Measuring the effect of air intake temperature in fuel economy

[Buzo]

Project Description:

The first thing I learned when tuning my car is that we need 60% more fuel to get our cars started/running in winter than what we need in summer. It is due to the cold temperature affecting the air density primarily. (It might be other factors like engine wear, but I'll ignore them for this experiment)

For those cars with the hot air hose disconnected between the manifold and the air filter, because the hose is missing, broken, or there was a modification in the intake system and the hose don't longer fit...

Can the fuel economy be improved just by reconnecting that hose? If so, How much?

Background:

From basic physics, we know the air flow entering our engine is defined by:

Air Flow (g/s) =0.058*MAP(kPa)*RPM/IAT(°K)

Where:
0.058 is a constant from the gas law applied to our 2.0L engine. If you don't believe me, its OK, it is just a constant anyway.
MAP is the manifold air pressure in kPA
RPM is the engine speed in revs per minute
IAT is the intake air temperature

If you see, the IAT is dividing the whole thing, so the most the temperature the less the air flow.

Then based in my datalogs from the whole year, my IAT is at an average of 45°C in February and 65° in August with the hose disconnected.

Applying the formula above, (assuming the engine idling at 1000 RPM and 35 kPa) I get an air flow of 6.0g/s at 45°C and 6.38g/s at 65°C - a 6% of difference.

If I connect the hose, I can rise the intake Air temp up to 100°C.

Re-applying the formula now I get an air flow of 5.44 g/s at 100 °C. - a 15% difference!

Then just because the air flow is 15% less, automatically you should need 15% less fuel flow.

If my fuel economy is currently 26 MPG, I should be able to rise it 15% - to 29.9 mpg just by reconnecting that one hose!

Next Step:

Modify my intake system so I can put the old air filter back along with its automatic air temperature control.
Run the car dataloging the IAT and all other factors, using the same routes and same tuning of course.
Run one tank with the hose connected, and another tank with the hose disconnected.
Repeat as many times as needed (like a mini mith-buster)

Compare fuel economy...
Learn...
Kill some time...
Have fun...

NOTE: for those concerned of the effect of too-hot air under high load, the actuator in the air filter case is vacuum activated, so the engine will get cold air under hard acceleration/load, because there is not vacuum built up at WOT.

Due date:

None! this is just for fun and I don't care if it takes me forever to get a result, or if the result is not as expected.
However, I appreciate this great forum for giving us the opportunity to document what we do to our cars!

[Dr_Snooz]

Curious to see what your results are.

[Buzo]

Things are going faster than I thought. I finished adapting the stock air filter to my TBI today. It was nice to see it back in the car.
I'm actually going to keep it like this regardless of the result of the study and will give it a black re-spray.

So, who's going to help me to recover the ability to put pics in the forum? (big size pics, not thumbnails)

[go2eleven]

I am curious to see these results as well.

I was under the impression that the greater difficulty starting in the winter was more due to the cold affecting things like oil viscosity. I was also under the impression that, driving habits being equal, colder temperature intake air would mean better gas mileage, because the increased density of the air means the engine gets more work per unit volume of air pulled in through the intake.

But I have no science to back this up.

[AccordB20A]

also interested to know. my cars all use more fuel at night. but you can feel its faster at night in the cold too...... im guessing the ecu knows its cold so it adds more fuel to the denser air make more bang and more power.

[2oodoor]

ON my carbed car, the pre-heater removed as well as the electric heater below the carb base, AND I do not have coolant running through the manifold (blocked off).
I am having significant issues with cold air and ambient tempurature causing the fuel to de-atomize under the carb. Fuel is turning back in to liquid and pooling up under the carb on the manifold floor. This really screws with my cold idle mixture until about 20 minutes of driving, even more if the ambient temp is below 46 degrees or so on that day.
The cold affected fuel theory can actually be seen on my car because I can stop the engine and open up the throttle and observe the puddles of gas on the intake floor.

Improvements to this condition on my car would be:
Re-connect the coolant hoses that pass through the intake manifold, this would heat it up some.
Add a spacer or even a swirl spacer under the carb, this would increase the air space and velocity of the atomized fuel to be used before it became liquid again.
Add an air pre-heater of some sort that would allow some hot air in the fresh air intake.

[Buzo]

Thanks all for your comments,

Something is right for both FI and carbed cars, the colder the air the more the power an engine can develop, because more air/fuel can be fit into the same cylinder cavity.

But there are limits like you guys said, too cold air will make difficult the mix of air and fuel and too hot air will lead to a lack of energy.

Once I start using the air heater, I HOPE my car feels less powerful (that means I'd be using less fuel). Remember I am doing this for fuel economy.
My daily trips are through boring streets of the city, a bunch of red lights and speed bumps. I can sacrifice some power I guess.

Got a new air filter and finished painting the filter case. Its not that cold yet here, so while I wait for the winter I am going to be plotting the temp of the air to have a background with the new position of the IAT sensor.

[gp02a0083]

also interested to know. my cars all use more fuel at night. but you can feel its faster at night in the cold too...... im guessing the ecu knows its cold so it adds more fuel to the denser air make more bang and more power.

your absolutely correct B20A, the colder the intake charge = increased density. In a ICE we need to maintain the proper air:fuel ratio to obtain proper operation and performance. This is a common issue that the supercharged nitro methane funny cars and top fuel dragsters contend with and why they typically put down record ET's and MPH records here in englishtown NJ on friday night qualifying sessions rather than Saturday afternoon qualifying sessions. Think of it this way each combustion chamber has about a max volume of 500 cc's. On a N/A engine, a colder denser air charge driven into the cylinder technically weighs more so its allmost like running a hot engine with a larger volume. (Buzo's comment is along the same line here).Thus to keep a 14.7:1 AFR,additional fuel must be injected or introduced into the combustion chamber to maintain this ratio. Feedback to the ECU via the O2 sensor helps maintaining this. FI systems have that advantage over Carbs that they adjust accordingly /quickly in addition to barometric pressure changes. Carbs are crude and require lots of adjustment. Also by introducing a colder intake charge you are minimizing the biggest downfall to a ICE (internal combustion engine), that is the by product of heat. Offsetting this will show some power gains , however if the intake charge is too cold it will run like crap.

2oodoor, i would suggest reconnecting the coolant lines. Even if it was FI, i would suggest doing the same. This would keep the intake air temp more consistent overall and help you tune your carb. the Pre-heater hose you can leave off, i did the same with my 442. I would think at first you may have other issues with the fuel atomizing. To me it sounds like you may not be getting enough fuel pressure and or the jets are not clean, causing the jet pattern to be directed differently.

[Buzo]

All the mechanical work is done.

Finished adapting the filter, put the new filter on and replaced the stock IAT sensor with the GM one.

Test fit over my TBI. The filter case had some time in the direct light of the sun...



There is a hole where the stock IAT sensor is installed, I filled it to the size of the new sensor



New filter



With this project I realized my car needs some brightness, so starting painting all the plastics I can remove from the engine bay.



Here is my car idling and getting hot air from the hose that goes to the manifold.

[Dr_Snooz]

Bro, start using Photobucket. Imageshack always gives those stupid thumbnails for some reason.

Looks pretty fly. You could even pass California emissions looking like that.

[Buzo]

Bro, start using Photobucket. Imageshack always gives those stupid thumbnails for some reason.

Looks pretty fly. You could even pass California emissions looking like that.

Thanks! Actually we do need to have emissions sticker here in mexico too, but it does not include a visual check.
My car recently passed right in the middle of the Max and Min in all the measurables. The tech was looking at the ARF in my clock when he was running the test, I'll say that he got impressed.

Thanks for the tip of photobucket. I already created my account!

[Buzo]

Want to share this initial picture of the normal behavior of my car in a normal morning of February.
It is from 2012-02-16_06.58.36. It was 18°C.

The red line is the RPM, its there so we know when the car is idling, accelerating, cruising, etc.
The white line is the duty cycle, proportional to the fuel injected to the engine.
The blue line is the coolant temp, once the car gets warm, it stays in an average of 85°C
The green one is the Intake Air temp, as you can see it varies with the normal driving, but the average was around 35-40°C.

The goal is to keep the IAT as constant as possible and see if we see a reduction in the duty cycle.

[POS carb]

where is your IAT mounted? On the intake tube, tbi, or intake manifold?

[Buzo]

where is your IAT mounted? On the intake tube, tbi, or intake manifold?

It is mounted right in the TBI.

I was planning to put it where the stock IAT is at (intake tube), but changed my mind and put it in the TBI to be as close to the fuel spray as possible-

[Buzo]

This is the first reading with the hot air hose connected.
It was about 40 minutes trip, (double as the initial image).
Slightly dissapointed that the air temp raised at a very low rate.

[POS carb]

are you using the stock 3geez intake? if you use the hot air intake with the vacuum opener I assume you are using some kind of vacuum thermovalve to open it or are you triggering it with the megasquirt?
Maybe the trap door doesn't seal well or your IAT sensor is getting heat soaked

[Buzo]

Yes, I'm using the stock intake (door/actuator/temp control system). I created a kind of 1" neck over the TBI, then screwed the stock filter case on top of that neck. I sealed the small uneven surfaces using some RVT.

The door actuator gets vacuum directly from the intake manifold, through a check valve and a "T" to the thermo-switch inside the air filter area, which directs the vacuum to the actuator when cold and to the intake when hot. I plugged off the rest of the unused ports in the filter case, but I have to double check everything. I recall this mechanism working very well before, so I assume its something I forgot to plug or something easy.

Its possible the IAT is not "seen" the air flow too, because it got slightly covered with the new filter case...

[Buzo]

It was at 8°C today and was able to measure how the IAT gets hotter slightly quick. I need to compare with my old datalogs in the same temp range to measure the difference.

I noticed the door is making the open/close cycles correctly, but the IAT sensor is getting soaked with the heat of the intake manifold. Once the car is hot, disconnecting the hose makes no big difference in the reading of the sensor.

The questions are, Why Honda put the sensor in the intake tube and GM in the TBI? Which is the real air temp?

[Buzo]

The questions are, Why Honda put the sensor in the intake tube and GM in the TBI? Which is the real air temp?

I think the answers are here.

...they typically put down record ET's and MPH records here in englishtown NJ on friday night qualifying sessions rather than Saturday afternoon qualifying sessions.

and here

...also interested to know. my cars all use more fuel at night. but you can feel its faster at night in the cold too......

The coolant temp is the same at day and night, due to the thermostat and fans controlling it, but still the performance change... Then the cold air of the night is what really makes the difference.

Since I didn't see a real gain with the IAT sensor in the TBI, I changed it to the "honda's way" and installed it up in the intake tube.

[Buzo]

I think I got something good.

Logged some data with the IAT's new location, as you can see now, the IAT (or MAT) follows the door open/close cycles, hot when closed, cold when open.

The colder temp is around 30 C and the hottest is around 65 C.

But I am excited because I found how that white line variable called gammae varies with the MAT. Gammae is a compensations that the MS calculates based in several factors, including the MAT.

In other words, I should be able to save some fuel with the same tune I have now, because that gammae decreases as the IAT temp increases.




And here some background on the gammae from the MegaSquirt site:

What MegaSquirt^® does is take this downloaded REQ_FUEL number and then multiply (or adds) values that scale this number, to come up with the injected pulse width [PW]. Therefore, pulse width is:
PW = REQ_FUEL * VE * MAP * E + accel + Injector_open_time
The "E" above is the multiplied result of all enrichments, like warm-up, after-start, barometer and air temperature correction, closed-loop, etc:

E = gamma_Enrich = (Warmup/100) * (O2_Closed Loop/100) * (AirCorr/100) * (BaroCorr/100)

and

• Warmup is the warm-up enrichment value from the table the user enters in MegaTune,
• O2_Closed Loop is the EGO adjustment based on the EGO sensor feedback and the EGO settings the user enters in MegaTune,
• AirCorr is the adjustment for air density (based on the intake air temperature), and
• BaroCorr is the barometric correction based on the ambient air pressure (usually taken at start-up, but a second baro pressure sensor can be added to MegaSquirt-II for continuous updates to the BaroCorr).

I'm pretty sure OEM cars also have their compensation for IAT, and carbed cars do it also, but automatically, as the air density is reduced so is reduced the fuel pulled through the venturis.

[Buzo]

And this a pic of the neck I built to be able to fit the stock filter in top of the GM TBI, I cut the top off of the GM filter case thing, drilled holes and screwed the honda filter case from the top. Here is also the location of the IAT sensor in a GM vehicle.



We all now where the stock IAT is at



And finally, installed in the car

[cygnus x-1]

IAT is a notoriously difficult phenomenon to deal with. The air temperature changes as it passes through the manifold (warms up), and also as it mixes with the fuel (cools down), so you will get different readings depending on where you take the measurements. The closer you get to the cylinders, the more accurate the temperature will be. However the sensor is also more likely to be affected by heat from the manifold and cylinder head (heat soak). So I think in the end, the OEMs just put the sensor where it's convenient and only use it for minor fuel trimming.


C|

[Buzo]

So far I was able to rise my intake air temp at the same rate as the coolant temp, but once the car is fully warmed up, the air temp remains almost constant regardless of the hose being connected or disconnected, because it is taking the air from the engine bay and not from the passenger's fender bucket. I can actually see a small rise in the IAT when the fans kick in.

I guess the MAP is the one that indirectly reacts to the air temperature once the air/fuel mix passed the throttle, because it measures the pressure which is proportional to the air temperature.

[Buzo]

Monday, I logged the data in my way to work with the hot air system enabled.
Tuesday same trip, same initial temp, and logged the data with the vacuum hose removed (and plugged off) from the hot air door actuator .

Here some highlights:
It takes 15 minutes to get to the max temp (55 C) with the hose disabled.
It takes 10 minutes to get to same 55 C as above with the hose enabled and 11.5 minutes to get to the max temp of 75 C
It takes 10 minutes for the coolant to get fully warm (89 C)
The total trip length is 25 minutes.
The average IAT while cruising is almost the same. 49.7C enabled, 47.48C disabled.

Last night I noticed my car knocking during a couple of hard accelerations I intentionally made (hot air enabled). I never heard a knock with my old set up before. Need to repeat the same tonight but with the hot air disabled and discard if the knocking is due to the change in the air filter system or due to the higher air temp...

[Buzo]

I didn't know I could compare two datalogs, but here I just found how to. Here is how the IAT sensor gets warmer with the time when the hose is connected. It should save some fuel in that ~8 minutes time frame, I just don't know how much and if it worth.