Alternative Fuel Injectors

[ischmitz]

Yeah Rick - I saw that one. Scheider ******** is actually right here in Santa Barbara. And Henry is a nice guy.

I don't know what happened but this thread was a little bit far out there as others pointed out. It's sad when things deteriorate to personal attacks for no good reason.

I still hope Loren will share his endless knowledge - this is what makes this board IMHO. He does know a lot about the Motronic and related items. Why not enlighten others......

[3.2 CAB]

I have seen some replies already about the Ford injectors being used in various BMW engines. I know many V12 owners that have also gone to the "blue oval" injectors. The applications that I have seen in use are the Bosch injectors. I am still running the factory injectors in my V12-750iL, but if I have a need to replace, I too will be going to the Ford injectors.

[3.2 CAB]

As far as things go with Loren, I have gleaned a vast amount of information and knowledge from his experience on many electronics subjects.

[ianc]

Why not enlighten others......

It costs him business, and we can't have that, can we?

ianc

[ischmitz]

Originally Posted by ianc

It costs him business, and we can't have that, can we?

ianc

ianc, I beg to disagree with your statement. Market acceptance is a weird phenomenon. I would consider it a group-dynamic thing. Especially, nowadays with the internet it does not take much to tip the scale one way or the other and once a collective opinion is formed it is very difficult to change that. Followers are everywhere and things only solidify with time. Once you are in the collective “dog-house” it is much more difficult to revert this status.

I see asking for this sort of information more like throwing a bone. In the end I’ll figure it out anyway. In this example with some digging I finally found one of the original designers of the Motronic back in Germany. It helps that I speak the language.

It’s pathetic that certain individuals “sit” on information that they themselves have obtained somewhere while at the same time mocking others for not knowing. How f.cked up is that.

I’d like to mention individuals such as like Steve Wong, John Walker, Steve@Rennsport as well as many others in the business as very different examples. They are very open and forthcoming with information. And this by all means has increased their business rather than taken away from it I would figure.

Cheers,
Ingo

[ianc]

ianc, I beg to disagree with your statement

It’s pathetic that certain individuals “sit” on information that they themselves have obtained somewhere while at the same time mocking others for not knowing. How f.cked up is that.

Funny, sounds like an agreement to me...

ianc

[s5uewf]

Has anyone been able to come to a reasonable conclusion about which alternative injectors are acceptable to use?

- the 944 model injectors?
- the Ford SHO injectors?
- Chrysler 6 cylinder injectors?
- the "360's"?
- th3 "159's"

Or should we stay with the stock injectors at 3x the price?

[dfink]

Here is the summary as best I can tell.

There are many injectors that will fit in and squirt fuel. All the injectors fire at once. They are not sequential. So I would think anything that has the same flow rate would work for the normal driver.

The 159s have been used and commented on as working OK.

So far no one has posted any hard tested numbers on fuel ratios after changing.

You will not find any parts house listing a direct replacement other than the injector remanufacturing (cleaning) places.

Replacement of low impedance with high requires a modifcation to the cold start circuit.

[ischmitz]

Just an update from my recent work on a 911 DME:

The 911 DME injector output stage is specifically designed to handle low impedance injectors. The injection duration signal from the CPU is converted to a peak and hold signal in a custom IC: For about one millisecond the injectors are driven by the full (peak) current. This allows them to open quickly. Then a high-frequency signal with defined duty cycle is applied to reduce the current to the hold amount for the duration of the opening time. Since the injectors are not able to "follow" the rapid on/off sequence the net result is a reduced current flow while they remain open. Finally, a flyback stage takes care of the transients caused by switching off the inductive load.

In case of the 911 DME the peak current and the hold current are defined by passive components inside the DME. I know from other experience (Megasquirt-II) that insufficient hold current levels will make a car idle fine but run like crap under load. So it is vitally important that the injector characteristics match the design.

In case of Megasquirt the entire peak&hold injection signal is generated on the CPU. Opening time (peak current duration) and duty cycle (hold current) can be adjusted by software changes and allow precise matching to a variety of injectors.

Ingo

[rick-l]

Originally Posted by ischmitz

Just an update from my recent work on a 911 DME:

Finally, a flyback stage takes care of the transients caused by switching off the inductive load.

Anyone notice the subtlety of what T404 driven off of pin 5 of the fuel injection controller ic along with D401, the 47 volt zener diode does?

When you are in hold mode T404 is on. It lets the current from the injector coil recirculate and maintains the energy stored in the magnetic field. If all you wanted to do was provide a path for the injector coil current a diode (like D1) would do that just fine.

When you want to collapse the magnetic field in the injector coil (to make it close quickly) you hold pin 1 at ground, leave T404 off, and the energy in the coil is dissipated in the zener @ 47V * I coil.

Think that is the way it works? I can’t imagine any other reason for T404.

[ischmitz]

Rick, I think D401 (47V Zener) is an active flyback clamp. It clamps the flyback spike to something less than the Zener voltage by turning T402 on once the avalance voltage is reached.

T404 in combination with D1 is the flyback damping. See this schematic below "borrowd" from the Megasquirt site. It think it is the same principle shown here from Bowling & Grippo:



For the full description go here. http://www.megamanual.com/ms2/pcb.htm They realized the active clamp with its on transistor TIP42 and use the same flyback damping with the diode FR302 and transistor TIP125 controlled by the processor.

[ianc]

Very interesting and makes me realize how dumb I am.

What does this esoterica mean in terms of using an alternate injector such as the 159?

ianc

[ischmitz]

Inac, I think the bottom line is that the Carrera Motronic uses a peak&hold scheme to operate low-impedance injectors wired up in parallel. Once you use high-impedance versions you don't risk damaging things inside the DME. However, you might end up with lean conditions under load if for some reason the injectors do not stay open as long as the DME "commands" them to. As long as you use replacements that have similar dynamic characteristics and similar DC resistance you are probably O.K.

Ingo

[rick-l]

Originally Posted by ischmitz

Rick, I think D401 (47V Zener) is an active flyback clamp. It clamps the flyback spike to something less than the Zener voltage by turning T402 on once the avalance voltage is reached.

This is what I was thinking.

• RED T402-on +12 V on top of coil, ground on bottom CHARGE
• GREEN T404-on T402-off +12volts on both sides of coil HOLD Energy in coil decays only due to resistance in coil and driver
• BLUE T402-off T404-off Pin 1 totem pole output at ground. +12 on top of coil, +47 on bottom CLOSE INJECTOR , collapse field

Just a guess at how it works

[dfink]

A couple of more thoughts. When I started this mess I was under the assumtion that the system used sequential port injection. My assumtion was the Peak and Hold was required to get enough fuel in short time given.
I believe it has now been fully understood that the system fires all injectors at once. This gives 5 shots of fuel just laying around waiting for their turn to be sucked in with the last and final shot. (Which by the way has anyone figured out is there any real timing with fuel or does the system not care when that last shot is delivered.) My point is why did Porsche use the higher end system to open and close quickly when they have 6 shots to get enough fuel for the intake cycle. Or was it just "THE THING" to do at the time.
Also instead of a fancy DME could you not just use some sort of Pulse Modulation circuit that was controlled by RPM to squirt fuel. Then use some sort of crank fire system for the spark. (Ha perhaps that is what the DME is.) When you guy are testing the DME does the frequency of the fuel circuit change or does the frequency stay the same with the % open and close during that pulse changing. Just curious now.

[scarceller]

Originally Posted by dfink

This gives 5 shots of fuel just laying around waiting for their turn to be sucked in with the last and final shot.

dfink,

I'm not sure what you mean by 5 shots of fuel followed by the last and final 6th shot?

My understanding of this system is that it injects 2 shots to EACH injectors per complete power stroke. Meaning 1 shot of fuel for each turn of the crank and all injectors get fired at the same exact time. I also am under the understanding that the fuel delivery timing is fired off of the Top-Dead-Center sensor on the crank. So one shot of fuel every time the crank hits the TDC sensor mark.

This design allows each injector to deliver exactly 1/2 the fuel volume each time the injector fires. This is because the injectors fire 2 times per power stroke.

This is simply my understanding of the Motronic system. I could be wrong, but from everything I've read this is how the system works.

Could you clarify what you mean by 6 shots of fuel? Are you speaking about all injectors? and counting each injector individually?

[scarceller]

Originally Posted by ischmitz

Inac, I think the bottom line is that the Carrera Motronic uses a peak&hold scheme to operate low-impedance injectors wired up in parallel. Once you use high-impedance versions you don't risk damaging things inside the DME. However, you might end up with lean conditions under load if for some reason the injectors do not stay open as long as the DME "commands" them to. As long as you use replacements that have similar dynamic characteristics and similar DC resistance you are probably O.K.

Ingo

Hi Ingo,

I agree with you, Hi-Imp injectors will not damage the DME. I'm a EE by education and I reviewed the schematic with some other EE co-workers and we all agreed no harm in running Hi-Imp injectors.

But will they work as well as the stock Lo-Imp ones? that's another story and would need much testing.

[ischmitz]

dfink,

The DME is based on a pretty simple principle: It calculates the amount of required fuel per engine cycle to match the mass of ingested air. It essentially needs to meter the amount of O2 molecules (air mass) and deliver an exact amount of fuel to match this mass of air for optimal combustion. First, the barn-door senses the volume of air.

The fuel delivery is metered by the following assumption: Fuel injectors seeing a fixed differential pressure (fuel pressure - intake plenum pressure) will deliver an amount of fuel equal to their opening time. Fuel = flow rate/(opening time %)

The DME squirts at a fixed number of times per engine cycle. In other words if the engine rotates twice as fast it needs twice as much fuel and ingests twice as much air (under ideal conditions). So the injection events will happen twice as often. Is that what you where asking for?

However, there is a lot of fine tuning:
- Use the air temperature to concert the air volume into air mass. It is the mass that counts.
- Correct injector opening times by battery voltage since lower voltage will result in slower opening (less fuel)
- Use the O2 sensor to check and trim the amount of fuel if required
- Use the engine RPM and load to account for different fill efficiencies (tables)
- Use the head temperature to decide whether to richen the mixture during cold start
- Use the throttle position to decide whether to use best power or best fuel economy
- Use the throttle position to decide whether to idle

Since the fuel has nowhere to go once being injected into the intake it will "wait" there until the intake opens. This assumes nice hot intake runners and all that. On the downside the finely atomized fuel clouds have time to condense to bigger droplets. So from an emissions point of view injecting just before the port opens has advantages. However, from a design point of view the sequential injection required 6 times more discrete components and was at the time more expensive to realize. Today with integrated VLSI and thickfilm power devices costs are not an issue any more.

The peak & hold vs. high impedance decision is made for another reason. It is the dynamic range of the injectors. There is a minimum time and a maximum time the injectors can stay open. The maximum time is something like 90% duty cycle and the minimum time depends on injector design (peak & hold) have lower minimum times.

For a given application you want to make sure you do not exceed the 90% duty cycle (defined by RPM, maximum amount of fuel required, fuel pressure). This leads to the selection of the flow rate of the injectors. On the other end the minimum amount of fuel an injector can deliver is given by its minimum opening time, the fuel pressure and its flow rate. If the minimum time is too long it will be impossible to tune for proper idle since there will be too much fuel or no fuel. So this is one of the reasons peak & hold was used for high-HP applications. They react faster and their opening times are about 1msec. High-impedance are slightly higher. Of course today this might have changed to where newer designs are better.

Am I an expert: Hell no, but read here to educate yourself: http://www.megamanual.com/v22manual/mfuel.htm Despite all its perceived disadvantages Megasquirt is one hell of a learning experience. And that in itself made it worth my while when implementing it on a turbocharged Type I motor


Hope that helps
Ingo

[stlrj]

So far the Mopar 3.3 Hi-imp (saturated)injectors I'm running on my 86 3.2 are excellent. Idles smooth, great low end response from tip in off the line all the way to red line.

Joe Garcia
86 3.2 w/ stock chip, Mopar ignition cables cap and rotor, 0.060" plug gap ( who needs MSD when the stock ignition system is just as capable?)

[scarceller]

Originally Posted by stlrj

So far the Mopar 3.3 Hi-imp (saturated)injectors I'm running on my 86 3.2 are excellent. Idles smooth, great low end response from tip in off the line all the way to red line.

Joe Garcia
86 3.2 w/ stock chip, Mopar ignition cables cap and rotor, 0.060" plug gap ( who needs MSD when the stock ignition system is just as capable?)

Hi Joe,

Yes I remember us talking about these. What injectors are you using exactly? From what car? I would like to obtain a set for testing in the near future. I currently have a set of Bosch 320 injectors (Lo-Imp) that I plan to test soon and would like to test the ones you have as well.

I have a WideBandO2 setup in my car and I plan to record AFRs against RPM at WOT in 2nd gear. I'm still tuning my WOT map with the stock 158 injectors to get a good baseline, once I have this baseline I'll start testing the 320s.