time. Only loser is exxon-mobil, but I can live with that

⬐ hamburgerwah

Here is what will happen with this technology... nothing. The company is an IP play where they have locked up this concept in IP and will try and force it on jurisdictions by various means of legislation to ensure they can get the highest prices possible for said IP. Incumbents will resist choosing to continue with current public domain technology. In 30 years when the IP expires it might possibly come into more common use. The current IP regimes are a complete disaster and are hamstringing human progress.

PS: Also scroll compressors are substantially more energy efficient than rotary and various means of staging is even more efficient yet.

⬐ nuccy

This technology (the idea of linear compressor directly driven by a coil) is used for at least two decades, particularly in LG compressors in AC and refrigerators [1].

1. https://en.m.wikipedia.org/wiki/Linear_compressor

⬐ joshuaissac

> will try and force it on jurisdictions by various means of legislation

Have they said this is their strategy?

There are IP licensors (like ARM and the Fraunhofer Society) that sell licences to their IP without trying to force jurisdictions to adopt their technology.

> Incumbents will resist choosing to continue with current public domain technology

This would depend on the pricing of the IP licences and the competition within the market. If the licences are cheap enough and the market competitive enough that using this tech would provide an advantage, then incumbents would license it.

⬐ colechristensen

ARM sells... "intellectual property" that I might try to come up with a distinguishing definition from the IP of "one clever idea I want a monopoly on for a while".

A lot of IP boils down to an idea that you could describe in an elevator and a few good engineers could turn into a product in a few hours to a year.

ARM sells things that represent... thousands? hundreds of thousands? of engineer-years. You can't just whip that up in a short time, it totally makes sense to pay someone else for their trouble to make something... but if you wanted to and had a billion dollars to spare, sure you could do it yourself.

The problem with "I thought of it first" IP protections is the phenomenon of simultaneous discovery. A whole lot of ideas just have their time come, and tend to have multiple people working on them and coming to similar results in short periods. Making this a race and letting the winner sell out to groups that either want to extract rent out of the idea or to hide it in a safe to prevent competition for 30 years is not conducive to progress.

⬐ WalterBright

I patented a couple things just for the lulz, but in general I'm opposed to patents. I should have patented a couple others just because I kind of resent others claiming they invented it years later. But I can't complain too much, as I've also invented things that I later discovered I was late to the party.

⬐ c_o_n_v_e_x

Were you ever approached to have your lulz patents licensed?

⬐ WalterBright

The people who paid for all the patent paperwork owned the patents, and they licensed them, but I don't think anything ever came of them.

One was a method of texting that didn't require looking at the phone, the other was a way to display lo-res graphics data without pixelation.

⬐ TeMPOraL

> One was a method of texting that didn't require looking at the phone

Oh that's easy, every phone before iPhone could do it - you need a) a physical keyboard, and b) a soft real-time OS. Tactile feedback of physical buttons + stable UI configuration with learnable delays = you could learn to operate the phone without taking it out of your pocket.

:)

Joking aside, what was your invention? Could you link to the patent?

⬐ WalterBright

The idea was to put a rocker switch on the side, to rock between "dot" and "dash", and then use Morse code. To "read" Morse, the phone could vibrate, or at least decode it as text to read. You could text without walking into a tree (which I have done.)

The graphics one was pretty straight forward. Instead of dividing up the picture into rectangles, divide it up into triangles. Set a color for each triangle based on an average of the color pixels in it. Then, smoothly shade the rendering of the triangle based on the colors at each vertex. This will prevent any banding or that ugly blockiness. The number and placement of the vertices depends on the where the complexity in the image is, and of course would be adjustable.

The browser one is to have the browser rendering prioritize where the mouse cursor is. Great for slow connections.

⬐ scatters

I'm sure I read the Morse code one in a '90s science fiction story, although I think that was described as pure haptics; not sure if that would count as prior art for your patent.

⬐ WalterBright

Certainly, Morse code itself isn't patentable. It's more the specific combination here.

⬐ TeMPOraL

> The idea was to put a rocker switch on the side, to rock between "dot" and "dash", and then use Morse code. To "read" Morse, the phone could vibrate, or at least decode it as text to read. You could text without walking into a tree (which I have done.)

I can imagine this working. Kind of like CW for HAMs, except with a 3-state switch and couple dozen more layers of abstraction between you and the wire :). I wish my Morse code wasn't so rusty though - and it makes me wonder, are there better text encodings that could be used for such purpose?

The method for texting without looking I described was only limited to writing, but it's also something I've done in practice before smartphones, in particular on a Sony Ericsson K800i. This was possible because of good-quality keyboard (and joystick), and good UI decisions. I don't know if they put an actual soft-realtime OS in there (I suspect so), but all the UI delays were always the same - so I naturally memorized sequences like "unlock, menu, texting, wait ~0.5 second for it to load, open thread, type in message, send". It probably takes teenage-level or business-professional-level of use to get that kind of muscle memory, though.

> Instead of dividing up the picture into rectangles, divide it up into triangles. (...) This will prevent any banding or that ugly blockiness.

Is the insight here that a triangle-based subdivision like this will make areas of detail not align on horizontal/vertical lines, unless the underlying image has its details aligned on a band? Or am I misunderstanding this?

⬐ WalterBright

I've never attempted to learn Morse, but it looks like a few hours of practice should do it. Anyone can learn it, but I expect younger people would be more willing to do it.

> the insight

The vertices will be on pixel addresses. The edges of the triangle will have no discontinuities themselves, and the triangles sharing an edge will have no discontinuities. Hence, no banding or blockiness. A naive renderer would just render the colors of the triangle as a straight linear interpolation. A more advanced one could look at the adjoining triangles, and render it so the first derivative would also be zero across the line. I.e. curve fit it.

The selection of where the vertices go would likely be best selected where the colors are shifting fastest, i.e. an edge in the picture.

The neato thing is one could render a low res image on a high res display and it'll still look good. You wouldn't have these stray pixels scattered around the text like jpg does. The other neato thing is more vertices can be put in the complex areas of the image, rather than wasting bandwidth on the parts of it that are smooth.

Progressive sharpening of the rendering also works, as more vertices can be added at any time.

⬐ srean

Hexagonal, equivalently, equilateral triangular tessellations indeed have many advantages over using squares. Your scheme reminded me of Delaunay triangulation and its dual, Voronoi tessellation.

I have a faint suspicion that your familiarity with finite element models had something to do with your choice of triangles.

⬐ dharmon

The method actually reminds me of a simple h-refinement scheme for FEM.

⬐ WalterBright

Triangles are used by game engines for good reason. Hexagonal cannot be mapped without discontinuities. Equilateral triangles are not necessary.

I've done finite element models, but they were pretty unsophisticated.

⬐ srean

That's right, simplicial complexes of the right dimension would give one the most fidelity in terms of continuity. Triangles happen to be the one for 2D. For 3D one would get similar benefits using tetrahedra.

I was alluding the fact that barring boundary effects hexagonal tessellation and equilateral tessellation are essentially the same.

⬐ andrepd

> The method for texting without looking I described was only limited to writing, but it's also something I've done in practice before smartphones, in particular on a Sony Ericsson K800i. This was possible because of good-quality keyboard (and joystick), and good UI decisions. I don't know if they put an actual soft-realtime OS in there (I suspect so), but all the UI delays were always the same - so I naturally memorized sequences like "unlock, menu, texting, wait ~0.5 second for it to load, open thread, type in message, send". It probably takes teenage-level or business-professional-level of use to get that kind of muscle memory, though.

Wow, this threw me back. :) Indeed this was not only possible, but actually easy. I don't think it took a lot of effort or any conscious effort even, actually, the UI was so good, but most importantly there were no unpredictable delays and no unpredictable inconsistencies!

Why is modern software so bad in comparison? Sure it can do more things, but it does them worse. The interface of everything is constantly changing for the unfounded whims of designers who simply think "this looks prettier to me", things are not laid out logically, adverts permeate the interface of most products from phones to TVs...

⬐ WalterBright

Patent 7831208 Wireless mobile phone having encoded data entry facilities

Patent 7812993 Lossy method for compressing images and video

Patent 7711748 Method and apparatus for simplified access to online services

Patent 7028033 Method and apparatus for simplified access to online services

Patent 6897977 Lossy method for compressing pictures and video

Patent 6850782 Wireless device with vibrational communication capabilities

Patent 6657647 Controlling the order in which content is displayed in a browser

Patent 6418323 Wireless mobile phone with Morse code and related capabilities

AFAIK, none of these have been made use of. Though every time I see banding and blockiness in video, I think "they should implement triangles!"

⬐ TeMPOraL

Thanks! I've skimmed the ones related to vibrations/Morse code, I'll look at the rest later.

At the risk of forever poisoning myself for related work, I guess. Whose bright idea was with that "you look at it, and you and everyone you know is now liable for 3x more damages for accidental infringement" law?...

⬐ WalterBright

> Whose bright idea

Mine, obviously. All my ideas are Bright ideas.

The patents may have expired, anyway. One of the reasons I did the patents was I was always pestering people with those ideas, as I wanted someone to implement them. So they said "patent them, which will get the ideas out there."

⬐ aetherspawn

Mitsubishi already have a (relatively) low-cost scroll compressor for air conditioning applications that works very well.

⬐ xyzzy21

Also the refrigerants matter - unfortunately, the MOST efficient ones also happen to be really bad for the ozone layer so they've been "banned" (there are still plenty of units out there using them - just can't use on new systems).

So there's always a trade-off. You pay for this or you pay for that. And if its energy, you have to consider the material and energy supply chains or you are probably bullshitting people and yourself.

Ultimately an air con is a heat engine so Carnot's law applies. No possible magic beyond what thermodynamics allows.

ALL TOO MANY "inventions" summarily . You even have UC Berkeley getting into the con-game with the Water Seer (which claims violates thermodynamics). You'd THINK UCB wouldn't ever push physics-violating ideas but NOPE. A stain on the engineering school reputation!

⬐ mech422

My dad was making circular screw compressors back in the 70s and 80s...

I would think most of this stuff would have prior art...

⬐ michaelcampbell

SawStop tried this with table saws. Put such a bad taste in my mouth that I will never buy one.

⬐ bsder

Are you sure that isn't saw manufacturer propaganda?

I thought the issue with SawStop wasn't that the licensing was bad but that nobody was willing to take on the liability.

ie. If someone cut off their hand with a table saw, it was settled law that it was their own fault. If someone cut off their hand and the SawStop didn't fire, the manufacturer now had a gigantic liability.

⬐ frosted-flakes

SawStop is or was trying to get the government to ban the sale of saws without flesh detection technology. Guess who holds the patents for said technology?

Bosch came out with a superior design that doesn't use a brake at all, but were barred from selling it in the US because it infringed SawStop's patents.

⬐ bsder

> SawStop is or was trying to get the government to ban the sale of saws without flesh detection technology.

Yeah, that's bad. When you resort to lobbying, you're no longer on the good side of the ledger.

⬐ slownews45

Tell that to the kids whose hands are saved.

https://www.swnewsmedia.com/shakopee_valley_news/news/educat...

⬐ bsder

We have a SawStop in our space for exactly that reason.

However, I don't find forcing everyone to have it by law to be particularly palatable.

I will tell you, however, that it goes off about once a month like clockwork and, as far as I can tell, it hasn't yet fired because someone was about to get injured (this is a good thing--it means our people aren't being stupid).

It's a $100-$200 hit every single time it fires. So, we have an ongoing expense of about $1500 every year for SawStop. That's a new saw every year.

Should that really be forced on people?

⬐ whoknowswhat11

Absolutely not - no where did I say that anyone should be forced to buy the saw.

I was merely pointing out that saying you would never buy a sawstop for whatever crazy reason (ie, other companies wouldn't pay them the $3-$10 or whatever) seems a bit ridiculous.

If you have a fair number of folks using the saw, have them touch materials to saw first with saw off to get a read on it or list the ones not recommended.

⬐ dheera

I think it's completely reasonable to require it but also mandate that the health insurance companies pay for it, and all cartridge replacements. It's in their best interest to prevent accidents, and the cost is miniscule compared to even a single ER visit.

⬐ whoknowswhat11

Especially in SF bay area

$18K for a bottle of milk for a baby etc (trauma activation fees are crazy).

⬐ LgWoodenBadger

If it’s happening that frequently, and it’s not due to contact with flesh, then you’re cutting material that you shouldn’t be, or you should be disabling the protection feature.

⬐ bsder

> you’re cutting material that you shouldn’t be, or you should be disabling the protection feature.

I will absolutely fight you on this tooth and nail.

1) I don't want anyone to get into the habit of turning off safety features. Ever.

2) If the material can be cut on that saw without the SawStop, it should be able to be cut with the SawStop. If it can't, the fault is with SawStop, not the user.

I'm happy to pay the extra money for the extra safety. However, I will NOT allow you to blame the user for the fact that SawStop has not uncommon failure modes.

⬐ whoknowswhat11

Uhh - the manual will tell you not to cut aluminum, very wet wood etc with sawstop feature on.

If you turn off feature you are back to your regular saw - hopefully you use a stick etc in all cases because as you note replacing cartridges is expensive.

One thing I like, it'll put feature back on automatically for the next time machine gets turned on.

⬐ bsder

The user's definition of "wet wood" and SawStop's often differ quite significantly. Hope you're keeping a moisture meter nearby.

You also have to wait for the blade to come to a complete stop before touching anything or you will fire the system. This can be annoying when you have the motor turned off, retracted the blade, and start setting up your fence only to have the system fire because it wasn't quite stopped.

Was there a sticker on something that was slightly metallic? Oops.

I'm happy to have a SawStop on my saws, and I would even recommend it. However, given how often it false fires, I'm not sure I would mandate it by law.

⬐ hellbannedguy

Every piece of wood at Home depot has a staple in it for the upc code.

⬐ whoknowswhat11

Which does not trigger the device unless you are ALSO touching the staple or you cut with staple side down so you ground blade through staple into table.

⬐ slownews45

THIS is actually a great suggestion. Have some lower speed that sawstop does not trigger below. It really does not need to trigger once its (very) slow unless you have a heavier dado blade maybe?

Or have a noncartridge saw brake on the saw, when you come off saw for more than 3 seconds put some light pressure on to slow it more quickly.

I think a fair number of false fires are from this rather than the "wet wood" worry.

⬐ Dylan16807

So if we assume people are cutting appropriate material, your statement simplifies to "If it's happening that frequently you should disable it." And this is your argument in support of mandating it??

⬐ Bud

Lobbying is simply the practice of petitioning the government to adopt a given policy.

That's a foundation of democracy, really: the right to petition the government. I think you might want to re-examine your view a bit.

⬐ phkahler

>> Lobbying is simply the practice of petitioning the government to adopt a given policy.

Adopting a policy is different from having them mandate what people buy.

Hey, have people buy epipens and keep one at the school for each kid with allergies. Then later we will raise our prices 1000 percent. That's not influencing policy, its fucking people all the way to the bank - with government assistance.

⬐ specialist

I've done some unpaid policy work. Defending democracy by advocating election integrity. No one is writing books about my mendacious activities.

FANG has a phalanx of former electeds and staffers paid to hustle their former colleagues. To defend their grift and dismantle civil society. Grist for an endless torrent of outrage porn.

While pendants will argue both me and FANG are "lobbying", petitioning our governments, our respective efforts hardly seem comparable.

⬐ slownews45

This is actually totally false.

Sawstop dropped blade below table. Bosch also dropped blade below table. So the invention was very fast acting sensing and then dropping blade. Sawstop added brake.

Sawstop ALSO did a brake. The brake made a pretty reasonable difference in terms of damage.

Plenty of youtube videos exposing this saw.

"SawStop uses a different mechanism to drop the blade, but also employs an aluminum brake. In our testing on both top and front strikes, the blade stopped with less damage to my hand than Bosch’s Reaxx."

https://www.youtube.com/watch?v=B3JsUGwt_Mg

⬐ frosted-flakes

What is false? Did not Sawstop sue Bosch for patent infringement?

⬐ whoknowswhat11

"Bosch came out with a superior design that doesn't use a brake at all, but were barred from selling it in the US because it infringed SawStop's patents."

The design was not superior and the infringement was not based on a brake or lack of one. The infringement was because of copy of the skin contact sensing / blade drop technology as what were called underlying elements. I think the case may still be ongoing though? This one went a long while.

⬐ frosted-flakes

I never said that the infringement was related to the brake. I thought I was pretty clear that the patents were about skin-sensing technology.

And "not superior" is subjective. If the SawStop tech has enough false triggers and destroys as many blades as it apparently does based on what I've read (probably depends on material being cut), the increased cost almost certainly isn't worth it unless you're a high school shop class.

⬐ foerbert

I don't think that was clear. You mentioned the Bosch system had a patent issue and immediately mentioned it didn't even use a break. While it's possible to read that as the Bosch system being superior because it lacks a break, the main topic was the patent issue, so it's more natural to read it as being related to that.

Meanwhile there wasn't even a mention of the sensing technology.

⬐ frosted-flakes

I literally said "flesh detection technology" in my original comment.

⬐ slownews45

The sawstop story is interesting.

They basically tried to license it to the other saw companies.

The other companies said hell no. We have the market.

So sawstop started building their own (great) saws. They really are pretty to very high quality. The saw stop feature is also very effective.

It's so effective that if you are running a woodworking shop in a school, or ever want to let your kids use your saw, or have staff using your saw, you really owe it to them to ALWAYS buy a sawstop.

I've worked with a few owners who took the approach you did. Inevitably someone loses a finger - and after the dust settles and folks ask why they didn't put this VERY simple and not that expense solution in place - they usually regret it. Either they feel guilty for chisling on some $ and having someone lose function in their hand or if they are here in CA they pay out BIG TIME for their stupidity.

⬐ HumblyTossed

> The other companies said hell no. We have the market.

The other manufacturers would have had to immediately stop selling and possible have pulled all the other saws off the market. For liability reasons, you can't sell a line of saws that chop off fingers next to a line that doesn't. You're basically admitting the former is defective.

⬐ epistasis

What is the basis for this claim? It doesn't sound particularly plausible compared to other safety features, such as car airbags, which are not offered across all products in a brand.

⬐ Johnny555

For liability reasons, you can't sell a line of saws that chop off fingers next to a line that doesn't. You're basically admitting the former is defective.

I'd think the opposite is true - you can't sell a $250 saw that won't cut someone's finger off next to a $200 saw and expect it to sell. No one thinks it'll happen to them... otherwise Saw-stop's own line of saws would dominate the market. Some people will pay the premium, but I'd bet that most will not.

⬐ cbozeman

This is nonsense. Saws that chop off fingers aren't defective - the user is. It's a simple argument to make.

Saw Y doesn't have safety features and you use it at your own risk of injury. Saw X will stop cutting the moment it detects a body part. You also use it at your own risk of injury.

⬐ imglorp

I'm not convinced about that legal theory. There's another precedent in motorcycle ABS: unlike cars, not all bikes have it yet, so you can buy a Ninja both with and without ABS. The feature is even more decisively important for bikes than cars because it keeps them upright and stable under hard braking, so it's not just stopping distance. It adds several hundred to the price.

Or maybe you're right and we'll see an "admission of defect" suit sometime soon.

⬐ slownews45

A series of meetings followed, in which Gass "negotiated with major players such as Ryobi, Delta, Black & Decker, Emerson, and Craftsman" in an attempt to license his invention; he followed those negotiations with a February 2001 presentation to the Defense Research Industry, a trade group for attorneys representing the power-tool industry.[7] That presentation immediately preceded one from Dan Lanier, Black & Decker's national coordinating counsel; Lanier's presentation gave Gass the impression he was unlikely to succeed in convincing major power tool manufacturers to license SawStop technology.[7]

The license fee was going to be around 3% of wholesale (so about $3 on a $100 wholesale saw that goes for $200 retail). I think he came very close to licensing to ryobi?

Anyways, if you think $3 for potentially incredibly QOL saving tech is too much for an inventor that actually invented something (and tested on his own finger) then I don't know what to tell you.

That said, once someone's parents (if you run a woodshop in a school) or employee sues you - you may start valuing this invention a bit more.

⬐ hamburgerwah

$3 was at least 100, if not 1000 times more than these manufacturers would have ever been willing to pay in license. That was just the license, the technology itself added significant cost, weight and complexity.

It's no surprise that sawstop can make a nice saw for 4 to 17x the price of competitors.

⬐ None

None

⬐ WalterBright

On the other hand (!), maybe the onus for the lost fingers is on SawStop for demanding those licensing fees.

⬐ manigandham

As opposed to... giving the tech away for free? What alternative are you proposing is better?

⬐ WalterBright

I give away tech for free that I work full time on every day. (The D programming language). So does everyone else who works on it. It's how Open Source works.

⬐ manigandham

Good for you. This isn't open-source though. It's capital intensive product development and people want to be paid for their efforts.

Are you seriously suggesting that they should just work for free?

⬐ WalterBright

Not at all. They're free to manufacture and sell sawstop saws. They can also supply the necessary hardware to other saw makers, saving them the trouble of developing it themselves. It'd be likely cheaper to just buy the hardware than reverse-engineer it.

⬐ Danieru

Congratulations! You are today's 10,000th person to learn of: <Perverse Incentives>

⬐ snet0

How is it an example of a perverse incentive? If we don't reward inventors who make genuinely useful things, there's no incentive.

⬐ hellbannedguy

To certain people.

⬐ Danieru

I was replying to a comment which suggested the investor should be punished for not offering their invention for free.

The poster wanted to punish inventors for inventing. If that isn't a perverse incentive I don't know what is.

⬐ WalterBright

> The poster wanted to punish inventors for inventing.

Please try reading what I wrote again.

⬐ Danieru

> On the other hand (!), maybe the onus for the lost fingers is on SawStop for demanding those licensing fees.

Done.

Do you disagree that your writing says "the onus for lost finger is on SawStop"? Your writing seems very clear to me. I do not know why you are confused.

⬐ WalterBright

I said "maybe", as in who is the unfair party here? Nowhere did I say anyone should be punished.

⬐ GuB-42

Safety features are not free and no one should expect them to be.

If some safety feature is judged too important not to have, as it is common in cars, the government steps in and makes it mandatory, but it doesn't mean there are no more licensing fees. At best, the licensing fees are limited to a fair price.

⬐ WalterBright

> Safety features are not free and no one should expect them to be.

There's a difference between the cost to provide something and the license fee. Note the Moderna licensed its covid patents for $0.

> a fair price

What's a fair price? What is the rule that says one party is greedy and the other is reasonable?

⬐ foepys

Did anybody use Moderna's patents? As far as I know nobody did because Moderna licensed them for a limited time in which nobody would be able to ramp up a completely new production line.

Almost always those 0 cost licensing agreements are just PR stunts.

⬐ robertlagrant

Walter:

> On the other hand (!), maybe the onus for the lost fingers is on SawStop for demanding those licensing fees.

Also Walter:

> What's a fair price? What is the rule that says one party is greedy and the other is reasonable?

⬐ ianlevesque

There's a whole precedent for this with FRAND (Fair, Reasonable and Non-Discriminatory) licensing. It's really common for standards.

https://en.wikipedia.org/wiki/Reasonable_and_non-discriminat...

⬐ infogulch

All patents (really, all licensable IP) should be subject to this.

⬐ michaelcampbell

Like the polio vaccine.

⬐ meigwilym

Not that this proves or disproves anything, but Volvo famously gave away the patent for the three-point belt systems in cars.

https://www.forbes.com/sites/douglasbell/2019/08/13/60-years...

Sometimes companies and individuals think that it's wroth losing financially for the sake of wider safety.

⬐ erik_seaberg

It's good that the license was only $3, but the end user also has to buy at least one disposable $80 cartridge (probably more, apparently false alarms happen) for a problem that can be solved with a push stick.

⬐ AmVess

Push sticks have been turned into formidable projectiles, even by the most experienced and careful of user.

⬐ whoknowswhat11

Good point.

You can send in your activated cartridge to sawstop - if it was skin activated then they may replace it for free.

If you are concerned about a trip you can touch blade to piece with saw off and it'll tell you if would have tripped. Or do a first cut with system off. Or if you have a wet piece turn it off.

That said if you have a nail or something in piece there are situations that could trip it for sure.

⬐ kens

I was at a shop that had a "wall of shame" with sawstop cartridges that had tripped when people tried to saw wet wood, metallized plastic, and other conductive materials.

⬐ jjav

The detection can be disabled for the case of cutting conductive materials.

⬐ slownews45

It can also be used to sense if the wood is too wet - go to bypass, then do a cut in wettest part and you'll get an indication.

⬐ c_o_n_v_e_x

Vapor compression is a technology ripe for disruption. I reckon the IP holder knows this and is trying to monetize this while they can.

I think tech based on magnetocaloric cooling is one option. There was a company I was following a years ago (I've since forgotten the name) that was researching advanced thermoelectric cooling technologies that would be offer much better performance than peltiers.

⬐ Abishek_Muthian

Exergyn has been working on shape memory alloy based air-conditioner[1] which doesn't require refrigerant, Since the metals return to position depending upon the temperature it effectively serves as a heat pump. It is supposedly 30-40% cheaper than current ACs and being solid-state stack it's definitely more efficient(Not sure how it compares with the calimed efficiency of Magtor).

But what's more important is they seem to be already gotten a deal from Carrier and automobile industry which suggests they might not be interested in IP swatting but actually build stuff.

[1] https://www.economist.com/science-and-technology/2021/07/21/...

⬐ BulgarianIdiot

Your hypothetical scenario requires everyone being irrational.

The company probably doesn't want to sit on their IP for 30 years and not yield any profit. How does that benefit them.

They'll try to bend some arms, and if this fails, they're most likely to drop licensing prices until they meet market demand.

And yes, there's demand for efficient AC. I mean, come on.

⬐ robinsoh

> Incumbents will resist choosing to continue with current public domain technology. In 30 years when the IP expires it might possibly come into more common use. The current IP regimes are a complete disaster and are hamstringing human progress.

I am curious. What explains the success of Microsoft Windows as the dominant "technology" for PCs and laptops? That's not due to "intellectual protection rights" is it? I am not a software engineer, just a display engineer, would I be correct if I alleged that "IP regimes" are hamstringing operating system progress? My gut instinct is no, that would be an absurd claim. Perhaps equivalent to me claiming that "IP regimes" are hamstringing search engine progress? I am genuinely interested to find out what people think about this.

⬐ kragen

https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor....

Yes, IP regimes are definitely hamstringing operating system progress. https://en.swpat.org/wiki/NetApp%27s_filesystem_patents goes into some details about why Linux doesn't have automatic transparent instantaneous backups like the NetApp fileserver I was using a quarter century ago.

⬐ beambot

The federal government can and should suspend patent coverage for key climate-related innovations -- just like what was done for COVID vaccines.

⬐ beervirus

Yes, removing the profit motive will surely incentivize such inventions!

⬐ beambot

I have two dozen patents to my name and know dozens of others who hold as many or more than I do. I also know just as many successful entrepreneurs.

Not one had ever said "I would commercialize <x>, but for the patents", nor have any said "I would stop doing innovation if patents ceased to exist".

I concede that there are some field where it might make sense (eg pharmacology), but most of the time, patents are a tax -- hugely expensive for small innovators and weaponized by mega-corps.

Further, in times of crisis (eg COVID or war with radar), patents are suspended for certain topics to promote speed of execution. We are at that point for climate change & energy efficiency.

⬐ beervirus

Yes, anecdata.

⬐ hellbannedguy

There are a lot of inventors out there who can afford the fees.

I would love to see patent fees based on income level, and the more you get; the more they cost.

⬐ TeMPOraL

I would like to have a breakdown of individual vs. organization-backed inventors. Quite a lot of - but I don't know how much, relatively - patent applications are filled by individuals with help and in the name of their employer; there are companies who are proud of the number of patent applications they file per year - and those are mostly bona-fide inventions, not patent trolling.

(It's entirely expected, if you think about it - a big company employing skilled specialists doing actual[0] engineering is bound to become a patent mill, as those workers have to invent their way out of new problems.)

I suspect that this might be the primary way patents are applied for and awarded, and it's different from the usual narration, that patents are either small inventors wanting to ensure the market rewards them for their hard work, or evil megacorps buying them up as a part of peace-through-MAD balance with other evil megacorps.

--

[0] - Traditional, "trad", not-software engineering. I know this is just a stereotype, and this[1] series of articles does a great job debunking it, but... well, I haven't fully internalized these conclusions just yet.

[1] - https://www.hillelwayne.com/post/are-we-really-engineers/; thanks 'Twisol for bringing it up (https://news.ycombinator.com/item?id=28192650).

⬐ WalterBright

Or it could just buy those patents, using eminent domain powers.

⬐ vxNsr

Using eminent domain still requires paying close to fair market value… determining that will find quite a few lawyers’ French chateaus.

⬐ ajross

> PS: Also scroll compressors are substantially more energy efficient than rotary and various means of staging is even more efficient yet.

PPS: Also the efficiency of a heat exchanger cycle has very little to do with the individual losses of the compressor components. Engine losses matter at the compression ratios you see in a combustion piston, but at the scale coolants run at (especially ones like in an AC that involve a phase change) it's almost noise when compared with the fundamental thermodynamic limitations. You can do a little better for sure; 33% lower friction losses seems reasonable.

But you can't do much better: the linked article is surely not talking about 33% better overall cooling efficiency, that sounds insane to me.

⬐ rainbowzootsuit

I agree about the 33% efficiency doesn't make any sense from a thermodynamic perspective. The discussion of the peak current for the compressor motor is quite strange. That is a fraction of a second that the system experience the Locked Rotor Amps (LRA). It is also solvable when a system has variable compressors or inrush current limiting.

One bit of reasonable efficiency gains to be had is by switching refrigerants to an R22 analogue. R22 is banned per the Montreal Protocol as of 2020.

However, R22 has thermodynamic properties that make it more efficient than the standard refrigerant of today's residential and small commercial units, R410A. This is particularly as the heat of the atmosphere being rejected into rises with some Coefficient Of Performance (COP) differences in the 10% range.

https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=86088...

There are some analogues to R22 without the ozone damaging properties that share or exceed the the thermodynamic performance of R22. R407C is one that comes to mind, but there's a handful that can be retrofitted or purchased new.

However, a trade off is that generally a newer Energy Star unit will be significantly more efficient than an older R22 unit. My point with the R22 is that the industry will slowly move towards some of these other refrigerants because of their better thermodynamics.