19 replies to this topic

[Nathan]

Me and a buddy from our SAE team were talking about variable length intake runners. We then came across the idea of moving fuel injectors. Basically the idea works in the same way, the injectors location would change with RPM (we would assume more distance as revs increase) and also the idea of when we are off throttle the injector would move to its minimum distance from the port to help aid throttle response. All the technical stuff to make it work aside, would such an idea be worth anything? What kind of gains could be possible?

[McGuire]

Great question. A far cheaper and simpler way to accomplish the same thing is multiple injector locations. Some applications involving boost and/or large plenum volumes employ that approach.

However, in your case I doubt if mobile OR multiple injectors would be worth the trip. The gains would be very incremental and unless you have a sophisticated engine R&D program you would not even be able to spot them, probably.

I'm sure I don't need to tell you that a major issue for efforts such as Formula SAE is the careful allocation of tightly limited resources, financial and otherwise. Here there is little or no bang for the buck. The time and money would almost certainly be better spent somewhere else, I believe.

[djellison]

It's amazing how many SAE teams spend longer on coming up with good overalls and livery design than working on the car

I suppose it could work with two injectors in the same way the Toyota engine from the Lotus Exige has two cams - a normal and a power cam.

Doug

[McGuire]

Originally posted by djellison
It's amazing how many SAE teams spend longer on coming up with good overalls and livery design than working on the car

Just like big-time professional motor racing.

[desmo]

Originally posted by djellison

I suppose it could work with two injectors in the same way the Toyota engine from the Lotus Exige has two cams - a normal and a power cam.

Doug

That's called, I believe sequential FI. One engineer in the sport I communicated with mentioned he'd been involved at some point with an F1 engine that tried that. One injector in the "showerhead" position above the trumpet and another in the intake port. At the time a lot of work was being done on what he called "fuel presentation" in F1. Interesting concept. As McGuire points out, probably not a whole lot of potential upside relative to the bother and expense for FSAE.

[Greg Locock]

Noo, sequential is the rather daft name we give it when the injectors are actually timed relative to the valves, rather than having each bank fire at the same time. I guess because the injectors fire sequentially. We'd been looking forward to SEFI for a long time, which we got with EEC V, I think. I'm not sure it is even enabled in our current calibration, the benefits are slight.

On their turbo road cars Lotus used to use additional upstream injectors, that gives better idle control for a given maximum fuel flow rate.

[voice_of_reason]

Moving the injectors is not the same as varying the length of the trumpets (inlet runners).

- Trumpet length is varied as a function of RPM to alter the acoustic behaviour of the inlet. Trumpet gets shorter as rpm rises to match tuned length of trumpet to (nth order of) engine firing frequency (gets higher as RPM rises).
- If the engine has a wide RPM range (eg a F1 engine) the trumpets may jump up to the longest length at an intermediate speed then descend again. In this case, the inlet is tuned to the next harmonic (n+1) of engine speed.
- All this is about getting the engine to breath better by using acoustic effects to make it swallow more air, thus improving power.


Moving the injector is about fuel/air mixing and charge preparation:

- For any trumpet length there is an optimum position for the injector. Often when the trumpets move, the injectors stay in the same place, chosen as the best performance compromise. A bettter compromise may be for the injectors to follow the trumpets, ie to maintain the injector position relative to the trumpet mouth.
- Injector position and spray form and timing are routine optimisations to be done for each aero-acoustic combination to get the best performance, but are second order relative to the fundamental engine tuning.


In conclusion: there may be a small performance benfit from moving the injectors as well as the trumpets, but to fix the trumpets and move the injectors would be a mistake.

[mini1275]

Originally posted by voice_of_reason


- For any trumpet length there is an optimum position for the injector.

So throttle position(pressure/flow...) has nothing to do with it?

[desmo]

Originally posted by Greg Locock
Noo, sequential is the rather daft name we give it when the injectors are actually timed relative to the valves, rather than having each bank fire at the same time. I guess because the injectors fire sequentially. We'd been looking forward to SEFI for a long time, which we got with EEC V, I think. I'm not sure it is even enabled in our current calibration, the benefits are slight.

On their turbo road cars Lotus used to use additional upstream injectors, that gives better idle control for a given maximum fuel flow rate.

My bad, I wasn't sure. Is there a proper descriptor for using multiple injectors per cylinder? Might present day F1 engines be less sensitive to part throttle FI mappings as a result of TC?

[McGuire]

One benefit of upstream injection....it produces some evaporative charge cooling, especially when using methanol. On the CART turbo engines fuel was sprayed into compressor inlet, as well as through the conventional port injectors.

[Nathan]

Wow thanks alot guys! I figured the same as you, but was curious if maybe it would be worth a punt. Often times the silliest ideas become something decent.

[McGuire]

Point of order on the techno-jargon...in the motorsports end of the biz, multiple injection is often referred to as "sequential," the idea being that fuel is added to the air charge at various points in succession, or "sequentially." What the OE industry calls "sequential," the racing bunch has always known as simply "timed fuel injection," since the Sixties at least.

With the OE outfit I toiled for we had:

MFI aka MPFI (multi-port fuel injection)
SFI aka SPFI (sequential port fuel injection)
TBI (throttle body fuel injection)
SPFI (single-point fuel injection)

...SPFI differing from TBI in that there is one central injector solenoid with tubes and check-valve nozzles for each port location, involving a plastic manifold assembly called a "spider."

To me these are all trade or corporate names for stuff, and not truly technical terms. Perhaps the SAE will eventually issue an epistle specifying the exact definition of all these terms, including appropriate penalties for misuse. :

[voice_of_reason]

One benefit of upstream injection....it produces some evaporative charge cooling, especially when using methanol. On the CART turbo engines fuel was sprayed into compressor inlet, as well as through the conventional port injectors.

McGuire is right, and though it is drifting slightly off-subject, I will continue the thread, as it is fascinating, particularly in the special case of methanol.


Some background:

In general, for performance, evaporating fuel by taking heat from the inlet air is good (charge cooling effect dominates air displaced by fuel vapour => improved vol eff combined with improved mixture preparatioin => better power)

In general, for performance, evaporating fuel by taking heat from port walls is bad (fuel vapour displaces air for no charge cooling benefit => worse vol eff => worse power).

In either case, improved mixture preparation from vaporised fuel is good for combustion stability (=> tolerance to lean mixtures => good thermal efficiency / fuel consumption).

Changing the density of the air in the ports changes the speed of sound in the port so changes the speed at which acoustic waves propagate which in turn changes the RPM at which the engine "tunes". (Higher temperature => lower density => higher frequency => higher rpm at which tuning is optimum).

Now some fuel properties:

http://www.eere.ener...s/fueltable.pdf

A key feature of methanol is that it has enormous latent heat of vaporisation relative to its heat value (approx x8 relative to gasoline for a given amount of air).

In the CART engines mentioned by McGuire, up to 45% of the fuel was injected before the compressor of the turbo. Charge temperature reduction of 80°C was typical - a huge charge cooling effect (intercoolers were forbidden in the rules).

The effects are overwhelmingly favourable: resistance to det, improved power, better fuel consumption, lower piston temperature, less fuel leakage past the rings (=>less fuel vapour in the blow by => better oil scavenge system behaviour).

The limit on the fuel injected pre compressor is the saturation of the charge and practical problems of mixture distribution in the plenum. Transport delay from injectors 1m away from the valve need looking after too.

A curious side-effect is that keeping the compressor wheel clean of track dust and tyre debris, maintains its performance and improves its life!

By varying the fraction of fuel injected pre compressor, the acoustic behaviour of the ports is modified and the tuned speed of the port adjusted. By this means, an effect similar to variable length inlet trumpets is achived. In general, CART engines with fixed trumpets would have less pre compressor injection at high RPM for optimum power.

I could go on, but you get the message.... its a potent tweak, particularly for turbocharged methanol engines where the rules do not allow intercoolers or variable trumpets!

[McGuire]

On the CART-type methanol turbo engines, this was commonly done with an annular injector rail around the turbo inlet with a complement of four to six injectors.

These engines also used nine throttle valves in order to hold up the plenum pressure.

[desmo]

Originally posted by voice_of_reason


By varying the fraction of fuel injected pre compressor, the acoustic behaviour of the ports is modified and the tuned speed of the port adjusted. By this means, an effect similar to variable length inlet trumpets is achived. In general, CART engines with fixed trumpets would have less pre compressor injection at high RPM for optimum power.

I could go on, but you get the message.... its a potent tweak, particularly for turbocharged methanol engines where the rules do not allow intercoolers or variable trumpets!

Damn, that's clever!

[voice_of_reason]

These engines also used nine throttle valves in order to hold up the plenum pressure.

A more exact description is the reverse of this comment from McGuire. In fact the 9th butterfly at the plenum entry was used to control the boost pressure, by eliminating over-boost rather than ".. to hold up the plenum pressure.."

Electrically operated with a servo-motor under closed loop control, the 9th butterfly trimmed the boost pressure by closing (the wastegates being set to give a slightly higher boost pressure than actually desired). The software controlling the 9th butterfly learned the behaviour of the pop-off valve and allowed the driver to run boost pressure as high as possible without tripping the valve. The wastegate top can pressure was automatically controlled in (much slower) in closed loop to get the desired level of pre-butterfly boost pressure. In addition, the butterfly closed abruptly across gearshifts to avoid boost spikes (caused by engine airflow falling abruptly with RPM across upshift while turbo tried to keep going!).

There was another device at the entrance to the compressor linked to the throttles. The IGV (for inlet Guide Vane) was a throttle plus pre-swirl device to help maintain turbo speed, for throttle and boost response for better driveability (to compensate to some extent the transport delays of injectors a long way upstream!).

There was a lot of good performance development work done on this generation of CART engines before the series imploded.

[Engineguy]

Instead of issuing each team a popoff valve, they should have just issued them a positive displacement supercharger, two pulleys and a belt, and been done with it. Oh yeah... and get Eaton as a associate series sponsor.

IMHO, Turbos

[McGuire]

Originally posted by voice_of_reason



A more exact description is the reverse of this comment from McGuire. In fact the 9th butterfly at the plenum entry was used to control the boost pressure, by eliminating over-boost rather than ".. to hold up the plenum pressure.."

By "hold up" I mean attenuate. Do we have an ESL issue here?

[McGuire]

Originally posted by voice_of_reason

There was a lot of good performance development work done on this generation of CART engines before the series imploded.

Yeah, and this development work played no small part in assisting in the implosion. The technology discussed here reaches right into the areas where boost limits were being subverted. These guys were playing way beyond CART's technical capability to police them, and of course in the CART rules paradigm they were supposed to be policing themselves. Bwahahaha.

[ray b]

I can never understand why they inject in to a port at all
why not just squirt in to cylinder directly like a diesel does
sure you need a better quality injector but this F-1
not a production car that you are trying to build on the cheap
in fact why not squirt at something very hot like a ext valve
to speed vaporization then in to a cool port

very odd idea squirt thru the ext valve to cool the valve and shoot vapor
not liquid in to the cylinder