Fuel injectors, duty cycle and how not to look like an idiot. **please read**

[PirateMcFred]

I've seen people mentioning duty cycles when others inquire about their engine builds. Some people are stating that "[you will] exceed 100% duty cycle if you use brand X injector." It's frustrating for me to read some of these posts as it's scary to see ignorant people offering advice to others and the newly-informed user re-states the information again later to other people. This spreads misinformation and prods one of my nerves. So in an effort to inform without belittling n00bs here's a short explanation of fuel injectors and duty cycle: An injector is just a vavle. Injectors are controlled by the ECU with an electrical signal that lasts a specific peroid of time called a "pulse width" usually measured in milliseconds. In terms of fuel injectors they're either open or closed. However, since you cannot simply open an injector all the way instantaneously due to simple physics it takes a small amount of time to go from 0% open to 100% open. It's this lag in the injector that causes problems at extreme ends of duty cycle. Duty cycle (DC) is a measurement of how long the injector is open relative to how long it is closed. 50% DC means that the injector is open half the time. At 100% DC it's just open all the time and delivers the maximum amount of fuel as the injector is rated for. You cannot have a valve open more than 100% of the time. There's no such thing as 110% duty cycle. If you exceed 80% duty cycle then you get problems with the timing of the injector opening and closing in relation to the intake stroke of the engine when you're telling an injector to open and close 18 times a second or more. 2%-80% duty cycle is fine for normal injectors. The problem of the injector opening and closing quickly enough at elevated DCs is realized as the time it takes the vavle to close and reopen gets shorter and shorter. Think of 90%-100% DC as "valve float" for fuel injectors. The time between the ECU's electrical signal turning the injector off and back on again is shorter than the time it takes for the physical injector solenoid to open and close. The effect of this is inconsistant fuel delivery. The injector is neither closed nor open and as such it's not delivering the proper amount of fuel. Considering that high duty cycle normally only occurs at elevated manifold pressures and means that your engine is running at maximum cylinder pressure. Essentially you run your engine lean at wide open throttle and you end up with detonation. The opposite is true with very large injectors at low duty cycle but without the damage from detonation (low cylinder pressure). If you have a small 100cc/min injector at idle/part throttle operation it's easier to control than a 1000cc/min injector in the same engine. If you have the 100cc/min injector running a 10 millisecond pulse width (.010 seconds) you get ~.017ccs of fuel. Example: 100cc/min / 60 seconds = 1.667ccs per second 1.667ccs/.010 seconds = .017ccs 1000cc/min / 60 seconds = 16.667ccs per second 16.667ccs/.010 seconds = .167ccs If you ran the 1000cc injector at the same pulse width you'd get ~.17ccs. That's 10 times the amount of fuel you need. So you'd have to crank the injector pulse back to 1/10th of 10 milliseconds or .001 seconds. It's very hard to control a large injector in such a short period of time. The ECU quickly commands the injector to open and close, but by the time the electrical signal ends the 1000cc injector has barely even opened. The effect of large injectors on low rpm operation is that you have inconsistant fueling, poor atomization, harsh driveability, and poor emissions. So, running larger injectors is generally a good idea if you have extensive modifications to your engine that might push the limits of the factory injectors. Honda likes to run their injectors pretty hard from the factory so there's not a lot of room for extra power. A short term fix for stock injectors is an adjustable fuel pressure regulator. With the adj. regulator you can artificially increase the flow rating of the stock injectors by raising the fuel pressure, but it should only be used until properly sized injectors can be sourced and retuned. Consider yourself informed. Pirate

This ad is not displayed to registered or logged-in members. Register your free account today and become a member on Honda-Tech!

[mgags7]

Pirate, just to add.

The pulse width is the raw number of milliseconds that the injector is open, and the duty cycle is just a calculation from that, when you add in rpm.

For instance:

Say you're at 8000rpm, in a 4 stroke engine, you fire the injector once every 2rpm, so take 8000/2 = 4000; 4000 / 60 = 66.66 rev/s; 1/66.66 = .015s or 15ms per revolution.

Now say that you need the injector open for 10ms to get enough fuel in.

10ms/15ms = 66.67% DC

Good info, I just noticed the guy on that G23 thread talking about something along these lines.

In that light, any duty cycle number you get from an online injector size calculator that is over 100% is just telling you that you need a bigger injector or more pressure. That can exist on paper, but not in the real world. Being over 100% duty cycle means that your injector needs longer than the length of one engine cycle at a given rpm to get enough fuel in.


Modified by mgags7 at 2:09 PM 11/21/2007

[PirateMcFred]

That's true, you also have to consider that any raw engine RPM has to be divided by 2 for it being a 4 cycle engine. The injector fires once every 720˚ crank rotation, so your 4000 rpm example the actual engine rpm will be 8000rpm / 2.

[mgags7]

Ahh good call Pirate. I'll edit that in.

[hybrid2007]

Nice one Pirate

Its good to see some good tech here, and good too see someone taking the time to help us noobs out

[A Blue Lude]

Duty cycle information in Crome tends to be wrong.

That's my 2c. I wish I knew it back in the day, maybe it will help out someone else

[mgags7]

Originally Posted by A Blue Lude

Duty cycle information in Crome tends to be wrong.

That's my 2c. I wish I knew it back in the day, maybe it will help out someone else

The same goes for Hondata. The datalogger consistently reads out erratic numbers and sometime goes over 100% as well.

[xxmastermindxx]

Excellent post, cleared up a few questions I had lingering

Definite FAQ material.

[PirateMcFred]

I never thought that the cells in the fuel tables of Crome and hondata were ms pulse width values I always thought of the indicated value as arbitrary "fuel values."

*shrug*

[A Blue Lude]

I think that's the way they're set up. Neptune's numbers are definitely arbitrary values, according to James/HRtuning.

[mgags7]

Originally Posted by PirateMcFred

I never thought that the cells in the fuel tables of Crome and hondata were ms pulse width values I always thought of the indicated value as arbitrary "fuel values."

*shrug*

I'm not talking about the actual fuel table, though Blue might be, I'm just talking about the datalogger's numbers. It spits out DC % for you, and is often very wrong.

[A Blue Lude]

Originally Posted by mgags7

I'm not talking about the actual fuel table, though Blue might be, I'm just talking about the datalogger's numbers. It spits out DC % for you, and is often very wrong.

I was referring to Crome's duty cycle table, which is claiming to tell you the actual duty cycle in %.

I found Hondata S200's DC log ok the one time I used it. It did spike to a 100% for no reason though

[ECX]

I stand corrected, thank you for the information Pirate. I myself was misinformed to begin with, as I kept reading about the dsm 450's in particular being risky of over working themselves at _hp level. But it does make perfectly good sense than an injector can't be over 100% duty cycle now that I think about it. However you didn't need to bash me or anything titling "how not to look like an idiot" that part I disagree with. The good information is what I'm after, not someone pointing fingers and such when I'm wrong, which is often because I still have alot to learn.

Anyhow, thanks for clearing this up for me. Looks like I'll be sticking with the 450's after all.

EDIT: Also, how exactly do you know when it's "time" for bigger injectors as far as upgrading things go? You run so much boost from one setup and are fine on one set of injectors, and then get something bigger and all of a sudden need bigger injectors? Duh, the answer is because you need more fuel at any given point in time, but what exactly determines that? And please don't call me a noob, atleast I'm asking legitimate questions here.

[98vtec]

added to the faq.

good stuff pirate and matt

[95greenlude]

how do you calculate what your current duty cycle in on your injectors??

[ECX]

I'm not giving a direct answer because somehow it'll be wrong and false information, but it's in there towards the top of Pirates first post.
Think open/closed

[95greenlude]

^no nowhere in his post does it state HOW to calculate what your injectors DC is currently at....

[Rosko]

you'd need to take the time between engine cycles at a given rpm and divide that by the corresponding pulsewith on your fuel map.

Matt's got it covered in his post where he states 15ms as the time between cycles at 8000 rpm, and 10ms being his pulsewidth = 66.67% duty cycle

[95greenlude]

ok...so you need a tuning device....

[mgags7]

Originally Posted by 95greenlude

ok...so you need a tuning device....

Yeah. You *could* rig something up on the wires and keep a stock setup, but you'd have to buy an oscilloscope or something comparable in order to measure the pulse width, and those things are not cheap by any means.

Another thought I just had was, you could most likely use a high-end AC voltmeter to read duty cycle, some of them will actually measure it directly. But, you'd have to multiply the # you get from it by two because it would be measuring duty cycle for all 2 rotations of the engine between injector fires, when the duty cycle we calc is measured over just 1 rotation.

[Rosko]

OR the quick and dirty way to just get close would be to use the calculator at RCeng.com. you can backfigure duty cycle by using horsepower, injector size, pressure, brake specific #'s to get a close # if you are just trying to determine where you are at on injector size. obviously just an approximation though.

[JT]

Good info guys. It's nice to see people take the time to share.

[ECX]

Originally Posted by ECX

I stand corrected, thank you for the information Pirate. I myself was misinformed to begin with, as I kept reading about the dsm 450's in particular being risky of over working themselves at _hp level. But it does make perfectly good sense than an injector can't be over 100% duty cycle now that I think about it. However you didn't need to bash me or anything titling "how not to look like an idiot" that part I disagree with. The good information is what I'm after, not someone pointing fingers and such when I'm wrong, which is often because I still have alot to learn.

Anyhow, thanks for clearing this up for me. Looks like I'll be sticking with the 450's after all.

EDIT: Also, how exactly do you know when it's "time" for bigger injectors as far as upgrading things go? You run so much boost from one setup and are fine on one set of injectors, and then get something bigger and all of a sudden need bigger injectors? Duh, the answer is because you need more fuel at any given point in time, but what exactly determines that? And please don't call me a noob, atleast I'm asking legitimate questions here.

Ahem.

[Rosko]

not exactly sure what you are asking but more power just needs more fuel so when the duty cycle starts to get up around 80% then its time for bigger injectors. an injector can only flow so much in a given amount of time so when thats not enough they just need to be bigger (flow more fuel) so they can flow more fuel in the same amount of time.

the injector calculator on RC's site works really well on determining how big of an injector you will need based off of h.p. goals and #of cylinders and fuel type used.

[Gravis20]

Hey man I am noob to this but you seem to know whats going on so here it goes, I have a 96 lude SRV OBD2, Busted balancer belt got an JDM H22A swapped oil pumps for crank position sensor, everything good there. I have pressure to the fuel rail 34psi, dizzy is fine (spark good) injectors not firing? I have a resistor box on order but not sure if thats the fix? What trips the injectors to fire , I mean how does the ECU know when to send the pulse, Please help!!!