23

New cars have stop-start system (even though this has been available for a long time). My question is: why are car companies emphasizing shutting the engine down at a traffic light, rather than making the engine run at close to zero RPM (but certainly not zero) at idle speed using the car's ECU?

My idea is that the car engine keeps running, and when the driver releases the gas pedal, the engine returns to a very low idle speed. In the manufacturer's current scenario, when the engine shuts down, the battery power is used and can put strain/stress on the battery.

I see cars have RPM meter that shows rotations-per-minute in thousands. Can't engine just run at 100 RPM, drawing very much less gas/petrol?

Others have similar concerns. And why don't companies make cars that have an engine that can stop just short of firing a cylinder?

| improve this question | |
  • 5
    You feel incorrectly. This is an old myth, much like the myth that turning off a light bulb for few seconds doesn't apparently save energy. The truth is, it may save the light bulb, but not energy to keep it on! – juhist Jul 17 '19 at 19:52
  • 6
    @juhist Well, it might be true if you include the energy required to make a replacement light bulb :P The same probably applies to engines. – Luaan Jul 18 '19 at 8:50
  • 1
    I think Harper's answer is right on: lubrication is one problem and also engines are not smooth; it feels smooth because of the higher rpm. More than 20 years ago, BMW had worked on some engines where some of the cylinders would be unused for cruising speed, so the concept is already out there, but I am not sure why it didn't take off; I just remember that from a magazine article. – Thomas Jul 18 '19 at 21:34
  • 3
    @Thomas: GM certainly tried this in the early 80s with the infamous Cadillac V-8-6-4 engine. Let's just say it didn't work out and leave it at that. – Bob Jarvis - Reinstate Monica Jul 18 '19 at 22:14
  • 1
    @Raj - Please don't add to a question after you've asked it ... this becomes very confusing to anyone who is reading it and the answers posted. If you need to ask a new question to clarify, please do so and link back to the associated question when you do so. Makes life so much simpler (and Stack Exchange as well!). Thanks! – Pᴀᴜʟsᴛᴇʀ2♦ Jul 19 '19 at 17:07
20

The oil problem

You can run any engine speed you want as long as you keep the oil pressure up.

Because of the impossibility of getting a roller bearing down a crankshaft, engine bearings are "plain bearings" with polished steel floating above babbitt (a soft metal). The oil pressure makes the bearings "float", so the dynamic forces of pistons being thrown to and fro is transferred by the oil, not hard steel-on-babbitt contact. The babbitt is only there for coldstarts and very occasional unintended bottoming-out. Frequent bottoming-out will destroy the bearings (it sounds a bit like dieseling).

So you must keep oil pressure high enough to float the bearings.

Super slow idling doesn't work on common off-the-shelf engines because they have a simple positive-displacement oil pump, whose flow is coarsely proportional to RPM. This means the lower the RPM, the less oil pressure available for "floating" the bearings. At very low speeds, there's not enough flow to keep up oil pressure, and you can bottom out bearings - especially at heavy throttle.

So, to idle at 100 RPM, you need to solve the oil problem. Large museum engines have an electric auxiliary oil pump, specifically to pre-fill the oil pressure lines (which have leaked back, and are now empty) so you get pressure immediately on engine start. But that is not enough to pressurize - for that, you need a much bigger pump. But it only needs to run at sub-idle.

That pump is going to be a problem. At too low a speed, it becomes an energy net loss, because you're spending more energy running the pump than the engine gives you. Which defeats the purpose of running the engine.

The power pulse problem

Let's say you solve the oil problem. Experience is that engine RPM must be 500-900 RPM for power pulses to average out enough not to annoy customers. You also don't want to hit a resonance frequency of engine, mounts, suspension or road structure (or airbag sensors, ha!) or it will tear them apart.

Others talk about this at length. I'm going to arm-wave it. Because I assume your engine will be direct-bolted to a large electric motor/generator, and you'll be able to use flux-vector control to actively dampen those power pulses: generating during power pulses and motoring the rest of the time, so engine RPM stays constant. You'll need to do that or this won't work.

| improve this answer | |
  • Another problem is that the rate at which gas burns is affected by temperature and pressure, which are typically expected to be reduced by the downward-moving piston before the fuel has been burning long enough for them to get too high. Normally, although a spark will fire before an engine reaches top dead center, the cylinder will be on its way downward by the time most of the fuel is burning. If the engine is burning too slowly, the fuel could all ignite before the crank shaft has turned far enough that the piston can exert its effort into turning the crank rather than stressing bearings. – supercat Jul 20 '19 at 20:42
24

You are right in the thinking of using less gas and that's exactly what the purpose was to have the start/stop feature. The main goal of doing it (at least here in the States) is for zero fuel consumption/zero emissions while the engine is stopped. This provides for better overall fuel consumption numbers to meet what the EPA has mandated for companies to stay in compliance with regulations. Considering what the government mandates, zero RPM will always produce zero emissions and have less fuel consumption than will 100RPM ... it all adds up.

WRT the start/stop causing wear/tear on the starter/battery which would otherwise not be there, you are accurate in thinking so. The manufacturer can do two things to help mitigate this that I can think of (whether they do them is another story):

  1. They can design in a more robust starting/battery system in the vehicle.
  2. They can (and do) engineer the engine so it stops just short of firing a cylinder. When the engine goes to start again, it fires this cylinder to assist the starter in getting the engine started, thus helping to eliminate the wear/tear on the starting system.

Solar Mike is correct about V8's vs 3-cylinder engines and relative speeds to allow it to operate. This is less a function of the size (displacement or # of cylinders) of the engine, but more so to do with physics. An engine has to have enough rotating mass to continue motion (the old mass in motion tends to stay in motion kinda thing). The larger the rotating mass, the easier it is to keep things going. If mass isn't present, it needs to spin faster to be kept in motion. Secondary to that is the # of cylinders and how many cylinders fire per revolution of the crank. A V8 fires four cylinders every revolution (every 90° of crank rotation), while a 3-cylinder only fires a cylinder every 240° of crank rotation (given both are 4-stroke engines). The reason I point this out is with larger engines, such as those which are used in large cargo ships, run at full speed ~120RPM. Idle would be something much lower. Point is, the rotating mass of the engine allows it to run much lower in RPM. You can't fit a Wartsila engine in a car, though.

| improve this answer | |
  • 4
    I've seem many large engines fitted into cars (Merlin into Rolls...) but the day a large cargo ship engine gets in a car will be a thing to see :)... When you see people standing on the piston crown.... – Solar Mike Jul 17 '19 at 18:33
  • @pᴀᴜʟsᴛᴇʀ2, can you please elaborate on "engineer the engine so it stops just short of firing a cylinder" – Raj Jul 17 '19 at 18:57
  • 1
    @Raj the ecu knows exactly where the engine is in its cycle at all times... – Solar Mike Jul 17 '19 at 18:59
  • 2
    @Raj: Stops with a cylinder ready to fire. (Full of fuel+air, and already compressed.) For N cylinders, there are N/2 points in each rotation where this is the case for 1 cylinder. So when it wants to restart the engine, it can fire that spark plug. – Peter Cordes Jul 19 '19 at 18:01
8

The engine needs a minimum rpm to idle smoothly - a v8 can idle smoothly about 500to 600rpm but a 1 litre 3 cylinder needs around 700 to 900rpm...

It is not the companies that really decided to do the start/stop, if you do some research the various governments have implemented legislation over the years that required drivers to turn off the engine when waiting at lights to reduce smog and fumes. The companies designed the start/stop systems because some drivers never bothered... it must be too difficult to turn the key sometimes...

| improve this answer | |
  • 1
    It's not only V8: my old 1.33 litre 4-cylinder in Toyota Yaris used to actually idle at 500rpm in many conditions. Not always, though: there was some kind of automatic idle speed adjustment based on conditions. – juhist Jul 17 '19 at 18:21
  • 1
    @juhist I know, used to work on many cars and two "identical" cars (same engine 1.5 litre 4 cyl, box etc) side by side and one would idle at 500rpm the other no chance... – Solar Mike Jul 17 '19 at 18:24
  • 5
    In California traffic if you had to turn your engine off every time you got stopped in traffic you'd get a repetitive strain injury @SolarMike. – GdD Jul 17 '19 at 18:25
  • 9
    The important design feature of the system is not that you don't need to turn off the key, but that you don't have to do several distracting additional tasks when you move off again, i.e. turning the key, listening to the engine to hear when it starts, releasing the key again, and then moving you hands back to the steering wheel. You should be concentrating on situational awareness at that point, not operating multiple controls. Depressing the clutch pedal automatically deals with the engine restarting procedure. (I guess the system works in a similar way with automatics.) – alephzero Jul 17 '19 at 18:28
  • 2
    @alephzero with my car there's a whole bunch of additional annoying things: readjust the radio's vloume (it will start pretty low after starting the engine), wait for (electrical) power-steering to come up again,... – piet.t Jul 19 '19 at 6:45
5

From physical point of view, lowering the RPM means that you need to store the energy required to keep the engine running (and overcome friction) with a smaller angular speed. And the kinetic energy is proportional to speed squared.

If you want to go from 600 RPM to 100 RPM, you will have (among other things) to make the flywheel 36 times heavier. Of course, this is a slight exaggeration as friction also goes down with the speed, but the useful work a flywheel has to do (compression, operating the valves etc.) doesn't change much.

By the way, stopping just short of firing a cylinder helps restarting the engine, but firing a single cylinder is not enough to get the engine from 0 to 600 RPM. Such a trick makes the job of the starter easier, but you still need a starter.

A few thoughts regarding the Stop/Start: it doesn't save fuel with every stop, but it does save fuel on average. Typical traffic lights work in cycles of 30-60 seconds, so on average you stop for half of that time, 15 to 30 seconds. There are cases where it doesn't make sense to stop the engine, like slowly moving traffic jam, or when you need AC and the AC compressor is driven solely by the engine. That's why you have that button to disable Stop/Start.

| improve this answer | |
  • You could consider increasing the mass... Think of some of the victorian slow speed engines with flywheels that are 2 metres in diameter. – Solar Mike Jul 18 '19 at 10:18
  • 1
    @SolarMike "to make the flywheel 36 times heavier" means exactly that. Or did you mean to say "You could consider increasing the diameter"? – Dmitry Grigoryev Jul 18 '19 at 10:21
  • 1
    @SolarMike You could use depleted Uranium :) I didn't actually mean to suggest that having a bigger flywheel is a good solution, I meant to illustrate why start/stop is used. – Dmitry Grigoryev Jul 18 '19 at 10:47
  • 1
    @Raj It's just an example of course, but I tried to pick a reasonable number. Different engines idle at different RPM, and even the same engine will idle at a different rate while it's cold. – Dmitry Grigoryev Jul 19 '19 at 6:51
  • 2
    @Raj A cold engine need to idle at a higher RPM because the oil more viscous. – Dmitry Grigoryev Jul 19 '19 at 12:12
4

Car companies focus on stop/start of the engine simply because

  1. There's a minimum RPM the engine can run at. You can sort of extend it using dual inlet/exhaust variable valve timing, but there are limits to that.
  2. If the engine is running, it's consuming some amount of fuel always.

Also, some companies go bit further than stop/start. They stop the engine when coasting, in a mild hybrid, typically using a 48-volt system.

Some companies go even further than mild hybrids, and have a full hybrid that can turn off the engine at significant speeds (up to 80 km/h or so), when the load demand is low, supplying the demand entirely from the battery.

The best hybrids have a separate electric coolant pump to supply cabin heat when engine is off, and an electric air conditioning compressor to cool the cabin when engine is off.

I'd say the future is clearly one of hybrids and electric cars. Your low RPM idea has no place in this future. Furthermore, at very low RPMs the oil pump might not be good enough because they are designed to operate at normal idling RPMs.

| improve this answer | |
  • can you please expand on "at very low RPMs the oil pump might not be good enough" – Raj Jul 17 '19 at 19:32
  • 2
    Well, the oil pump may not be pumping enough oil if operating at a very low RPM. The oil pressure can be too low for the engine to work reliably. Too low oil pressure, engine damage. – juhist Jul 17 '19 at 19:50
  • 1
    @Raj An electrically-driven oil pump will be just fine, but that's a fairly recent thing. A mechanically-driven oil pump (like mine) depends on the engine speed. Above a certain speed, a pressure-relief valve returns some oil back to the sump without going through the engine. This is actually the normal operating pressure, so the threshold speed is quite low: close to the designed idle speed. Slower than that, there isn't enough oomph to keep the designed operating pressure, so you may start to have metal-to-metal contact where there is supposed to be a pressurized film of oil. – AaronD Jul 18 '19 at 5:27
2

I have the same feeling that @dmitry-grigoryev has said,

it doesn't save fuel with every stop, but it does save fuel on average. Typical traffic lights work in cycles of 30-60 seconds, so on average you stop for half of that time, 15 to 30 seconds. There are cases where it doesn't make sense to stop the engine, like slowly moving traffic jam, or when you need AC and the AC compressor is driven solely by the engine. That's why you have that button to disable Stop/Start.

That is why I access if at the coming traffic light if Car will be stopping for long or for less than 10 seconds, if it is less than 10 seconds, i disable the Stop/Start. If more than 10 seconds, I let the car engine be stopped . The button that the car manufacturer are putting is on the side, if the button for Stop/Start is on the steering wheel that will be better.

But Question is right in raising the awareness about alternative means of saving fuel without hurting the Car's Battery or overall engine with frequent full stop/start.

Few points that worth investigating ( I understand future may be Hybrid/Electric), but

  1. Engineer the engine so it stops just short of firing a cylinder. When the engine goes to start again, ECU fires this piston to assist the starter in getting the engine started, thus helping to eliminate the wear/tear on the starting system.
  2. Making some Cylinders Stop
  3. keep oil pressure high enough to float the bearings using dual inlet/exhaust variable valve timing or electrically-driven oil pump
  4. When a Car stops at Traffic Light, as A cold engine need to idle at a higher RPM because the oil more viscous, so in the beginning the engine will be hot, So reduce the RPM by extra 10-15% and then after 30 seconds, bring the RPM higher ( as engine might have become colder)
| improve this answer | |
1

My son's 2014 328i x-drive stops itself, and restart upon lifting off the brake. My 2016 328D (turbodiesel) x-drive does the same. Both are 2L with turbo. I go thru a particular light that is red for 90 seconds in my direction, I appreciate not seeing my fuel mileage dropping for 90 seconds, and sometimes a 2nd cycle, while sitting there waiting to cross. During the cold part of winter, and when the AC is running in the summer, it seems to restart after just 60 seconds.

I'd much rather have the engine just stop vs trying to lug along at some very slow RPM. I have small plane with 360 cubic inch engine (I think that works out to around 5.9L), and it will idle around 500 RPM after landing with the throttle pulled out and rolling slow, but it makes one wonder if it's going to stop. I think having that big 86" propeller in front might help to smooth things out. It does sound much better at 700-800 RPM.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.