Pulsewidth vs Dutycycle (vg30et ecu)

[Bernardd]

Can someone explain to me the relationship between

pw and dutycycle? I have found a map that changes the duty cycle but I'm not sure if the pw is changed

by the same amount. I have the engine sim set at 300 ohms at the CHTS, idling at 800-850 rpm. I can change the duty cycle from 7% (stock) to 4.2% by changing this map.

Bernard

[TimZ]

Pulsewidth is just the amount of time that the injector is open for a given engine cycle. It is generally measured in milliseconds.

 

Duty Cycle is the percentage of time that the injector is open during a given engine cycle. It is the pulsewidth divided by the cycle time (the total time between pulses), times 100 to get a percentage value.

 

So, duty cycle is related to pulsewidth, but since the cycle time varies with engine rpm, duty cycle takes engine rpm into account, where pulsewidth does not.

[Bernardd]

Can someone with the JWT ecu measure the duty

cycle at idle?

[z-ya]

Actually...

 

Pulse width is the width of the pulse driving the injectors as measured in time. Typically the voltage of the "on" pulse is zero, because the injector driver provide a ground to the injector to energize them. Because engine RPM dictates the frequency at which these pulses occur, the pulse width is usually measure in milliseconds.

 

Duty cycle is the percentage of time the injector is open vs closed. If the "on" pulse width was 10ms, and these pulses occured every 100ms, the duty cycle would be 10%. It's basically a way to measure what percentage of the time the injectors are on.

 

Pete

[TimZ]

Originally posted by z-ya:

Actually...

 

Pulse width is the width of the pulse driving the injectors as measured in time. Typically the voltage of the "on" pulse is zero, because the injector driver provide a ground to the injector to energize them. Because engine RPM dictates the frequency at which these pulses occur, the pulse width is usually measure in milliseconds.

 

Yes, you are correct - pulsewidth is a measurement of the amount of time the injector is commanded to be open, not the amount of time it is actually open, which is always a bit less due to the injector's mechanical properties.

 

However, pulsewidth is independent of engine RPM in the context of the original question.

Duty Cycle has RPM as part of it's definition, pulsewidth does not.

 

The only linkage between RPM and pulsewidth is that the maximum allowable pulsewidth is determined by RPM, but again, this is because you can't have a duty cycle greater than 100%.

[z-ya]

Actually, engine RPM does dictate the pulse width. As engine RPM increases, so does fuel requirments, so the pulse width needs to increase.

 

As I wrote in my previous post, "engine RPM dictates the frequency at which the pulses occur". This is accurate. Engine RPM is typically less than 10K RPM (Hz), and the pulse width is some fraction of the pulse cycle (1/frequency). This is why width is typically measured in milliseconds. Otherwise you would be dealing with extra decimal places and or scientific notation to describle the pulse width in let's say nanoseconds.

 

Pete

[Bernardd]

Ok so if I use this formula 120000/rpm it'll give me

time per revolution? 120000/800=150ms per cycle. 7% of 150 = 10.5ms pulsewidth. And a 4.2% dutycycle gives me a 6.3ms pulsewidth. Does that make sense?

Bernard

[Pop N Wood]

I have found a map that changes the duty cycle but I'm not sure if the pw is changed

Yes. At any given RPM changing the duty cycle will directly vary the pulsewidth.

 

120000/rpm it'll give me

time per revolution

No. Time (in seconds) per revolution = 1/(60*RPM) since there are 60 seconds in a minute. For milliseconds, simply mulitply the above by 1000.

 

Note there may be several injector firings during each revolution of the crankshaft. To convert engine RPM to injector firings per second you need to know whether you have a 2 or 4 stroke engine and the number of unique injectors being controlled by the computer.

 

pulsewidth dictates the volume of fuel injected on one cycle

Well put. I hadn't thought about it but that states it very clearly.

 

Duty Cycle dictates fuel flow (volume per second).

To nitpick it is a measure of fuel flow rate.

 

Think of a fuel injector as a garden hose that is either fully on or fully off. When turned on, the hose passes water at a certain rate (say 10 gallons per minute). If you turn the hose on for 1 minute, you will fill a 10 gallon jug. Thus the total volume of fuel passed is the flow rate mulitplied by the injector "on" time.

 

Or you can fill that same 10 gallon jug by turning the hose on for 12 seconds, off for 48 seconds, then repeat 5 times. It now takes you 5 minutes to pass the same amount of water because you only have the hose on for 12 seconds out of every minute (the duty cycle). Sooo you have effectively cut your fuel flow rate from 10 gallons/minute to 2 gallons/minute (i.e. by a factor of the duty cycle).

 

What TimZ is saying makes sense. For a single intake stroke, the cylinder will inhale a certain volume of air. This volume of air will need a specific volume of fuel, i.e. a specific injector "on" time or commanded pulsewidth.

 

As engine RPM increases, the number of intake strokes per second increases, but not necessarily the volume of air being inhaled by each stroke. This is more dependent upon engine vacuum than RPM. Thus as the RPM changes, the pulsewidth (volume of fuel per stoke) is the same, but the pulse come at a faster rate, hence higher duty cycle.

[Bernardd]

So for my testing purposes I have decreased the fuel flow by X % equal to the X % reduction in duty cycle? I have not changed anything as far as vacuum or load the engine is seeing and the rpm is always 800. I should probably borrow a scope and find out exactly how much it changed.

[Mudge]

Bernard, I'm not sure what your saying, but as an example, if you lowered your fuel pressure you would be IN EFFECT rasing your duty cycles (which can be a bad thing, because this means less fuel injecting capability).

 

If you raise your fuel pressure, you have raised the ability to inject more fuel into the car, just like an injector upgrade, this would therefore LOWER your duty cycles meaning you have more headroom before you hit 100% duty cycles, which is the point at which you can not dish out any more fuel.

 

For blown/turbo/nitrous cars, 65% duty cycle is popular, for road race cars it is similar, so long as it doesn't affect your idle in a negative fashion, meaning 1ms would be more fuel than is needed at idle and resulting in a pig rich idle.

 

For a NA drag car, 85%-90% is approximately "correct", since you are not going to be at those high duty cycles for long periods of time, the injector drivers should not get too hot, and the injectors will have no problem dishing out the fuel.

 

Duty cycles are NOT a concern at 800RPM, pulsewidths may be, as I said above it is as low as they can go (1ms) and you are running overly rich, then it is an issue. Depending on how far that would put you off your desired mark, lowered fuel pressure, raised idle RPM, and/or new lower rate injectors would help or cure that, again depending how far you are off now.

[z-ya]

TimZ wrote:

 

I'm sorry, but this is not correct. Pulsewidth is (as it's name strongly implies) simply the time measurement of the width of the pulse. It does not have any provision (nor does it need one) to accommodate the number of pulses that happen per second. This is specifically what duty cycle is for.

 

Further, pulsewidth varies fairly linearly with the engine's intake manifold pressure, and it does so regardless of engine rpm. I know this doesn't sound right, but if you think about it it makes sense. For any given intake cycle, your engine takes in a pretty constant volume of air. The intake manifold pressure dictates how many air molecules are present in that volume of air, which dictates the amount of fuel that needs to be injected. The injector meters fuel by opening for a specific amount of time, so for a given manifold pressure, the injector pulsewidth stays pretty constant, regardless of RPM.

I guess I was trying to say that the pulse width is measured in time, and RPM is really a measurement of frequency If you add the injector off time to the injector on time (100% DC), and invert the result, you will have the engine frequency in cycles per second. Kind of long and convoluted, but that's where I was trying to go.

 

Thanks for the description about the relationship between intake manifold pressure and pulse width. It does make sense. Come to think of it, I've seen this behavior on my aftermarket EFI computer which displays injector open time in ms. If you look at the load band, the injector open time (pulse width) is almost constant over the entire RPM range. The frequency at which the injectors open increases with RPM, but the pulse width stays constant.

 

TimZ wrote:

As RPM increases, you are pulling in more air per second, but the amount of air per engine cycle is roughly the same (for a given manifold pressure). You are just pulling in that volume of air more times per second. Again, as the RPM increases, the spacing between the injector pulses gets shorter, so the duty cycle increases.

Excellent description. My problem is that I think of it from an EEs point of view, when there are a lot more to engine dynamics than electrical signals that control it.

 

Pete

[Bernardd]

The reason for me asking the question was to determine what cc injector I could use with the new settings in the ecu. I can lower the duty cycle further but only if I start messing with the entire load calculations and I don't want to do that at this time. If 270cc inj's have a 7% duty cycle @ 800 rpm, what cc inj would flow the same amount of fuel at 4.2% duty cycle? Is it a linear relationship? all other things being equal. I probably should have worded the question this way in the original post.

Bernard

[TimZ]

Originally posted by Bernardd:

The reason for me asking the question was to determine what cc injector I could use with the new settings in the ecu. I can lower the duty cycle further but only if I start messing with the entire load calculations and I don't want to do that at this time. If 270cc inj's have a 7% duty cycle @ 800 rpm, what cc inj would flow the same amount of fuel at 4.2% duty cycle? Is it a linear relationship? all other things being equal. I probably should have worded the question this way in the original post.

Bernard

Yes, it's a linear relationship, so long as the pulsewidths stay above the injector's minimum on-time - usually about 1 msec for low impedance injectors, and maybe 2 msec for high impedance types. In this case, at 4.2% and 800rpm, you should be at a little above 3 msec, assuming the injector fires every engine rev, 6 msec if it fires every other rev.

 

So, if you do the math a 450cc injector will flow the same amount of fuel at 4.2% as a 270cc injector at 7%.

 

However, it's not clear to me what you are trying to accomplish here - you do realize that the larger injector will flow that much more all the time, not just at idle, right? This equates to approximately a 67% increase in fuel flow, which will make the engine run stupid rich everywhere else, if it doesn't just die and foul the plugs.

[Bernardd]

The math is what I wasn't sure about. What I'm doing is building my own 300zxt to L28ET ecu conversion. At or just above idle are the only points I can test right now, but I can retune the entire fuel curve for larger injectors, larger maf, rev limit, timing etc.

Bernard

[TimZ]

Originally posted by z-ya:

Actually, engine RPM does dictate the pulse width. As engine RPM increases, so does fuel requirments, so the pulse width needs to increase.

I'm sorry, but this is not correct. Pulsewidth is (as it's name strongly implies) simply the time measurement of the width of the pulse. It does not have any provision (nor does it need one) to accommodate the number of pulses that happen per second. This is specifically what duty cycle is for.

 

Further, pulsewidth varies fairly linearly with the engine's intake manifold pressure, and it does so regardless of engine rpm. I know this doesn't sound right, but if you think about it it makes sense. For any given intake cycle, your engine takes in a pretty constant volume of air. The intake manifold pressure dictates how many air molecules are present in that volume of air, which dictates the amount of fuel that needs to be injected. The injector meters fuel by opening for a specific amount of time, so for a given manifold pressure, the injector pulsewidth stays pretty constant, regardless of RPM.

 

As RPM increases, you are pulling in more air per second, but the amount of air per engine cycle is roughly the same (for a given manifold pressure). You are just pulling in that volume of air more times per second. Again, as the RPM increases, the spacing between the injector pulses gets shorter, so the duty cycle increases.

 

This is a well documented behaviour for Electronic Fuel Injection systems.

 

If you like, another way of looking at it is that pulsewidth dictates the volume of fuel injected on one cycle. Duty Cycle dictates fuel flow (volume per second).

 

As I wrote in my previous post, "engine RPM dictates the frequency at which the pulses occur". This is accurate.

Yes, its accurate, but it has nothing to do with the pulsewidth itself, aside from the frequency determining the maximum allowable pulsewidth, which is due to the duty cycle (can't be > 100%).

 

Sorry if this sounds like I'm nitpicking, but this distinction is fairly important in understanding how the EFI system works.

 

Originally posted by Bernardd:

Ok so if I use this formula 120000/rpm it'll give me time per revolution?

That formula will give you time per 4-stroke engine cycle (i.e., 2 revolutions). This may or may not be what you need, depending on how your system fires the injectors.