42 replies to this topic

[MatsNorway]

http://www.speedhunt...orts-shop-tour/

Edited by MatsNorway, 11 July 2012 - 18:09.

[bigleagueslider]

Looks like more trouble than it's worth.

[Magoo]

Variable-length inlet tuning the hard way.

[Greg Locock]

This was in production for a few years

http://media.photobu.../DSCF2280-1.jpg


Switchover speed was around 3500 rpm, that's the vac actuator at the front.

[Kelpiecross]

This was in production for a few years

http://media.photobu.../DSCF2280-1.jpg


Switchover speed was around 3500 rpm, that's the vac actuator at the front.

What engine is this?

[bigleagueslider]

Variable-length inlet tuning the hard way.

It's not really a "variable" system. Instead it's more of a "binary" system. The relative change in tuned length between the two inlets is also a bit excessive. Even for a very high rpm F1 engine, the length of the lower inlets looks much too short to be of benefit. I think it's just a poorly considered design by some garage mechanic.

[Greg Locock]

The photos are a rerofit onto a pre 90s Ford (Australia) Falcon, the manifold was fitted to i6 Falcons from about 2002 onwards. I doubt we'll ever see 6 butterflies on a single spindle again!

I can't find the graph showing the torque curve with the butterflies open and closed. At the time I was working on engine noise, and it was with fear in my heart that I leant over one on the chassis dyno, and flipped the butterfly backwards and forwards , and decided there was no NVH issue! I'm still a bit surprised by that.

[Magoo]

It's not really a "variable" system. Instead it's more of a "binary" system. The relative change in tuned length between the two inlets is also a bit excessive. Even for a very high rpm F1 engine, the length of the lower inlets looks much too short to be of benefit. I think it's just a poorly considered design by some garage mechanic.

By far most variable inlet systems are "binary." Never assume that two tract lengths are tuned to the same wave. You are looking at an old Nissan factory setup, so that's what you get for thinking.

[MatsNorway]

As mentioned this is far from Backyard setup. It was made illegal for the offroad series it was supposed to run in. It probably got or was designed for a fairly wild cam?. Now the privateer owners will use it for drifting.

[Kelpiecross]

The photos are a rerofit onto a pre 90s Ford (Australia) Falcon, the manifold was fitted to i6 Falcons from about 2002 onwards. I doubt we'll ever see 6 butterflies on a single spindle again!

I can't find the graph showing the torque curve with the butterflies open and closed. At the time I was working on engine noise, and it was with fear in my heart that I leant over one on the chassis dyno, and flipped the butterfly backwards and forwards , and decided there was no NVH issue! I'm still a bit surprised by that.

Thanks for clearing that up. I thought the manifold looked distinctly Falcon-ish but the engine was a puzzle.

[kikiturbo2]

This was in production for a few years

http://media.photobu.../DSCF2280-1.jpg


Switchover speed was around 3500 rpm, that's the vac actuator at the front.

peugeot, as well as some other europeans had similar systems..

[Greg Locock]

peugeot, as well as some other europeans had similar systems..

"had" being the operative word, I suspect.

[bigleagueslider]

By far most variable inlet systems are "binary." Never assume that two tract lengths are tuned to the same wave. You are looking at an old Nissan factory setup, so that's what you get for thinking.

It depends on whether you're talking about race engines or production engines. Where racing rules permit variable intake geometry, the systems are usually truly variable designs. On the other hand, production auto engines mostly use simpler two position or two path designs.

Using a runner length tuned for anything other than first order harmonics is usually not worth the effort since higher order harmonics tend to be far less energetic. The only reason I could see for the very short duct is to provide interference where the operating frequencies in the longer runner are hurting VE. But then again, that's just my thinking on the subject.

[kikiturbo2]

"had" being the operative word, I suspect.

you are right, of course..... The trend lasted for a couple of years... Next version of the same engine had better breathing trough thinner valves and better port shape, used nice long inlet manifold with cast tapered throats and produced more torque everywhere..
Then came variable cam timing and complex inlets became part of history ..

[MatsNorway]

Please note that the engine in the picture is a Wankel.

Why a servo actuated tract like in the picture is not more common on sporty cars is a tiny bit surpricing for me. At least on a wankel who has no vaiable intake timing like VTEC and such. (as far as i know)

Edited by MatsNorway, 15 July 2012 - 18:51.

[bigleagueslider]

Please note that the engine in the picture is a Wankel. Why a servo actuated tract like in the picture is not more common on sporty cars is a tiny bit surpricing for me. At least on a wankel who has no vaiable intake timing like VTEC and such. (as far as i know)

MatsNorway,

That picture is the Mazda 4 rotor Wankel that won LeMans in '91. And I'd agree that the Wankel does respond very well to variable length intakes, due to the very pronounced pressure pulses produced by its porting arrangement.

Here's a nice video of a linear variable intake system on a race engine:

slider

[Magoo]

Using a runner length tuned for anything other than first order harmonics is usually not worth the effort since higher order harmonics tend to be far less energetic.

Really. How long does it need to be at 4000 rpm, do you reckon.

[bigleagueslider]

Really. How long does it need to be at 4000 rpm, do you reckon.

Beg your pardon, but how long does what need to be at 4000 rpm?

[Magoo]

Beg your pardon, but how long does what need to be at 4000 rpm?

Inlet tract. As you know.

[bigleagueslider]

Inlet tract. As you know.

Magoo,

Sorry, I understand your question now.

While the precise answer is quite complicated, the simple answer is as long as it takes for pressure wave to travel up the runner duct from the intake valve face (at INOP) to the trumpet lip, and back again to the intake valve face (close to INCL). The acoustic pressure wave travels along the manifold runner at sonic velocity. The negative pressure pulse is created when the intake valve opens. It travels along the inlet runner, with its relatively uniform cross section, until it reaches the trumpet lip. At the trumpet lip the duct cross section changes abruptly, and this causes a reversion of the pressure wave. The reflected pressure wave is now positive and travels back down the inlet runner. Ideally it arrives at the inlet valve face just a bit before it closes, and the higher pressure pulse forces a bit more air/fuel into the cylinder at the last instant.

The basic process is dependent upon the speed of sound at the inlet manifold conditions, and the valve event time duration. In most cases, the primary mode is the only one worth considering. So an isolated intake runner length tuned to take advantage of the primary mode at 4000 rpm should be roughly 1/2 the distance a pressure wave will travel at sonic velocity during the time between INOP and INCL at 4000 rpm. This time period would be around .013 seconds at 4000 rpm with typical race engine valve timings.

In reality it's not quite so simple, but I think you can get the basic concept. And of course I should also point out that I'm no expert on the subject. So if I've made an error in my explanation, maybe someone can correct me.

Regards,
slider

[Magoo]

Magoo,

Sorry, I understand your question now.

While the precise answer is quite complicated, the simple answer is as long as it takes for pressure wave to travel up the runner duct from the intake valve face (at INOP) to the trumpet lip, and back again to the intake valve face (close to INCL). The acoustic pressure wave travels along the manifold runner at sonic velocity. The negative pressure pulse is created when the intake valve opens. It travels along the inlet runner, with its relatively uniform cross section, until it reaches the trumpet lip. At the trumpet lip the duct cross section changes abruptly, and this causes a reversion of the pressure wave. The reflected pressure wave is now positive and travels back down the inlet runner. Ideally it arrives at the inlet valve face just a bit before it closes, and the higher pressure pulse forces a bit more air/fuel into the cylinder at the last instant.

The basic process is dependent upon the speed of sound at the inlet manifold conditions, and the valve event time duration. In most cases, the primary mode is the only one worth considering. So an isolated intake runner length tuned to take advantage of the primary mode at 4000 rpm should be roughly 1/2 the distance a pressure wave will travel at sonic velocity during the time between INOP and INCL at 4000 rpm. This time period would be around .013 seconds at 4000 rpm with typical race engine valve timings.

In reality it's not quite so simple, but I think you can get the basic concept. And of course I should also point out that I'm no expert on the subject. So if I've made an error in my explanation, maybe someone can correct me.

Regards,
slider

What you meant to say was not quite six feet.

[Grumbles]

...In most cases, the primary mode is the only one worth considering...

About fifty squillion drag racers (including myself) would disagree. Dimensioning the runner to catch the first bounce often makes it impossibly long (unless it's curled like a french horn) but even worse any advantage in wave amplitude is usually negated by losses in bulk flow. Increasing the CSA to offset these losses just weakens the initial jerk and therefore the pressure wave amplitude. No doubt higher revving engines would be a different kettle of fish, but for many engines that make peak torque somewhere around 6500 - 7500 the second harmonic often seems to be the best compromise.

[gruntguru]

There is a simpler way.
http://3.bp.blogspot...08-r6-motor.jpg

[Tony Matthews]

Year 2000 Ferrari 049 V10. The trumpets followed quite a sophisticated pattern, and one bank worked differently to the other, alternately.

[mariner]

Assuming the cross section of the Ferrair V - 10 Tony has posted is fully accurate two things look interesting.

- The use of cam followers on such a high reving engine with Pneumatic valve springs

- The shape of the exhaust ports opening up like megaphones.

Maybe the latter is just the blending of twin exhaust valve ports into one ?


Also there is quite literally no sump at all, the crank is so low its maybe 5- 10mm from the car underside

[desmo]

Assuming the cross section of the Ferrair V - 10 Tony has posted is fully accurate two things look interesting.

- The use of cam followers on such a high reving engine with Pneumatic valve springs

Finger followers became universal or nearly so around 10 years ago in F1. Here's John Judd on the rationale:

"I think you need finger followers in Formula One now. The use of air springs means that the spring itself is no longer a problem. With wire springs you got pushed into a corner between spring design and valve lift. You had to compromise badly whereas with air springs there is no real compromise - you just have to make sure that you don't overload the cam/tappet interface.

"As much as 17mm of valve lift is not out of court these days [1998]. To get the valve open to that amount of lift in the time available you need a very high acceleration and for that you need a large-diameter tappet. The steeper the rate of lift the bigger the tappet needs to be and there comes the point at which it is better done with a finger-type follower where you have some rocker ratio.

"Essentially you don't put the cam directly over the valve, you get some rocker ratio on it and you can get the lift you need with lower valve gear recipricating mass. If we were to build another Formula One engine I think we would have to look very seriously at finger cam followers."

[Tony Matthews]

Also, the pairs of valves (inlet/inlet) were splayed by a few degrees, so big problems for the cam/tappet.

[MatsNorway]

Heeey squash/squeeze pistons. Do they use that today too?

[Magoo]

Finger followers became universal or nearly so around 10 years ago in F1. Here's John Judd on the rationale:

"I think you need finger followers in Formula One now. The use of air springs means that the spring itself is no longer a problem. With wire springs you got pushed into a corner between spring design and valve lift. You had to compromise badly whereas with air springs there is no real compromise - you just have to make sure that you don't overload the cam/tappet interface.

"As much as 17mm of valve lift is not out of court these days [1998]. To get the valve open to that amount of lift in the time available you need a very high acceleration and for that you need a large-diameter tappet. The steeper the rate of lift the bigger the tappet needs to be and there comes the point at which it is better done with a finger-type follower where you have some rocker ratio.

"Essentially you don't put the cam directly over the valve, you get some rocker ratio on it and you can get the lift you need with lower valve gear recipricating mass. If we were to build another Formula One engine I think we would have to look very seriously at finger cam followers."

Indeed. Most everything in high rpm applications is finger follower anymore. Buckets are passe if not obsolete.

[gruntguru]

Judd's statement misses a key point:
You do not need to offset the cam from the valve centreline to create a motion ratio when using finger followers. The convex contact shape achieves that (see Ferrari drawing above) by angling the contact plane. The same can be achieved with "domed" bucket lifters or roller followers.

Another space advantage occurs because the finger follower can be as narrow as the cam lobe whereas the bucket follower gets wider as higher valve accelerations are required.

[GreenMachine]

Two questions about this engine:

In the RH bank, there are three fittings (immediately below three of the bolts holding the camshaft bearing caps), which appear to be some form of oil feed function? Can anyone shed any light on these?

What is the structure/mechanism in the 'V' between the cylinder banks?

Thanks in advance ...

[desmo]

Good questions, I wondered the same.

[Magoo]

The book in which the above illustration appears -- Ferrari Formula 1 by Peter Wright -- is one of the best technical books on F1 ever produced, and one of the most beautiful as well. This is due in no small part to the work of illustrator Tony Matthews.

[desmo]

Sidebar question for Tony who drew the illustration of the 049, how helpful or important is it really to understand the details functionally of what you are illustrating? Or can you just graphically describe what you see or are shown regardless?

Is the thing in the vee perhaps a valve for venting coolant into the airbox in the event of an overpressurization?

[Tony Matthews]

It's not always vital, but frustrating sometimes to draw something in sufficient detail for others to understand how it functions without understanding it oneself. Most things I can work out without being given a clue, others not. Some vital details are not made available, others are hidden completely. I was not allowed to see, let alone photograph and then draw, the vibration dampers on the rear of the 049 camshafts.

The three valves shown in the section of the LH bank of the 049 are for controlling the gas flow/pressure to the pneumatic valve springs. The item in the centre of the V is, as far as I know, to control the coolant pressure.

[desmo]

Hmmmm. Might there be some small but still useful degree of VVT possible from a putative camshaft TV damper? You've decoupled the direct connection between the timing gear stack and the camshaft, why not avail oneself of the opportunity given that it might well be difficult or impossible to prove it wasn't simply a TV damper from looking at the static parts? Then again would F1 even use VVT even assuming it were allowed?

[Greg Locock]

There are plenty of places you could put compliance into the drive of the camshaft, which is what i think you are getting at.

[Magoo]

I would think an engine like this needs a damper a lot more than it needs a cam phaser.

[desmo]

Maybe a device that discreetly accomplishes both would be better still than just the TV damper.

[Tony Matthews]

...how helpful or important is it really to understand the details functionally of what you are illustrating? Or can you just graphically describe what you see or are shown regardless?

I didn't give a very comprehensive answer first time, and I don't want to labour the point, but I ought to add that the colour cutaways are not so much about showing how things function, but to show where things are in as atractive a manner (according to taste) as can be. If showing how things work is the brief, then a different approach is needed, usually a simplified style, probably using annotation and arrows! In this case I would normally have been aided by a sketch and/or chat with an engineer, so that I knew - more or less - what to show, assuming I couldn't work it out on my own.

Most things are fairly obvious, others can be a complete mystery, including the elusive VDs. I was not allowed to show the Ilmor VDs, but unlike Ferrari I was at least able to see, hold and photograph them - but I don't know how they work, and I gather from various sources that designing them was a major headache to some very clever people.

Edited by Tony Matthews, 27 July 2012 - 07:20.

[GreenMachine]

The three valves shown in the section of the LH bank of the 049 are for controlling the gas flow/pressure to the pneumatic valve springs. The item in the centre of the V is, as far as I know, to control the coolant pressure.

Thanks Tony.

Love your work btw, I always learn something from it.

[Magoo]

Thanks Tony.

Love your work btw, I always learn something from it.

Below is a crop from Tony's latest rendering at Motor City Garage, the Honda Accord BTCC car. Check out the amazing detail. Thinking about it, here is a type of racing we may never see again.

See the entire painting and Tony's article about it here:

Tony Matthews on the Honda Accord BTCC | Motor City Garage

[Magoo]

Speaking of Tony Matthews engine cutaways, here (trumpets) is his latest at MCG, the Ilmor Chevrolet 265A IndyCar V8.


Tony Matthews on the Ilmor 265A Indy engine | Motor City Garage