There have been some mention of injector capacity and duty cycle for modded cars. But in looking, I couldn't find any good data, just anicdotal recantations that were second or third hand. So I did what some of us should have done in the first place, gone and measured it! But I ran into some funny stuff.....

I used BiM-COM to report both mass flow, and calculated injector pulse width. And I found something really funny, the ECU wants to hold the injector open for MORE than a complete rotation of the cam! WTF?!?! So I thought I'd done a bone head move, and sent the data off to Andy, and he couldn't find an issue, and also sent me a ref about this happening in the Audi/VW world...
http://forums.audiworld.com/s4/msgs/2186118.phtml
The consensus there was that while the ECU data is misleading, the A/F ratios really govern what one should be worried about...... But many were uncomfortable with the over 100% duty cycle.

So what's happening? No clue. But it's looking like JCW injectors don't really have the ***** for even a 15% pully, but the A/F ratios are still good.... Here's the graph....



Crap. Looks like more money.......

Matt

 

isn't this what causes the overrun burbling? I think they retard it on purpose...

 

Whatchu talkin' bout Willis?

 

Yup, very common on newer port fuel injected cars. They are calculating the DC base on valve opening time. In general, a little over 100% is just fine, (if the ECU calculates that way) because it is not really 100%. The actual 100%DC (using injection time versus time) is when the injector goes static, and dumps tons of fuel. This is bad because no more fuel can be added.

You can tell it is not that dangerous as when you hit 100% the AFR's don't get leaner and leaner as the boost and RPM goes up. Not that everyone has an AFR gauge. But a good tool to get none the less.

 

Who are you agreeing with (Yup)? And in what regard? (are the JCW injectors not up to snuff?) Pls elaborate.

Your post brings a few thoughts to mind...
I was of the impression that DC beyond 75% (actual) was bad because the injectors would heat up. I appreciate what you've said re: "goes static, and dumps tons of fuel." Are you speaking in context of non-stock injectors which have higher capacity/DC?

What has your experience been with stock MCS injectors? Your comment re: "AFR's don't get leaner and leaner as the boost and RPM goes up" has me thinking of my old dyno plots. At the time of those dyno runs my car wasn't that heavily moded - only catback, CAI, T/B & hose 15% & ECU. In the upper reaches of RPM my fuel went below 10:1, then between 6500 - 7160+ went back up to 11:5. It had been my understanding that this was the injectors 'maxing' out. See my gallery AF-T for what I'm talking about.

Thanks for your thoughts

 

If I understand what you're saying, the car thinks the cam should be open for a specific amount of degress of the cam's rotation. Probably a pre-programmed value because there doesn't appear to be any sensors in the head to monitor how long a valve stays open). It must use some sort of calculated duration open based upon the pre-programmed value and the RPM the motor is turning. So then it divides that into the duration the injector is spraying, which can give duty cycles of greater than 100% (contrary to the definition of duty cycle which says only values of 0% to 100% are valid... time operational / cycle time). So basically to get the > 100% numbers the denominator is less than the numerator.

Is the duty-cycle just a computational value that the car can report, or is the ECU actually changing other parameters based on this computed value? If so, changing the cam to have valves open longer would directly affect what the car is doing with those numbers.

But that 100% is not the same 100% as shown in the graphs above, right? Because that 100% is not computed according to the pure definition of duty cycle...rather it is a fuel injector operational time versus valve open time. What happens if the time is skewed, where the injector fires too soon (or too late) and it doesn't overlap the time the valve is open?

 

I converted Crank RPM to cam ms/rev and took a ratio of calculated injection time to the amount of time it took the cam to rotate. So over 100% means that the injector would be open for more than a full cam rotation, or a not physically possible scenario.

I haven't looked at the actual injector pulse width with an o-scope yet, I need a power inverter, as mine runs off of 100 VAC....

I do have a wideband, and would have to do some more logging to see exactly what A/F ratios are doing....

Matt

 

so whats this mean.. besides that jcw is just branded crap hahahaha :D should peeps like us be getting bigger injectors ?

 

To know for sure, I'd have to look at the real waveform driving the injector. But it's looking like the 440 crowd may be on the right track! Since I don't know exactly how this number is used by the ECU, right now it's just a mystery....

Any of the ECU hackers have a clue?

Matt

 

I don't actually know anything, but some random thoughts.
From what you have said, it sounds like the ecu calculates duty cycle in an odd manner--maybe only using part of the cam rotation cycle?
Also, from your graphs, it appears as though the two lines will intersect at some point. Mass delivery slope appears to be decreasing, while duty cycle slope is increasing. Perhaps where they cross is where the stuff really hits the fan?

probably won't help, but you never know...
dan

 

The ECU didn't calculate duty cycle. I did. It's based on the amount of time it takes the cam to make one full revolution. So anything over 100% isn't possible, 100% would be keeping the injector on all the time.

As far as when the graphs cross, that's just a figment of the units and the scales used. If the ECU really is firing the injector for the amount calculated, it's staying open all the time for all RPMs past 100% duty cycle, or about 6500 RPMs....

Matt

 

Doesn't the ecu start the rev-limit process at 6500 rpm? Coincidence...

 

I don't recall hearing about this. Even so, many folks have the rew limit raised - some up to the 8K range.