Recently several online car blogs were caught flat-footed, confidently declaring that the new Honda Type-R would come with a CVT.
I have no idea where this came from, but seeing as Honda has always said that it would be stick shift only, I felt that the subsequent pie in face moment when the rumour was confirmed to be false was well deserved.
But more interestingly, while there would be a predictable backlash against reducing driver involvement in using the least involved transmission type possible, from a performance perspective, some came out with voices of cautious approval. Having driven (and with some encouragement, flogged) a Honda CVT before, I do think they’d be well up for it.
So I’m reopening the query. Would you ever conceive that CVT could be appropriate for performance applications? Consider that combustion engines are under increasing pressure from a crop of sporty electrics…
Provided they can find people that can both program and actually know how to race/offroad/utility the CVT does show promise on paper. But from personal experience with a CVT in a SUV it was a little lacking in, for the lack of better words practical experience.
Well here in asia, 95% of drag race bikes are cvt based. So it does actually work. The only problem with cvt is really just the transmissiom efficiency right?
Wait. Yeah. Derp i mean the torque converter efficiency. Also that torque amount. Also another reason it’s used in drag bikes here. They don’t use a torque converter. They use centrifugal clutches
Well, electric cars have good performance potential because you have 100% of torque available from the beginning, and a CVT is designed to keep your engine at it’s peak torque, so I could see it’s feasibility. Plus without any conventional shifting, that’s a lack of interruptions in power delivery.
I really don’t like how any new cars that come out with a CVT use a “stepped CVT” which goes against the idea of using a CVT at all by having predetermined ratios set so the CVT will shift regularly jumping from one point to another rather than having a smooth transition.
Its designed to keep it at peak efficiency either by holding it at peak power for optimal acceleration or by keeping the engine at its most fuel efficient rpm no matter the speed.
I disliked CVTs for a good while, but my experience of them was limited to 1) my old Prius, which wasn’t a proper CVT and didn’t work like one, or 2) old japanese CVTs like those fitted to Nissan Micra K11s, which give you the full NEEEEEEEEE experience.
Assuming they were crap, I was by chance in a Subaru Outback diesel CVT for a while a few months ago, and I must say it was bloody excellent. Didn’t feel like a CVT at all other than the extreme smoothness, and while it had 7 “gears” it seemed to use those as a rough guide rather than being fixed to them. Didn’t scream it’s nuts off all the time either.
How it would translate to a performance application I don’t know, but if there were more defined modes (like above, plus the 7 rigid ratios) then I can’t see them being a particularly horrendous idea.
I have to say the turbo + cvt, I heard about it on paper and was like oh that has potential then I tried it and it really was very satisfying. The only problem is the throttle response lacks immediacy, obviously.
The preprogrammed ratios actually goes a long way to addressing this: when programmed well (and the Gen X Civic’s is), that problem pretty much vanishes. Whats the point of preprogrammed ratios in a cvt? Mostly versatiloty for when you do want preset ratios like gear braking or when you’re doing some spirited technical driving. This of course seems to undermine the original post of “cvt for performance”, unless somehow the cvt was programmed to keep the rpm steady even off the throttle. I dunno.
Though i can vouch, on a motorbike, a modern cvt is at least just as fast as a manual bike on a performance application.
Because despite it’s losses, keeping peak torque/power and no gear change makes for quite an experience on acceleration. But top speed is lower of course, and handling on corners is weird, you need to control the power by brake and not by throttle
Given electronic program, you have the option of having it stepped. So you can run it in CVT mode for efficiency and just planting your foot on it (which, for example, Honda Civic engineers probably assume you would most of the time), but if you want it to follow a specific rpm range instead of dropping to idle when you lift off, you can make it hold a ratio, which to me makes perfect sense if I want to drive through, say, a mountain pass in an enthusiastic fashion.
Hmmm wonder what a cvt would do in a drag racing context… i mean i use a 3sp auto but mates use 2 sp powerglides and a single speed box would help massively in case of wheel spin or having to lift off the throttle
Well yeah thats a bit of an issue with close to 700Nm lol. But for arguments sake a single speed gearbox that could regulate the rpm and torque linearly through out a quarter mile and control launch torque would make for very easy dial your own racing… but if they cant handle torque how would they work with big turbo/supercharger or electric motor applications?
It would have to be a bespoke system, the CVT torque limit mostly has to do with compromises for cost, economy and driveability. A thicker and tougher chain can be used but that limits the degree of ratios because it limited how small the pulley can go or it increases the physical size of the gearbox. Although this begs the question, if you have a CVT do you need big engines? Could you hook up a f1 style engine with 600+hp in the 15,000-20,000 rpm range but only making maybe 200-300nm up there.