SARD 63518 1000ccm injectors

[Mazi]

Looking for Bat voltage compensation table and setup...

[MarkMc]

Sorry Mazi,
Have not done this one.

[Dragon]

Sard injectors are tested at 2.5 bar of pressure, 12.5v.
They are specified as having 0.85ms.
Sorry, it is only a starting point, not a full table.

[Mazi]

Thank you Dragon

I know that they are low impedance injectors and I would like to avoid serial resistors. Will have to check them on test bench to figure out numbers.

[MarkMc]

If you have access to an osciloscope and a current clamp you can run a test yourself using the output test. Start with an injector current setting = 10volts/injector resistance.

On the scope you will be able to see when the injector opens by looking for the little "notch" in the rising current (see pic). The correct current setting should be double the opening current (current level at the notch) as long as it is not more than 8volts/resistance. Example: A 2.0ohm injector should not have a current setting of more than 8/2 = 4amps.

[Dragon]

Mark,

Can one use a similar method to find out the lag time of an injector at various voltages?

[MarkMc]

Theoretically yes but seeing as this is only an indication of when the pintle moves it is not an exact measure of when the fuel flows.

What we do is measure the fuel flow at a range of pulse widths, voltages and pressures. What you end up with is (say for a particular voltage and pressure) a line graph of flow to pulse width which will be basically linear but at the very low pulse widths the injector flow is eratic. If you extrapolate the linear part of the graph down to zero flow this will happen at a pulse width of more than zero...this is your dead time.

Another way you could do it is if you could vary the power supply in the car. Start with a 0 dead time table and hold the car at a steady voltage and tune to something easy like lambda 1.00 at idle. If you vary the voltage and leave the fuel tuning alone the mixture will change, simply use the Dead Time table to get your Lambda 1.00 back at each voltage. If you can do the whole table then you should be able to vary the voltage and always have lambda one. Repeat for different fuel pressures if you can.

[Mazi]

I'm using a little bit more expensive way. Laser interferometry. Very accurate measurement method. But unfortunately it is not always available. I will try to make some measurement next week to compare Mark's "current" method with laser one. I will publish the test results.

[Mazi]

Hm.. I can't find a glitch on current trace. .

Yellow trace is current measured with 1A/100mV 100MHz current probe. Green trace is connected close to injector apx 1.5m away from ECU.

Is it coincidence that current limiter is turned on exactly at 1280us?

[SprinterTRD]

The initial low time (1280US) is to make sure the pintal opens all the way. This time may change if the injector reaches peak current earlier and has a maximum timeout value incase the injector does not reach peak current. If you use negative numbers in the Injector current setting, the timeout value is allowed to be longer as some injectors build up current slowly.

This is described in the F1 help files.