Request for Data from TheHolyCow
 

Was wondering if you could gather (or report it, if you have it) some injector duty cycle data.

Specifically, was wondering what the duty cycle is for the same rpm but different gears and/or throttle angle.

I other words, would like to compare duty cycle if you're driving at 2000 rpm in 2nd gear versus 2000 rpm in 5th gear.

Not sure you have the capability for throttle angle measurement, though.

you can always send him a PM. I think you would get a faster response that way.

well, at the same rpm's intake air flow should be the same regardless of gearing, so I'd have to figure the duty cycle would also be the same.

Intake air flow will vary based not only on RPM, but on throttle opening too. Duty cycle should mainly follow air flow, with minor variations, and at some times larger variations.

Your request is very timely, actually. I've got an ELM327 coming in the mail, and I've acquired an old laptop to keep in the car at all times. A couple days ago, while waiting for my ELM327, I started thinking about fumesucker's hookup logging fuel injector pulses with his sound card. A project to combine that data with OBDII logged data may be far beyond my abilities but I'm going to try.

I will have to try to test and observe for those things. I have been concerned mainly with speed, acceleration rate, throttle opening, and gear, but not so much with RPM. My VW consistently does better at lower RPM under every condition or circumstance so I haven't had much reason to test same RPM with different conditions; instead, I just keep the RPM as low as possible.

I can answer one piece, though. I previously was able to log OBDII data with an expensive VW-specific tool, so I was able to see throttle angle; its DBW is usually obedient, giving me throttle angle proprtional to gas pedal position. So...
Same RPM, different throttle angle: As you might intuit, fuel injector duty cycle goes up with the throttle more open -- UNTIL you get over a certain point. I'm not exactly sure what that point is; it's somewhere between 50 and 80 percent throttle. At WOT and the same RPM, it actually uses less fuel than at ~70% throttle. I believe this is not a universal behavior, but I'm not sure.

I'm not sure what you are after here, perhaps a comparison with the same RPM but different aerodynamic drag?

I can watch injector duty cycle on my car, and throttle position is easily watched by the TPS (throttle position sensor).

-Bob C.

What I'm after is this: is there a difference in fuel flow rate at the same RPM but different speeds (i.e. gears)? Further, how does the flow rate vary with throttle angle while holding rpm constant?

Since we can't measure flow rate directly, the next best thing is FI duty cycle.

Another reason I'm wondering has to do with P&G technique - how can I improve my mileage at lower speeds (city driving)? Have I hit a lower limit? If the flow rates are pretty much the same at any given rpm, my commute mileage will never get better (one-off long highway trips will show better mileage, but my main concern is daily driving). 2000 rpm is a handy number to kick around for me, as I use that as my 'cutoff.' When accelerating I try not to exceed 2000 rpm, I can cruise at highway speeds at 2000 rpm without much problem, etc.

Theoretically, the answer should be "it depends" due to the effects of BSFC. I'm trying to confirm it ... ... within reason. For ME I would need to setup a similar gage system as HolyBovine has, but at the moment my curiousity is killing me.

I don't have data to prove this but my thinkings are, at any given rpm you are going to use X amount of gas as a minimum and that will go up with throttle opening and load on the vehicle. I would say that your MPG (miles per gallon) would change but your GPM (gallons per minute) would stay constant with constant rpm regardless of gear. that is given every other factor being the same (and aerodynamic drag falls under engine load).

given this, you want to lower your rpm level during city driving. possibly use pulse and glide (in neutral) or even EOCing (engine off coasting) these techniques requre some research to make sure they won't damage your transmission but several people on here are claiming big differences after these techniques are implemented.

there again, I have no hard evidence to prove this and I have been wrong before and will be wrong again (unless I die a very untimely death)

Hmm. Fuel varies approximately linearly with airflow. So if you have a certain RPM but more airflow (more throttle position), yes there will obviously be more fuel flow (higher DC%).

I have all of the tools for this on my "other" car, but it'll be a few weeks before I'm ready to drive that (lapsed inspection sticker etc.) so I can't get you a nice pretty graph in the near future. Maybe in 2-3 weeks.

The unfortunate part about the whole low-RPM, low-throttle position thing is this: When coming off idle, a very slight touch of the throttle will quickly increase the amount of airflow- if you move the throttle angle from 3degrees to 4 degrees that will probably double your airflow. However at higher throttle positions, a lot of throttle movement only makes a small difference- going from 45 degrees throttle position to 90 degrees throttle position (WOT) only increase airflow about 10% even at high RPM. It's one of those obnoxious things where I wish I could modify the laws of physics a bit.

I was going to type a whole bunch of long boring paragraphs about engine load, volumetric efficiency, air/fuel ratios, etc., to attempt to answer your real question, but I'll let HC answer first because a picture could be worth a thousand words.

-BC

Okay, I experimented with 5th vs. 4th using cruise control for constant speed. It was as I expected.

In 5th at 50mph going up a long somewhat shallow grade in a torrential downpour, my meter was reading ~5.5. In 4th it read ~7.5.

In 5th on level ground at 70mph, it read ~9. In 4th, ~12.5.

I didn't get a chance to try lower speeds in consistent conditions, but in my experience, the observations above apply to any speed and any two usable gears. The effect is probably more pronounced between any two other gears, since 4th and 5th are nearly the same in my car.

What about rpm? Using the cruise keeps speed constant, so the ECU varies rpm to maintain speed.

Could you try this but look at constant rpm versus gear?

Again, if my car uses "X" amount of fuel at 2000 rpm (regardless of gear), I'm cooked in city/local driving due to the low speed of the car.

I drive an automatic. I checked the owner's manual about flat towing and it is not allowed. No EOC for me. I do P&G. Am wondering if I can improve my mileage numbers for my daily work commute. On long highway drives I get great mileage with P&G ... not the case with city/commute driving.

I have been driving in the city with a very light foot, etc. As I mentioned earlier, I am also curious if I've hit a "limit" where I cannot achieve better mpg.

As I said, RPM and throttle position together determine fuel rate. At 2000rpm and 20% throttle it will use the same amount of fuel regardless of gear.

I'm not sure how to do the experiment for constant RPM / different gear. Speed will be different. Aerodynamic drag and rolling resistance will be different. Importantly, I'll have to find a long stretch of flat, consistent road with ZERO traffic so I can go along at different speeds without bothering others or being interfered by them. I may be able to do it in the morning using 4th and 5th again, since they're so close together.

The more I think about it and analyze it, the more I'm interested in what the result would be. I've never tried to consider constant RPM, as I have no use for that -- RPM is always a means to an end for me. I do know a lower gear at a given RPM will be less efficient, but I don't know which way the fuel rate and throttle position will go (though I do know that fuel rate and throttle position will both move in the same direction).

Until you measure that, you're making an assumption, no?

That's the point.

If my car burns 1 gal/hour at 2000 rpm, then at 25 mph in 2nd gear my mileage is 25 mpg. If I'm in 4th gear and traveling at 60 mph, then my mileage is 60 mpg.

According to the BSFC charts, this shouldn't be the case because fuel flow rate is a function of engine load.

See where I'm going with this?

I eagerly await your results.

another scenereo is to go with other mods to help the p+g that you are doing. aero mods (I know they don't work as good at lower speeds) work on rolling resistance and possibly regrease your wheel bearings.

I think their is a limit to what city mileage you can get but there are still some things you can do. you do also reach a point where the end doesn't justify the means. you could do a full engine/trans swap on it which will take time and money. only you could decide if that would be worth it.

This does make sense to 2000rpm you are using 1GPH so at so at 60 MPH you will travel a little over twoce as far and will still use that 1GPH and your MPG will be a little over twice. Just what your data shows. Sure there are minor detail differences- throttle position, MAP, etc but at either cruise speed you are requiring the same amount of power and the same amount of fuel.

Okay, tried 2000 rpm on level ground in 4th and 5th.
4th: ~5.5
5th: ~7
Therefore, throttle position must have been more closed in 4th than 5th.

I stand by this statement:
"At 2000rpm and 20% throttle it will use the same amount of fuel regardless of gear."
Fuel rate is not related to gear, only to RPM and TPS. Exceptions: any lean/rich condition (should not happen), IAT difference (doesn't apply to this question), atmospheric pressure difference (again, doesn't apply).

Fuel rate WILL differ, because throttle position will differ due to the different gear. A taller gear increases load, which means larger throttle opening for the same RPM, which increases fuel rate for a given RPM. Load is pretty much tied to throttle position.

I really think you're looking at this question the wrong way, using the wrong tool. Fuel rate is fine for choosing a gear to use at a given speed, but if you want to know which gear is more efficient at a given RPM you need instant MPG, not fuel rate. Why do the math to figure MPG when you could just use a much more common MPG meter? The ScanGauge (or any OBDII tool), while unable to measure fuel rate, can calculate it accurately enough to answer the question.

Interesting. When I can cobble a few sheckels together, I'll get a HF multimeter...

I have an automatic. I can't 'pick a gear' realistically speaking. Also, if my car is using the same amount of fuel at the same RPM regardless of gear, then I am at a limit in terms of mpg improvements (with regards to my commute). I cannot attain better mileage than what I'm getting now unless I can further reduce RPM ... which is probably not going to happen as I'm driving like grandad already.

As far as instantaneous mpg, I started thinking about this concept when I was on a business trip last month and rented a car that had an instantaneous MPG readout. I noticed that at the same RPM but different speeds I was getting different MPG readouts. I started to think about it, did some dimensional analysis, and figured out that the vehicle's speed is key in determining MPG.

Further, the reason I asked you for data is that you are closer to measuring the actual fuel flow rate than a SG is.

Thanks for the info.

When I setup my doohickey like yours, I'll post some info.

another idea to help (if your environment is hilly) the engine load plays a big part in MPG. I have found that if I see a hill coming, I will give it gas and gain maybe 5MPH before the hill putting me at roughly 5 over the speed limit and once I get to the hill lose 10MPH to the crest of the hill putting me at 5 under the limit. this way I don't have such a big load on the vehicle going up the hill even though I will not get as good of gas mileage before the hill.

the net MPG for the entire run will be better. you can set up a scangauge to tell you engine load. that may help you too. also, don't be afraid to upset some people with slower driving. it is a speed "LIMIT". I still don't like going much under it but if the limit is 55, I will go 50 all day long (that is where I personally get my best mileage).

side note: I still consider myself a young person at 28 so this isn't advice coming from and old fart that is just saying to slow down

But it's not. RPM != fuel usage. It's only one part of the calculation.

There was a thread about clearing up that question, actually. Here's the thing: Speed affects distance. Fuel rate affects fuel volume used. So, of course speed (Distance Per Time, MPH) is key in determining instant MPG when the other variable you have is fuel rate (Volume Per Time, GPH). You strike "hour" out from both of those and now you have MPG (given that you remove "hour" by proper mathematical rules). MPH@GPH = MPG.

For this purpose, the SG's calculation (or that of any car's built-in DIC) is fine and far less work. Fuel rate is important to know other times -- for example, when you're stopped, you still have GPH but you have no MPH.

Does anyone know if the SG calculate DFCO into its average or not? I know it can't directly detect DFCO, but open loop @ closed throttle should always mean DFCO, right?

It is, as you said, if the throttle position is the same.

For most driving, my perception is that the throttle angle is damn close to being the same. That's enough for me, although a HF gage setup like yours will actually provide data versus guesstimating on my part.

I know. I did the dimensional analysis (the math part you are referring to) and figured it out.


I agree, and someday I'll probably buy a SG. But for now, it's enough to know that the rate of fuel burn is primarily a function of RPM, regardless of speed or gear. Since I pay very close attention to my tach, this is a good way for me to think with regards to improvements to MPG. Until I get a SG, that is.

It is most likely that your perception is very wrong.

The ELM327, if you have a laptop you can use in the car, is far cheaper than a ScanGauge and will get you the information you need, as well as logging it. Perhaps that can help you find the best practices for your car? They go for under $30 shipped on eBay.
https://search.ebay.com/search/search...e=260258247681

For sure.

"The easiest person to fool is yourself." - Richard Feynman

Hell, I might just do that as well.

I see they're about the same price on the ELM website. What obviousness am I missing?

Maybe I missed something, but I thought the ELM website only sold the chip, not complete ready-to-use built units.

Also, I had funds in PayPal. :)

honestly, a scangauge was one of the first FE mods I bought. it will tell you what the TPS (throttle position sensor) is saying and also the load on the engine. I would bite the bullet and get the SG.

I don't know a lot about the other device you are talking about but don't you need a laptop for that? not that it is bad but it is a lot easier to have the little 1x4 inch scangauge on the dash than a laptop. there again I may be in left field

Uuuhh, yeah, that could be it.

Yup, you need a laptop, or if you spend a little more you can use it with a PDA or some cell phones (even then it's still a fraction of the cost of a SG). No matter what it's not as convenient or as nicely packaged as the SG.

It does offer additional functionality (logging), though, and I think I may be able to combine it with real fuel injector pulse width data.

Just chiming in late on this very interesting discussion.

I think you're getting right to the heart of the matter with that statement, but I think confusion is invited because your phrasing is ambiguous.

When you say the "same amount of fuel," I think you're leaving it unclear as to whether you mean 'same amount of fuel per unit of time' as compared with saying 'same amount of fuel per unit of distance.' The former is true. The latter is not.

In other words, "2000rpm and 20% throttle" will always use the same number of gallons per hour, regardless of gear (and vehicle speed). But if I change gear (while maintaining 2000rpm and 20% throttle), vehicle speed will change. Since I am changing MPH while holding GPH constant, MPG is going to change.

I agree with you that logging could potentially be very helpful, although I'm still trying to figure out the best approach to that. I also think you have a very good idea about integrating injector data somehow, so I look forward to hearing more about that.

Well, my idea for integrating injector data begins here:
https://www.gassavers.org/showthread.php?t=7608

I just have to find a way to make the sound file into a stream of simple numeric data describing pulse width/duty cycle, and maybe average the amplitude for each pulse (though that should be pretty consistent). I don't know if it's beyond my abilities or not. Once that's done it should be a piece of cake to combine it with OBDII data. I'm not really sure why, though; I don't think there's much more for me to learn about my car by doing it...it's just an idea I have.

"I'm not really sure why"

The point of looking directly at the injectors is that it might be much more accurate than the data OBD-II provides. But I think this would only be an issue on certain cars, under certain conditions. So I can't claim it's a universally valuable thing to do. I just don't know for sure.

Oh, I know that's a good reason why to look at the data, but I already do that in realtime with the meter. What I'm not sure of is why I want to log it and lace it together with OBDII data. I guess that'd be so I can look at it compared to the other data, to answer questions like the one at the top of this thread.

That's the root of what I want to know.

If I'm driving in my development at 20 mph in low gear at 2000 rpm, I'm burning the same volume per unit of time as I am if I'm driving at 60 mph at 2000 rpm in top gear.

No, because throttle position will be different.

This morning I actually was thinking about this thread a little as I accelerated, going from gear to gear at WOT. As I entered each gear at the same RPM and same throttle angle, I watched the meter, and it used the same fuel rate.

I'm still waiting for my ELM327 to arrive from China. The PCMCIA USB card I ordered arrived but doesn't work in the laptop I planned to use; I need to try it in another to find out if it's bad or if the laptop is bad. :(

Sure. According the the BSFC charts, I should be burning *more* fuel. Thus the question. You should see it in your data, assuming you can correlate duty cycle, throttle angle, and rpm at the same time.

I would think, though, that one could do it "by the seat of your pants" to get a feel for what's happening.

BSFC charts describe how efficiently the engine produces power when you're talking about a fixed amount of work, but don't take into account the different amount of work that you need done at different speeds. You can be burning more fuel while still being more efficient, because you are getting more work done.

The same car at 60mph requires a lot more power to move than at 20mph. There's aerodynamic drag, rolling resistance from the tires and wheel bearings, and drivetrain drag from everything between the transmission and the wheels.

In your example, 60mph would use more volume of fuel but be more efficient.

As thc said, that's true only if the throttle position is the same.

Yes, that's exactly what I noticed when I had my meter connected. (And just to be clear, we mean fuel rate per unit of time, not unit of distance.)

Here's how I look at it. Fuel rate per unit of time is directly proportional to duty cycle. Duty cycle just means how much time the injectors spend being open, as compared with how much time they spend closed. The ECU is constantly judging what the duty cycle should be, and it's basically taking into account two things: engine speed, and throttle position.

So if those two things are held constant, then fuel rate (per unit time) is also going to be constant, even if you change gears (and, therefore, vehicle speed).

There are some other factors that matter but we're assuming that those are also held constant. I'm talking about things like engine temperature, fuel temperature, air temperature and atmospheric pressure.

Actually, I'm pretty sure the ECU uses the MAF and O2 sensors to decide. RPM and TPS, in this context, are merely a way for us to predict/express MAF, because we know that air flow depends on RPM and TPS.

We could just talk about MAF, but that wouldn't really help us decide on gears and speed and what not.

Yes, but units of SFC are lb per hp-hr. Which also means that I should be able to very easily compute my vehicle's fuel consumption (lb per hr) based on engine load and rpm.

I'm not worried about transients, I'm interested in steady state comparisons at different speeds.


This is why I'm interested in this subject and data ... there should be a way to connect the BSFC charts to the data measured in operational settings.

Back to my example. The answer is that I don't know (if I knew I wouldn't be asking). If I'm driving with the same throttle angle and rpm but different gears, then I would assume the FI duty cycle would be the same in each case.

If the rpm were identical in both cases but the throttle angle smaller in lower gear, I would assume the FI duty cycle would be longer, as the engine would be operating with a poorer SFC due to pumping losses, etc.

If the rpm were identical in both cases but the throttle angle larger in the higher gear, then I would assume the FI duty cycle would be shorter as the engine would be operating with a better SFC.

The logic I'm using comes right from the "iso-curves" on the BSFC chart that correspond to differing engine loads.

Let me add something that might have been what you were getting at (and is where I've gone wrong)

I could be driving at 2000 rpm, however in 2nd gear I could have a throttle setting of 1/4 and in 5th gear could have a throttle setting of 1/2.

In each case the engine is at 2000 rpm however the difference in throttle angles results in more power with the 1/2 throttle setting versus the 1/4 setting. Which means if I go to my BSFC chart, I would select the SFC corresponding to the correct rpm and load. (I'm assuming prior assertions that "load" corresponds to "throttle setting" are correct. In other words, 50% load is the same thing as 1/2 throttle.)

The next step, to compute the actual fuel consumption, is to multiply the BSFC value by the HP produced by the engine at that rpm and throttle setting.

This is where I go wrong (and to be honest where I came to a quandary several weeks ago, however I'd since forgotten about it)

Question: To figure out HP for a certain throttle and rpm, can I take a WOT dyno reading and multiply it by the fraction representing the throttle setting? In other words, if at 2000 rpm and WOT the hp is 100, can I figure the power produced at the 2000 rpm and 1/4 throttle is 100hp*0.25=25hp?

Sorry I didn't elucidate this earlier, as I mentioned I came to this several weeks ago while poring over the BSFC and BMEP/SFC charts and totally forgot about it.

In which case the lower throttle setting may or may not have a longer FI duty cycle depending on the engine's characteristics (power produced and BSFC ratings for throttle settings (i.e. BSFC isocurve)).

(it sure would be nice to have an BSFC chart for my car rather than the generic one that's floating around out there)

False. While the engine is operating less efficiently, it's producing so much less power that the inefficiencies do not add up to the difference in power. So maybe you're asking for 50% of the power, and paying 5% pumping losses; you're still using 45% less fuel.

Since aerodynamic drag goes up as a square of speed, often lower speeds are more efficient -- but of course the sweet spot(s) can vary due to BSFC map, gearing, etc.

No. At a given RPM, FI duty cycle will move in the same direction as throttle -- wider throttle, higher FI duty cycle; restricted throttle, lower FI duty cycle.

This is one very troublesome semantic failure in fuel economy discussions...it's hard to know when someone (or a document or chart) is using load to mean technical engine load (throttle position / vacuum), the amount of weight on a vehicle, the drag, etc...stupid language...

I have no data or theory to back it up, but I am almost sure that the answer is no. That is a great question and I hope someone has a more authoritative answer.

In another thread, someone mentioned that throttle is not linear; so, the difference in power between 2/10 and 3/10 throttle is way more than the difference between 7/10 and 8/10.

No. If anything you might be able to compare injector duty cycle %, but even then you have a ton of other theoretical issues. (What is HP? Total heat generated? Flywheel output? Rear-wheel output?) I suppose you could subtract WOT duty cycle from idle duty cycle (at the same RPM), then graph that out and figure out your HP output based on WOT DC% - idle DC% or something crazy.

Like I've tried (poorly) to explain before, airflow is NOT a linear relationship to throttle position. I believe it would be an inverse-logarithmic curve, but I'd have to think that one through. To explain simply, the first 1/8" of pedal travel might increase HP by 30%, but the last 1/8" of pedal travel might increase HP by 2%.

Agreed - Engine "load" has to be one of the most annoying discussion concepts that I've come across.

-BC