Sequential Injection?

[foliage]

Do these boards use sequential injection at any time, or are they batch fire? If so does anyone know the details of how the injectors fire?

Just curious.

Thanks.

[Torque]

As far as I know from the Z32 manual the ECU uses batch mode while warming up and
later switches to sequential.

[PL]

It's nothing to do with the boards - it's how the ECU was originally programmed to run.

It varies between engines too. Easy to hear when bench testing as the injectors make a very distinctive sound when they go batch fire.

Here's what I've found:

All SR20DET (except RNN14) and 16-bit Z32 are batch fire at all times.

All ECU's go batch fire if they go into limp-home mode.

CA18DET is sequential at low loads and goes batch fire at high loads.

All others are sequential all the time.

This is just what I've found during bench testing of the commonly NIStuned ECU's. There will be others that obey/break these rules but generally the above holds true.

PL

[Torque]

Here from the Z32 (1990) manual.

Maybe I got it wrong and it is indeed batch?
Hmmmm ..

I always considered batch injection inferior and to be from the dark ages of injection systems ..

[PL]

I think they just copy/paste the same section in every Nissan manual! S15 manual has the same stuff but I can tell you those injectors batch fire all the time on the bench.

PL

[Torque]

Hi Pete,
Sure I believe you ..

And you say that also the 16-Bit Z32 ECU does this?

So I wonder what all the fuzz about latency is about when the ECU fires all injectors at the same time anyhow.
(It will relentlessly spray against fully closed intake valves (except for cylinder one))

Good thing that the ignition is not batch fire as well

Time for some reading, I always skipped the chapter 'batch injection'

[PL]

Latency is still perfectly valid. All the injectors fire simultaneously, but the time they take to open and close is still a required parameter whether sequential or batch fired.

Yeah, I was amazed when the 16 bit Z32 ECU batch fired the injectors but I checked everything over and over - even pulled the NIStune board out and put it all back to std. There was nothing I could do to make it fire sequential and it wasn't in limp mode. This is the only 6 cyl ECU I've ever seen do this!

PL

[foliage]

Interesting stuff pete!

[foliage]

Someone over on the SAU forums thinks the ECU boards only have 2 injector drivers? If they are sequential they must have 6 surely.

[PL]

Yeah, over on SAU they think a lot of things...

It's been awhile since I took a detailed look but I'm pretty sure they use injector driver chips with 4 drivers each. So a 6 would use 4 elements from one and 2 from the other.

There's sure to be somebody who knows the ECU circuitry better than me but I'm pretty sure that was the deal.

I can definitely tell you the RB20/VG30 is sequential cos I've got 6 injectors hooked up to my test rig and you can distinctly tell that they're firing sequentially.

PL

[foliage]

That is what I thought, thanks pete.

[foliage]

Think I just found where the misinformation came from:

(a) R32 RB20DET ECU with external VCT controller (Buddyclub etc) and Type 2 board with modified TPS Idle firmware

Required to swap injectors 4 and 6 since these are different between RB20 and RB25 looms

As I pointed out in the other topic, swapping injectors 4&6 is not necessary. RB20 ECU only has two injector drivers, 1+3+5 and 2+4+6.. If you look inside the ecu, you'll see that injectors 4 and 6 are the same signal.

Also for Z32 ecu, the injector pins are correct from the start, so the swapping suggested in the pdf shouldn't be done..

Interesting about the RB20 injector wires swap. I've been watching a local tuning place using R32 ECU's on RB25's (with RB25 loom) for some time now and after seeing some engine failures on those cylinders they started to wonder. So they did an experiment one day and ran a car up before and after swapping the wires to see if it made any difference. They actually got a substantial power gain when the wires were swapped.

I'll have a look at the wiring myself when I get a chance but the evidence points to a wiring change being required.

PL

You can check it without removing the ecu, just measure signal wires from the injector connectors.. 1+3+5 and 2+4+6 should beep straight to each other.

http://www.nistune.com/docs/NIStune_Typ ... lation.pdf

Above pdf has a RB20DET ecu image at page 5.. There you can clearly see the two green capacitors, at left lower corner. RB26DETT or Z32 ECU has six of these, as seen in page 8... They are part of the RC circuit, one resistor and capacitor for each injector driver. I know this is a bit lame proof, but easy to check.

Think his test was flawed, if they use a common ground and the injector drivers are not powered when he does the test then they could well be all joined together when the multimeter is hooked up if they are powered from the same bank of injector drivers in the same chip.

[PL]

Dammit - now I'm gonna have to pull an R32 ECU apart again and trace some tracks!

PL

[foliage]

Just want to be 100% that it is sequential as those posts make it sound like it may not be, also if it is sequential if pins need to be swapped between two injectors. Matt later said it wasn't necessary? Do you know why he said this?

Haha you are awesome pete, thanks for looking into it!

If the pins do need to be swapped would explain the idle misfire me and another member have. I just put it down to running lean at idle.

[PL]

I just wish I had a better memory...

PL