EFI University

[Hugh]

Guy's,

I have Bosch 1600 cc injectors on my engine and tried hard to create a matched set of injectors when I first purchased them by obtaining two sets and having them flow tested then split into a low and high flow set.

Following Paul's thread on Dynamic Injector Flow testing http://efi101.com/forum/viewtopic.php?t=2610&highlight=injector+dynamics I wanted to find out a little more on how well matched the injectors really were and also to benchmark them properly for different fuel pressures.

See discussion on how I measure differential fuel pressure on a turbo engine here http://efi101.com/forum/viewtopic.php?t=2873&start=0&postdays=0&postorder=asc&highlight=injector+dynamics

I have allot of other injector flow data to share with you, but for now I would like to show you the flowrate performance of the injectors at different duty cycles and voltages.

This first picture shows you a really interesting profile at different voltages and injector duty cycles.



Alarmingly the injectors flow more at 1.5 milliseconds than they do at 1.65 milliseconds.

Paul explained that the spike is caused by the injector pintle bouncing off it's stops and partially closing again.

Although there are not many options you have in the Motec software to account for the variable flow rate across the range, you can at least apply an injector dead time in the Motec “General Setup” > “Fuel” > “Battery Comp” which dials out the largest part of the error.

Here is the actual flow deviation from the tests



Here is the Correction that we were able to apply to the ECU



Many thanks to Paul for the patience he showed while I bombarded him with questions and for preparing the ECU file with all corrections applied ready to slot into my car when the injectors make their way back to me through the Christmas postal system.

More to follow ...........

Cheers

Hugh
_________________
R34 GTR Nissan Skyline + twin HKS 3037S
OS Giken 3 litre block + OS Giken sequential
M600 ECU - still learning after 3 years

[Gadgeroonie]

out of interest how closely matched were all 6 injectors at different pressure's etc

[APEX Speed Technology]

Paul's been very good to us too. We've had some long conversations about injector performance and how relevant some things might/might not be. I really appreciate the analytical approach that they take....

Regards,

Neel
_________________
Neel Vasavada
Apex Speed Technology
2947 S. Sepulveda Blvd
Los Angeles, CA 90064
310-314-2005
www.apexspeedtech.com

[gt0311]

P Yaw is.
80% motec scientist
10% racer
10% UFC fighter !
100% customer service .

Hugh , your are on your way to make 1200hp .
best wishes.

[alexander]

Is he using a MoTeC as a test rig, and are those pulsewidths coming from a datalog? I'd imagine the injector driver (e.g. flyback design) also impacts the linearity of the injector, so each ECU design could require slightly different corrections.

[Hugh]

Gadgeroonie,

I have some graphs of the injector performance against each other that I will post up shortly.

You told me a while ago that have tuned a big power GTR, what size of injectors did it use?

Neel,

I agree, Paul was fantastic to deal with and was keen to make certain that he presented the information in the easiest manner to understand.

gt0311,

Agreed, great service and 1200 BHP would be good, nut probably more than I can aim for just now.

Alexander,

I will ask Paul if he will post an explanation of how the tests are carried out and how the injector driver circuit he uses will be matched to whichever ECU his customer has.

Cheers

Hugh
_________________
R34 GTR Nissan Skyline + twin HKS 3037S
OS Giken 3 litre block + OS Giken sequential
M600 ECU - still learning after 3 years

[VR6]

Great subject! Good reading...

[Hugh]

The reason I wanted to have my injectors flow rate tested is that the inlet manifold to fuel rail differential fuel pressure on my car varies considerably.

I have two Tomei, in tank, high flow rate fuel pumps running at 100% duty feeding through to an SX billet fuel pressure regulator.

For those not familiar, the Tomei in tank pumps are similar / slightly better in flow rate when compared to Bosch '044 pumps.

Looking at the data log below of a ¼ mile run, you will see the top trace shows RPM. The first hump is the launch RPM with a dip when the clutch is dropped, the car then bogged, the vertical blue line is first gear change and so on.

The graph below shows how the differential pressure across the injector fluctuates.

I need to do some work to confirm the accuracy of the Motorola differential pressure sensor I use, but there is no doubt that there is a variable fuel pressure and therefore a variable amount of fuel being delivered from the injectors.

The engine was running just over 2 bar boost when up in the rev range. The fuel pressure varies by 12% between when the engine bogged and when it was flat out in 5th, and since flow rate varies by the square of the pressure means that there must have been just under 6% actual flow variation.

In reality the actual fuel delivery is also probably very closely linked to the injector duty cycle and will therefore be consistent for a given set of mapping.

However, being able to dial out this error should keep things more consistent, especially if you get some pump wear or other malady that interferes with the fuel pressure.


_________________
R34 GTR Nissan Skyline + twin HKS 3037S
OS Giken 3 litre block + OS Giken sequential
M600 ECU - still learning after 3 years