Max flow for MFI injectors?
 

What is the maximum flow rate and/or hp rating (assuming 6 of them on a 911 motor) for MFI injectors? I think I've seen references to conversion of stock injectors to RSR or turbo specs? Is that needed to support 350-400 hp? Thanks.

Scott

Scott, I just recently finished modifying a 013 pumps space cam and flyweights along with readjusting the governor springs for a 2.8 RSR spec motor. Using stock injectors I was able to achieve a flow rate of 68ml - 70ml/1000 strokes/per injector @ 3750 pump rpm with WOT. So adding up all 6 would be 408ml-420ml per 1000 strokes. This would produce around 1,640ml per min. if my math is correct.

Then at twice the test rpm would that be apx twice the fuel delivery quantity?

Doesn't 1000 strokes per min equal 6000 RPM?.
After all....1000 (each injector) X 6 Cylinders = 6000/min.
Just a thought.
Bob

The MFI pump is driven off the cam so it turns half the engine speed. All testing data from Bosch is given at the MFI pump speed of half the engine speed. So the 3750 rpm on the pump test equals 7500 engine rpm.
As for the 1000 strokes, all the test data for proper flows is recorded per 1000 strokes at pump rpm's from 400 to 4000 rpm's. If you are testing the flow for idle @ 400 rpm and zero degree throttle the test period will take 150 sec. or 2 min. 30 sec. to achieve the 1000 strokes of fuel. Testing at 1750 rpm would be 34.3 sec./1000 strokes; testing at 1000 pump rpm is 1 min. to get your 1000 strokes of fuel. Hope this helps.

Just curious... does anyone know the duration of the injection event? I would assume relevant units would be camshaft or crank degrees I assume?

Rough order of magnitude would due.... I'm curious if the event is short enough that changing the advance/retard would be useful for trying to tune out reversion with narrow lobe cams (ie 906 grind).

t

Rumor has it the MB MFI injectors have more headroom than the Porsche ones.

As far as the injection event time, it would be degrees and RPM that decide, right?

Interesting question.

I wonder if the duration is bassically fixed but the pressure increases?

If duration dose not stay bassicaly the same I would suspect the finnishing point of the duration probably stays the same and the pulse justs starts sooner.

Just a guess.

Remember the injection event comes from a piston activated by a cam that pushes out a given quanity of fuel. Thus, the event would start slow, increase to a peak, and slow near the end of the cams ramp.

Thus the end point of the injection piston should alwas be at the same point.

MFI injectors open at a fixed pressure - since liquids do not compress, the expectation is that the duration is fixed, in degrees.

My thinking is the injection pump piston and stroke are a fixed in size, the supply pressure is constant and that the variable is how much fuel is allowed to enter the injector cylinder.

If this is right my vote would be that the duration increases but begins earlier and that pressure during the event also increases to a degree.

Also, at higher rpm each injection event has less time so the pressure will also increase with any rpm increase and assuming the TQ level is the same, the duration will actually shorten.

Again, interesting question.


I would be wrong if the stroke length of the injector pistons changes. I have never seen the barrels and injector pistons. I guess it would make more sense for the volume above the injector piston to change.

Bassically I have no clue but the question tickeled my brain.

Keith, Barrels & pistons............

http://forums.pelicanparts.com/uploa...1276039768.jpg


http://forums.pelicanparts.com/uploa...1276039788.jpg

So how dose it work?

Dose the piston move up and down on the push rod to change the volume of the injection stroke?

Staring at my pump innards, it seems that the piston stroke is fixed and the main rack rotates the barrel around the piston increasing or decreasing the volume.

Since everything is all fixed, pump cam/cam/crank I should think one could advance or retard just like the valve cam shafts.

Would it be a matter of just measuring the degree/lift of the roller crank in the pump? Since it is a roller design I would think the cam is really aggressive (fast lift).

t

I'll do the best I can. Your correct on the fixed stroke and fuel supply. How does it work! Well, each piston assembly is connected to roller tappet and rides on a cam lobe. The cylinder is fitted in the upper section of the pump and is immersed in fuel and there are small holes on the sides of the cylinders for fuel to enter and flow through the cylinder. The inlet fuel hole is larger than the exist hole in order to keep the pistons top lands full of fuel. As the plunger is forced up by the cam lobe, fuel contained on the top piston lands is forced out through a check valve into the injector lines and the injectors. The amount of fuel is controlled by the main rack which is connected to the plunger via a toothed clamp which is engaged on to the main rack. Movement of the rack forward (towards the front of the car) rotates the plungers and opens up the metering lands which allows more fuel to enter delivery plunger. Pictures below may help in what I was trying to explain. This was a very fast explanation and only 50% of what "how it works"

http://forums.pelicanparts.com/uploa...1276045996.jpg

http://forums.pelicanparts.com/uploa...1276046018.jpg


http://forums.pelicanparts.com/uploa...1276046037.jpg

356rs:
Nice write up. I had to go dig out my box of disassembled pump and look - it's been a while :). Rack rotates outer barrel which intern is fixed to the piston which is spiral cut. Thanks for the correction.
Eyeballing the roller crank it looks long duration with low lift. So I would think, given we are working with reasonably incompressable liquids, that the injection event will be on the order of the intake valve event. Thus short of regrinding the roller crank all we can do is move the injection event so that it starts after the exhaust closes (or in that ballpark).

How does one actually set the relative positions of the motor and pump? Is there timing marks on the MFI pulley?

So does the spiral cut change the volume inside the barrel chamber or does it just "throttle" the fuel that is allowed to enter through the little port hole?

tadd: There are timing marks on the engine crankshaft pulley as well as the MFI pump and MFI pulley.
Flieger: Good question. As I see it, the spiral cut allows more fuel to collect in the piston land when the plunger is rotated toward more throttle. I will try to post a better picture on the plungers today. The 2.0 - 2.2 plungers are different than the 2.4 - 2.7RS plungers. The later plungers have a much larger spiral cut in them.

I do seem to remember that the Porshe engineers tried varying the timing on the pump on the 917.
If with drastic changes in timing, the result was about the same in HP.
Is that correct?
Bob

Thanks for that info. Mark. Is the 70 ml limited by the pump or the injector in this case? Understood that the pump needs to be modified to supply a 400 hp engine. Going on that assumption, do stock injectors need to be modified too?

Scott

Is the barrel just a simple cylindrical bore or is there some interesting geometry to work with the spiral piston groove to vary the amount of chamber volume. Unless the spiral groove causes the piston to move up or down relative to the follower, I do not see how the volume of fuel delivered can change with a fixed stroke.

Great pictures and info!!!


Some more guesses.

The area inside each barrel is probably always being pressurized with fuel by the fuel pump until the piston starts to move up and pressure in the barrel starts to increase activating the check valve.

The spiral cut in the shaft probably determines how far the piston is allowed to recede in the barrel. The more it recedes, the more fuel that enders the barrel. The more fuel in the barrel, the more that will be pushed out to the injector with the cam action.

As the piston comes down the back of the cam, this will deactivate the check valve and help suck fuel in for the next event.

Just a guess.

For those of you who are still confused, let me confuse you some more... Think of this pump as how a turbo charged 2 stroke engine works. In this case, the turbo is the electric fuel pump, and the throttle valve(s) are the barrels that are turned by the main rack, and the pistons are the same driven by the crank (cam in the MFI).

While in a 2 stroke motor the air/fuel/oil mix goes into the crank case, and the piston going down forces that mix into the cylinder, the MFI relies on the electric pump to supply the pressure/volume for the fuel to enter the piston while its down and the intake port is exposed. Once the piston rises past that intake port, the fuel is now sealed in the cylinder and pressure is now increasing as the piston continues upward, forcing the check valve to open (this is the 180-220PSI pump operating numbers we hear about), and fuel is on its way to the injectors.

Scott: I think the stock injectors might have a problem after 70ml/1000 strokes @4000 rpm. I don't know if they can be modified. The RSR injector is physically the same but has a different part number and most likely NLA. The bigger problem is that the electric fuel pump is about at it max when the MFI pump is delivering 70ml at each injector.
Flieger: Simple bore with the exception of the fuel port holes. The cylinder is fixed and the pumping plunger is able to rotate from idle to full load. At full load the spiral grove on the plunger is at it's largest opening inline with the fuel port hole in the cylinder. Also at full load the angle of the spiral grove closes the fuel inlet port at a later time which allows more fuel to be delivered.

Edit: Lots of the information gathered on the injectors would indicate they will flow more than 70ml/1000 strokes @4000 pump rpm. Good news.

Edit: Don, Great explanation.

Got it.

So how is the volume changed?

Is it just by opening the intake to the plunger more. It seems there has to be a change of volume inside the cylinder.

911st, again, think of it as a 2 stroke. the size of the cylinder does not change. The piston stroke does not change. The amount of air/fuel mixture you allow in (based on throttle) is what changes. This is how you regulate what goes into any engine or pump for that matter.

So when the main rack/barrels are at a low volume position, the port is at its smallest (port, throttle, gate, valve, whatever you want to call it), only allowing a small amount of fuel in based on a set pressure/volume that the electric pump supplies (~14psi & ~33L/hr). When the port is at is largest (WOT, higher revs, cold starting conditions, etc) more fuel is allowed to enter the chamber, thus more volume.

First picture shows the relationship between the cylinder fuel inlet port and the plunger spiral angle "metering land" at a point where as the plunger moves to TDC the fuel will be cut off sooner that it would compared to the lower picture which shows the plunger rotated more towards full load. You can see the spiral angle now at more of a full load cuts the fuel off later as it goes to TDC.
http://forums.pelicanparts.com/uploa...1276112212.jpg

http://forums.pelicanparts.com/uploa...1276112227.jpg

I can see that its hard to get your head around because petrol isn't a compressible liquid like the example I gave with the two stroke engine. But don't let that confuse you. We're talking very small change in amounts of fuel per cylinder (in the mfi pump)... and keep in mind RPM will be the biggest factor of increased fuel volume (just as a regular engine pushes a lot more air at high revs than low revs).

No seals and .0001 tolerance. But there is a sealing ring of fuel. Below the spiral metering land on the plunger you will notice that little slot. That slot is opened in the middle and connected to the center of the plunger which has a small bore drilled up to the top. Fuel is allowed down into the port and then to the slot. In the cylinder there is a very small grove cut into the bore below the fuel inlet port, about 20mm down from the top of the cylinder. The grove is about .3mm wide in the cylinder bore. The purpose is to allow pressurized fuel to fill the grove when the plunger is at BDC. Is this some sort of seal? Maybe helps lube the bore? Oil lubes the plunger below the cylinder.

I think I have got it.

It is not that the amount of fuel that enters the barrel is limited or metered.

The barrel's fuel intake port's effective roof height moves up or down (thinking 2 stroke).

Thus, if the roof is low the piston will close off the barrel's intake port early and there is a large amount of fuel delivered.

If the roof is high, the piston will push fuel back out the intake port at first and until the port is later effectively closed. Then the lesser remainder amount of the fuel in the barrel cylinder is delivered past the check valve and to the injector.

Is that it?

911st, your on the right track. The roof is the spiral slot on the side of the piston (adjusted by left/right), and there is a bleed port in the cylinder that works with that spiral, the bleed off port stays in the same spot. every time the piston goes down, it fills with the max amount of fuel. At a low fuel supply setting (low rpm, idle, etc), the piston heads to the top of its stroke, and the piston's spiral lines up with that port the whole time- bleeding off as much fuel in the cylinder until it passes that port, sending the rest of that fuel into the lines. when that piston is turned to a high fuel setting (high rpm, wot, etc), the piston heads up, and the bleed off port is closed off sooner- sending more fuel through the lines.

Thank you Don and Mark for sticking in there with me.


Then we should expect the following"

The fuel delivery event is always completed at the same time relative to crank shaft and intake valve position. However, it starts earlier at higher delivery quantities.

At some point if a larger piston is needed to increase delivery, it would shorten the duration and should increase the delivery pressure. (Same fuel in a shorter time.)

The more open the metering function is, the more fuel we send, the earlier the duration begins relative to degrees of crank shaft position or relative to the intake valves opening.

The higher the rpm rate, the higher the delivery pressure may become (same quantity in a shorter time).

The injector's pop or opening threshold being the minimum.

The rate the cam's ramps, injection piston size, and rpm would all be factors that determine if the quantity of fuel delivered is such that it might be to much for the injectors.

I suspect the first response or test would be elevated pressure at the injector as the fuel will have no where else to go.

Plastic injector lines could thus have a function as a dampener. If so, this will lengthen the delivery duration to a later point relative to intake valve opening and crank angle.

What is the injection timing relitive to crank shaft position?

I suspect we would want it to be finished delivering its fuel near 90 deg ATDC on the intake cycle where intake air speed is at its greatest for better mixing but do not know.

I do know I once had my pump 180 deg out. Ran fine at idle, ran fine at high rpm. However, it had a major stumble in the middle.


Also, is the pump cam a big offset circle or an egg shape?

The plastic lines are really thick, and are rated for some serious pressures. There won't really be much delay, as we are working with a hydraulic system... there is pressure in the lines due to the injectors, and we can see problems if those injectors are not holding to spec. As soon as the check valve in the pump opens, pressure in the line is already high, I don't think its going to take much but the volume to get the injector to open.

I also would like to see a macro side shot of the fuel piston cam.

OK, thanks 356RS, I now get it. The fuel volume is metered as it enters the chamber using the spiral groove and the port hole, my second theory.

Thanks. :)

Requested photos:

Space cam:
http://forums.pelicanparts.com/uploa...1276120686.jpg


http://forums.pelicanparts.com/uploa...1276120713.jpg


http://forums.pelicanparts.com/uploa...1276120729.jpg


http://forums.pelicanparts.com/uploa...1276120743.jpg

Camshaft:
http://forums.pelicanparts.com/uploa...1276120775.jpg

Injectors
 

Injector maximum flow...

The MFI injector will deliver whatever the MFI pump FORCES through it. Do not confuse its operation with an EFI injector where pulse width determines fuel delivery. Metering is acomplished by the MFI pump plunger assy. The injector is just an atomising assembly.

Assuming you could "overload" the injector....it would still deliver what was being forced through it, just the pintle no longer chatters on its seat and remains constantly open, and atomisation suffers. Your next point of concern would be pintle "choking" resulting in enormous line pressures and the resulting mechanical failure (rupture) of a line, fitting, pump barell, pump cam etc.


The difference between the Mercedes Benz/ Aston Martin and Porsche injectors....

Early Mercedes (and Aston Martin DBS V8) injectors (DC 8 C 45 R1) used a 1.8mm pintle - these were superceded by DC 8 C 45 R2 which used a 1.2mm pintle (the same pintle as the Porsche EP/DCC 45 R3 ) The body lengths of the MB/AM injectors are shorter than the Porsche injector, and they also use a M14x1.5 mounting thread instead of the M12x1.5 of the Porsche injectors.

The AM DBS V8 is 5.34 liters......667cc per cylinder....equate that to 6 cylinders and you get 4.0 liters.....so you can see that your MFI injectors have ample headroom for your MFI engines (they even have sufficient capacity for your turbo MFI engines ala 935) By headroom, I mean the pintle's capacity to effectively atomise at the requisite rate of fuel delivery.

Thanks John for the information on the injectors. Now we know.

Metering and sealing (RSR style)
 

Here's a snippet of a great document posted by David E Clarke that explains metering and the "oil block"

The oil block was used to seal and lubricate the racing pump plungers, but not on the 911 road cars.

http://forums.pelicanparts.com/uploa...1276123634.jpg
http://forums.pelicanparts.com/uploa...1276123650.jpg
(C) Robert Bosch GmbH

This reminds me of BP's "top kill";)

Force oil up at a high pressure to prevent fuel getting down. The grooves serve as an expansion chamber to collect and discharge the fluids after fulfilling their "dynamic sealing" duties.

That's a perfect explanation Max.