Hypermiling a diesel

[Diesel_Dave]

Originally Posted by slowmover

The UG should also help me determine (and monitor) the optimal throttle position for efficiency for those tests as well.

Yup. "Real time" on a heavy pickup appears to make the gauge a necessity.

I don't have a Scangauge, so I don't have an instananeous read-out of MPG, just the average trip mpg from the lie-o-meter. What I do is reset the lie-o-meter at the beginning of my trip.

Here's two tips for using the lie-o-meter:
First, if you have a regular route (like commuting) what you can do it set "checkpoints", try different techniques and see how they affect your trip mileage up to that point. For example, I check my lie-o-meter at the bottom of the first "hill" out of town. On a good day I'm at about 30 mpg, on a bad day (like today) it was 25 mpg. Obviously it's not an exact science and there's a lot of variablity with weather, etc. but it will give you an idea of how well you're doing. You can try different techniques and see what the outcome is. From my experience what will happen is one day, you'll get it right and will suprise yourself at how well you did--then it's a matter of figuring out what made the difference and repeating it.

Second, when learning to P&G just compare your avg mpg at the bottom (or top) of each P&G. When pulsing your avg mpg will obviously go down, but note whether the avg mpg at the end of this glide is better than it was at the end of the last glide.

[Diesel_Dave]

Originally Posted by TX_Dj

In my case, I just see no point in spooling the turbo (much)... more boost == more fuel... by planting your foot to P&G, are you not spooling the turbo and increasing fueling?

I've always been a skeptic of P&G (even though I've seen evidence to say it works) because well, Newton says you can't get more out of it than what you put into it... and if you put X into it by speeding up fast and coasting down, getting X as the result, that's great... but seems to me that if you use the technique of getting on the torque peak, mashing the pedal and spooling it all up for a hard pull, X > Y output... because well, it takes more power to accelerate quickly and your glide will be relatively constant from any speed all other factors being the same.

What seems to be working best for me in my truck (granted, an 8000 lbs dually is a bit of a different game than a 1.9 TDI) is to keep boost and pyro low, RPM near or just under peak torque and cruise steady rate while allowing the vehicle to slow on inclines in order to stay below target boost/pyro numbers. Then again, I'm definitely getting nowhere near as good as Dave, and he's admitted to P&G techniques. His truck is also lighter, shorter, more aero and less rolling resistance than mine.

I'll admit that I was skeptical of P&G at first and it took me a while to get the hang of it. FYI, it depends on traffic a lot. I'm fortunate that a significant portion of my regular commute is a fairly lightly traveled 4 lane divided non-interstate highway. You can see a description of my basic technique in this thread (Post #9):
http://ecomodder.com/forum/showthrea...ear-18337.html

I'll tell you the story of one of the things that convinced me that P&G worked. Before I moved, I had 2 routes to choose from. One was the interstate, the other was a parallel highway with quite a few traffic lights. I found that even if I slowed down on the interstate, there was no way I could match my FE on the highway, even at the same speed. What I figured out was that when I was timing the red lights on the highway (by coasting in neutral coming up to them if they were red, then accelerating on green) I WAS P&G without noticing it. So what I learned to do was basically do the same thing (accelerating then coasting in neutral) more or less all the time (as traffic allows).

And no one's claiming that P&G somehow defies the law of physics. I think you're think about it a little bit the wrong way. You think the pulse is bad because it burns more fuel and burnig more fuel = bad. That's not necessarily true. Burning more fuel overall is bad, but burning more fuel at any given moment in time is not necessarily bad. Accelerating does use more fuel, but the conversion of that fuel to mechanical energy will happen more efficiently (if done correctly, because your engine is at higher load). At the end of the pulse you will have burned more fuel than if you were cruising steady. But, during the glide you will make up for it.

I've made up the numbers but here's a quick axample, your engine may be 35% effecient at near full load and 25% efficient at part load (fairly reasonable numbers). Let's say you're at 30 mph. You can pulse up to 50 mph at 35% efficiency, then coast in neutral (burning almost no fuel). By the time you glide back down to 30 mph (and gotten the pulse energy back), you will be better off than if you had run the whole time steady at 40 mph and only 25% efficiency. Obviously, I'm simplifying things, but hopefully that makes sense.

[slowmover]

Second, when learning to P&G just compare your avg mpg at the bottom (or top) of each P&G. When pulsing your avg mpg will obviously go down, but note whether the avg mpg at the end of this glide is better than it was at the end of the last glide.

Anyone following me the first time I tried this would have been laughing . . the way I was punching up the overhead readout to re-set it must have looked like I was using a ceiling mounted walkie-talkie conversing with a near-monosyllabic boss asking me constant questions.


the conversion of that fuel to mechanical energy will happen more efficiently (if done correctly, because your engine is at higher load)

Thus the Scanguage (if I am correct) as another has mentioned that this (or similar) gives a readout of engine load as a percentage. With so much free-wheeling -- or danged close to it -- and all the mass/weight of a 7,400-lb pickup to consider some sort of real-time device is needed.

The usual close-coupling of engine speed against gear selection is missing for tactile driver feedback over longer stretches of pavement. Just guessing is out the window, IMO, as the forces needed to react to a problem are high, and the distances never inconsiderable (as one may be loaded, towing heavy; both).

Some read-out to know "where we are" as against other gauges of things (visual, mainly) would provide some relief to the driver, IMO.

So much of this is happening in an in-between state: We are neither accelerating, nor slowing (per se) but we are not maintaining a steady speed either. It's a conundrum, and it's about control that a pickup truck (or larger work vehicle) faces that an air-conditioned go-kart driver doesn't face.

On one of those just stare at the brake pedal a split-second, and blink. Full stop achieved. Not so when grossing 26k.

In other words, we've taken one of the moments of uncertainty -- a change of state -- and widened it into a fetish.

I don't want DD to think I'm getting on him (I tried for some clarification on this post on CF; how to analyze this) as he is doing what I have yet to find elsewhere. That he has a lab (stable commute) to use daily is huge, IMO. I've re-cast what I've been doing as a result of his participating here and elsewhere.

But these trucks, solo/empty/non-revenue is not their highest & best use. Forgive me the need to drag it in. There is a larger context, and FE means even more when tenths of a mpg pay a man thousands of dollars annually on big trucks, and one or two mpg on a pickemup in revenue miles.

When I have 13-speeds to choose from it is easy to find the right one ease along, to glide . . . but when I am relegated to but six foward speeds, it's a heckuva lot harder to make the trade-off. A truck driver has to literally be able to read the pavement, and to use every foot of it to advantage. Thus, a longer period of uncertainty (in a manner of speaking) needs an aid for corrections.

Instant mpg is kind of over-rated. Constant current mpg is useful only on a daily or per-load basis (as per linked post). Thus I hope that one of these gauges will show engine load percentage as I think this is the right step for this vehicle type (not otherwise covered) to cover those long pauses between different states.

Pulse or Glide in 5th Gear

Thankfully the 400-lb transmission in these trucks has a Direct that is capable of handling nearly any road speed above about 30-mph (25 if empty and all other inputs lightest) when not towing much. Even a decent bed load doesn't affect this very much when on the flats.

.

[Diesel_Dave]

Anyone following me the first time I tried this would have been laughing . . the way I was punching up the overhead readout to re-set it must have looked like I was using a ceiling mounted walkie-talkie conversing with a near-monosyllabic boss asking me constant questions.

Haha

I don't actually reset the overhead every pulse. I just reset it at the beginning of the trip. What I was saying is this: Let's say you're cruising along at and your overhead is reading 25.0 mpg. Do you pulse and the avg mpg may go down to 23.0 mpg. As you glide down it should tick up and up to say 25.3 mpg. Do another pulse and it'll go down to 23.3, then glide up to 25.6 mpg, etc, etc. So compare your avg mpg number at the same point in the pulse/glide (i.e., 25.0 vs 25.3 vs 25.6).

Obviously, your avg mpg numbers will change more rapidly at the beginning of a trip. Like I said, it's not an exact science, but that's how I do it.

[Diesel_Dave]

That he has a lab (stable commute) to use daily is huge, IMO. But these trucks, solo/empty/non-revenue is not their highest & best use.

Agreed

[Piwoslaw]

Originally Posted by Vekke

Some will drive in the bumber and usually in that case I start to slow down so much that person will pass me. usually 10 km/h speed under the limit is enough. So I courage people to pass me in those cases.

They're saving fuel by drafting you! Though in your case, your wake is probably too small for anyone other than a bicyclist to get any benefit

[TX_Dj]

My wake could probably swallow a bus.

Wish I could put some miles on my bicycle instead, but not an option due to safety and weather reasons. Just have to get back on track for the EV build and then I can worry less about my diesel bill.

[euromodder]

Originally Posted by Piwoslaw

They're saving fuel by drafting you! Though in your case, your wake is probably too small for anyone other than a bicyclist to get any benefit

You don't gain by drafting a small car, unless you're really riding on its bumper.

[TX_Dj]

Dave, Thanks for that explanation. In light of other things I've learned in the past few days, it does make sense... and for the same reasons, I now know that any time I've tried that in the past, I've done it wrong.

Unfortunately for me, there is no BSFC for my engine... it is not stock, nor are the programs I run on it.

For example, my main "eco" program takes me down to about 140 HP / 250 ft-lbs (wheel), and my main "not eco" program takes me up to about 500 HP / 1000 ft-lbs (wheel). Everything on the intake/exhaust sides of the engine is not stock, and with extra air comes a need for extra discipline (or, the "nanny mode").

If I were to P&G on my "fun" program using 6th gear and the 75-95+% throttle position figures I've seen around here, it'd turn the skies black behind me... not something I like to do, because even though diesel smoke is black, we know it's really green... wasted money, that is. The "fun" program takes a different methodology for clean burning power than burying the skinny pedal, one must ease into it to "light" the turbo, then you can crank up the load. Obviously, this program isn't where I'm getting my higher number tanks.

On the eco program, though, there is very little rhyme or reason to its behavior. Keep in mind, this program wasn't designed to be eco, I just use it as my nanny to keep me within more efficient ranges of operation. The manual says "Use this program when you absolutely, positively cannot make any smoke." And it won't, no matter what I do. But, because they didn't tune the program to be for economy, per se, or for anything really other than "no smoke", its drive-ability suffers greatly. In order to rev out enough (500 rpm) so that I hit my target RPM on my next shift point, I have to progressively increase throttle position gradually from maybe 10% to 60% from 1500-2000. Timing is incredibly advanced, ensuring that all the fuel in the cylinder burns completely before exiting. Because the tuner made stupid decisions on the timing map, running *any* amount of load below 1500 causes the truck to complain, partly from timing knock and partly from gear rattle, both a result of too much advance at too low an RPM without enough throttle to warrant it.

My UG will be here sometime next week. It's my hope that in using it, I can better determine where the peak efficiency is either on the stock program or this "half power" nanny-mode, and perhaps then I can start to P&G my truck effectively. I would drop the nanny-mode entirely in favor of stock, but even then I manage to slip up and shoot 200-300F past my EGT targets, or 5-10 psi over my boost targets.

My long-term goal is to switch over to EFI Live so I gain total control over the engine's behavior, and the ability to tune the maps exactly where I need them to be for best eco running. With this tool, one has total access to the entire chain of Throttle Position commands X Fuel, X fuel @ Y RPM commands Z rail pressure, Z pressure @ X fuel commands A duration, and Y RPM @ A duration commands B advance. With creative planning, one can build the tables to not over-advance or over-pressurize at low loads and throttle positions and low rpm, enhancing drivability and smoothing out the overall engine performance and efficiency.

Don't get me wrong - the stock program does quite nicely by itself, but my engine will tend to fall a bit outside the BSFC given by cummins due to the hardware modifications, so now it's time to build the programs that will adapt engine running to those mods and find the efficiency from them.

[Diesel_Dave]

Unfortunately for me, there is no BSFC for my engine... it is not stock, nor are the programs I run on it.

Yes, you're right about that. Cummins doesn't release BSFC maps publically for proprietary reasons. That being said, you don't have to know what the BSFC map is for P&G to work. Most BSFC maps for modern turbo-diesel engines have more-or-less the same basic shape. Of course there are differences but the general shape is what allows P&G to be effective--principally the fact that the engine is far more efficient at high loads than at load loads. Same goes for non-stock programs--you're not going to change the overall shape of the BSFC map significantly with a program.

Everything on the intake/exhaust sides of the engine is not stock, and with extra air comes a need for extra discipline (or, the "nanny mode").

Assuming your turbo is still stock, I don't think your intake & exhaust mods will have a huge effect on the air flow until you're running up near max power. When your flows are low, it's not going to make very much difference. I can see how your intake and exhaust mods would have an impact on the max power of your engine, but personally I don't see it having a huge effect on things when you're dring for FE (when your air flows will be relatively low most of the time because your rpm's will be low).

On the eco program, though, there is very little rhyme or reason to its behavior. Keep in mind, this program wasn't designed to be eco, I just use it as my nanny to keep me within more efficient ranges of operation.

I'm not sure where you got this "eco" program but, honestly, it doesn't sound that great to me. From what you're describing I think you'd be better off just sticking with the stock program. Maybe add a little timing and/or rail pressure to it, but that's about all. Sounds like the program messes with the transient response. You probably want that for your "fun" program, but you won't need it for FE. Sounds to me like this is why you're having smoke issues. Remember that smoke is primarily from running a low air-fuel ratio. If you're sending in too much fuel before the air is there you'll get more power (because more fuel is burned), but that fuel won't be burned as efficiently (smoke is unburned fuel). Advancing the timing and increasing the rail pressure will generally reduce smoke and help FE, so I'm guessing you're problem is air-fuel ratio. Since you have a 5.9L engine, it was designed to run quite low smoke without a DPF-several times lower than the visible smoke limit. Something about your program is increasing smoke relative to stock. My guess would be it's low air-fuel ratio. Either that or your timing is just way, way too advanced. That can give you smoke too, and hurt your FE.

When you P&G, it sounds like you may also be pulsing a bit to hard as well. The other day I timed myself on a level stretch of road with no traffic. I pulsed from 27 mph to 50 mph (all in 6th gear) in about 12 seconds. My glide back down to 27 mph took about 1 min 10 sec. That gives you some idea of how I do it. I'm not claiming that my way is the absolute best way, but that's what I do and it works for me. You might want to give it a shot and see what happens. You might have to slow your pulse down slightly because you have a 5.9L which has a wastegate turbo, as opposed to the VG turbo I have in my 6.7L--it might not spool up quite as fast.

Hope that helps...