Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
Please note that it is not permitted to have referral links posted in your signature. Keep these links contained in the appropriate forum. Thank you.
https://www.boards.ie/discussion/2055940817/signature-rules
https://www.boards.ie/discussion/2055940817/signature-rules
Changing Gears at a High Rev Count?
Options
Comments
-
-
-
By the way, this is completely wrong. You want to use plenty of RPM below max torque as well as above, to keep the average torque over a certain time as high as possible. And power is an archaic term that has nothing whatsoever to do with acceleration (besides the fact that they're both derived from torque).SouperComputer wrote:an engine is at its most effective if gear is enagaged at max torque and changed up at max power, this is what is referred to as the powerband in racing.
Considering this will leave you nowhere near your example torque peak at 3,500rpm, you'll be no-where near maximum acceleration. In reality, 5th and 6th gear would still drop you below 3,500 so you can make use of all torque.as road speed increaces, empahsis is no longer on acceleration due to frictional and aerodynamic losses, so the gears become "shorter". Eg 5th and 6h may only drop to 4500 and 5000 RPM respectively when engaged.
Okay, this is going to need some proper explanations. Just completely ignore power for a moment. Torque is turning force. Using wheels, we can convert turning force to linear force. Let's make a ridiculously arbitrary example car with a wheel of radius 1 metre and an engine that produces flat torque of 100Nm (Newton-metres) throughout the entire rev range. Ignoring any frictional forces (and there are LOTS), if the engine were directly connected to the wheel with no gears etc. (i.e. crankshaft = axle), it would be applying a forward force (F) of 100N (Newtons are the unit of force) on the car. Since our magic theoretical car weighs (m, for mass) just 100kg, the acceleration (a=F/m) would be 1m/s/s, "one metre per second, per second)". So the car would be going 1m/s after 1 second, 2m/s after 2 seconds, etc.
If the engine has a maximum speed of 10,000rpm (~167 revs per second), then the car will accelerate constantly up to 1047m/s (167 wheel revs per second times 6.28m circumference of the wheel) and it will accelerate no more, because torque above 10,000rpm is nil.
So what if we want to go faster than this? We connect the crankshaft to the axle via gears. If we have a big gear on the crankshaft turning a smaller gear with half the circumference (or half the teeth, as gears are usually measured), then at 10,000 engine rpm, the road wheels will be turning at 20,000 rpm and the car will be traveling at 1047m/s. However the gears also half the torque applied to the road wheels, thus halving the linear force applied by the road wheels to the ground, and thus halving the acceleration of the car (down to 0.5 m/s/s).
It would be great if we could get the best of both worlds, and this is pretty much what a gearbox allows us to do. We use 1st gear (the gear on the crankshaft is the same size as the one on the axle) to accelerate all the way up to 10,000 rpm, then we change to 2nd gear (crankshaft gear is twice the size of axle gear) and the engine drops to 5,000rpm to continue accelerating (albeit at half the rate) all the way up to 20,000 road wheels rpm. We can slap on a third gear (crankshaft gear is three times larger than axle gear) to get us all the way up to 30,000 road wheels rpm (3141m/s) but by now we're only accelerating at 0.33m/s/s.
Let's stop thinking about torque and rpm and all that rotational stuff for a moment and think merely in terms of linear thrust. The engine can get the wheels to push us forwards with a constant 100N of thrust. Let's reintroduce friction back into the equation. There are lots and lots of frictional forces but the biggest single frictional force (at speed) is air resistance. Another big one is friction inside the engine, but let's pretend that our torque curve already takes this into account, so we can ignore it, and the rest of the frictional forces are (for the purposes of this explanation) negligable. Air resistance is a function of speed (and air density and the car body's coefficient of drag and its frontal area and stuff, but let's say all that is constant) alone. Let's say air resistance is equal to the square of speed. At 1mph (notice we've changed units now. we've changed cars too! bet you didn't notice that! it doesn't matter though) air resistance is a 1N force working in the opposite direction to thrust, so it's basically sapped 1N off our 100N of available available thrust, leaving us with a net forward force of 99N. Our 100kg car is now accelerating at 0.99m/s/s. However up at a whopping 10mph, we're facing 100N of air resistance. This saps our entire available thrust, so the net forward force is 0N. Since a=F/m, we're no longer accelerating and stuck at a constant 10mph, regardless of the car's weight.
Think of thrust (which is directly derived from torque) as income. Air resistance (directly derived from speed) is expense. Then acceleration (directly derived from net forward force) can be considered profit
As long as our income exceeds our expenses, we're making profit, and accelerating.
Now let's make things more confusing by considering a more realistic engine, that has a non-flat torque curve. If the engine of our original example car (the one with three gears) produces peak torque at 5,000rpm, then it produces peak thrust at the road wheels at 5,000rpm, regardless of anything else. Let's pretend that torque drops in straight, diagonal lines either side of the peak, so we get only 90Nm at 4,500 and 5,500rpm. we get 80Nm at 4,000 and 6,000rpm, 70Nm at 3,500 and 6,500rpm, etc. Draw it on graph paper if you must
In first gear (1:1 ratio) it will produce peak thrust and thus peak acceleration at 5,000 engine and road wheels rpm. Once the engine and road wheels are up at 9,000rpm, we're getting just 20Nm of torque and hence 20N of thrust and hence 0.2m/s/s acceleration in our 100kg car. However, in 2nd gear (2:1) at 9,000 road wheel rpm, the engine is doing 4,500rpm, and so is outputting 90Nm of torque! Remember the gearing halves the road wheel torque, but even so, there are still 45N of thrust (0.45m/s/s acceleration) at the wheels. This is way more than the measly 20N we were getting at the same speed in first gear, so let's upshift! SouperComputer, Note that we're upshifting before the torque peak, and still getting better acceleration. In fact, we're upshifting too late. At 8,000 road wheel rpm, in first gear, the engine is turning at 8,000rpm and outputting 40Nm. At the same road speed, in second gear, the engine is turning at 4,000rpm, outputting 80Nm, which is reduced by the gearbox to 40Nm at the road. At this precise speed, we'd be accelerating at exactly the same rate in both first and second gear. Anything below this speed, and first gear will give us better acceleration. Anything above this speed, and we'll be accelerating more quickly in second gear.
So just to sum up this particular gearchange. For maximum average acceleration, we change from first to second gear at 8,000rpm down to 4,000rpm. Notice that 4,000rpm is not peak torque, which is where souper computer would have you change down to, and more importantly, notice that i haven't mentioned power anywhere in this tutorial. That's right, power has nothing to do with acceleration. So you'll see that finding the best acceleration and gear change points isn't as simple as looking at peak figures. You have to see the full graph of torque vs rpm (preferably wheel rpm. you can work this out with engine rpm and a list of your car's gear ratios) and just find out where the curves for two adjacent gears overlap (just like causal sorta said).
I'll give a lollipop to the first person to figure out the optimum RPM at which to upshift from 3rd to 4th gear 0 -
-
-
Advertisement
-
AMurphy wrote:However, there are such objects are “close-ratio” gearboxes, why would you want (not want) one.?
Here we go. With a normal gearbox, the shift points and gear ratios are designed to keep the engine revving between the cyan lines to keep the average engine torque fairly high, i.e. by the time you reach the upper cyan line, the next ratio is putting more torque to the road, so you need to change up there. They're also designed to keep you happily cruising at a comfortable and healthy 2,500-3,000rpm at 70mph.
A close ratio gearbox has the ratios closer together, so by the time you reach the upper green line, the next ratio is already putting more torque to the road, so you change up much sooner and the result is keeping the engine torque much higher on average.
The downside to this is that you run out ratios (and hence top speed) much sooner than with a normal gearbox. This is generally okay when you're racing, because you care about accelerating flat-out between corners, and will never be able to reach the maximum speed provided by normal ratio boxes even on a 1 mile back straight.
You can also overcome this problem, to an extent, by stuffing more total ratios into the gearbox. Most close ratio gearboxes are 6-speed. But eventually you reach a trade-off between keeping the engine at optimum torque and torque lost due to the time spent changing gears.
And this is where infinitely variable ratios come in. These are something I dreamed up as a teenager, and would love to have got the patent on, even though I had no idea how to practically implement it
They use just one gear, but you can actually vary the size of the engine-side gear in relation to the wheel-side gear, so you get an infinite number of ratios between 3.x and 0.x. The idea is that you can now keep the engine at the exact torque peak (on the yellow line) through the entire acceleration. In passenger cars, they do this if you floor it, but for normal driving, they can keep the revs at the exact point where the engine is most fuel-efficient throughout the entire acceleration, which is also very useful.
The downside with these is that in their current implementation, they sap even more torque from the wheels than even a regular automatic transmission, and this is definitely not good. In addition, because of the prominence of the 5-speed gearbox, engines have long been designed to have nice broad flat torque curves so you don't have to change so often to find acceleration. In these cases, the infinitely variable transmissions offers little if any improvement over a normal 5-speed. However, they do open the door for engine developers to focus on designing very "peaky" (that is, very high maximum torque but over a very small rev range) engines, because with infinitely variable transmissions, the engine is never away from the absolute peak. So I wouldn't buy one today, but eventually they should be a lot more useful than what we're used to.
Notice that my diagram didn't take into account that changing from 1st to 2nd happens at different rpm to changing from 4th to 5th. It would have been too cluttered :P0 -
-
-
-
-
Advertisement
-
-
-
-
causal wrote:I don't know what F1 engines are made from - alu-kevlar-carbon-magnesium - anyone know?
This is the best I can get ya:5.5 Engine materials:
5.5.1 The basic structure of the crankshaft and camshafts must be made from steel or cast iron.
5.5.2 Pistons, cylinder heads and cylinder blocks may not be composite structures which use carbon or aramid fibre reinforcing materials.AMurphy wrote:A F1 engine may have smaller diameter crankshaft/conrod bearings than a std car of similar displacement, why?.
The piston bears a greater load, so if it were a tractor/lorry where piston loading is also very high, large bearings are used. Why not in a F1?
Each cylinder of an F1 3.0l V-10 will be smaller than a road-going 3.0l V-6. The stroke is shorter, so the crankshaft must be smaller diameter.
Piston loadings in a formula 1 engine is not that high. Much of its torque is generated by the frequency of explosions at 17,000rpm compared to the truck's torque generated using fewer (1,500rpm) explosions but each one being much more powerful and causing more thump on the piston.
Yes, there's obviously lots of force inside an F1 engine, but they design everything to be as light as possible (low mass = low inertia = change direction very quickly = high rpm) while still meeting certain strength requirements. F1 engines have very small tolerances in their timing (intake, spark and exhaust) at huge speeds to keep torque coming well past 18,000rpm without needing massive force per ignition. Also, cost is less important, so they probably use smaller but better designed bearings. Dunno much about that, though 0 -
-
Balfa wrote:This is the best I can get ya:
Each cylinder of an F1 3.0l V-10 will be smaller than a road-going 3.0l V-6. The stroke is shorter, so the crankshaft must be smaller diameter.
Piston loadings in a formula 1 engine is not that high. Much of its torque is generated by the frequency of explosions at 17,000rpm compared to the truck's torque generated using fewer (1,500rpm) explosions but each one being much more powerful and causing more thump on the piston.
Yes, there's obviously lots of force inside an F1 engine, but they design everything to be as light as possible (low mass = low inertia = change direction very quickly = high rpm) while still meeting certain strength requirements. F1 engines have very small tolerances in their timing (intake, spark and exhaust) at huge speeds to keep torque coming well past 18,000rpm without needing massive force per ignition. Also, cost is less important, so they probably use smaller but better designed bearings. Dunno much about that, though
Bearing diameter, not crankshaft diameter. give it another shot.
even if the cylinder were a dead one, the forces of turning the piston, etc about at 18K is still to be considered, like you indicate keep everything light, very light, probably too light for the abuse expected in the average mom-mobile.
I'd expect the components are manufactured and selected to much tighter tolerances than your average Punto.0 -
AMurphy wrote:....
I'd expect the components are manufactured and selected to much tighter tolerances than your average Punto.
Puntos have tolerances???
(only kidding) 0 -
Pretty easy to do in a vacuumCapt'n Midnight wrote:that's about 5,853mph ! :eek:
Whoops. misunderstoodBearing diameter, not crankshaft diameter. give it another shot. I dunno, really. Let's put it down to manufacturing tolerences, though
That's where NASCAR gets all their power. Granted, they use big 5.x litre engines, but they still get well over 100HP per litre while using 1950's technology. No fuel injection, no DOHC, no variable timing, etc. AND they have huge intake restrictors! They just have extremely tight fabrication processes. It's an impressive feat 0 -
Balfa wrote:Pretty easy to do in a vacuum
Whoops. misunderstood I dunno, really. Let's put it down to manufacturing tolerences, though
That's where NASCAR gets all their power. ... No fuel injection, no DOHC, no variable timing, etc. AND they have huge intake restrictors! They just have extremely tight fabrication processes. It's an impressive feat
NASCAR can play about with valve lift, porting, valve overlap and "light" as you pointed out previously.
As for the bearing size.
The film of lubrication in the bearing is subject to shear. The higher the relative velocity between the rotating shaft and the stationary bearing the greater the shear and the more detrimental to the life of lubricant and bearing.
So, to lower the relative velocities of fixed and rotating bearing, it has to be made smaller to insure it will survive.
So lets say you have your car with 2" bearings at 8K, you'd need 1" bearings at 16K to maintain the same bearing surface velocity.0 -
-
Advertisement
-
-
css wrote:Those 1.8's in the passats are ancient, so you'd be wise not to rev them too much, they might fall apart alright..
Like was said, all engines are different. No harm in revving any jap engines once warm!!!
What are you on about. The same basic block is in the Mk2 GTI. Pretty much bomb proof.0 -
-
This discussion has been closed.
Advertisement