Fuel Economy and Smoothness without Compromise

So I've already got a self-mapping fuel tuning system, and I even have it setup with an economy map that I can switch to while riding, but I haven't been completely happy with it. My economy map is still a compromise. I have to be careful about what areas I lean out and how lean I go, because there is some overlap in areas of the map used for steady cruising and for light acceleration. I can have awesome cruising fuel economy, but stress about whether the bike will hesitate and dump me on the ground mid-corner at an intersection. I found that I only really use the economy mode if I'm going to be on the freeway for a long time, so I'm missing out on a lot of fuel economy opportunity any other time that I'm steadily crusing for shorter periods of time.

What I really need is something like the stock closed-loop mode. I can't use the stock closed-loop mode because I have a wide-band O2 sensor in place of the stock sensor (used for self tuning). My bike is always in a closed-loop mode, but targeting varying customized fuel mixtures as opposed to the stock closed-loop mode targeting a specific lean (14.7:1) mixture only while steadily cruising. Adding the stock sensor back into the exhaust along with the wide-band sensor would just wreak havoc on the self-tuning system.

I thought my only option was to try to convince the maker of my fuel tuning system to add a cruising fuel economy mode (my efforts failed). That is until recently when I started learning to program microcontrollers (for my kid's Power Wheels upgrade).

My solution: program a microcontroller to monitor the throttle position and engine speed, determine when I'm steadily cruising, and automatically switch between the 2 maps on my fuel tuning system rather than requiring me to press a button to switch. Now I'll be able to lean out my economy map exactly where and how I want it for steady cruising only. I'll use less fuel because I'll automatically be in economy mode whenever I'm cruising, but I won't get any lean hesitation when accelerating.

I have an initial version of the code written already, but I'm waiting for the microcontroller to be delivered before I can put the hardware together, wire it up to the bike, and start testing. I'll post more details as I make progress, and pictures when I get something done beside writing code (I guess I could upload the source code if anyone is really dying to see it).

Like I said,"I'm not in your league!" :blink:

Good luck and keep us informed! :yahoo:

Very cool, are you kind of taking the stock system's theory of operation and taking it to another level as far as closed loop leanness when cruising, and then switching back to your acceleration map when you start to vary the throttle more? What kind of mileage figures do you get between the different maps / riding? Good luck.

Cool. And, you have too much time on your hands!

That's very cool.

I wish I had seen your eariler threads before buying a PCIII. Thanks for taking the time to keep posting your fuel mapping journey.

Joe

I don't really follow a whole lot of what you're saying, but are you talking about one of these?

https://www.youtube.com/watch?v=SjxfAjAyp5k

carlson_mn said:

Very cool, are you kind of taking the stock system's theory of operation and taking it to another level as far as closed loop leanness when cruising, and then switching back to your acceleration map when you start to vary the throttle more? What kind of mileage figures do you get between the different maps / riding? Good luck.

Click to expand...

That's the basic plan. I'll be sampling the throttle position and RPM about 16 times a second, looking at the average change in both over the past couple readings, and considering it to be steady only if the rate of change in throttle and RPM are below a certain threshold. If it remains steady for a certain period of time, I then consider it to be cruising, and switch to the cruising map. The major difference is that I have an entire steady cruising map to customize the cruising AFR differently at different RPMs and throttle positions. The stock system simply avoids going into closed-loop cruising mode if RPMs and throttle position are not within a certain defined cruising range, and when it is in closed-loop mode, it can only target 14.7:1 AFR.

I did a comparison between my two maps one time, and I think the results were something like 36mpg vs. 42mpg. I might get more of an improvement after this project is complete because I'll be able to lean my cruising map out more, because it'll be guaranteed to only be used while cruising.

joebiodiesel said:

I wish I had seen your eariler threads before buying a PCIII.

Click to expand...

This Motty AFR Tuner is not plug-and-play. If you're not comfortable with splicing about 15 wires into the bike's ECU wiring harness, then you would have chosen the PCIII or PCV anyway The PCV has the ability to store 2 maps and switch between them, so my creation would probably work with a PCV if you were able to get a custom map tuned for pure cruising economy. The PCIII could also take advantage of this, but you need to purchase the expensive LCD screen add-on to be able to store and switch between 2 maps. The difference is that it would be an abrupt change to the cruising map. Mine will be setup so that the base fuel injector map of the cruising map will be a copy of that from my smoothness map. When switching to the economy map, the closed-loop system will take a few seconds to adjust the injector map values to reach the target AFR, so it will be a smooth transition.

Then again, with a PCIII or PCV without AutoTune, you can just keep the stock O2 sensor plugged in for stock cruising fuel economy.

lharrell79 said:

I don't really follow a whole lot of what you're saying, but are you talking about one of these?

https://www.youtube.com/watch?v=SjxfAjAyp5k

Click to expand...

It's basically a smaller version of that.

If you can manage it, a third data channel that monitors intake vacuum would be a powerful tool to determine how hard the engine is working.

I have ridden my FJR with a vacuum gauge monitoring the intake vacuum, DMM Velcroed to the handlebar monitoring the TPS voltage and my eye on the tach and it was highly educational. One thing that I learned right away about my fuel economy was that my running low rpms was actually harmful to economy. The engine was loaded by low rpm and high gearing, forcing me to hold the throttle open wider to maintain speed. By going to a lower gear the engine moved into a better power band resulting in a significantly smaller throttle openings. When I had low rpms and wider throttle openings the vacuum gauge would show low vacuum (closer to atmosphere) and by snicking down one gear the vacuum would noticeably increase (indicator of economy) and the throttle was less open so there was less fuel being used. The dynamic range of vacuum is ~10-12 in/Hg at idle at sea level and ~1-2 in/Hg at WOT.

FWIW per an earlier conversation, I did reconnect my o2 sensor with my PC3, zeroed the map @ 20% and below throttle since I had no low end issues stock. Getting fine gas mileage for my style of riding (40mpg), mostly hard sport riding. 45mpg on a highway cruise.

ionbeam said:

If you can manage it, a third data channel that monitors intake vacuum would be a powerful tool to determine how hard the engine is working.

Click to expand...

I could probably manage getting the signal. There's already a vacuum sensor on the bike, so I would just tap into the wire where it goes to the ECU. The problem would be deciding how to interpret that signal and how it would influence the decision of whether I am steadily cruising or not.

ionbeam said:

I have ridden my FJR with a vacuum gauge monitoring the intake vacuum, DMM Velcroed to the handlebar monitoring the TPS voltage and my eye on the tach and it was highly educational. One thing that I learned right away about my fuel economy was that my running low rpms was actually harmful to economy...

Click to expand...

Smaller throttle opening does not equal better fuel mileage. As you pointed out, there is more vacuum at lower throttle opening, which means more of the engine's power is lost to overcoming that vacuum. Cruising at a lower RPM below the power band is more efficient because of this. Best fuel economy should be found in the highest gear that you can cruise in at a given speed without lugging the engine, because you will have the throttle open more and the engine can suck air into the intake more easily. For example, some cars run a coolant line through the air box (hot coolant from the engine) to purposely warm up the intake air so that it is less dense and requires the driver to give more throttle to maintain speed. This is a cheap trick to improve fuel economy at the loss of torque/power at full throttle.

UselessPickles said:

Smaller throttle opening does not equal better fuel mileage. As you pointed out, there is more vacuum at lower throttle opening, which means more of the engine's power is lost to overcoming that vacuum.

Click to expand...

From a Franz Hofmann paper:

"The single biggest robber of engine efficiency is the simple throttle! As an example, in a large 7 liter engine, each cylinder is almost one liter in size. With the throttle almost closed, (and with the engine intake manifold vacuum at 18" HG, or approximately 62% of absolute vacuum or zero psi absolute pressure) imagine trying to lift a plunger 4". You would be lifting 208 lbs.! (Assume a 4.25" bore, equaling 14.17 square inches, times 14.7 lbs/sq/in air pressure at sea level.)

Take this one step further, and multiply it times the number of cylinders, and you have some kind of idea how much power it takes to just spin an engine over without any combustion process taking place!"

This is the idea behind the BMW automobile's "Valvetronic" variable valve lift, used in place of the standard throttle plates, to control intake air volume.

My understanding is that it is supposed to minimize the pumping losses of the high intake vacuum.

I got all the components soldered to a board:





Not too bad for my first time

It's pretty basic:

• A prototype board from Radio Shack so I can solder everything down nicely without going through the effort of designing a printed circuit board and paying for one to be made.
  
• Red and black wires are +12v (from the bike) and ground (from bike).
  
• A couple capacitors and a 5v regulator on the left to provide consistent 5v to the microcontroller.
  
• That's a socket for an 18-pin chip in the middle. I can easily replace the chip if I accidentally fry it with static shock. It also allowed me to work on the board without needing to be excessively careful about static. The chip I'm using only has 14 pins, so 4 of those pins on the socket will go to waste (it's what I had already).
  
• A very small ceramic capacitor to the right of the socket smooths out high frequency fluctuations to voltage right at the point that it is supplied to the chip.
  
• The grey ribbon of wires is for loading the program onto the chip without needing to remove the chip from this board and place it into a programming board. I can also use those wires to debug the program while it's running in the hardware (pause the program, look at values in memory, etc).
  
• Blue wire will tap into the camshaft position sensor wire to keep track of RPMs.
  
• Green wire will tap into the throttle position sensor wire.
  
• Yellow wire will connect to my button that currently serves as my map switch button. The button will now allow me to disable the cruise detecting and force it to stay in fuel-wasting smoothness mode.
  
• White wire is the output signal to my Motty AFR Tuner that tells it when to switch maps.
  

I still have probably another evening worth of work to do before I can find out if I soldered everything correctly. Hopefully I'll have more progress to report tomorrow night.

More progress!

I got a 6-pin connector on the wires coming off the board:



And trimmed the board and slightly modified a small project box from Radio Shack for a nice clean package:







Now I just need to splice those wires into some of the wires on my Motty AFR Tuner, pop a microcontroller into the socket on the board, load my program onto it and start testing.

I was able to confirm that the circuit on the board seems to be correct. My wife bravely held the positive and negative wires against a 9v battery while I tested several points on the board with a multimeter to confirm that 5v was being supplied in all the right places. I think I might be in the clear with the hardware side of things, so hopefully any problems I have will simply be software bugs that just need some tweaking of the code.

UselessPickles said:

The PCIII could also take advantage of this, but you need to purchase the expensive LCD screen add-on to be able to store and switch between 2 maps. The difference is that it would be an abrupt change to the cruising map.

Click to expand...

Close but no cigar. The expensive LCD display is not needed .. but a hub and switch is. I think I paid about $120 for that.

The switching between maps is immediate, but I haven't noticed anything abrupt about it.

My intention was to do just what you are doing, keeping the zero map for good economy but a smoothness or performance map available too (I am OK with manually switching it). The switch does that. But the problem I've found is that the zero map has a lean surge so when I want to cruise and save fuel the speed varies all the time and royally pisses me off (and the folks behind me) so I never use it.

SkooterG said:

Cool. And, you have too much time on your hands!

Click to expand...

Jeff is one fart smucker! I'm continually amazed with his antics

Useless Pickles, you could use the PC V with autotune and select gear mapping, allowing for a lean map in higher gears or particular rev range and normal map for lower gears for crisp throttle.

Andy

Andyfjr said:

Useless Pickles, you could use the PC V with autotune and select gear mapping, allowing for a lean map in higher gears or particular rev range and normal map for lower gears for crisp throttle.

Andy

Click to expand...

That would still be a major compromise, because I wouldn't be able to fully lean out the higher gears for cruising purposes only. I'd have to leave it rich enough for acceleration too. Besides, there's no way I'm switching to the PCV with AutoTune after what I went through to install the Motty AFR Tuner:



I'd also be giving up the data logger if I made the switch, and I'd be giving up the ability to have self-tuning AND switchable maps at the same time (unless the PCV has had a firmware upgrade that changes this, the digital input for switching maps automatically becomes an on/off switch for the AutoTune when AutoTune is used).