Why are most petrol V6 engines so terrible? | Auto Expert John Cadogan

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
the most common v6 petrol engines are terrible and i'm looking at you kluger santa fe sorrento but why are they so emphatically bad details next i'm john hogan from autoexpert.com dot hey you and i get new cars cheap for buyers here in australia dirt by sea all let us rejoice etc website for that obviously the saving not the gert by rejoicing bizzo or you can just click the card that's uh up there now dude like do it now otherwise the universe might end in a flash of entropy and if that happens none of us will get to see the second coming of jesus which if my calculations are correct is scheduled for next wednesday at tea time in [Music] missouri as usual provided you clicked here is a multifaceted crap v6 question inspired by a dude just like you only named remy you have spoken about how v6 engines tend to be high revving engines i.e peak power occurs at higher revs compared with other configurations is there something inherent about the v6 that lends itself to high revs and prohibits better performance at low revs what is different about a v6 i would appreciate your five cents worth firstly and frankly this is the biggest factor v6s the petrol ones are generally old engines manufacturers are moving away from v6s and into fours usually with turbocharging and direct injection and incremental improvements in combustion management between when that v6 was designed and today mean that they can get away with a four-cylinder engine and still deliver the performance required for the particular platform and it does tend to be cheaper to make four-cylinder engines obviously because they only require half the number of heads and half the number of camshafts and two-thirds of the number of pistons and less machining in the block and less oil pump capacity and fewer losses through the driving of the cams and things of this nature etc even when you add a turbocharger and an intercooler and a direct injection system it's still going to be heaps cheaper to make a four-cylinder and it's so much easier to package that in the engine bay especially if your engine bay if you're out there in r d making a new car for this or that especially if your target engine bay needs to accommodate a diesel and a petrol engine because the v6 petrol tends to need a wider space than the diesel's inline four typically this characteristic of v6s okay being gutless at low revs and needing the crap revved out of them to perform is mainly due to them generally not having direct injection and also lacking a turbo it's less to do with the intrinsic configuration of the v6 a multi-point atmo v6 is pretty much a brute force option of existing componentry using an old engine and saving the big bucks in r d that would otherwise need to be spent developing a brand new turbo four that would in many ways be superior the cost of engine development is stratospheric right so there's a a lot of latency in this proposition out there in the new car market there's lots of old v6s especially in seven seat suvs for example the sorrento 3.5 v6 petrol engine which in my view urgently needs pensioning off and replacing with the turbo 4 and its cousin over at hyundai the santa fe it's pretty typical of the breed it's port injected which is old tech and atmo induction and it makes 200 kilowatts at a staggering 6300 rpm most people rarely rev a vehicle such as that quite this high so that power is largely hypothetical in vehicles such as that mazda's 2.5 turbo 4 is direct injected and turbocharged and it makes 170 kilowatts at 5000 rpm which is so much more accessible you can actually do a pretty neat engineering beer garden type trick with these numbers what you need to do is get out yo calculator and divide the peak power in kilowatts by the capacity in liters and then divide that by the number of thousand revs at which that power occurs okay for the sorrento it's going to be 200 kilowatts divided by 3.5 liters of capacity divided by 6.3 000 revs which is where peak power occurs and the number that falls out on yo lcd display is 9.1 and i have to admit they're pretty dodgy units kilowatts per liter per thousand rpm but for the mazda which has a turbo and direct injection it's 13.6 kilowatts per liter per 1000 rpm it means the 3.5 v6 in the sorrento is generating 9.1 kilowatts for every liter that it displaces for every thousand rpm required to turn it at peak power and the mazda 2.5 t is therefore doing basically 50 better there's this thing in engineering sort of engine design and it's called bmep brake mean effective pressure and i've spoken about it before on this channel and this calculation that i just outlined is essentially a bastardized version of that without all the conversion factors and the caveats like the ghetto engineering equivalent of bmep accessible to anyone engine designers have a few options okay to meet the target output they can just spin the engine faster to make more power because power equals torque times revs so more power is available to you if you just spin the engine faster provided you can maintain the torque at that elevated rpm but this approach is admittedly quite hard on mechanical components and it makes the power less accessible for ordinary drivers like non-pathological rev heads and it does tend to hurt low end performance like low and mid rpm power delivery which is something that really matters out there in the real world driving daily or a designer can just make the engine bigger that's pretty obvious as well because bigger engine makes more torque because bigger engine equals more airflow equals more combustion per cycle equals more torque at the crank that tends to be fairly inelegant also as solutions go especially in a regulatory and community sentiment environment where efficiency and emissions are tightly regulated and increasingly highly regarded alternatively they can make fundamental improvements to combustion dynamics and combustion management and volumetric efficiency and things of that nature direct injection does the combustion management voodoo and a turbocharger improves volumetric efficiency so essentially what you're looking at with a v6 is a window into the past like 20 years into the past in many cases and the superior performance especially at low and medium revs where most people drive of smaller more modern turbocharged direct injection engines is due to fundamental improvements in technology i am assuming that this is by design and that the characteristics are invariant of the number of cylinders but this is not reflected in engines out there in service while v6s are almost invariably high revving many v8s particularly diesels are not like the current land cruiser v8 diesel okay so quick point of order on this you cannot compare petrol and diesel engines directly because the different fuels have vastly different combustion dynamics diesels have significantly higher cylinder pressures during compression all modern diesels are turbocharged and direct injected and in fact diesel engines adopted these technologies before petrol plus diesels are intrinsically more thermally efficient than petrol because the mixture tolerates greater compression and more compression equals expansion through a greater range during combustion equals more thermodynamic efficiency like 20 more more or less all old petrol multi-point atmo engines were bad in comparison to modern engines like the honda 1.8 i four cylinder inline engine in the civic is a relic and it's emblematic of honda's embalming during the gfc right that engine makes 104 kilowatts at six and a half thousand rpm which is 8.9 in bastardized bmep which is slightly more awful than the current v6 sorrento santa fe engine and the honda engine needs to rev even higher to get the job done emblematic of that old tech let us go back in time now to something like a model year 2000 v8 holden commodore 225 kilowatts out of the box summon ats here we come 5.7 liters spinning at 5200 rpm with atmo induction and multi-point injection and yet just 7.6 in bastardized bmep if you run the numbers even if you mod it so that it revs to 6000 as a kind of thought experiment and you maintain the torque production at peak power you'd notionally get a proportional increase in peak power to something like 260 kilowatts but the efficiency would remain the same you'd still be stuck at something like 7.6 which is comparatively crap let us step up the summonet's rankings somewhat and go to 2005 hsv grange it was crap too at least thermodynamically crap went okay six liters atmo induction multi-point injection 297 kilowatts 6000 rpm burnout master class here we come but 8.25 in bastardized bmep which is awful efficiency frankly the reason they typically didn't get criticized for needing a big rev and for performing so badly per unit volume frankly was a nearly 300 kilowatts is kind of a big band-aid when it comes to media criticism b motoring journalists typically are not mensa candidates or that technically cognizant and c criticizing holden and or hsv is the motoring journalism equivalent of apostasy just ask me i've done it on national television a few times pro tip okay you have to be super careful doing this bastardized bmep kind of analysis with petrol engines tuned for different octane ratings like different rom numbers this is generally because higher octane engines typically deliver more compression because octane rating is all about resistance to uncontrolled combustion and that greater compression yields greater expansion and higher efficiency by virtue of intrinsic combustion dynamics this v6 thing is really just an age artifact like a technology artifact it's like a cd compared to streaming your music or something big atmo multi-point engines are dinosaurs it's not their cylinder size or geometry that's bad the cylinders in the vast majority of passenger engines are in the ballpark of half a liter to 0.6 liters and the optimal size is half a liter per cylinder so it's common to see 1.5 liter three cylinders two liter fours three liter sixes and four liter v8s at least it is these days my guess is that dimensional constraints which are usually the cause for a v configuration in the first place are prohibitive to the length of the stroke thus necessitating an over square design to achieve the desired displacement and thus meaning the engine has to go comparatively faster to make power i'm going to respectfully disagree with you there remy the big saving from v configurations is length length of the block seeing as in a v6 the cylinders are more or less side by side but if they were in line they'd be lengthwise so you save a hell of a lot on length not all that much on height the cylinder banks are even canted over by virtue of the 60 degree v right or the 90 degree v if the design is especially once over lightly so no dude the stroke is not especially constrained in a v6 because that extra height is over there like that in the v right most v6 petrol engines like kluger sereno santa fe etc they are over square however in the ballpark the boar is in the early 90 millimeters like 93 or 94 or something and the stroke is in the early 80s like about 84 versus the mazda 2.5 t at 93 by 92 so it's pretty square what we're talking about here is roughly nine millimeters in stroke difference okay of which only four and a half millimeters is going to be the height of the block like above the center line of the crank and it's candid over at 30 to 45 degrees per bank so this is hardly a packaging consideration it's less than a quarter of an inch miracle it's much more likely to be delivering the high rpm performance required in order to achieve the revs that are required to deliver the target output power okay so i think remy's proposition here is a little chicken egg kind of reversal i am at a loss because big medium speed diesels as used for motive power in ships and diesel electric locomotives are of a decidedly under square design and at full power at as little as 800 rpm i realize this is a different kettle of fish but still the same principles apply the spatial constraints are different interestingly i did my cadet ship in mechanical engineering in the railways so i have worked in the factories rebuilding those huge v12 and v6 diesels and i've walked through operating locomotives cranking out their maximum power uphill with those huge roots superchargers whistling their tits off on two-stroke gm v16s anyway if memory serves and that is a real hair on the back of the neck thing to do it's pretty awesome stuff right even if the engines are only doing 900 rpm or something it's pretty impressive when you're inside the locomotive the reality is however that only small engines can do big revs without turning themselves into scrap metal in other words engines aren't scalable you can't run a 200 liter engine at the same maximum revs as a 2-liter engine it'll break much sooner than that i'm just guessing this next bit i also suspect there will be flow dynamics type considerations in play as well unless it happens to be an inline 400 cylinder engine or something which is a nice idea but self-evidently absurd huge engines in ships and locomotives make peak power at lower revs because they simply will not tolerate spinning any faster the rotating parts are huge dude don't get in their way if they let go so in a nutshell when you look at a v6 engine in a passenger car like an suv or something you are really just looking deeply into the past plus car makers do work hard to appease america and not to stereotype but moroccans do love their large engines and they are kind of addicted to gasoline in a sense they really don't appreciate what they're missing out on and what the rest of the world has been doing for quite some time now
Info
Channel: Auto Expert John Cadogan
Views: 205,628
Rating: undefined out of 5
Keywords:
Id: GYARuoa_xvg
Channel Id: undefined
Length: 17min 55sec (1075 seconds)
Published: Mon Nov 08 2021
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.