Batch injection with siamese ports, injector timing

Converting To Delco ECU From Carby Or Other Injection Systems
VK_3800
Posts: 526
Joined: Sun Jul 04, 2010 3:15 pm
cars: SS Torana
Location: NZ

Batch injection with siamese ports, injector timing

Post by VK_3800 »

Hi all

Thought I'd better create a new topic to continue this discussion:
https://pcmhacking.net/forums/viewt ... 5&start=40

Specifically we're talking about a VN V6 setup on a Holden grey engine (with $12P). Its already been physically set up with 3 injectors in the manifold near the ports, using the V6 coil pack with the original crank angle sensor. Its running with a factory tune and some injector size adjustments, not particularly well but nothing worse than I'd expect from a completely un-tuned engine.

It was mentioned in the link above that Mini engines are pretty much a no-go with the siamese ports and batch injection, however I think that's a worse case than this due to the small number of cylinders and firing order - it will always have the cylinders sharing an intake port firing one after the other making it very difficult to supply the correct fuel to the second. With the six and a firing order of 1-5-3-6-2-4 this isn't the case. If we label the ports A (1 and 2), B (3 and 4) and C (5 and 6) for argument's sake, then the intake port order is:
1-5-3-6-2-4
A-C-B-C-A-B
with respect to the cylinder firing order. To my mind this means that with double fire injection (once per revolution rather than once per cycle), then each intake port has a fresh injection event unless the injection is for cylinder 5 (C-B-C) or 2 (A-B-A) - then the port needs to be used twice in one revolution. But at least not twice in succession, and every other starting position does not involve using the same port in a revolution.

Does the above make sense?

I know that the ECU doesn't have a position reference and would normally start injecting based on pulse inputs at any position. Being run by a V6 ignition module which does use a position (so it can fire the correct coils), does it solve this problem by starting the signal to the ECU at the right point, or is this just passed straight though? Unfortunately I'm thinking its the second which doesn't help. Is there any way to get around this, so it always starts the input to the ECU at #1?

When exactly do the injectors fire in relation to the spark? Tweaking the timing of this somehow may also be an option to improve the above problem, to stop the first cylinder in a shared intake port from removing too much fuel.

What is the limiting factor for double fire injection? Its not exactly a high revving engine.

Thanks for any input.
v6bucket
Posts: 761
Joined: Tue Jul 21, 2009 10:02 am

Re: Batch injection with siamese ports, injector timing

Post by v6bucket »

Can you give some back ground about the setup, what type of CAS etc.
"Definition of a Drag Racer: A person who spends money he doesn't have on parts he doesn't need to impress people he doesn't know." - Grumpy Jenkins----- Poverty---
Ken
Posts: 206
Joined: Tue Dec 17, 2013 1:05 am

Re: Batch injection with siamese ports, injector timing

Post by Ken »

Just a quick thought I had, not sure if it's a viable option, thought I'd throw it out there anyway..

If you're only running one injector per runner, couldn't you still connect both cylinder injector signal wires to the one injector with diode's wired in with each signal.
That way you're still running a mixture(distribution) similar to a STD carb setup anyway where 2 and 5 will still be slightly leaner because of the shared inlet runner.
Something like in this pic.?
Or are there problems I'm not seeing with such a setup
TwinSig.PNG
TwinSig.PNG (2.4 KiB) Viewed 7576 times
Probably no reason you couldn't do that with a mini either.
Last edited by Ken on Wed Jul 04, 2018 3:57 pm, edited 1 time in total.
User avatar
antus
Site Admin
Posts: 8237
Joined: Sat Feb 28, 2009 8:34 pm
cars: TX Gemini 2L Twincam
TX Gemini SR20 18psi
Datsun 1200 Ute
Subaru Blitzen '06 EZ30 4th gen, 3.0R Spec B
Contact:

Re: Batch injection with siamese ports, injector timing

Post by antus »

I know of a tuning shop in Adelaide had a classic with this configuration. It was 50/50 chance of the engine being on the right stroke when they started it. They'd key it on and it'd either run fine, or run like crap and they'd key off and key on again. They were looking for a solution but I wasnt aware of anything off the shelf. However I think if you were to do it then you could some kind of a set up that interupts the reference signal until its seen some kind of signal for the beginning the of the revolution from the cam. Then leave the reference connected until key off. It needs to be triggered by something which can identify position and stroke.

Ken, im not sure what your talking about. In MPI all injectors fire at the same time, and its a digital signal which opens them together for a given amount of time. An offset is added to the calculated amount of time compensate for the physcial injector opening time after the electrical signal is on (the dead time). The bit to get right is on the ignition side so of it, so it injects, then one of the siamese ports open, take the fuel, stroke happens and then it closes, then there is another injection event and the fuel goes in to the other cylinder. If your 50% out of whack the injection happens when both cylinders on the port are closed and then the first one to open gets double fuel and the other one gets none when its its turn. At least that was my understanding when it was described to me.
Have you read the FAQ? For lots of information and links to significant threads see here: http://pcmhacking.net/forums/viewtopic.php?f=7&t=1396
Charlescrown
Posts: 1831
Joined: Sat Aug 06, 2011 7:58 am
cars: V8 VR Commodore BT1
LB Lancer 2L turbo & Delco
Starion TBI with Delco
Mitsubishi Lancer EVO4 track car
NA MX5
3 vintage motor bikes
Location: Padstow NSW

Re: Batch injection with siamese ports, injector timing

Post by Charlescrown »

I don't see a big issue. The carbie was a compromise with the port setup so the EFI will be the same. I would think about using only 3 injectors of a higher flow rate ans accept the exhaust gas dilution in some cylinders. It would only have an effect at idle unless you have a big overlap cam and you get that problem anyway. The benefit of the Delco is you can have a lot more control over timing so put more in the low end idle area to improve the idle quality.
VK_3800
Posts: 526
Joined: Sun Jul 04, 2010 3:15 pm
cars: SS Torana
Location: NZ

Re: Batch injection with siamese ports, injector timing

Post by VK_3800 »

v6bucket wrote:Can you give some back ground about the setup, what type of CAS etc.
Pretty much just factory VN, only three injectors are connected and the factory trigger wheel is bolted to the front of the balancer. Still using the V6 ignition module and coils.
VK_3800
Posts: 526
Joined: Sun Jul 04, 2010 3:15 pm
cars: SS Torana
Location: NZ

Re: Batch injection with siamese ports, injector timing

Post by VK_3800 »

antus wrote:I know of a tuning shop in Adelaide had a classic with this configuration. It was 50/50 chance of the engine being on the right stroke when they started it. They'd key it on and it'd either run fine, or run like crap and they'd key off and key on again. They were looking for a solution but I wasnt aware of anything off the shelf. However I think if you were to do it then you could some kind of a set up that interupts the reference signal until its seen some kind of signal for the beginning the of the revolution from the cam. Then leave the reference connected until key off. It needs to be triggered by something which can identify position and stroke.
That's exactly it, by my reckoning we have 2/3 chance of getting it right but same idea. Was kinda hoping the V6 ignition module was actually doing this already but probably not.
yoda69
Posts: 1215
Joined: Sun Mar 15, 2009 10:20 am
cars: 2004 VYII Acclaim Wagon V6 Auto LPG/Petrol
2004 VYII Berlina sedan V6 Auto
2005 VZ Monaro CV8 manual
Location: Geelong, VIC

Re: Batch injection with siamese ports, injector timing

Post by yoda69 »

The problem is at idle where you have smaller pulse width.
In EFI fuel is pulsed in with correct amount on average to suit the air going into the engine, a carby is constantly fuelling so a more homogeneous mixture in the intake manifold when compared to efi.
Efi advantage is by the time the air and fuel has entered the combustion chamber it has swirled and mixed with the precisely measured fuel that was injected.
Not sure if 11P still has bank fire, but if you could use that and 3 x 2 groups of injectors you should be able to get it to work fairly well.

EDIT: would need bank and single fire for this to work
Last edited by yoda69 on Wed Jul 04, 2018 5:00 pm, edited 1 time in total.
VK_3800
Posts: 526
Joined: Sun Jul 04, 2010 3:15 pm
cars: SS Torana
Location: NZ

Re: Batch injection with siamese ports, injector timing

Post by VK_3800 »

Charlescrown wrote:I don't see a big issue. The carbie was a compromise with the port setup so the EFI will be the same.
Not quite the same with batch injection as the injection event isn't constant, or even for every cylinder fire. It relies on fuel collecting in the manifold and there is much more potential for the first cylinder to steal it all from the port and leave nothing for the second.

I have also read about a system where the injector is aimed directly at the valve for the lean cylinder and sized as large as possible with appropriate timing - this allows it to almost direct inject the lean cylinder while the rich one gets enough from collected fuel. Could be tuned either way but getting tricky to work with and seems like its overkill. If there were some way to solve the starting reference signal position like Antus described that would eliminate the 1/3 chance of getting it wrong. I wonder if it could be as simple as triggering power for the ignition module at the right moment with some sort of latching switch triggered by a position sensor? At cranking RPM it can't be too hard to miss?

I should really map the valve opening roughly with respect to injection to see where we're at I suppose. Can anyone fill in the gaps for me on exactly when the ECU fires injectors?
VK_3800
Posts: 526
Joined: Sun Jul 04, 2010 3:15 pm
cars: SS Torana
Location: NZ

Re: Batch injection with siamese ports, injector timing

Post by VK_3800 »

yoda69 wrote:Not sure if 11P still has bank fire, but if you could use that and 3 x 2 groups of injectors you should be able to get it to work fairly well.

EDIT: would need bank and single fire for this to work
Hmmm, not overly keen on adding injectors but I wonder if that's the answer to the position signal somehow? Changing ECU would require some re-wiring but if that forces it to always start injection at the same point it solves the problem.
Post Reply