7 ms·

The reason Mazda struggled with the rotary engine is due to what their application was. Rotaries work most efficiently and reliably when run for long periods of

The reason Mazda struggled with the rotary engine is due to what their application was. Rotaries work most efficiently and reliably when run for long periods of time (hours) at a consistent RPM. Vehicles with traditional manual or automatic transmissions (what most Mazda rotary cars had afaik) do anything but keep consistent RPM. Mazda subsequently struggled with keeping emissions down, and fuel economy up, while prolonging the life of the engine. If you tie a rotary to a generator, or true CVT transmission, they're much better suited to the task.

koolba 10y ago
That sounds like a good choice for a hybrid that uses the rotary ICE to periodically charge the battery.
- lttlrck 10y ago
  Or a range extender generator like in the BMW i3
  - thephilsproject 10y ago
    I first saw this on the Audi A1 e-tron concept; it was a 250cc wankel for the range extender. I thought that was the future when I saw it, and here I am now with a Vauxhall Ampera (Chevy Volt)!
  - phineastcat 10y ago
    Mazda actually had a Mazda2 EV prototype with a rotary range extender that they unveiled in 2013. Not sure what ever happened with it, though.
alistairSH 10y ago
While that's certainly true, others have developed similar small Wankel derivatives and had little luck moving beyond a prototype phase. LiquidPiston themselves have developed other engines with no market success that I've seen. There may be a market for these in small UAVs, but I don't see anything here to indicate LiquidPiston is any more likely to "win" than anybody else.
JPKab 10y ago
How is their maintenance/durability? My understanding on rotary engines is that in order to maintain compression, they have to have a solid seal on very fast moving surface contact points, which is incredibly hard to do. Naturally, the material that produces the seal is going to wear down over time, and to replace it the entire engine will have to be taken apart. I love the concept of a Wankel engine, and it would be awesome if they've overcome that particular aspect.
- Derpdiherp 10y ago
  Perhaps they could come up with some system like what we see in traditional piston engines piston seals - a replaceable seal that goes around the rotary triangular err piston for lack of a better word. If I remember correctly all the other wankel / rotary engines have just had a lump.
  - alistairSH 10y ago
    Mazda's automotive Wankels have spring-loaded seals on the apex of the rotor and on the sides. See this image: https://hackadaycom.files.wordpress.com/2016/02/wankel-inside-kart-engine.jpg https://hackadaycom.files.wordpress.com/2016/02/wankel-insid...
    - Derpdiherp 10y ago
      Didn't know that - thanks for the information. Was it wear against the side of the casing that led to the issues then?
      - Grishnakh 10y ago
        I don't know, but just looking at that photo, and comparing to a traditional piston engine with its piston rings, I can see a bunch of problems: 1) there's a bunch of irregular shapes here to seal: each "chamber" has shorter seals in two places (the vertices), and two longer seals on the sides. So 4 separate seals per chamber (two of which are re-used for adjacent chambers). In a piston engine, each chamber has usually 2 or 3 seals, with one being the main seal (top of the piston) and the others being either a backup or an oil ring (for scraping off most of the oil). A circle is a much simpler design. 2) 4 seals means 4 places where there's going to be gaps where gases can escape between chambers, causing extra emissions. In a piston engine, there's only 1 gap in each piston ring, and you can rotate the rings so the gaps are not aligned. The sealing on piston engines is extremely good. 3) The piston engine is pretty simple really, as far as sealing and lubrication: oil is either squirted (some engines) or splashed up on the backsides of the pistons and drawn up on the piston walls to lubricate the rings as the piston rises, and then the oil is almost 100% cleaned off by the rings as it travels on the downward stroke. This mechanism works so well that modern piston engines burn a truly negligible amount of oil. Oil burning has always been a problem on rotaries, and looking at that photo it's fairly obvious why. Mazda (and others) has been working on rotaries with very smart engineers for literally decades, and they still haven't fixed these problems. I don't think they ever will; these problems seem pretty fundamental to the layout of a rotary engine. They had some real promise and advantages, but like many technologies, the small details made it impossible to beat incumbent technologies. Piston engines were bad too, ages ago, before they figured out a lot of small details like how to seal them well, how to optimize intake and exhaust valve design and timing, how to best design combustion chambers to maximize compression ratio and eliminate hot spots causing detonation, etc. But some problems just can't be overcome.
mschuster91 10y ago
> or true CVT transmission What I never understood: why don't do it like on a ship, attach a generator to the fossil-fuel engine, add a battery and supercap pack, and four small-ish wheel electric motors? That should save the weight of a gear-shift box, complex mechanics on FWD cars (because the axle transmitting the movement energy must be flexible), and it would allow the motor to run at its most efficient RPM count, as well as providing full torque from 0 km/h. In essence, a Tesla with a fossil-fuel engine in the frunk...
- prodmerc 10y ago
  I'm assuming all that weight still gets better mileage than what would basically be a petrol->kinetic->electric->kinetic convertor. Modern engines are pretty efficient, adding an electric stage would likely reduce that efficiency.
  - mschuster91 10y ago
    It's not only about the weight. An electric intermediate circuit can: a) absorb braking energy - that's why I mentioned super-caps, because batteries can't ingest the vast amounts of energy that braking could theoretically produce (usually the brake power is 4x engine power, which means for a 250 kW motor a minimum brake power of 1 MW, and for sportscars likely muuuch more) b) keep the engine, if not in idle (i.e. you're stuck in traffic), at its most fuel-efficient RPM range - this is something any cheap-ass scooter does, and every diesel-electro locomotive and ship, but no car! Not to mention it isn't just about fuel-efficiency, but also many exhaust gas treatments only work at peak performance under very specific driving conditions (esp. temperature). The efficiency loss by conversion (97% and better in efficiency class IE4, see https://de.wikipedia.org/wiki/Elektromotor#Wirkungsgrad_und_Effizienz https://de.wikipedia.org/wiki/Elektromotor#Wirkungsgrad_und_...) is, taking especially the recovered braking energy into account, smaller than the efficiency delta gained by loss of hundreds of kg of weight (batteries are heavier, yes, but they can also replace "dumb weights" needed before to keep the car on the ground in the "elk test"). Also, less parts that are prone to (dirt) failure, e.g. the power-loaded joints on FWD cars, clutches, bearings, the entire gearbox system, differentials (in 4WD/multi-WD systems!). And you save on tires because in curves the individual motors can adjust speed. If done well, it should be pretty cost-efficient (less stops in the shop), too.

koolba 10y ago

That sounds like a good choice for a hybrid that uses the rotary ICE to periodically charge the battery.

lttlrck 10y ago
Or a range extender generator like in the BMW i3
- thephilsproject 10y ago
  I first saw this on the Audi A1 e-tron concept; it was a 250cc wankel for the range extender. I thought that was the future when I saw it, and here I am now with a Vauxhall Ampera (Chevy Volt)!
- phineastcat 10y ago
  Mazda actually had a Mazda2 EV prototype with a rotary range extender that they unveiled in 2013. Not sure what ever happened with it, though.

alistairSH 10y ago

While that's certainly true, others have developed similar small Wankel derivatives and had little luck moving beyond a prototype phase. LiquidPiston themselves have developed other engines with no market success that I've seen. There may be a market for these in small UAVs, but I don't see anything here to indicate LiquidPiston is any more likely to "win" than anybody else.

JPKab 10y ago

How is their maintenance/durability? My understanding on rotary engines is that in order to maintain compression, they have to have a solid seal on very fast moving surface contact points, which is incredibly hard to do. Naturally, the material that produces the seal is going to wear down over time, and to replace it the entire engine will have to be taken apart. I love the concept of a Wankel engine, and it would be awesome if they've overcome that particular aspect.

Derpdiherp 10y ago
Perhaps they could come up with some system like what we see in traditional piston engines piston seals - a replaceable seal that goes around the rotary triangular err piston for lack of a better word. If I remember correctly all the other wankel / rotary engines have just had a lump.
- alistairSH 10y ago
  Mazda's automotive Wankels have spring-loaded seals on the apex of the rotor and on the sides. See this image: https://hackadaycom.files.wordpress.com/2016/02/wankel-inside-kart-engine.jpg https://hackadaycom.files.wordpress.com/2016/02/wankel-insid...
  - Derpdiherp 10y ago
    Didn't know that - thanks for the information. Was it wear against the side of the casing that led to the issues then?
    - Grishnakh 10y ago
      I don't know, but just looking at that photo, and comparing to a traditional piston engine with its piston rings, I can see a bunch of problems: 1) there's a bunch of irregular shapes here to seal: each "chamber" has shorter seals in two places (the vertices), and two longer seals on the sides. So 4 separate seals per chamber (two of which are re-used for adjacent chambers). In a piston engine, each chamber has usually 2 or 3 seals, with one being the main seal (top of the piston) and the others being either a backup or an oil ring (for scraping off most of the oil). A circle is a much simpler design. 2) 4 seals means 4 places where there's going to be gaps where gases can escape between chambers, causing extra emissions. In a piston engine, there's only 1 gap in each piston ring, and you can rotate the rings so the gaps are not aligned. The sealing on piston engines is extremely good. 3) The piston engine is pretty simple really, as far as sealing and lubrication: oil is either squirted (some engines) or splashed up on the backsides of the pistons and drawn up on the piston walls to lubricate the rings as the piston rises, and then the oil is almost 100% cleaned off by the rings as it travels on the downward stroke. This mechanism works so well that modern piston engines burn a truly negligible amount of oil. Oil burning has always been a problem on rotaries, and looking at that photo it's fairly obvious why. Mazda (and others) has been working on rotaries with very smart engineers for literally decades, and they still haven't fixed these problems. I don't think they ever will; these problems seem pretty fundamental to the layout of a rotary engine. They had some real promise and advantages, but like many technologies, the small details made it impossible to beat incumbent technologies. Piston engines were bad too, ages ago, before they figured out a lot of small details like how to seal them well, how to optimize intake and exhaust valve design and timing, how to best design combustion chambers to maximize compression ratio and eliminate hot spots causing detonation, etc. But some problems just can't be overcome.

mschuster91 10y ago

> or true CVT transmission What I never understood: why don't do it like on a ship, attach a generator to the fossil-fuel engine, add a battery and supercap pack, and four small-ish wheel electric motors? That should save the weight of a gear-shift box, complex mechanics on FWD cars (because the axle transmitting the movement energy must be flexible), and it would allow the motor to run at its most efficient RPM count, as well as providing full torque from 0 km/h. In essence, a Tesla with a fossil-fuel engine in the frunk...

prodmerc 10y ago
I'm assuming all that weight still gets better mileage than what would basically be a petrol->kinetic->electric->kinetic convertor. Modern engines are pretty efficient, adding an electric stage would likely reduce that efficiency.
- mschuster91 10y ago
  It's not only about the weight. An electric intermediate circuit can: a) absorb braking energy - that's why I mentioned super-caps, because batteries can't ingest the vast amounts of energy that braking could theoretically produce (usually the brake power is 4x engine power, which means for a 250 kW motor a minimum brake power of 1 MW, and for sportscars likely muuuch more) b) keep the engine, if not in idle (i.e. you're stuck in traffic), at its most fuel-efficient RPM range - this is something any cheap-ass scooter does, and every diesel-electro locomotive and ship, but no car! Not to mention it isn't just about fuel-efficiency, but also many exhaust gas treatments only work at peak performance under very specific driving conditions (esp. temperature). The efficiency loss by conversion (97% and better in efficiency class IE4, see https://de.wikipedia.org/wiki/Elektromotor#Wirkungsgrad_und_Effizienz https://de.wikipedia.org/wiki/Elektromotor#Wirkungsgrad_und_...) is, taking especially the recovered braking energy into account, smaller than the efficiency delta gained by loss of hundreds of kg of weight (batteries are heavier, yes, but they can also replace "dumb weights" needed before to keep the car on the ground in the "elk test"). Also, less parts that are prone to (dirt) failure, e.g. the power-loaded joints on FWD cars, clutches, bearings, the entire gearbox system, differentials (in 4WD/multi-WD systems!). And you save on tires because in curves the individual motors can adjust speed. If done well, it should be pretty cost-efficient (less stops in the shop), too.