Fuel Injector Selection

[timlothian]

Hi there,

I am trying to source some fuel injectors for a 550cc motorbike engine.

I want a 20 degree conical spray angle, as found on a Bosch EV6 injector, due to the length of the intake port within the head. ie: in order to spray onto the back of the valve, and not all over the walls, i need quite a tight spray angle.

However, I cant find one that is low impedance. The EV6 injectors are all high impedance. Now, because I am using a small engine with injectors that are designed for larger engines, I will require finer control at low rpm to keep the pulse width low. Somewhere in the order of 2 milliseconds I think.

Now, the question. With the EV6 high impedance injector will it be possible for me to control this injector at low rpm using an M800 ECU or do I need to find a low impedance model? If so can anyone recommend an injector?

Thanks for your help

Tim

[OmarK]

Are you able to run Hi/Lo injectors in this particular application? So in a 4-cylinder motor you have a total of 8 injectors.

[timlothian]

I have considered this much, but for simplicity I am hoping to only use the two port injectors in the low position. This is the first development done on the engine, (Aprilia 550 for FSAE for anyone wondering) and I want to leave further refinement to later.

I was thinking that if I found that the injectors were giving too much fuel at idle that I could switch from a 3 bar reg to a 2.5 bar reg. But from what ive read a pulse width of 2 ms seems to be achieveable.

From peoples experience, have you had any troubles getting a pulse width of 2ms controlable from a high impedance injector?

I know the theory says a low impedance injector is easier to control at low pusle width, but to what degree is this? Is it important for 2ms, 1ms......?

Further, does anyone know of people who publish injector spray angle figures? Motec...I know you have test figures for injectors but does this include spray angle or just flow?? At the moment I am limited to 1 or 2 bosch injectors that I could purchase (if i find them in store) because bosch seem to be the only people who can give information.

thanks again for the help

Tim

[MarkMc]

Hi Tim,
How controllable an injector is is generally a function of its mechanical make up, there is obviously a mechanical lag from when the ECU pulse starts to when the fuel is actually flowing. There is even a non-stable period at very low pulse widths where you cannot guarantee that the injector is repeatable from firing to firing, this is not really a problem with the ECU as it can give an accurate repeatable pulse.

I do not think specifiying a particular pulse width is going to help you, unless you can give me a bit more detail of what you are thiniking. What you are more worried about is if the injector can give you the best top end performance, the engine is for a race car after all.

You can head two ways with injector sizing. The first is basically a "street car" approach where you choose an injector that is just big enough to be useful at the top end, this injector will also be the smallest to suit the situation and mean that you have good control at idle because you are not near the non-stable opperation of the injector (can depend on how much HP you want to make though). This sizing method can give the best compromise and the only thing you may loose it the ability to change the inection timing at high engine load/rpm as the injector is close to 100% duty cycle. With not room to move on the injection timing you may miss out of a few HP so you are not getting the ultimate performance out of the engine.

The second method is the "race car" approach where the low end is irrelevant and in this case you would choose an injector that is much bigger than neccessary, this means you have a low duty cycle at high load/RPM giving you plenty of room to move the injection timing around and get the maximum HP the engine can give....the downside is obviously the injector may not work well at idle and you have to work out if the top end benefit is worth having a rough idle, possibly needing the thing to idle high to keep the injectors working at a useful pulse width.

as you have stated you could drop the fuel pressureand effectively make the injector smaller but this may affect atomization. As I said, if you can give me a bit more detail on what you are wanting to achieve I can give you more advice. I hope the information above is useful.

[timlothian]

Mark,

Your response is very useful and thank you for replying. I was one of the students who did the ECU course with you at UNSW in 2008 (i think thats when it was).

With regards to my approach, my thoughts were that because im sizing injectors each for a 225cc cylinder, (ie 550cc twin cylinder engine) that most of the commercially available injectors will provide bigger flow than I could ever require. As such, I dont think ill run into the troubles of sizing like a street car where I lose the ability to increase duty cycle. Therefore my goals were to ensure I didnt run into problems at the other end of the scale as I am likely to have an injector bigger than I need by way of most available injectors being made for cars. My initial thoughts were that if I had an injector bigger than needed, I may run into 'impossible' idle timing requirements. ie: requiring a pulse width lower than the minimum injector lag time.

The other consideration I have to make is that I would like injectors that have data attached to them such as spray angle. (rather than just saying: "this will fit into your commodore") With ports as deep as 110mm from the flange and two valves of only 26mm diameter I need to spray the fuel quite a long way without it spreading too much. This was why I came up with the Bosch EV6 injector with a 20 degree spray angle as it best fits the geometry to spray to the back of the valve. The trouble is, Bosch seem to be the only people who make available data such as spray angles and so I am limited in my selection. Further, I havnt found many with an angle of 20 degrees so I am even more limited in flow choices.

I consider the spray geometry to be my critical design point here, and hope that the flow rate that comes with that injector will be within a controllable range. (I can control the pulse width, but not the geometry). The EV6 injector I have found satisfys the 20 degree spray angle but has a flow of 310 g/hr. My hope is that this injector is not so big that I cant control it at idle. While idle is not so important for a race car, it does have to be able to do it, especially with our limited time for development (Formula SAE teams) I dont want to spend so much time trying to get an oversized injector to idle.

Do you think this High Impedance, 310 g/hr injector would be able to handle the pulse width I require at idle? (say 2ms)

Thank you again for your help.

Tim

[MarkMc]

Without testing the injector I could not be 100% sure but looking at the other smaller Bosch injector dead times you would expect somewhere around 0.5-0.6msecs dead time. I am sure that at 2msecs on a small injector you will be fine. The low imperdance injectors seem to be about half of this on average.

[timlothian]

Thank you again for your help, the more I looked into this, I too compared similar injectors and began to convince myself that it would be fine.

For anyones interest this is the table I read when I talk of similar injectors:
http://www.sardracing.com/images/produc ... torapp.jpg

When you say the low impedance injectors seem to be half this, I presume you are talking about them being in the order of half of the 0.5-0.6ms dead time.


I hope this thread has been helpful to more people than just myself.

Tim

ps: Mark: Some of the new members of the SAE team are keen to attend a Motec ECU and I2 course again this year. Do you think youll be running a course again in Sydney this year, and if so would you like us to host again? It looks like well be expecting similar numbers to attend as last year. You can contact me at z3158563@zmail.unsw.edu.au if you like.

[Holmz]

Thank you again for your help...

I hope this thread has been helpful to more people than just myself.

ps: Mark: Some of the new members of the SAE team are keen to attend a Motec ECU and I2 course again this year. Do you think youll be running a course again in Sydney this year, and if so would you like us to host again? It looks like well be expecting similar numbers to attend as last year. You can contact me at z3158563@zmail.unsw.edu.au if you like.

Yes it is useful.

So is the right answer to use a small set of injectors for the idle and low speed operation, and the switch to a larger set so that each set of injectors runs at duty cycles that are some percentage down from 100%?
And would that %'age be somewhere around ("the duration of the cam" / "720 degrees")

...
ps: Mark: Some of the new members of the SAE team are keen to attend a Motec ECU and I2 course again this year. Do you think you'll be running a course again in Sydney this year, and if so would you like us to host again? It looks like well be expecting similar numbers to attend as last year. You can contact me at z3158563@zmail.unsw.edu.au if you like.

I am interested too - if the dates become available a PM would appreciated.

[MarkMc]

Just to clarify for everyone. The dead time of an injector is the time it takes to open. Think of it as a gate you want to go through, you start to push but because the gate mas weight it takes a while before you can get it open enough to go through. The same with an injector, the MoTeC ECUs have a dead time or Battery compensation which simply adds the injectors dead time (found from testing) to any pulse width you ask for, meaning that anything you ask for in the main fuel table is actual fuel flowing time, the ECU has already, automatically added the little bit extra pulse to get the thing open. You will notice that if you put zero in the fuel table and had the engine turning you still get some pulse to the injector, this is the dead time pulse only so you get no fuel injected into the engine.

Now, a little bit further on the gate analogy, lets say you have a number of people wanting to get through the gate but the gate can only be open for a small amount of time, maybe you get four people through one time and three the next and on the odd occasion five. This means you have not got prefect control of the gates ability to flow people from opening to opening, the same with an injector. It is not until you have overcome the dead time and the non-linear time do you get a repeatable amount of fuel. So an injector that has a dead time (no fuel) of 0.5msecs may not be that nice at idle until about 0.8-1msec. You can see that if you have a big injector you are more likely to run into the non-linear time (which is quite short) on a small engine so this is bad for idle. Runing a small injector that covers your top end as well is the best compromise but you may loose a few HP because you cannot get the injection timing just right.

For a street car as long as the injectors are about 85-90% at the top end you will be fine. The duty cycle does not really have anything to do with the cam duration because not inlet cam is open for 90% of 720 degrees, the fuel sort of backs up in the inlet manifold making a sequentially injected engine effectively group fired at the top end....kind of why a race car will loose idle quality for the benefit of a big injector giving them low duty cycles (50-60%) at the top so they can get the injection timing right. Always a balancing act.

As was mentioned, having two sets of injectors can give you the best of both worlds.

[MarkMc]

As for seminars, if you have the numbers and the venue all you need to do is give me a rough date and we will be happy to do it.

Have laptop, will travel.