My EFI setup

[taunusrainer]

Wow, in fact I am surprised by the terminology. To me it is a bit harder as I am not a native speaker.
I will write down what I understand so far.


Batch: The same as alternating?

Sequential: The injector is injecting only while the intake valve is open. The amount of fuel is PWM controlled by the computer.

Simultanous: All injectors are parallel, injecting the same amount of fuel at the same time, PWM controlled.

Alternating: The injectors of one cylinder bank are parallel, the cylinder banks are fired alternatly.

Continuous: The injector is always open and passive, giving a constant stream of fuel into the air. The amount of fuel is varied by an external device.

That right?
Thx
ray

[FirstYearDeek]

Batch: The same as alternating?

Strictly speaking, batch simply means more than one injector is controlled by a single output, meaning more than one injector fires at a time. Some systems fire one bank then the other, some fire half of the injectors on one bank then half the injectors on the other bank to cut down on fuel rail pulsing.

Sequential: The injector is injecting only while the intake valve is open. The amount of fuel is PWM controlled by the computer.

Almost; Sequential means that the only injector that fires is the one for the cylinder that's on its intake stroke. Unless the motor is direct injection (like a modern diesel or very recent gasoline engines) then the injector simply does not have time to get the required volume of fuel in the mixture ONLY when the valve is open.

Sequentially fired motors are truly sequential only at very low loads and at idle. Once the load or RPM climbs to about 25% of the engine's total output, the system is essentially the same as a bank or batch system, where the fuel is injected before the intake valve opens.

These systems are PWM controlled as well. PWM control is simply a method for reducing the total current supplied to an open injector, therefore reducing heat and extending component life.

Simultanous: All injectors are parallel, injecting the same amount of fuel at the same time, PWM controlled.

That's right.

Alternating: The injectors of one cylinder bank are parallel, the cylinder banks are fired alternatly.

That's also correct. This is another definition of a Batch system. It's very popular with the V-motor crowd.

Continuous: The injector is always open and passive, giving a constant stream of fuel into the air. The amount of fuel is varied by an external device.

You got it. These are exclusively throttle-body injection systems, and are basically replacements for carburetors as they are "normally" used. (Not one per cylinder like our freak motors)

-Deek

[FirstYearDeek]

I should say that I don't know whether the system you have is alternating or simultaneous... I'm almost certain it's not sequential but I've been wrong once or twice before.

-Deek

[socrace]

Deek, thanks for the reed valve info, I feel a little better now about ignoring them! The basic idle air design I am trying to copy has been used for years by Kinsler Injection (which uses individual throttles, no reed valves) http://www.kinsler.com
They don't really lay it out too well on their web site, but the components are shown in their catalog: http://www.kinsler.com/handbook145.html

Ray, ok have to admit that I went back and took a closer look and you are right about the SEI and LTD manifolds not matching up to the GL1100 heads..
The intake manifold bores are 32mm and the head bores are 35mm. That leaves a ridge of 1.5mm, which may also be enough to affect the o-ring seal.
Think what I'll have to do here is grind a 45deg edge on the manifold flange, to take off the 1.5mm ridge, then use paper gaskets.

IMO, this really does just add to the case for trying out a central throttle body with original manifold elbows. The larger air flow area could make significantly more power than using the LTD/SEI hardware (or carbs also maybe), another goal worthy of all this efi effort..

Bob D

[taunusrainer]

Deek, thank You for clearing the mist...terminology is not my thing, I am electrician and therefore I have to learn all the combustion engine specific words.
I can software set the megasquirt computer to alternating or simultaneous mode. Guess I will go alternating, to lower the current draw.
Jesus it went so cold outside, -9°C, I cannot do the fuel pump plumbing on the bike now.
Wife won't allow bike sit in the living room for "small repairs"

Maybe You can help me out with another bug:
I have an Opel/GM fuel pump from a 1996 2.0 EFI four cylinder car. The pump is inline with first a pressure regulator and then a fuel filter cartridge. It seems GM did not go the usual way that the pressure regulator sits on the end of the fuel rail, bleeding excessive pressure to the fuel return hose. I don't have the car here, so I cannot look up.
Shall I simply leave this setup as is, connect the pump intake to the gas tank and the filter outlet to the fuel rail? No return hose then...

I really have no idea, but trying might destroy something.

Thank You!
Ray

[bladredhead]

sweet project!

Maybe i can't see it, or it's not an issue.. but how are you holding the injectors in place? I was part of a project in school that retrofitted an EFI setup on a 600cc suzuki motorcycle engine. We didn't restrain the injectors and eventually they worked loose and the fuel rail came off, it was an exciting revalation

[FirstYearDeek]

...The pump is inline with first a pressure regulator and then a fuel filter cartridge. It seems GM did not go the usual way that the pressure regulator sits on the end of the fuel rail, bleeding excessive pressure to the fuel return hose. I don't have the car here, so I cannot look up.
Shall I simply leave this setup as is, connect the pump intake to the gas tank and the filter outlet to the fuel rail? No return hose then...

I really have no idea, but trying might destroy something.

Thank You!
Ray

I'm not familiar with that particular setup but I do have some experience with "returnless" systems. If you can find a shop manual online or scan the manual page describing the fuel system, I may be of more assistance.

Typically, in a returnless system, there is some pump motor control going on. The pressure regulator probably has a vacuum hose attachment? This regulator will adjust the system pressure using that vacuum to a point, (to keep pressure constant) but there is always the risk of blow-by and cavitation, which is bad in a fuel system.

If you hook that pump up to 12v with no modulation, it may not survive... if it's the type I'm used to.

I would try to find a traditional pump-> filter -> injectors -> regulator ->return system.

-Deek

[FirstYearDeek]

...Maybe i can't see it, or it's not an issue.. but how are you holding the injectors in place? I was part of a project in school that retrofitted an EFI setup on a 600cc suzuki motorcycle engine. We didn't restrain the injectors and eventually they worked loose and the fuel rail came off, it was an exciting revalation

Yes, holding down those injectors is very important. You can empty a fuel tank in a hurry at 90+psi. Not to mention the potential flamethrower you've just created!

-Deek

[taunusrainer]

Maybe i can't see it, or it's not an issue.. but how are you holding the injectors in place?

Gents, You have eagle eyes, You are right, there is no special holder.
I have machined the weld on tubes a little undersize, with a groove inside for the o-ring.
It was really hard to press them in, I had to use brake piston lube and soften the o-rings with a hair-drier.
But then they really kinda snapped in and are sitting tough now.
As stated far above, I found (too late) a better solution for weld-on bungs w/bolts.

Deek, I will go lookin for tech info on that car and will post a pic of my pump assy tomorrow.
Probably I will alter my "fuel rail" to the usual design with regulator-to-return-hose.

Will have to go to bed now it is close to 11pm and tomorrow starts at 5.30am
CU
Ray

[badorderbob]

Glad to see your finally working on your design you talked about for a long time ago!! Very impressive keep up the good work Ray, Bob

[bladredhead]

Maybe i can't see it, or it's not an issue.. but how are you holding the injectors in place?

Gents, You have eagle eyes, You are right, there is no special holder.
I have machined the weld on tubes a little undersize, with a groove inside for the o-ring.
It was really hard to press them in, I had to use brake piston lube and soften the o-rings with a hair-drier.
But then they really kinda snapped in and are sitting tough now.
As stated far above, I found (too late) a better solution for weld-on bungs w/bolts.

Deek, I will go lookin for tech info on that car and will post a pic of my pump assy tomorrow.
Probably I will alter my "fuel rail" to the usual design with regulator-to-return-hose.

Will have to go to bed now it is close to 11pm and tomorrow starts at 5.30am
CU
Ray

you have more faith in o-rings than i do, i'd keep a fire extinguisher handy

[guy plain]

hi...ive been following this with some interest.....i like the idea and hope it really works out....i have a 75 gl1000 with the kick start....do you think the FI will work with the kicker?

[FirstYearDeek]

...do you think the FI will work with the kicker?

It depends on how you have it wired, and what your trigger mechanism is. The stock software for MegaSquirt puts the injectors at 100% current during starting to help with low voltage. It knows to do this because the start button is pressed in. You could conceivably wire the start button to tell the MS you're starting, but not actually turn the starter.

All other systems will operate as usual, providing you're triggering off the ignition system. (You're the starter, not the starter... the MS doesn't know the difference other than the injector setting I mentioned above)

-Deek

[DJ]

IS THERE A HOW-TO BOOK or some Hot Rod Magazine primer on fuel injection for those of us who do not have the time to study 25 different sources or the time to pick up a clue at a time here and there?

I remember all the time I spent studying turbocharging books only to discover at the dragstrip that I was very over-prepared, and needed only a fraction of what I wasted so much time studying.

How do you re-program factory controls?
Do you have to buy aftermarket controllers and then spend lots of time on a computer inventing your own fuel curve to see what works and what doesn't?

Is there a simple system that can decide for itself by using the inputs from oxy sensor, temp, air pressure etc, or is there still a lot of fuel curve tweaking left to do?

Where can a person find the answers without going into a year long study of many many sources of small clues to add together into a jigsaw picture??

[FirstYearDeek]

IS THERE A HOW-TO BOOK or some Hot Rod Magazine primer on fuel injection for those of us who do not have the time to study 25 different sources or the time to pick up a clue at a time here and there?

I remember all the time I spent studying turbocharging books only to discover at the dragstrip that I was very over-prepared, and needed only a fraction of what I wasted so much time studying.

In my opinion, the only reference you need is at http://www.megasquirt.info

Even if you don't end up using the Megasquirt controller, you'll get the basics of fuel injection theory and practice, which is applicable to many systems.

How do you re-program factory controls?

By factory controls, I assume you mean the factory ECU (or PCM) that takes inputs from the sensors and outputs fuel and/or spark?

The answer to that 99% of the time is; you can't. OEM's simply cannot allow access to the code in these "black boxes" due to liability issues. Several systems have been "cracked" however, and "piggyback" controllers are also available. So are "mod" chips that replace the stock chip and provide some programming capabilities. These systems are few and far between and the modifications required to make them programmable are expensive and usually very kludgy.

Do you have to buy aftermarket controllers and then spend lots of time on a computer inventing your own fuel curve to see what works and what doesn't?

For the most part, yes. With an unlimited budget, just call up Edlebrock or F.A.S.T. systems or the Big Stuff folks and tell them what you want. You'll get all the parts, pieces, and maybe even a base fuel map that won't flood your bike. From there it's up to you and their support people. (Which typically aren't very helpful) Time on a computer is optional; there are people who get paid good money to tune EFI systems. (I used to be one of them but frankly, I got tired of dealing with pompous a$$ holes with more money than brains that thought that they could throw money at someone and end up with a finely polished turd.)

Is there a simple system that can decide for itself by using the inputs from oxy sensor, temp, air pressure etc, or is there still a lot of fuel curve tweaking left to do?

The initial setup of the software in most systems will get you a drivable motor out of the box. Engine metrics like Volumetric Efficiency, Displacement and base timing curves along with a few basic enrichment settings (for cold starts, etc) will get you 80% of the way there. That last 20% is deep into the realm of "what's good enough?" There is a point of diminishing returns when you're talking about tuning, which is normally a lesson in trade-offs.

When last I was exposed to the "pro" aftermarket controllers, there wasn't a single one that had a good "auto tune" feature. MegaSquirt and the MegaTune software alone have got this right so far. You hook up your tuning device (usually a laptop) start recording a log and go for a drive... the log is analyzed and where conditions result in an over-rich or lean condition, your map is modified.

There are also a few systems (MegaSquirt is included) that have a "closed loop" feature where fueling is adjusted automatically based on the feedback from a wide band oxygen sensor. If you let the software have total control over the map, it will correct any lean or rich areas without your intervention. This is not an ideal solution; correction on the fly like this can result in some odd engine behavior during accel and decel.

Where can a person find the answers without going into a year long study of many many sources of small clues to add together into a jigsaw picture??

Year-long study? To really understand a system, it will be a year if you're very, very intelligent, are intimately knowledgeable about the components you're using, and have a great support team behind you. Starting with unknown components or lacking a great support team are recipes for failure, if you don't already have the knowledge.

Most first-time DIY builds for someone starting at ground zero never get completed. There are some steep learning curves to climb and most people go to what they know, which is usually carbs. Seriously, who wants to work on a bike that is otherwise rideable? The first-time DIY builds for someone starting at ground zero that do complete usually take every bit of a year to select the components, install and configure them and tuning can sometimes take weeks after that. True comprehension, unless you're some kind of savant, will take many more months or years and much more than reading a one-stop-shopping type of reference.

Honestly, it sounds like you're looking for an easy way out. Don't be offended, but DIY engine management is not for the lazy or faint of heart. (or poor ) If you "do not have the time to study 25 different sources or the time to pick up a clue at a time here and there" then you don't have time for DIY and should consider a turn-key system where the leg-work has been done for you.

-Deek