Alternative Fuel Injectors

[scarceller]

First, I respect that you disagree and I will be the first to say I am sometimes wrong but here is what I see by looking at the DME schematic. I do not know much about how or if Hi-imp injectors will work in our Carrera and I agree that much testing would be required. However, I disagree with you on how the stock injectors are wired to the DME, from what I see they ARE wired in parallel. Please read on.

It clearly shows that pin 14 & 15 of the DME harness are the feed lines to the injectors. One pin feeds the 3 injectors on one side while the other feeds the other 3. But in the schematic it also shows pin 14 & 15 tied together and being supplied only ONE signal from the drive transistor.

Here is the entire schematic: you can also get it here in PDF form:
http://members.cox.net/widebody//DME/Schematic/911_an.pdf
http://members.cox.net/widebody//DME/Schematic/911_dig.pdf
Look at the 911_an.pdf (Analog Board) and then look at how and what pins 14 & 15 are connected to. My conclusion is that ALL 6 injetors get the same exact signal. Furthermore, it is driven by only 1 transistor T402 that is a Darlington style transistor part #RBDT65A.

If after reviewing the schematic you still feel I am not correct I'd love to discuss this. I am not always 100% certain and am open to furthering my understanding. Can anyone else help here?

Here is the schematic but it's best to download it in PDF version:



Here is the wire diagram for how the injectors are wired to the DME, this is from the DME-Test Plan:

Originally Posted by dfink

I have to disagree with your calculations. In this case you can not consider that there are 6 injectors in parallel because only one circuit at a time is being activated. Also because of the short frequency you would never see anywhere near that type of current draw. Granted the high impedance injectors are going to draw less current in a steady state. People say they work but have not tested with AFM. Keep in mind the original idea behind low impedance injectors. Hit the injector with a very high pulse to get it open quickly then back it off to modulate the correct flow. It all happens quickly but is still controlled. I doubt however that you would hurt the DME by installing the higher impedance injectors. Here looks to be a good look at current flow for different type injectors.

http://www.msdfuelinjection.com/ecu_compatibility.html

[rick-l]

Has anyone figured out what the IC S400 0127 is?

I suspect it is a peak and hold driver that senses current through R409, a 0.1 ohm resistor.

[scarceller]

Rick, I have not tried to find the specs on that IC S400 but I can see that it drives the Base on the darlignton transistor T402. That IC is the driver that creates the pulses for the injectors. I also do see the feedback loop you refer to, it does seem to feedback the Injector signal to the same IC. I agree that your guess is most likely correct, in that it somehow is a current feedback loop to keep the Darlington from becoming toast.

Do you also see that the 6 injectors are in parallel? as I do? I have measured the resistance of my injectors at 2.3 ohm wich gives a resistance 0.38 ohms for all 6 injectors. Then I see resistor R409 in series with the .38 ohm load. But how do you know that R409 is 0.1ohm resistor? If it is then we have a load of .48 ohms. This type of load at 14.7V would yield a current of like 30amps. But of course the signal is pulsed duty cycle so the current would never be that high for long. But even at a pulsed duty cycle this current would still be rather high. For this reason I think your notion of a current limiting feedback loop makes sense.

I still want to know how you found the R409 value? Did you look at the DME board?

Thanks

Originally Posted by rick-l

Has anyone figured out what the IC S400 0127 is?

I suspect it is a peak and hold driver that senses current through R409, a 0.1 ohm resistor.

[rick-l]

Originally Posted by scarceller

I still want to know how you found the R409 value? Did you look at the DME board?

The schematic said its value was R10 and I assumed the R was the decimal point.

[scarceller]

Rick, yes I see that R10 now but maybe it's a 10ohm 1 watt resistor? That would make more sense. If it is 10ohm then we have total load of 10ohm plus the injector load of .38ohm for a grand total of 10.38ohm which would draw about 1.4 amps at full tilt. I like that number much better.

I currently have my DME open since I'm still tuning and burning chips, so I could easily look for that resistor and confirm.

My bet is it is a 10ohm 1watt resistor.

But it's best to check.

Originally Posted by rick-l

The schematic said its value was R10 and I assumed the R was the decimal point.

[dfink]

I think you are missing a few pins. I don't know about yours but I believe the connector to my injectors is 6 or 8 pins. Are you sure the drawings are all from the same year.
Common sense dictates that you can not send a single pulse. If you did there would fuel going into every cylinder nearly all the time. There is no mechanical distribution system so it has to be controlled electronically. I know the germans liked to supply full power and then control circuits using the ground. (by the way not a good idea) If all the injectors are supplied with a continuous 12volts then the ground is controlling the flow. Going strictly from the diagram shown the circuit appears to be controlled by a couple of cascading transistors with a resistor and a capacitor. In essence a one shot timing circuit. (This could possibly give the high flow initial pulse and then step up to higher resistance to lower the current draw. What is missing is the circuit in between that provides distribution.

[safe]

Originally Posted by rvanderpyl

Have a look here, you might find some info
http://www.msefi.com/viewtopic.php?t=2576

I have found the data I was looking for after posting that reply, but thanks. They are apparently 440cc @ 3bar. That is maybe 30-40% to big for my motor, but perhaps perfect for using ethanol as fuel

I would still like to fabricate a setup to flow them myself under some different fuel pressures. It would be great if the injectors suited both gasoline and ethanol since I will probably use an ECU capable of changing maps by flipping a switch.

[rick-l]

Originally Posted by scarceller

My bet is it is a 10ohm 1watt resistor.

Here is an obsolete part that explains how the peak and hold works. (not the one in the DME)
L584 data sheet source

10 ohms wouldn't make much sense. If you have six, one ohm injectors in parallel the 10 ohm current sense would dwarf the injectors. It would also limit the peak current to under an amp and a half (0.23 amps per injector).

[ianc]

I thought that with the DME, one bank of 3 injectors (R or L) Was fired at the same time?

I'm way out of my depth here having barely squeaked through 'Circuits and Systems' in college over 20 years ago, but thought that might go toward proving or disproving one point of view?

ianc

[scarceller]

dfink,

I was also surprised that I see all the injectors firing at the same rate. But remember this is how CIS works and it works well. As far as I can see I see all injectors connected to same transistor. I hope others will chime in here.

The scanned DME Test Plan drawing shows only 4 pins of that connector on the firewall being used. So even if the connector has more pins only 4 are used: see this picture:

Originally Posted by dfink

I think you are missing a few pins. I don't know about yours but I believe the connector to my injectors is 6 or 8 pins. Are you sure the drawings are all from the same year.
Common sense dictates that you can not send a single pulse. If you did there would fuel going into every cylinder nearly all the time. There is no mechanical distribution system so it has to be controlled electronically. I know the germans liked to supply full power and then control circuits using the ground. (by the way not a good idea) If all the injectors are supplied with a continuous 12volts then the ground is controlling the flow. Going strictly from the diagram shown the circuit appears to be controlled by a couple of cascading transistors with a resistor and a capacitor. In essence a one shot timing circuit. (This could possibly give the high flow initial pulse and then step up to higher resistance to lower the current draw. What is missing is the circuit in between that provides distribution.

[dfink]

Where do pins 1 and 2 go from the firewall connector. Is that a common ground. Doesn't look right to be comming from power. If that is a common ground then you have no power going to the injectors if the DME is a ground circuit.

[scarceller]

dfink,

Pin 1&2 go to that main junction point right under the DME relay that junction is not ground, it is feed from pin #87 on the DME relay and that pin is 12volts. All the wires at that junction get 12volts. This is how all six injectors get 12 volts on one side of each injector.

In the end there are only 4 wires that feed the injectors through that connector. Pin 1 & 2 are 12 Volts, then pin 5 & 6 goto pins 14 & 15 on the DME respectively. But then in the DME schematic you see pin 14 & 15 are tied together. Then if you further trace the line from 14&15 in the DME you find only one single source which is the Darlington Transistor T402. That is how I see things. I just am reading what I see on the wiring diagram and in the schematic.

The wiring diagram (showing how the injectors are wired to the DME) was scaned right out of my "DME-Test Plan" doc which covers 84-89 Carrera.

The DME schematics are for an 87 911 straight from Bosch. I have tons of docs like this. I tend to just collect this stuff.

Also, I agree that the fuel injector harness on the firewall behind the Heater Blower has more pins than 4 but it appears only 4 are used. This is often the case in many connectors, not all the pins are used.

If I'm wrong I hope someone else chimes in. Please. I would hate to pass on bad info.

Hope that helps.

Originally Posted by dfink

Where do pins 1 and 2 go from the firewall connector. Is that a common ground. Doesn't look right to be comming from power. If that is a common ground then you have no power going to the injectors if the DME is a ground circuit.

[dfink]

I was mentioned earlier that this would be similar in operation to CIS but didn't CIS have a fuel distributor?
I may have to take a trip to see "donner mause" and take a look. I have the service manuals for the 84-89 so I can take a look at those and see whats up. They won't have the DME circuit but will show the schematic for the car.
I think something is being overlooked. I still say they can't all turn on at once.

[dfink]

Well just looked at my service manual and you are correct in that is what it shows. There still must be more to the story because I just can't get my mind around the injectors turning on to every cylinder three times every rotation.

[rick-l]

Originally Posted by dfink

I was mentioned earlier that this would be similar in operation to CIS but didn't CIS have a fuel distributor?
I may have to take a trip to see "donner mause" and take a look. I have the service manuals for the 84-89 so I can take a look at those and see whats up. They won't have the DME circuit but will show the schematic for the car.
I think something is being overlooked. I still say they can't all turn on at once.

A Continuous Injection System (CIS) Continuously sprays fuel into the intake.

I think what he meant was the DME Continuously meters fuel into the intake by Pulse Width Modulating the on time as a percentage of off time of all the injectors (simultaneously). This happens regardless of intake valve position.

[scarceller]

dfink,

I had the same feeling but from what I see that's exactly how they work. The 84-89 Carrera system fires all injectors at once.

Also, this is not that surprising. The prior CIS cars fired all at once and they ran great.

Then in the 90s we got into multiport injection systems that controlled injectors in pairs or a 1 by 1 base. But these Motronic DMEs seem to run them all at once.

It's really nice to have the ability for us to share this info in a constructive manner.

And as I always say "It's about the learning" so if I'm wrong I always want to know it. And I'm willing to listen.

Thanks.

Originally Posted by dfink

Well just looked at my service manual and you are correct in that is what it shows. There still must be more to the story because I just can't get my mind around the injectors turning on to every cylinder three times every rotation.

[scarceller]

Rick, you are Correct.

My point on CIS is this: regardless of how the injectors are controlled (open and spray) the point is that in CIS all injectors sprayed the same amount of fuel at a given time. Meaning that one injector DID NOT spary diffrent amount of fuel than the other, they all worked together.

And in the Motronic DME it's the same thing. They all behave together exactly the same. Sure they use pulse signal to open and close to vary the amount of fuel sprayed. So yes at low load they spray less fuel than at high load, but they all spray the same amount with NO regard to cam timing. They are in NO way timed with the intake valve, that's my point. All 6 injectors get the same EXACT signal.

I'm actually happy with this mode of operation, it makes our cars that much simpler.

Hope that makes sence.

Originally Posted by rick-l

A Continuous Injection System (CIS) Continuously sprays fuel into the intake.

I think what he meant was the DME Continuously meters fuel into the intake by Pulse Width Modulating the on time as a percentage of off time of all the injectors (simultaneously). This happens regardless of intake valve position.

[scarceller]

Rick,

I also just finished reading a section in a book titled "Bosch Fuel Injection & Engine Management" on the operation of the injectors and right from the text it reads:

"Since about 1979, L-Jetronics regulated the current in the control-unit final stage, so do not need separate series resistors"

They are refering to the the firing of the injectors here.

So you are correct in assuming current limiting circuitry in the drive circuit for the injectors, at least this book confirms this.

Also in the same book they explain that the injectors are fired (all at once) 1 time per crank rev. This means 1/2 the fuel needed for combustion is injected with each crank rev. Remember the spark plug fires 1 time per 2 crank revs. This implies that the fuel just sits behind the valve till the valve opens and needs it.

I will scan these pages if you would like to read them.

[rick-l]

Originally Posted by scarceller

I will scan these pages if you would like to read them.

If that is the blue book by Probst with the heated wire anemometer on the cover I have it.

[scarceller]

Yes that's the book see section 3 page 10

Read "Fuel Injectors" and also "Fuel Metering and Pulse Time" sections.

Originally Posted by rick-l

If that is the blue book by Probst with the heated wire anemometer on the cover I have it.