injector sizing
Turbo Mini Forums

Anyone know the formula for working out injector cc against power?

I thought it would be relevent

To make life easier - some companies supply a page of maths - and a fuel injector calculator for your needs:

http://www.rceng.com/technical.htm

Dont forget the table I put up in the faq/knowledge section some time ago as well giving the flow rates of all the available OEM injectors and supported HP.

It's here http://www.turbominis.co.uk/forums/index.php?p=vt&tid=6400

Peter

Peter

Editor

MAGA7INE
Seven Magazine
The brand new quality Classic Mini Tech and Journal
Issue 1 out late 2027.

So... if I've done my math right(and I probably haven't), a 440cc injector w/ a 2.5ms injection window(port injection, 6 or 7000 rpm, whatever Marcel figured, I dont' remember) would only give you enough fuel to combine with about 50 cc's of air and give a good burn... how far off am I?

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

you asking, or telling ????

I was asking because I know that can't be right...
...and I think i've actually figured out what I was doing wrong. Do not confuse crank degrees with cam degrees when trying to figure out how long your valves are actually open, and also, when you need to double your injector duty cycle for a single port because 360 crank degrees happens in 1 port with 1 injector...

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

ah - that sounds more like it.

I'm starting to think that my 800c/min injectors might not be quite enough for the power and rev's I'm gunning for....

Still - where I've fitted two (per inlet tract) I'm sure I can fit three with a re-work.
Sticking to two for the time being though!

Hmm... soo... what are your goals?
I'm trying to see how small I can go to get it to idle well... and you're trying to use 2 or more injectors per port that are twice the size I'm gunning for.... heh..
going on that, are you going for 250hp or more?

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

Well, the BSFC must be assumed to be on the 'upper end' of the 'forced induction' for a turbo A-series (0.65 assumed) - and I want 7K rpm capability. 200+ hp is also desired.
All of this is adding up to bloomin big injectors!

Edited by TurboDave16V on 11th Jan, 2006.

Actually, you may even want to go higher than that on the BSFC.
I came up(using rc's calculator) that the MPI mini in stock form has a BSFC of 0.56 which is well above the range of a normal naturally aspirated car. I don't know how accurate those calculations are for s siamese port head, but from a "black box" type view, it seems to make sense.

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

Interesting stuff - I guess we'll see in time.

To put in three injectors per tract is overkill until some testing has been done (i believe)... All I'd need to do is machine some new bosses and do a bit of hacking and welding.

To be honest, I think i'll be very happy with a 175hp and 6500rpm - if this all these (four) injectors can deliver. At that point, it's not 'failure'; so much as another stage in an ongoing development...

Edited by TurboDave16V on 11th Jan, 2006.

With two 800cc injectors per port with a BSFC of 0.65 and 40% duty cycle, I compute that you could get about 187hp. If you were to use a 400cc and a 1600cc injector per port (which would be good for driveability and top end) with the same BSFC and duty cycle, you would be able to go to about 235hp. If you go to 50% duty cycle, the numbers go to 235 and 293 respectively.

With the staging code I included in the MSII siamese-port code, you could use these 2 different sizes and have the larger injector come in at 3000rpm or even 2000rpm and use a single VE table for tuning and a single timing point for mid-pulse timing.

Also, I don't think that the revs will be an issue since the desired duty cycle is about 40-50%. For this to be a problem you'd have to be revving the engine such that injector opening time is more than the 50-60% duty cycle available. With an injector opening time of about 1msec, that would be around 30000-36000 rpm which is a bit more than what an a-series engine can do :)

Regards,
Jean

http://www.jbperf.com/

ah - thanks for stating the obvious about the reving situation Jean... I wasn't seeing past the actual window we have and its relation in ms to the 'full' window of a conventional 4-port...

Edited by TurboDave16V on 11th Jan, 2006.

So... I seem to have noticed something. According to marcel's page, he says 210 degrees is 58% of 360 degrees, and it is. Then later he used 58% as the Max IDC, but I've always been under the impression that IDC was based on the amount of time it takes for 2 revolutions of the crank, or 720 degrees. So unless marcels cams are rated in cam degrees, his max IDC seems to be double. If they're rated in cam degrees, how can the intake valve be open for 210 degrees? That would leave no time for the power or compression strokes.

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

Injector Duty Cylce is IDC in case anyone sits there for a minute trying to figure that one out (or is it just me!).

That is a good point though - I've always worked on Duty cycle being 360 crank degrees - most Aftermarket ECU's however fire the injectors once every 720 crank degrees.
Marcel will (no doubt) have plotted the actual % as a value of the crank degrees.
I can't think about it any more. Need to get a coffee and wake up!

Of couse, the intake valve is open for 210 degrees out of 720 degrees but the port sees 2 intake valves and the injector has to inject twice per 720 degrees so the 58% max IDC is correct.

Since we're working with a siamese port we have to work with a duty cycle with respect to 360 degrees.

http://www.jbperf.com/

Well, my point was, he seems to use 58% as the amount of time available to supply a single cylinder w/ fuel. This is why all of my calculations that I've been questioning against his have been way off.
And we still have to conform to industry standards when sizing the injectors. We need enough fuel in 29% of 720 degrees at max RPM for 1 cyl. So that's the how we need to find the time ceiling so we know if our injectors can react quickly enough.

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

I don't know what standards you're taking about but for sizing the injectors, you need the amount of power the engine will produce, the number of injectors on the engine, and the percentage of time the injector will be open (and BSFC). The number of cylinders has nothing to do with it.

Unless you're using 2 injectors per port with each injector timed for one cylinder, your 29% is not correct.

Assuming a single injector per port, the injector will be open for a maximum of 58% of the time independent of the fact that it's per 360 degrees or 720 degrees or 1 minute.

However, again for a single injector per port, this 58% is the effective injector open time and you're right that this will be different than the duty cycle that the ECU will have to command because of the injector opening time. But as I was pointing to TurboDave just a few posts up from this one, this is a non-issue with typical injector opening time and the rpm that can be reached by the a-series engine.

Of course this assumes that the mode of injection used is the single injection per 360 degrees where you inject on the outside cylinder open valve. If you want to inject only on open valve for the outside and inside cylinders then it's going to be less but it's not 29%, it's much more complex to compute than that. You have to find what the time of the injection window is at you max rpm, deduct the injector opening and closing time from that and put that in percentage of 360 degrees (because you have 2 windows and injection pulses per 720 degrees). You'll then have your duty cycle for your injector sizing. But I don't know of an ECU that allows you to time the injection that way so unless you have one then this is irrelevant. Otherwise, the 58% quoted by Marcel is correct.

By the way, I would not use this 58% to size the injectors even if it's the correct max duty cycle. You want to give yourself some margin (for imprecise injector opening time, imprecise injection timing, overrev, ...) and if you have a bigger cam where the overlap is larger then the window may be even smaller. I would use something like 40 to 50% duty cycle just to be safe.

I hope I'm not just stating the obvious here and that I misunderstood what you were saying.

Regards,
Jean

http://www.jbperf.com/

Another way to think about this 'percent' thing is to try and consider 0% and 100%.
At zero % the valve is fully closed, hene the flow is the same wether you inject every 360 or 720 degrees. Nice and simple.

At 100% the injector is fully open.
If you basically never turn off the injector, it doesn't matter if you fire it every 360 degrees or 720 degrees - it'll deliver the same amount of fuel in two whole crank revolutions...

So at 50% - firing once every 720 degrees, it'll be held open for the required time to inject 50% of the maximum fuel that would normally be injected firing once overy 720 degrees.
Putting some numbers into it - say 100% of 720degrees is 200cc's, hence 50% is 100cc's every 720crank degrees.

Now consider firing once every 360 degrees, again it'll be held open for the required time to inject 50% of the maximum fuel that would normally be injected firing once overy 360 degrees. Again- doing the numbers - 100% of 360 degrees will be 100cc's. So 50% of that is 50cc's per 360 crank degrees.
Which equals a total of 100cc's for every 720 crank degrees.

Edited by TurboDave16V on 14th Jan, 2006.

I believe industry standard is to rate injector duty cycle on a 720degree engine cycle. And unfortunately I don't think margin is really something we have alot of so I'm trying to hit the mark as best I can. Most people who have gone down the port injection route have one of two problems: the engine won't idle(fuel injectors too large), or it won't rev past 3000rpm(randomly picked number, injectors too small)
29% may not be the max the injector sees, but it will be the maximum amount of time we have to deliver enough fuel to the outside cyl. That's why I feel it's much more important than the 58% value.
I understand wehre the number came from, but looking at marcels numbers on this don't quite make sense.
Let's take these two quotes:
"...start when the inside valve is 0.050” from the seat leaving an injection ‘window’ of 210°, or 58% of a revolution"
and:
"It needs to be big enough to get all the charge into one cylinder in one shot in less than 58% duty cycle and yet be small enough to be controlled during idle."

but I've never seen an IDC rated on a 360 degree revolution. IDC, as I've always understood it, and as most injector manufacturers use it, go for a 720 degree engine cycle. So you really only have 29% of the 720 degree engine cycle to get all of your fuel to a single cyl. It doesn't matter that the injector can be on for 58% of 720 deg. The trick is controlling an injector throughout the engines rev range that can be fast enough to get the fuel into the cyl in 29% of the time at high RPM, yet still be small enough to idle properly. If it can do that, it can run for 58%IDC as well. That is not a problem.

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

I also want to point out something else that leads me to believe the 58% has been used incorrectly.
http://www.planet.eon.net/%7Echichm/efi/size.htm
He calculates that the injector size of the MPi is probably 220cc/min, but I've seen elsewhere that the injectors are either 440cc or 480cc. Conveniently, if you look at his last formula and replace .58 w/ .29, you get 440cc.

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

Injectors are rated in cc/min or lbs/hr. From this rating you get the amount of power the injector can supply when combined with the BSFC and the IDC. The IDC is just the amount of time the injector will be open during this minute or hour, irrespective of engine cycle or number of cylinders. The injector can be supplying 1 or 4 cylinders and the engine can be running at 3000 RPM or 8000 RPM, that will not make a difference.

I agree with you that there are problems with the computations but for a different reason. If you take 220cc, 58% (or 440cc, 29%) and BSFC of 0.5, you get that the maximum power would be around 49HP which is less than what the MPi engine can produce at the crank.

I don't know what mode of injection is used in the MPi. Is it injecting in a pulse-pulse-wait-wait method or injecting once per 360 degrees? If it's the first one, then the maximum effective IDC certainly won't be 58% because you have to include the injector opening time within the injection window. But it will not be 29% but some other value that has to be computed as I mentioned in the previous post. If it's the latter method, then the 58% is valid since the opening time has plenty of time to occur outside the window in the 42% remaining time. If you look at the 58% and 440cc, that would allow a maximum power of about 98HP at the crank which doesn't look bad if you consider the losses and a certain margin the manufacturer would want to have. Also it fits pretty well with what has been produced with modified MPi engines with stock injection which have produce aroud 85HP max at the wheels if I remember correctly.

By the way, the 2 problems you mention about injector sizing are the reason for injector staging. Of course they require twice the number of injectors but you can have one injector that will handle idle and low load and another one that will handle the top end (with the help of the smaller one). However, with a good ECU and correct setup, you can make the engine idle reasonably with rather large injectors. What size do you come up with for your setup if you take around 40-50% IDC?

Jean

http://www.jbperf.com/

Jean - MPi is fully sequential; ie the injector fires when the valve is open. Hence, it'd be (180 theoetical degrees apart) AABBAABB...

Do you know if they change mode at high RPM to do a single injection pulse that overlaps both cylinders, as shown in the patent?

Also, I assume it's not really as the theoretical 180 degrees apart but there is some overlap such that A starts before B ends and B starts before A ends. So it's more like:
|__A__|__A__|_________|__A__|__A__|...
_|_________|__B__|__B__|_________|__B__|...

If not, the effective IDC is quite low (under 50%) and there's a lot of injection window time wasted opening and closing the injectors.

http://www.jbperf.com/

Jean, not sure what you mean really.
Fig 4 in the patent clearly shows that injector 1 (cyl1&2) opens soon after the valve opens (cyl 2) and shuts off before the (cyl 1) valve opens.

Also, refer to the paragraph on the bottom of page 5... This makes reference to an injection 'period' - a window, if you like - but no mention of an injection 'event'...

In reality - you can't draw or assume too much from patents diagams or descriptions. I know the patent attorneys at my company do selective 'dumbing down' and are carefully vague so as not to give too much away. They also 'expand' the description so it encompasses more than the original disclosure was for, preventing anyone from getting round the patent by altering a single system or process.

All I reckon you can deduce from the Rover patent is the following:

1) That there is definately a break between the two adjacent injection periods.

2) That the 'start' of the injection period (NOTE - NOT NECCESARILLY THE INJECTION EVENT) begins somewhere between TDC and 40 degrees ATDC - OR whenever the closing of the inlet valve actually occurs (MPi cam figures coming soon!)

3) The (second) injection period ends somewhere between BDC and 40 degrees before BDC.

Again - note the carefull use of the term periods (aka window) rather than 'event'.

What it DOES say, is stay within these boundaries, and you're infringing the Rover Patent - which is clearly the area that we need to be concentrating in. If you're injecting fuel outside of these periods, then you're probably not doing the right thing...

Edited by TurboDave16V on 15th Jan, 2006.