Hypermiling a diesel
 

Hi,
is there any differences in hypermiling techniques when driving diesel car? I made search and did not find such topic.

Maybe some tips are in petrol only, some for diesel only? Would like to know. Lets discuss these differences here.

Q1) Does pulse and glide (in neutral) make same % difference compared to steady speed drive than in petrol car. I think it probably does not give same amount of difference beacause diesel does not have throttling losses.

If I am lucky, I have new car tomorrow evening...diesel :)
Audi A4 Avant 1.9TDI 81kw, 1997

Meelis

On diesel you have to make the Push with full throttle (97%). RPMs should start the push at torgue peak or rigth before it. On a 1.9 TDI that is about 1800rpm. Use the shortest gear that meets the RPM reguirement. Usually 4:th gear.

If you want to get low figures on your TDI I would recommend to buy 175/80R14 tires to front and 60/40 lowering spring kit.

On petrol gas position is about 70%.

I P&G with my diesel and see probably a ~25% benefit vs steady speed. Granted, I also have a 3+ ton truck so it's not exactly the same as a passenger car.

You're right that diesels don't have throttling losses, most there is still a sharp dropoff in BSFC at light loads. That's due mainly to factors other than throttling, such as heat transfer and the theromdynamic effects of lower temperatures and pressures which reduce efficiency. So the concept of running higher loads as a means to higher efficiency is still valid with a diesel.

Secondly, one advantage that a diesel has when P&G is that since you don't have throttling losses, your "glides" are relatively more fuel efficient than a throttled engine when you're coasting in neutral (idling). That being said, idle FE can vary significantly from vehicle to vehicle.

Here's an old thread on the topic:
http://ecomodder.com/forum/showthrea...esel-9664.html

Hi there,

that does not work for me (2.0 L TDI in VW T5 Van); I use much less fuel when accelerating if I do this from around 1200 RPM (as soon as the engine can deliver enough torque to accelerate).

Generally spoken, I have the impression that I use less fuel if I avoid full boost pressure, which comes around 1700 RPM. Since my setup does 1400 RPM at 56 mph (90 km/h) which is my preferred cruising speed, I am usually well below that. Upshifting early (<1600 RPM) does the rest.

BUT: I usually do not P&G, so things may be different when doing this.

so long,

tinduck

So driving techniques are almoust same in gasoline and diesel cars, but probaby nut behind the wheel in gas car can make bigger difference (in FE %) compared to "normal driver"?

Pulse and glide is a waste of fuel for me. If you just cruise in your in between your peak torque and the half way point to peak power rpm you'll be just fine. For my car my peak torque is about 50mph at 2300rpm and my peak power is around 100 mph at 4250 rpm. At 70mph my rpm is 3000 which hits right dead center of my upgraded turbo's compressor map.

The main reason other than safety why I don't pulse and glide is it will plug my DPF.

Its trickier to get as effiecient burn in a petrol because gas position is different, and you may have to tweak it during acceleration (correct if I am wrong). In a diesel simple pedal to the medal attitude brings best fuel efficiency if certain facts are met. So in a diesel you dont have to adjust the nut so much...

What I wrote applies almost all turbo diesels, but some in some cars the torgue peak is higher. Steady throttle and highest gear if drived steady speed is also a efficient choice but still the accelerations to target speed/posted speed limit should be:

1. Done at torgue peak or close to it
2. Using shortest gear possible so without revving it over 2500 rpm or 3000 rpm. This depends on the car. On my lupo 3L the electromanual gearbox in eco mode car computer uses gears so that it starts the acceration at 2000 rpm and shifts bigger gear in 3000 rpm. On my lupo almost nothing happens under 2000 rpm. I believe that engineers protect in my lupo the gearbox in lower rpm range. You can hear and feel when engine reaches 2000 RPM. Noise changes from baaam to BAAAAAAM if you know what I mean ;).

On the lupo if I am doing P&G the speed drops to 74 km/h the engine will put the third gear in because in fourth the RPMs drop under 2000 RPM where nothing happens. At 83 it already changes to fourth. So in that speed range manual would be better because it also takes fuel at least in the lupo to change gears.

Effect of the driving technigue can easily be tested on short test route. It does not have to be long 6 miles or so. Your 09 VW will show quite accurate readings with different driving technigues. Keep the test speed close to 50 MPH so aero does not play so big role.
1. First steady throttle like you drive at the moment from 0-50 and keep it there
2. Second trip that with rapid acceleration to target speed and steady speed
3. third rapid acceleration and to target speed +7 MPH coast so that you lose 16 MPH and again a push.

My estimation of the results are
1. Is worse
2. is 4% better
3. is 15% better

BR:Vesa

The above is only applicable to TURBO diesels, naturally aspirated like mine smoke up a storm and waste fuel near WOT, best at about 75% ish

Cheers
Ryan

Yes, black smoke is wasted fuel. I try to accelerate quickly with "Big Red", but I try to minimize the smoke. Unless someone is "drafting" me. :)

How did you manage to get a DPF on a '85 vehicle?

Because I keep my 5-speed diesel Beetle in low rpm's it is only feasible to glide on the down hill sections and slight declines. My diesel has enough torque to drive in 5th hear at 30 mph @ 1000 rpm's,it has the ability to pick up speed smoothly at that rpm/speed.
If at faster speeds and on a decline I can put it neutral my rpm's drop to a idle and I coast/glide gaining speed or slightly losing speed,usually the former.
at 60 mph my rpm is just under 2000 rpm this is where my car needs help aka a taller fifth gear or another final drive.
it seems pointless to glide when my rpm's are in the low range. (almost always) I about never drive my Beetle over 2000 rpm..

On a side note I do coast on declines even if my rpm's are low to begin with,coasting makes them even lower. So I coast, hmm I'm a coaster, just coasting thru life taken it easy..in the slow lane, Wow sounds boring but to the contrary I find it interesting enough!

Oddly enough, some of those came with DPFs, but they were all recalled and removed because they plug with no way of cleaning them out. If his car is still equipped, it would be hurting the fuel mileage in all conditions; removing it would enable further mileage gains with P&G.

I just finished a trip and a tank using some advice from this thread. Specifically, I throttled harder during the pulse phase. The vehicle is a, currently not very aerodynamic, diesel pick-up truck with manual transmission. The engine puts out some serious torque in a 1000 rpm band so the truck accelerates quickly even in steep inclinations when flooring it. Therefore I used p&g consistently even in inclinations despite the gliding phase suffering. Previously I used to keep a steady speed uphills.
Anyway, the net result was a very good improvement. I got 6.5 liter/10 km one way and 6.0 the other way. Previous figures for this run was about 7.4 when driving with a gasoline style p&g. For comparison, running on cruise control yields 8.7.

Now the question:
Since this vehicle is "a brick" and does not glide well, there are a pretty large number of acceleration phases for a given distance. The engine and turbo is larger than your average car so the turbo lag is significant. It takes a couple of seconds until the torque wave kicks in, especially in highest gear. This means I am flooring it without getting much in return a lot of the time.
Is this really the best way of applying power?

The built in fuel meter only averages consumption over longer time so I can't study this in real time. Anyone with knowledge on modern diesels that can say something about the typical engine mapping and efficiency before the turbo starts producing?
Should I gear down to reduce lag and have shorter pulses? (more rpm and transmission losses and more wear on tyres etc.)
Should I apply power more progressively and not floor it at once?

HAHA, very good results!
Probably you should try one gear down (4th instead of 5th?) so that engine is in right rpms from the beginning of the pulse. I just got my diesel so I am not expert at all, but Vekke writed that and he is getting really good hypermiling numbers.

The engine is in the right rpm range for full torque but there is still turbo lag. The turbo needs a good flow of exhausts to spin up. Once it does, there is plenty of power even in the highest gear. The question is what is going on with the efficiency during the lag-phase...

I'm the scientific type of guy so I'm looking for actual numbers and figures from real measurements. I'd love to do my own experiments but my lack of real-time instrumentation makes it difficult in this case.

HAHA, upload the details of your truck so we have more information.

When driving my diesel truck I don't floor it when I pulse, I just use a fairly smooth acceleration (so there's little turbo lag). That being said, my truck has a VG turbo so there's less turbo lag anyway.

Trying to increase post count so I can use links and pictures...

The vehicle is the Mitsubishi L200, named the "Triton" in some locations on our insignificant little planet.
The engine is a 2.5 l, VG turbo, 400 Nm at 2000 - 2800 rpm. Vehicle weight is about 2000 kg. (now you get a chance to practice some metric units...)

I found some power curves on a site belonging to an aftermarket chip trimmer. The green curves are for the stock engine. The second graph is a bit unusual but it illustrates the turbo lag as the time it takes from throttle until torque. If one is to believe the graph, the lag is 2.27 seconds until 90% torque although it doesn't say from where in the rpm-range. This approximately correlates to the driving experience at a speed matching 2000 rpm. (smooth clutch engagement and then almost full throttle)

http://www.sttemtec.com/1/1.0.1.0/14...8c38861e95.jpg
http://www.sttemtec.com/1/1.0.1.0/14...f4f96e6f95.jpg

This graph didn't make it in the previous post ...

http://www.sttemtec.com/1/1.0.1.0/14...f4f96e6f95.jpg

Chip tuning helps for sure to the turbo lag.

- What gear did you use? You should use the smallest gear which starts from 2000 rpm.
- What speed did you use the P&G

If the push and glide times are the same push 7 seconds and glide 7 seconds I dont bother to do P&G:ing. IMO there is no gains to be made in those circumstaces. That speed when this happens depends on your Cd and trucks weight and power(torgue) to ratio and few other factors.

For turbo lag you can also remove the intercooler. Specially in the nordic winter unless you dont tow something very heavy with your truck often.

As you can see from my numbers, P&G pays off well in this case. The last trip was on highway at about 90 km/h average speed. I like to find myself a big semi behind which I can pulse & glide into the draft. This gives you some idea about the average speed. The highest gear (5) results in about 2000 rpm at that speed.

Interesting about the intercooler...

The chip trimmer company mentioned above actually achieves the results partly by replacing the intercooler with a larger one.
See: Power Kit L200 Triton 2,5 DI-D
They refer to less pressure drop so maybe the same effect is had by removing it altogether.

The question remains...
What happens with efficiency during the lag phase?

There is no black smoke or anything so perhaps the ECU does the best of it.

It was the very first vehicle to have them. Its a Kalifornistani model.

Just adding some information for completion...
I communicated with STT Emtec about the Mitsubishi L200 engine mapping and the perceived turbo lag. I learned the following:

The lag is not really a turbo lag but rather the computer not supplying enough fuel to get things moving. This is done to control the emissions (regulations on diesels are very strict these days). The trim box is able to overcome the lag by injecting more fuel in the beginning of the acceleration phase.

I also got the information that the engine delivered peak efficiency at slightly above 2000 rpm and ~300Nm torque (~85 hp), which equals a fuel consumption of about 200g/kWh.

I also learned that installing the trim box will cause the fuel instrumentation to miscalculate because it is not actually measuring the flow and the box is piggybacked on the ECU (south). The "optimizations" are active only at more than 75% throttle so for normal driving, this won't be an issue. For p&g eco-driving, the instrumentation would become pretty useless since you alsmost only use the engine at 75+%.

Anyway, from this I draw the conclusion that the lag during initial acceleration is pretty harmless from the point of eco-driving.

Don't remove the intercooler. It aids in fuel economy too. By reducing the combustion gas temperatures it reduces heat lost to the cooling system allowing more energy to be extracted by the piston.

What comes to intercooler off, temps on my Lupo 3L are now with steady throttle and no grill block at the moment:
80 kmh = 40 celsius
100kmh = 48 celsius
120kmh = 70 celsius
Outside temperature 13 celsius and wet.
On winter if you have a intercooler those intake temperatures will be on minus for the first 10 kilometers and your engine will warm up slower. There is no power gains to be made if you feed minus celsius air to your engine at least with my current knowledge. You should try to get the engine warm up not froze it.

They have propably reached those power figured with bigger intercooler for sure, but if you are hypermiling your acceleration times will be so short that intake temps wont have time to raise so high that you would need it. The intercooler propably also flows better but if you take it all out its har to beat that in flow efficiency. Every bend in your pressure pipes makes little drag and depending on how many bends you have the turbo has to work harder to be able to produce the same boost level. Usually the pressure drop is about 0.1 to 0.15 bar.

On summer in a big truck you propably still need it but that can be tested with fuel consumtion meters like ultragauge etc where you can see the intake temps.

Oh and in Finland and Sweden the winter temperatures can be as low as -35 degrees celsius or -31 degrees fahrenheit. So the biggest problems are to first get the engine running and after that make it warm so that you dont need freeze inside the car.

I definitely think that hypermiling in a diesel makes a much bigger difference than in a gasoline car. I own both. If you check out my cars on Fuelly, you'll be able to tell.

Good luck and remember, if it sounds too complicated, the person who's saying it probably doesn't know what they're talking about.

The 1.9 TDI 110 cv engine is one of the best efficiency ones, but maybe some ECU ecotuning will reduce your fuel consumption between -10% to -15%, you could try if any change is made also reducing coefficient of drag (covering full or the most of the upper grill, remember Audi A2 grill?) and fine tuning your camshaft timing. Try some "search" in Fred's TDI Page. TDIClub.com. VW TDI Enthusiast Community.
I did -13% at 82km/h to -18% at 122km/h using cruise control, fine tuning camshaft timing and with aerodrag reduction,
http://ecomodder.com/forum/showthrea...3-a-19233.html
http://ecomodder.com/forum/showthrea...ape-19191.html

I had a manual of how to fine tuning camshaft timing in my other 1.9 TDI 110cv AHF, but i did'nt find now.
I will try doing P & G while data logging using VAG COM OBDII.
But, max efficiency of my engine is about 0.4 and using cruise control in fifth gear, .32 is worst efficiency maintainaing cruise control in fifth. teretically 20% less fuel consumption was possible, i think easier with larger gears, bu i Will try.

The ECM not supplying "enough" fuel is most probably due to emissions contraints (probably smoke). Typically smoke comes from too rich of a mixture. Keep in mind that smoke levels from the stock engine are probably seceral times less that the visible smoke range. The chip tuner may actually be increasing smoke quite a bit, but you still might not be able to see anything. Another possiblity is that the chip tuner is advancing the timing to get more power. This probably increases the NOx above the stock level. At the same time it probably lowers the smoke too (more advanced timing generally lowers smoke). That's just my guess..

This TDI powered Audi A4 is really good! Dont know yet how accurate is onboard trip computer but this weekend I made some driving. Trip is about 106km one way, resetted computer, engine was warm. When I arrived to destination, computer trip average was 4.0L/100km (58.8mpg) :) Made some city driving (many short trips) next day, still was 4.5L/100km, not bad at all!

Driving back 2 days later, resetted computer again, I managed 4.1L/100km (57.37mpg). Route was little bit longer, total trip time was 6 minutes longer.

Driving style: I was driving around 90-100km/h, made some P&G in neutral if it was possible (no cars behind me)

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.

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 resistant than mine. And he has a better haircut . . . .

I'd believe that!:cool:

In a diesel it's not going to give as spectacular a result as in a gasser.
If - like me - you don't want to shut down the engine during the glide, the net result is likely going to be marginal.
I've tried P&G (engine on), and while it looked good on the ScanGauge, the tank average nor actually refueling showed anything out of the ordinary.
I've found it to be rather annoying to other drivers as well - one's hardly ever alon on the road over here.

I've resorted to pulsing pretty hard, then coasting (engine on) to transition points along my usual route - turns, curves, stops, lights, anything requiring a substantial drop in speed.
I've experimented with the speed and place combination that I need in order to arrive at the next transition point with a suitable speed.

On longer stretches of road, I still use a constant speed.

I've reached the same conclusion with my VW diesel with automatic transmission. The Scangauge lies about P&G so I don't do it anymore; I just coast in neutral when I want to conserve momentum and DFCO when I have to slow or stop.

The Scangauge thing is really annoying actually. It is really lousy feedback when it comes to coasting and engine mods. I am keeping it only for the xgauges, the code scanner, and to test future aero mods.

If you calibrate your scangauge to full P&G technigue it will show correct. I was not able to get correct figures to other/both techniques. So you can calibrate it to what technique you use the most.

Great info. I'll play with it some more once I've got my ultragauge installed and calibrated correctly... maybe I'll see some effects in the end.

I know that on my most hilly trip, which I take a few times a year, I do see a difference if I DWL up the hills and coast neutral down the hills rather than maintaining a target speed... so I'm not saying that techniques like P&G are snake oil, I just have trouble seeing how they can be as effective as some folks say they are unless your transition points make use of terrain so you can convert potential energy into kinetic...

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

I was doing some P&G where the SG reported very optimistic mileage, and when the tank was filled the mileage ended up worse than constant speeds and coasting. That's the main reason I don't do it, other than it annoys other drivers, and nobody needs to do that.

The traffic I drive varies from day to day, so over the course of two weeks between fill-ups my SG tank mileage will be off by +/-5% or more, so calibration from tank to tank is useless. My driving technique is as constant as it can be. If the silly SG could just figure out how much fuel is being consumed a bit more accurately, it would be a very useful tool. As is, not so much.

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.

Nearly the same over here. I use p&g only on slightly downhill parts of the road to extend coasting in neutral (engine on). On a flat road, it seems to be useless with my combination of TDI and large CdA. On steeper declines, I DFCO to stay inside the speed limit.
ScanGauge wise, this seems to work but I have yet to see a record tank while driving with SG. Unfortunately, the cold season is ahead, so a comparison between good summer/autumn tank mileages and using SG over the winter is rather unfair, I guess the real effect (if any) can be seen next spring when conditions get better again.
Furthermore, the gas stations seem to have switched to winter diesel now, especially my uphill fuel usage has risen by at least a litre/100km according to SG, and flat road fuel usage is up a bit too.

so long,

tinduck

In finland there is so little traffic that its not a problem most of the time. I can confirm if there is even one car behind me its not so fun to do it specially if the car behind be dont keep safe distance. Some will notice what I am doing and keep enough distance so it does no harm to them. 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.

Bigger truck are not the same case as small hatchbacks that is sure but on downhills you should see some gains with P&G. Next week I will be in a 60 tonn truck trailer combination to teaching hypermiling techniques so I will learn also myself more about the situation with big trucks.