Tiff Needell's 1980 Rover V8 S SD1 Group 2 on our rollers
Discussion
Part 4
Compression Ratio
Increasing an engine's compression ratio makes it more thermally efficient. More of the energy contained in the fuel is released to do useful work. This translates directly into extra torque per litre. However if CR is taken too high on a given fuel octane then detonation can result and this destroys engines very quickly. The resistance of a given engine to detonation is determined by a number of things including combustion chamber size and shape, residual gas fraction, proximity of parts of the piston crown to the chamber surface, crevices where gas can lurk that is hard to ignite, turbulence, burn speed and squish. The same factors which make 4v heads scavenge well and burn fast also make them more resistant to detonation than average 2v road heads. In general a 4v engine in a similar state of tune as regards camshaft, induction and exhaust systems can take about 1 point more CR than its equivalent 2v engine.
Torque from higher CR is also a topic that can be quantified quite well because a simple equation gives us a good approximation of the thermal efficiency at a given CR. For spark ignition engines the equation is as follows:
Thermal Efficiency = 1 - (1/[CR^0.35])
There are many families of road car engines, mainly dating from the 1980s through to about 2000 that contain both 2v and 4v variants on the same basic block and crank so most of the engine geometry and other parameters don't change which makes them a very good comparison tool for the effects of just swapping to a 4v head and valve train. The VW Golf 4 cylinder engines in both 1800cc and 2 litre variants, Peugeot XU 1905cc and also the TU 1.6 litre. The Ford Zetec was a development of the 2v CVH engine although a few more things did change than just the head.
If we take the CR of an average 2v variant of one of those engines as 9.5:1 and the CR of a 4v variant as 10.5:1 we can calculate the increase in torque per litre just from the change in CR alone.
Thermal efficiency at 9.5:1 = 0.545
Thermal efficiency at 10.5:1 = 0.561
This gives a percentage change of 561/545 = 2.9%
So now we can put together the whole story about why stock 4v road engines produce about 10-14 percent more torque per litre than stock 2v ones of a similar family. Volumetric efficiency improvements which are mainly down to better chamber scavenging give about 4% to 5% gain. Combustion efficiency improvements from the faster burn and lower ignition advance give a similar amount and finally thermal efficiency from higher CR adds another 3 percent.
So far we haven't had to take frictional losses into account because these don't change much between 4v and 2v derivatives of similar road engines. However getting very high torque per litre from race engines means minimising friction losses as much as possible. We'll try and quantify those later to give an idea of what current state of the art design can achieve when we get deeper into how tuned engines differ in torque per litre terms from stock ones.
Compression Ratio
Increasing an engine's compression ratio makes it more thermally efficient. More of the energy contained in the fuel is released to do useful work. This translates directly into extra torque per litre. However if CR is taken too high on a given fuel octane then detonation can result and this destroys engines very quickly. The resistance of a given engine to detonation is determined by a number of things including combustion chamber size and shape, residual gas fraction, proximity of parts of the piston crown to the chamber surface, crevices where gas can lurk that is hard to ignite, turbulence, burn speed and squish. The same factors which make 4v heads scavenge well and burn fast also make them more resistant to detonation than average 2v road heads. In general a 4v engine in a similar state of tune as regards camshaft, induction and exhaust systems can take about 1 point more CR than its equivalent 2v engine.
Torque from higher CR is also a topic that can be quantified quite well because a simple equation gives us a good approximation of the thermal efficiency at a given CR. For spark ignition engines the equation is as follows:
Thermal Efficiency = 1 - (1/[CR^0.35])
There are many families of road car engines, mainly dating from the 1980s through to about 2000 that contain both 2v and 4v variants on the same basic block and crank so most of the engine geometry and other parameters don't change which makes them a very good comparison tool for the effects of just swapping to a 4v head and valve train. The VW Golf 4 cylinder engines in both 1800cc and 2 litre variants, Peugeot XU 1905cc and also the TU 1.6 litre. The Ford Zetec was a development of the 2v CVH engine although a few more things did change than just the head.
If we take the CR of an average 2v variant of one of those engines as 9.5:1 and the CR of a 4v variant as 10.5:1 we can calculate the increase in torque per litre just from the change in CR alone.
Thermal efficiency at 9.5:1 = 0.545
Thermal efficiency at 10.5:1 = 0.561
This gives a percentage change of 561/545 = 2.9%
So now we can put together the whole story about why stock 4v road engines produce about 10-14 percent more torque per litre than stock 2v ones of a similar family. Volumetric efficiency improvements which are mainly down to better chamber scavenging give about 4% to 5% gain. Combustion efficiency improvements from the faster burn and lower ignition advance give a similar amount and finally thermal efficiency from higher CR adds another 3 percent.
So far we haven't had to take frictional losses into account because these don't change much between 4v and 2v derivatives of similar road engines. However getting very high torque per litre from race engines means minimising friction losses as much as possible. We'll try and quantify those later to give an idea of what current state of the art design can achieve when we get deeper into how tuned engines differ in torque per litre terms from stock ones.
One nice detail on my (5v) camshafts is the varying lobe width. The base circle (no lift) runs on a strip about 5mm wide, which broadens out to the full lobe width before it starts depressing the valve. I don't know how much friction it saves, but I haven't personally seen another cam like it, although I'm sure it's not unique.
No reason why you couldn't do it to a 2v cam either.
No reason why you couldn't do it to a 2v cam either.
Years ago we had a bad dude cam (as The Donald would say). It was a design needing wider diameter followers than our standard B ones and the edge of the follower dug into the cam just each side of the nose 
. We could have put an MGC follower in but we decided to junk the cam as only fast road spec. Talking to the late Bill Nicholson, at a tuning seminar where I was answering questions etc, he said there was a work around which was simplicity itself. Tuftride the cam lobes and followers and it can't dig in, he said this worked to perfection when he was working at BSA. Bill used Jack French cams....another blast from the past! It just reminded me about friction on followers and made me wonder if the tuftriding would reduce frictional losses twixt cam and followers?
Peter
. We could have put an MGC follower in but we decided to junk the cam as only fast road spec. Talking to the late Bill Nicholson, at a tuning seminar where I was answering questions etc, he said there was a work around which was simplicity itself. Tuftride the cam lobes and followers and it can't dig in, he said this worked to perfection when he was working at BSA. Bill used Jack French cams....another blast from the past! It just reminded me about friction on followers and made me wonder if the tuftriding would reduce frictional losses twixt cam and followers?Peter
I believe Diamond Like Carbon (DLC) coatings are the in thing for reducing friction. No as simple as applying the same coating to both surfaces. There are different grades of coatings and there needs to be a careful selection as to which works with which and under what circumstances.
http://www.f1-forecast.com/pdf/F1-Files/Honda/F1-S...
http://www.f1-forecast.com/pdf/F1-Files/Honda/F1-S...
AW111 said:
One nice detail on my (5v) camshafts is the varying lobe width. The base circle (no lift) runs on a strip about 5mm wide, which broadens out to the full lobe width before it starts depressing the valve. I don't know how much friction it saves, but I haven't personally seen another cam like it, although I'm sure it's not unique.
No reason why you couldn't do it to a 2v cam either.
Is that 5mm width central to the main part of the camshaft lobe or on one side of it?No reason why you couldn't do it to a 2v cam either.
PeterBurgess said:
A good read and lots of google links thanks. Has anyone on PH had a cam and followers done and how much did it cost?
Peter
Probably have to pay most poeoples minimum charge for just one set. Don't need to do the cams too, have those REM superfinished. Followers can be superfinished too as a cheaper alternative to DLC. Sandwell UK are good for super finishing. Makes a massive difference on torque required to turn a cam.Peter
227bhp said:
That looks worse than the bottom of a 50yr old council house oven.
Japanese import (half-cut).The only stamp in the book was it's first service, but the oil filter was non-genuine, so I think it had at least one oil change since then.
The view in the photo meant that I had to strip & clean the engine before fitting it, but it cleaned up ok.
AW111 said:
Roughly central. Were you thinking of it as an aid to rotating the lifter?
Indeed I was 
However even thinning the contact area down acts as an aid to rotating the lifter because it removes that part of the lobe which acts closer to the centre of the lifter or even on the "wrong" side of it. I would suggest it has absolutely no purpose in reducing friction which is near as dammit nil when the valve is not open.
Mignon said:
GreenV8S said:
Well, that's a long-help misconception dashed.
Is it still the case that in practice, bigger (valves) is still better (engine breathing) for 2V? I have always assumed that a better flowing intake will mean less need to extend the valve duration to get volumetric efficiency up and hence less at the mercy of pulse tuning effects.
Total flow determines power potential i.e at what rpm the engine can still breathe efficiently. It doesn't much affect how much air the engine can take in at a single gulp though. When the cylinders are full they're full and that's determined by scavenging and pulse tuning.Is it still the case that in practice, bigger (valves) is still better (engine breathing) for 2V? I have always assumed that a better flowing intake will mean less need to extend the valve duration to get volumetric efficiency up and hence less at the mercy of pulse tuning effects.
If we were talking about liquid then full is full, but not so with air or gas.
Two inlets are better than one as I described it earlier, it's because the velocity and volume are higher. If everything is timed correctly then as the inlet valves are closing then air is still flowing in. If the ports are large and lazy then you could take a snapshot at a certain point (say BDTC) of 1 bar of external pressure and an in cylinder equivalent pressure of 1 bar, if they are proportioned properly (along with everything else being in harmony too such as camshaft spec and CR) then you could get an in cylinder pressure higher than 1 bar at BTDC due to Inertial ramming.
If it was the case that it was the exhaust valves and ports which were one of the governing factors as opposed to the inlets then we would be fitting bigger or two exhaust valves and one inlet and gaining power, yet we don't.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff