Fuel Injector Recommendations (GENERAL EFI INFO)!!!
Turbo Mini Forums

*******************************************************
This post was started initially about injectors, but ultimately went onto becoming a pretty good read about the problems and issues when trying to inject an A-series - especially a turbocharged one.
For that reason, it has been made a 'sticky' to act as an easy-to-find reference.

Thanks,
TurboDave
*******************************************************




I'm planning on doing FI on my turbo. However, I don't know what size injectors to use. My first plan is to place them directly on the intake manifold (one on each port), as done by BigAl. However, I don't know what rating I should use. Too big, and I won't have the sensitivity to changes. Too small, and I won't be able to put enough fuel into the throat.

Suggestions are welcome!

Edited by TurboDave16V on 5th Sep, 2006.

**************************************
1967 Cooper 's' Turbo build @ www.drjing.com

It's going to be suck-it and see, as you're plannign on a wet manifold. With the injectors being so far away, I think your available injection window is even less than if you had them at the manifold face, due to the air velocity dropping off sharply (whilst your fuel is still leaving the injector)...

If you assume a 50% duty cycle, then for 200hp you're up to 350cc/min at a 45psi fuel rail...
It really depends on BSFC, the injector response, etc, etc, etc...
So - what i'd do is start with a set of 250 or 300cc injectors and see how it goes. You can get a baseline to work from then. If you buy a 2nd hand set, get them perfectly balanced, then you have a starting point to work from.

What we REALLY need however, is a non-propertional fuel regulator... Basically, one that will be typically 3-bar static. Ideally, it'll pull vacuum, such that it drops to 2.5 bar or so at tickover, meaning the reduced presure will not flow as much fuel - avoiding overfuelling.
Then, as boost rises, the fuel pressure wants to rise above the increase boost pressure - eg a stock 3-bar regulator at 1 bar boost will be up to 4 bar in the rail. If we could find a unit that increases this to (say) 5-bar, we'd be ramming more fuel through the injector for the same duty cycle - around 15% more by my basic-est maths... This Would solve a lot of problems!!!

If anyone can find/reccomend a suitable unit - I'll buy you a beer!!!

Edited by TurboDave16V on 27th Oct, 2005.

Thanks, guys.

1071, I saw that one in another thread, but it's greek to me. I'm not sure what bhp I'll have, nor how much fuel delivery I'll need at %, etc. I was hoping that someone had a ball-park figure/model injector to work with. Maybe I'll try a couple of used WRX ones to start...

Dave, how does one go about balancing the injectors? Is there a company that does this service, or is it a do-it-yourself deal?

**************************************
1967 Cooper 's' Turbo build @ www.drjing.com

Edited by TurboDave16V on 27th Oct, 2005.

If you can sort us something for athens, that'd be magic...
All very interesting stuff what you posted... we just need to get injection working in the first place, as anythign like that would require a balanced mixture... Same for ignition timing; does the A-series lend itself to different advance figures for the inner and outer cylinders - I suspect so...

I already have traction control based on 1st gear ratios/undriven wheel sensor in my DTA. It's got PID configuration, but i've never managed to get it to work satisfactorilly - probably because the friction between the front wheels changes so much - and the pretty firm spring rate in the front of the mini means the weight change (and hence friction twixt rubber/road) is pretty high during a launch.
Hat's off if you solve that one!

I've got an athens log on you can use dave

valid for 3yrs. free with my uni course. may help you out. if it does send me a message.

Hi Dave!

look at this homepage:
http://www.vortechsuperchargers.com/produc...l/fuel_fmu.html

They sell rising rate pressure regulators with different ratios! I think thats what you are after?!
Harry

Bimmer Twinky headed and turboed A-Series:
http://www.minifreunde.at/harry/projects.htm

Good one Harry!!!

US sourced aswell

uh- looked at it again... 6:1 rise????? WTF???

Edited by TurboDave16V on 30th Oct, 2005.

i think these doobry's are used by ricers in teh states when they ad a "blower" between teh "header" and "muffler" without having to mess wiht the "stock" ecu..

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

I think the ratio means:
e.g. 4:1; for 7psi boost the fuel pressure is increased 4 times the boost, which gives 28psi fuel pressure increase!

Bimmer Twinky headed and turboed A-Series:
http://www.minifreunde.at/harry/projects.htm

aaaahhhhh - That sounds more feasible!

I was thinking that 150+psi at the fuel rail sounds a little high!!!

hehehe - don't forget to take the fire extinguisher with you on your first trip around the block when running that fuel pressure!

...but I think the injectors will stop working for you anyway at that pressure...

BTW my datalogger is fully working! Pressure sensors measuring TIP and Boost now ... and 3EGTs fitted.

Bimmer Twinky headed and turboed A-Series:
http://www.minifreunde.at/harry/projects.htm

Nice one Harry... We need to talk

Did you ever sort the 'overboosting' problem with your turbo?



Incidentally - I've just emailled Vortech about the SFMU and told them exactly what i want; and can they supply a unit that'll do the job.

Even so - I'm thinkin that for 1.5 bar boost, a 2:1 rise is going to need a 90+psi fuel pump. A lot of pumps flowrate drops off RAPIDLY after 70psi it seems... This is gettign even more complicated!!!

If I fix it all - I'll be coming back to AvonPark before too long though - so watch out!!!

Dave,

two pump numbers for big hp and high pressure:

Bosch 0580254909 150l/h @5bar
Bosch 0580254053 175l/h @5bar

My overboosting problem hasn't gone, but I will change to a bigger turbine A/R with ported wastegate bore next year.

Bimmer Twinky headed and turboed A-Series:
http://www.minifreunde.at/harry/projects.htm

ok, back to the drawing board!


Spoke to RC Engineering - a well respected injector supplier.



A few interesting points:

Response time of a Low-impedance 500cc injector is around 1.1ms, Response time of a high impedance is around 1.25ms at 4Bar.
> One myth put to bed then.

Upping the rail pressure past 80psi yields very little gain in injector flow.
> Good to know.

Pulling pressure away at idle makes a significant difference to flow.
> Obvious, but good to hear.

More importantly, doing som number crunching after - we'll need a HUGE injector for a 220hp, 7000rpm A-series.
SO,
It needs to be two injectors per inlet tract. I'll initially be trying to control two large (high impedance) injectors per inlet tract. DTA has 2d.p. on the injection timing, so hopefully that'll help me. Alternativly, it'll have to be a 'staged' (switch on 2nd injector, and half the fuel count) injector setup (need to ask Graeme nicely as DTA doesn't do this till after 50% duty cycle - ie USELESS!!!).

It's likely i'll still use a vortech regulator. Will also swap to a MG Metro camshaft to open out the injction window - but mainly to revert to a nicer 'road' camshaft.



Still got to make a better clutch first though!!!

Edited by TurboDave16V on 5th Nov, 2005.

I did the same calculations like you Dave - and came to similar results. It is impossible to do a full sequential injection with just one big injector per inlet tract (for our hp targets)
Not even with some nice rising rate regulator.
There would not be any useful idle AFR. The Lucas injectors have a little less minimum opening time - but I don't know if they are still available with big injector ccs.
Do you have the formula for fuel flow vias fuel pressure? So you could do the calculations regarding the rising rate reg yourself:

New injector flow = (Root (New fuel press/old fuel press))*old injector flow

Bimmer Twinky headed and turboed A-Series:
http://www.minifreunde.at/harry/projects.htm

Yeah, Harry - done the math - and upping fuel pressure (and lowering it at idle) is definately the key.

I'm going to pursue using two injectors, along with a 'softer' camshaft, and the 'rising rate' fuel regulator adjustment.

I've just found out the DTA DOES support dual injectors (of the same size) by switching the second set in and instantanously halfing the fuel count. That basically means i can run (per inlet tract) one injector at idle, and two at full hit.

It'll be interesting to see just how far i can get with this setup - end of the day, I'll start with 'oe' 440cc injectors. If i 'run out' i can easilly get 550cc injectors that still have a good resolution for idle...

You know the answer - You been drinking?



As you can see - earliest time the fuel can enter the outside valve is (no less than) 30degrees ATDC. Latest it can shut, is (no later than) 40degrees ABDC.

If you inject too early, the first drops of fuel out of the injector will find themselves down the inside valve; If you hold the injector open too long, the excess fuel will sit in the port, collected next time the inside valve starts to open.

Assume 100% DC is '360 crank degrees', and (using marcels figures for a 256 cam) then we have 190/360=0.52% duty cycle - I assume he adjusted it a little talking into account his 'measured' values - or maybe for a 'stock' camshaft...

ps - running sequential might not be fully needed... what definately IS needed however is to be able to swing the 'start' of injection event - assuming you're firing the injectors once every 360 degrees that is... You could also do with having the math ready to determine when you're 'overshooting' the injection window.

THIS HAS JUST GOT ME THINKING THOUGH - UH-OH, IT'S EVEN MORE COMPLEX THAN I FIRST THOUGHT (HOT OFF THE PRESS)

Hovever - there is a possibility, that (when firing every 360 degrees) something nasty might happen - especially if the injectors are close to the head (which is what we want)...

Think about the inside valve now... The injector for this 'cylinder' starts firing at 150 degrees Before TDC on the inlet stroke! ALL the large amount of fuel is being sprayed into a 'closed' inlet port. This isn't a problem normally - but it is highly likely, that some of the fuel will venture down the port - and when it meets the divider, will split nicely - some resting on top of the inside valve (fine) but crucially, some might rest on top of the outside valve.

Oh a nat-asp, this 'excess' might simply be able to be 'sucked' off the outside valve, and into the inside cylinder. In Forced-induction however; very, very unlikely...

Solution? Fudge knows! Wet manifold? No thanks - but clearly mounting the injectors too far forwards, MIGHT present a problem...

Now - if you can fully sequentialise it, that means you ca run a decent injector position - AND - get spot-on fuelling.
But bear in mind - this isn't a 'normal' sequential - no, you're doing something much more specialised!!!

Typing this has prompted me to make an animation explaining the problems when trying to pursue a decent injector position.
I can really see a potential for an injection controller that you just wire into your injecotr 'outs' from your ECU, that sorts all this out in one hit...

Edited by TurboDave16V on 6th Nov, 2005.

Did you see the japanese fuel injected turboed Mini of Mr. Gouda in the last Miniworld?
Any comments? Any ideas/asumptions?

Bimmer Twinky headed and turboed A-Series:
http://www.minifreunde.at/harry/projects.htm

Nope - no-ones sent it to me yet?

Are you offering to scan and mail?

Lots to digest here! I'm enjoying it.

**************************************
1967 Cooper 's' Turbo build @ www.drjing.com

You appreciating how difficult it is yet?

As far as i'm concerned, there is no point doing a 'halfway house'.

Equally, there's no point doing it if it needs such large injectors that it can't fuel at tickover.

How about if the outside port was mildly pressurised higher than the inside port with a short squirt of air at the same time of the injection event- would that not encourage the fuel to take the path of least resistance?

well, now you've got to sort something out such that you inject the 'air' at thr right time - if you're clever enough to do that, then you'll be able to sort out sequential fuelling easilly enough!