driving style

[Old.Tony]

I sorta think it does give us part of an answer - the rise in fuel usage over RPM is exponential, not linear.

I couldn't get a graph of the info I wanted over time - I'm working on that. I want speed + RPM + fuel vs time, and then overlay two graphs on each other - one for light throttle, one for heavy.

 

[KraftyPg]

But even a blind man with a deaf seeing eye dog could come to the conclusion that more RPM = more fuel use. What is not being shown is whether there is a certain figure at which taking off slower becomes less efficient than taking off faster or vice versa.

 

[Old.Tony]

Yeah I know that's not being shown - unfortunately the software I have doesn't let me gather that data. I'm writing to the author and seeing if he can add the feature or point me at the menu item that I've missed. I thought it was interesting that the RPM vs Fuel wasn't a linear progression, but an exponential one.

 

[mgermasi]

IMO to be on par with what we are trying to find out the graph should include either speed or distance traveled.
Also very important is the fact that I'm talking about a AUTOMATIC transmission where you do not have control of which gear to select.
I noticed that with a very heavy auto the car takes let's say:

)- 1 Km @ 2300 RPM to reach the speed of 100Km/h and then I can back it off to 1900RPM
Similar effect or worst for a speed of 80 Km/h

as opposed to:

)- 500 m @ 3000RPM to reach the speed of 100Km/h and then I can back it off to 1900RPM
To get to 80Km/h even shorter distance is required

I will confirm distances in the next few days but this is more or less what I was trying to say/ask

Thank you guys for your thoughts

 

[Old.Tony]

I'd still need to graph two things against time, and I can't. The software only allows one line graph against time, or two items in a scatter chart like the one I produced above. Let's see what the software author can do for us.

 

[mgermasi]

Yeah I know that's not being shown - unfortunately the software I have doesn't let me gather that data. I'm writing to the author and seeing if he can add the feature or point me at the menu item that I've missed. I thought it was interesting that the RPM vs Fuel wasn't a linear progression, but an exponential one.

It is very interesting that graph Tony, also because I was some how expecting to see the fuel flow rate to actually decrease.
I always thought that the mixture should get leaner as the RPM increase.
I was wrong!

 

[Old.Tony]

Being a scatter graph it's actually difficult to spot. Fuel flow rates definitely increase dramatically as RPM rises. For example, the test I conducted with the engine @ 4800rpm for the 2-stroke thread saw fuel flow rates up towards 50 litres per hour.

I am now waiting for a response from the develop of Torque. I've explained what we're trying to get and told him that even if it was just recorded as CSV data it'd be good enough for our purposes. Fingers crossed!

 

[Old.Tony]

I have the answer, I've added the data for logging, I've taken the car out and have TWO results worth noting.

I logged engine RPM, GPS speed in km/h and Fuel Flow Rate (LPH) against time. I did two runs on a level road, the first keeping the rpm around 2000 and the second using up to around 3000rpm.

With the log results in CSV I imported this into Excel (actually Open Office) and did a couple of calculations. I then calculated fuel flow rate times 1000 (to make it in ml) times the number of seconds from the last reading divided by 3600 (seconds per hour) which gives me the number of millilitres of fuel consumed in that time period. It's really a stab - it's looking at the fuel flow rate at the moment the data is collected which is cyclic as the program goes through all the OBD PIDs grabbing data over the bluetooth connection, but it's not too bad and a hell of a lot better than what we had before!

So, I accelerated gently - peak RPM was 2042.75 - to 48km/h which took 32.33 seconds and consumed 82.86ml of diesel. Peak fuel flow rate was 12.5LPH.

I then turned the car around and accelerated harder, peak RPM was 3138.5 - to 53.13km/h which took 9.47 seconds and consumed 32.06ml of diesel. Peak fuel flow rate was 35.4LPH.

I'd like to do several runs to gather more data so that I can get some median values. This would remove the error of the pinpoint rates being off-norm. I'd also like to make sure they're all performed under near identical conditions - so I'd like to take off from the same stop sign travelling down the same road at different rates.

So at the moment, it looks like a MODERATE burst is actually better than a sustained slow wind-up. Significantly, too - but it's early days, this was one reasonably unsteady drive up the street and back with other traffic playing a minor role.

I'd love to log some data from a standing start up to freeway speeds so I can see how things go as torque converters lock, torque levels rise, wind resistance begins to play a role etc etc.

I have a new toy to play with!

 

[mgermasi]

Hey Tony I really like to read your researches, you put 100% of yourself at them.
Very well done!
As I was kind of suspecting when the car is taking off harder uses about 33% more RPM but for just a third of the time I just could workout if it was or not the best way to drive it.
It would be even more interesting to see if the same conditions are applicable when towing because I believe in that case you'll also have a lot of torque converter slippage.
Thanks for your time in testing out this values for us.
Cheers Max

 

[bods]

Hey Tony I really like to read your researches, you put 100% of yourself at them.
Very well done!
As I was kind of suspecting when the car is taking off harder uses about 33% more RPM but for just a third of the time I just could workout if it was or not the best way to drive it.
It would be even more interesting to see if the same conditions are applicable when towing because I believe in that case you'll also have a lot of torque converter slippage.
Thanks for your time in testing out this values for us.
Cheers Max

As far as the torque converter goes playing a part, you would need to know the stall speed of the converter. If you can find that out an keep the rpm around that point, you will have no or very little slippage which should give you better economy because you're not wasting fuel for no acceleration benefit

 

[nakedape]

Once you have enough data you should be able to build a decent model.

 

[KraftyPg]

Miranda? Elle? Jen?

I'll take one of those if someone is building one.

 

[Pete22]

Tony - seems that some folks improve fuel consumption in an automatic by drivig it in manual mode. Would be interesting to see if this approach works with the D40

http://www.fuelly.com/tip/571/Shifting-Automatic-Transmissions-manually

 

[Old.Tony]

Yeah they're talking tiptronics - autos that you can manually shift higher and the gearbox will obey - ours won't.

I like the idea of the stall speed. Our stall speed is supposed to be between 2700rpm to 3100rpm according to the manual. The replacement TC from Wholesale Automatics lowers that to 2400rpm.

This might actually explain why the harder acceleration had such good results. I will be doing some more testing soon.

 

[mgermasi]

Wholesale wants around $2K fitted = around 1300 liters of diesel @ $1.49/l How long does take to save them?
Is it better off to get the torque converter from the stx550? For sure it must have a lower stall speed the car has almost 60% more torque than mine.
What do you reckon?

 

[Old.Tony]

I bet the 550 TC won't fit the 4-cyl gearbox. They are very different boxes, and my money's on Nissan making a small enough change that there's no way you'll get one into the other.

You'd be right though. With the 550 doing about 300rpm less in top gear to achieve the same speed, the torque range is shifted down enough to warrant a lower stall speed on the TC so investigating it is a great idea. I wonder if the TCC operation in the 550 is the same?

(Note: I am deliberately baiting the '550 TC won't fit' issue because with my luck, it WILL and I'll lose the bet, which I am actually counting on because then I'd consider a 550 TC for my own car!).

 

[mgermasi]

I'll try to get to nisswreck next week (no commitment) and see if they have both and measure them up.
As far as your bet goes don't worry I'm bribale you can just PM me an offer, LOL!

Cheers