Fuel injection theory.

[RomanV]

successful fuel delivery is dependant on: (among other things)
injector size,
fuel pressure,
injector duty cycle,
and air velocity into the combustion chamber.

I was thinking 'why dont all cars run big injectors?'

Obviously if you increase the injector size, you have to reduce the injector duty cycle to acheive the same A/F ratio.

So whats wrong with doing this?

Is there could be a point where you reach a duty cycle percentage that is too low?

eg. The fuel would come out in large bursts with a long interval, rather than smaller, rapid bursts. So the fuel is not mixing with the air as well, leading to a less powerful combustion. Even though the same air/fuel ratio is being acheived.

So, I suppose my long winded question is this:
Do large injectors have a detrimental effect on performance/economy as outlined above, and is this the reason some tuned cars run out of injector duty cycle on standard injectors?
OR is there a much more simple explanation for why cars dont run big injectors from the factory.
eg. Larger injectors are more expensive.

[CozmoNz]

I would suspect its more the volume of fuel, since they *open and close* at a rate dependant on Rpm...

so... wouldnt they simply open and close quicker (than stock injectors), to inject the same volume of fuel, to maintain the same a/f ratio?

why must you confuse me with such long winded questions!

[QikStarlie]

too big and you'l have idling issuses. at idle my starlet sits on 0% (rounded down to the nearest number) duty cycle with 450cc injectors..

[matt dunn]

Yes it would make sense to build all cars with 700cc injectors as they all must cost the same to make,

but,

large injectors are not very accurate at low duty cycles,
and smaller injector have a better spray pattern at low duty cycles too,
as fuel economy is relevant to the way the fuel and air mix, so if an injector has a better spray pattern at idle is will help fuel economy and emissions which is a big thing now.

Matt

[RomanV]

Aaah, thanks Matt!

Thats what I thought.

Cheers.

[wde_bdy]

Also, some cars are purposefully restricted, XR6 Turbo for example. One of the first upgrades you need to do is injectors so you can't mess with them much without blatantly stuffing your warranty.

Callum

[Argon]

Good one Roman, It's food for thought. Correct me if I'm wrong, but this is what I reckon is going on.

My understanding about the set up is that injectors can have their duty cycle increased by running an adjustable fuel pressure regulator and winding the fuel pressure up and effectively using the force of the fuel pump to push as much fuel through the injector as possible.

To decrease duty cycle, it would involve the same fuel pressure regulator, this time using it to decrease the fuel pressure in the fuel rail. In doing this, bigger injectors could be used to supply a comparitively smaller amount of fuel.

However, because the fuel being delivered the injector nozzle is at a lower pressure, it would dribble out of the injector and not atomise correctly in the inlet manifold.

This could be likened to holding a garden hose in front of a large fan with the nozzle on the hose set to squirt a singular jet of water. The water in the hose is running through the hose at a minimal pressure and once it leaves the nozzle it does not disperse evenly infront of the fan. Turning the nozzle to the spray option increases the water pressure in the hose, as there is more resistance being provided by the reduced nozzle outlet size. this means that as the water leaves the hose it will be forced through and will form an evenly distributed mist, which will in turn mix well with the stream of air from the fan.

In terms of the fuel injection system, high fuel pressures and small injector apertures allows the fuel to be delivered and mixed with air optimumly, giving the most performance from the least amount of fuel possible.

However, it can also be seen that cars with large injectors ( race and hard tuned street applications ) have a tendancy to not idle so great or run very well at low rpm. This can be also attributed to large cam durations and aggressive timing, but also part of the equation is the fact these cars are running rich air/fuel ratios at low rpm, due to the large injectors they must run to supply enough fuel to make enough power high in the rev range. A rough idle and sluggish throttle response is common along with poor economy and big fuel dumps.

For those not familiar with it, a fuel dump is when the throttle goes from only slightly open to wide open and the injectors temporarily over supply the engine with fuel, causing the A/F ratios to go rich and dump a large amount of unburnt fuel into the exhaust causing a large black cloud out of the exhaust when the car takes off. A good example is the Signal GTR Skyline on High Octane Overboost, once it takes off from crusing RPM, 10 bucks of gas pretty much falls out of the exhaust untill the engine catches up in demand with the amount fuel being supplied.

There are a few other issues going on with it that I can see, but basically using big injectors will allow more fuel to be supplied, but will make the car more of a gas hungry pig with little response and minimal power output at crusing speeds. If this was what we are looking for, a Austrailian made V8 would be the ticket.

[matt dunn]

Good one Roman, It's food for thought.

Correct me if I'm wrong,

I will

injectors can have their duty cycle increased by running an adjustable fuel pressure regulator

No. Injecor duty cycle has nothing to do with fuel pressure.
Injector duty cycle is the percentage of time the injector has electrical currect flowing through it.

In terms of the fuel injection system, high fuel pressures and small injector apertures allows the fuel to be delivered and mixed with air optimumly, giving the most performance from the least amount of fuel possible.

That part is correct

cars with large injectors have a tendancy to not idle so great or run very well at low rpm. This can be also attributed to large cam durations and aggressive timing, but also part of the equation is the fact these cars are running rich air/fuel ratios at low rpm, due to the large injectors they must run to supply enough fuel to make enough power high in the rev range.

No not quite. They run a rich mixture because not all of the fuel mixes with the air properly and gets burnt so to get enough fuel to mix, you have to put too much in to start with.

For those not familiar with it, a fuel dump is when the throttle goes from only slightly open to wide open and the injectors temporarily over supply the engine with fuel, causing the A/F ratios to go rich and dump a large amount of unburnt fuel into the exhaust causing a large black cloud out of the exhaust when the car takes off. A good example is the Signal GTR Skyline on High Octane Overboost, once it takes off from crusing RPM, 10 bucks of gas pretty much falls out of the exhaust untill the engine catches up in demand with the amount fuel being supplied.

That is the equivilant of the accelerator pump. The way you expilan it it sounds like the ECU is just not capable of suppling the right amount of fuel. That is wrong as the extra fuel is put in there on purpose for a reason, to get throttle response.

There are a few other issues going on with it that I can see, but basically using big injectors will allow more fuel to be supplied, but will make the car more of a gas hungry pig with little response and minimal power output at crusing speeds. If this was what we are looking for, a Austrailian made V8 would be the ticket.

A car cruising at 100kmh, the injectior size will have f all to do with fuel economy unless you have rediculous size injectors like 800cc+ and even then is may not have an effect.. It's only at idle and just above that injector size will have an effect on fuel economy.

Matt

[RedMist]

Two things to consider. The fuel atomisation issue has already been covered. But to make optimum power you want as small an injector as possible to just scrape in at a max injector cycle of around 85% at your projected HP. Bigger injectors in the same engine will supply the same amount of fuel at a reduced injector druation of lets say 60% and will not atomise as well, and will not produce as much HP as the smaller injectors. This is why there is so many different flow rated injectors, not because a manufacturer is attempting to stop modification.


Secondly it takes time (MS) to open and close an injector. So if you have big injectors, at idle they are attempting to slam open and close as quickly as possible, and they still over supply the engine with fuel, hence the offbeat idle. I have 300degree 415 thou lift cams and the engine idles at 700RPM with a sound like stock.

[fivebob]

But to make optimum power you want as small an injector as possible to just scrape in at a max injector cycle of around 85% at your projected HP. Bigger injectors in the same engine will supply the same amount of fuel at a reduced injector druation of lets say 60% and will not atomise as well, and will not produce as much HP as the smaller injectors.

I agree that you need as small an injector as possible, but sizing for 85% duty cycle at max HP is, IMO, not correct, unless peak HP and peak rpm coincide. You need to take into account how far past peak HP you intend to rev the engine, and size the injectors accordingly. Typically this means sizing your injectors for 65-70% at peak HP. Reduced duration by itself does not imply poorer atomisation, if you used increased base fuel pressure to provide the additional flow the atomisation may well be better, though larger injectors at the same fuel pressure tend to have poorer atomisation.

Secondly it takes time (MS) to open and close an injector. So if you have big injectors, at idle they are attempting to slam open and close as quickly as possible, and they still over supply the engine with fuel, hence the offbeat idle. I have 300degree 415 thou lift cams and the engine idles at 700RPM with a sound like stock.

True, but if you use peak and hold injector drivers, have injector drivers that take care of the voltage spikes that occur on injector closing, and have a method of increasing resolution at low load, then big injectors can be made to idle just like stock ones. However these options come at the cost of a more expensive ECU and I doubt any manufacturer would fit them as stock because they’re not really required, and additional expense for no real benefit does not make sense.

In the arena of aftermarket ECUs it also comes down to the factor of cost, most cheap ECUs don’t have peak & hold drivers, and of those that do very few have any sort of flyback control, nor do they have any method of improving resolution at low load so often suffer from poor idle on big injectors. Motec does have all these features, and is often praised for it’s ability to control very large injectors, but as has been seen many times on automotive forums a lot of people are put off by the cost

[RedMist]

I agree that you need as small an injector as possible, but sizing for 85% duty cycle at max HP is, IMO, not correct, unless peak HP and peak rpm coincide. You need to take into account how far past peak HP you intend to rev the engine, and size the injectors accordingly. Typically this means sizing your injectors for 65-70% at peak HP. Reduced duration by itself does not imply poorer atomisation, if you used increased base fuel pressure to provide the additional flow the atomisation may well be better, though larger injectors at the same fuel pressure tend to have poorer atomisation.

Should have clarified my thought perhaps. I was thinking in regards to calculating injector size. Most online calcs use max hp to derive injector sizing.

In the arena of aftermarket ECUs it also comes down to the factor of cost, most cheap ECUs don’t have peak & hold drivers, and of those that do very few have any sort of flyback control, nor do they have any method of improving resolution at low load so often suffer from poor idle on big injectors. Motec does have all these features, and is often praised for it’s ability to control very large injectors, but as has been seen many times on automotive forums a lot of people are put off by the cost

The Link LEM uses two BUK555 FETS as peak and hold injector drivers . One would presume that these FETs would require some form of protection from flyback even if they use a DR straight to VBATT flyback. Is this the cheapest ECU you were referring to?

If so how does the Motec flyback differ?

[matt dunn]

Hate to interupt you two but,

I have not had much to do with the V5 LEM, but I was under the impression that the link does not have current control of low resistance injectors, hence an external resistor is required,

where the motec does have current control allow is to apply full 12v across a low resistance injector to give it a 'kick' to get it open and then the motec will limit the current through the injector so it will not burn out the coil inside.

Is this what you two are on about? And does the new LEM have this feature?

Matt

[fivebob]

Should have clarified my thought perhaps. I was thinking in regards to calculating injector size. Most online calcs use max hp to derive injector sizing.

Indeed they do, however that does not make it correct. Even the 85% rule isn't entirely accurate, a better way of calculating injector size would be to factor in the injector open+close time to determine if the injection event can take place in the available time. Of course that requires far more information about the injector than most people have available, so it's easier to go with the 85% rule. Doesn't work to well on very high rpm engines though.

One other thing I forgot to mention is that by running a high duty cycle you lose some of the advantages of sequential injection, the injectors spend a lot of their time spraying fuel onto the back of closed valves, which doesn't help atomisation.

The Link LEM uses two BUK555 FETS as peak and hold injector drivers .

Yes but it requires the use of 6ohm resistors so if I remember my high school physics correctly V=IR, that means a peak current of approx 2 amps, compared to the Motec's peak available current of 8 amps (and it's ability to vary the peak portion) that means faster opening times. Granted that this only applies to very low resistance injectors which most people wouldn't use (again due to cost), but it still helps when controlling these injectors at low pulsewidths.

One would presume that these FETs would require some form of protection from flyback even if they use a DR straight to VBATT flyback. Is this the cheapest ECU you were referring to?

Using the DR straight to VBATT flyback isn't really an option for low resistance injectors, but, being a GenBoard user, you already knew that

BTW I never referred to the cheapest, just the cheap ECU's, and there are certainly a lot that are cheaper than the LEM

If so how does the Motec flyback differ?

I'm not privy to Motec's design philosophy so I can't say how it differs, nor am I very proficient in electronics. However I can say that just using a diode as a recirculation circuit will prolong injector close times, even a low(ish) voltage zener diode (placed backward) will dramatically decrease the close time, but this requires high wattage diodes if they are to survive for long periods.

I was going to address the issue of bounce in detail, but you've edited that from your post, suffice it to say that the bounce is mechanical, not electronic, and comes from allowing the injector to close too fast by not controlling flyback decay properly.

[vvega]

just some injestor info for those following that dont understand the differance between the 2

Electrically, there are two different types of electronic fuel injectors. One type of injector,
characterised by a high coil impedance (> 12Ω) is known as saturation injectors. The other
sort of injector, typically with coils of less than 6 ohms impedance, is known as peak-andhold
injectors. The names are derived from the current waveform that accompanies the
injector when it is switched on.
Saturation, or high impedance, injectors, require a simple switch to operate. When a 16 ohm
injector has 12V applied to it, the current that is expected to be drawn is 0.75 amps (by Ohm's
law). However, the current in a coil cannot change instantly. When power is applied to this
injector, the current builds in the coil over a period of around half a millisecond, until it
reaches its 0.75 Amp maximum, i.e. saturation. The coils remain in saturation until power is
switched off at the end of injection.
In order to reduce opening time, the coil impedance can be reduced. Very low impedance
would result in very large continuous currents, resulting in a great deal of heat and waste of
energy. However, a much smaller current is needed to hold the injector open than is needed to
initially switch it on. Thus, such an injector is described by two currents: a peak current for
opening, and a hold current for maintaining the fuel flow. These peak-and-hold injectors are
particularly good for high rpm engines needing large fuel-flow rates.
Peak-and-hold injectors come with a wide variety of current requirements. Most peak-andhold
injectors need 2A to switch on with 0.5A hold current (2/0.5A peak-and-hold), or are
4/1A peak-and-hold. Some very high performance injectors need as much as 6A to switch on,
especially under high fuel pressures.



the haltech also has a 4/1 pwm outputs for injectors
peak of 4 amps and a hold of 1 amp

hope this helps

v

[vvega]

and one on injector bounce

The flyback voltage clipping point is set based on the injectors electromagnetic coupling and mechanical spring rate.

The stored energy of the magnetic field around the injector winding is used to control the pintle closing speed. If the pintle is allowed to close too fast, the pintle and seat will soon become pounded out and begin to leak fuel. A fast closing rate will also cause the pintle to bounce, allowing extra fuel to be delivered to the engine. This extra fuel can't be accurately controlled, so the engineer must adjust the energy held within the flyback voltage to control the closing rate of the injector. This is accomplished by placing a zener diode across the PCM driver (transistor or MOSFET), as illustrated in Fig. 6. Note: Some fuel injection systems do not clip the flyback voltage.

As the magnetic field falls back into the injector winding, the energy loops through the clipping circuit. This allows the current to diminish at a set rate. The lower the voltage level set by the zener diode, the more enorgy that will be allowed to loop through the circuit. If an ordinary diode were used instead of a zener diode, it would allow the largest amount of stored energy to loop through the circuit.

A diode will allow the stored energy to loop until it reaches source voltage-in this case, 12 volts. This would allow the injector the longest period of time to close. The higher the zener diode voltage, the shorter the period of time allowed for the injector to close. This is due to the energy looping through the circuit being cut off early by the voltage rating of the zener diode.

.....snip


bounce is not a mechanical problem as per say
it is like you said fivebob a failure of the ecu to successfully control decay times


i guess it all depends on how you look at it, but any mechanical device will fault if the control electronics are not controlling it correctly :S

v

[vvega]

and because inknow you love to read articals fivebob heres the rest of that snip i put up before...http://www.findarticles.com/p/articles/mi_qa3828/is_200501/ai_n9468886

its a bit of prat becasue it missing all the diagrames and waveforms but he seams to have a good grasp of what he disccusing

v

[fivebob]

and because inknow you love to read articals fivebob heres the rest of that snip i put up before...http://www.findarticles.com/p/articles/mi_qa3828/is_200501/ai_n9468886

its a bit of prat becasue it missing all the diagrames and waveforms but he seams to have a good grasp of what he disccusing

LOL I already read that one, here's a link to the original(?) complete with diagrams
http://www.automotivetestsolutions.com/ ... eforms.htm

[fivebob]

For anyone that's interested in collecting the required data to do proper injector sizing calculations, I've hosted a copy of an Excel spreadsheet I found several years ago that lists most Japanese injectors, their resistance, open+close times (Invalid Injection time column), and their flow rates at various base fuel pressures.

http://users.actrix.co.nz/richardm/Page ... tors_j.xls

For those non Microsoft office users amongst you, you'll just have to wait until I get time to convert it into HTML

[RedMist]

and because inknow you love to read articals fivebob heres the rest of that snip i put up before...http://www.findarticles.com/p/articles/mi_qa3828/is_200501/ai_n9468886

its a bit of prat becasue it missing all the diagrames and waveforms but he seams to have a good grasp of what he disccusing

v

This is the original with associated pruddy pic's

http://omni.cc.purdue.edu/~tlrobert/pin ... rticle.pdf

[RedMist]

One other thing I forgot to mention is that by running a high duty cycle you lose some of the advantages of sequential injection, the injectors spend a lot of their time spraying fuel onto the back of closed valves, which doesn't help atomisation.

We could now get into the debate of the benefits of sequential over batch firing... But I think thats a subject better kept for another day!

Using the DR straight to VBATT flyback isn't really an option for low resistance injectors, but, being a GenBoard user, you already knew that

Bugger, damn your memory to hell
However of interest I'm not willing to risk either my race reliability or the engine mechanicals to the Genboard. Its still running LEM 4. Genboard is simply a serial MUX and datalog collector. Saying that Trees tend to be more detremental to my finishing rate than power reliabiliy.

I was going to address the issue of bounce in detail, but you've edited that from your post, suffice it to say that the bounce is mechanical, not electronic, and comes from allowing the injector to close too fast by not controlling flyback decay properly.

Actually I edited the post twice, both times to reduce its content dramatically. I wished to respond with something of a better researched reply, however I havent the time at the moment. Race car is still in pieces and I'm attempting to race this coming Sunday.