Recovering lost heat energy in engine coolant...
 

In other words, a coolant-based hot air intake.

I have removed the cold-air intake tube from my truck, and have re-installed the stock induction setup. I purchased the smallest transmission oil cooler I could find, and have mounted it to the filter box, after the filter element. This will be the intake heater. I decided against placing the intake heater before the filter element, as I do not think that the intake heater should serve as a pre-filter, nor do I think that the intake element would like being heated up to well above ambient.

The coolant lines have been routed out of the intake box, but they have not yet been connected to the coolant system. I plan to use a 4-way heater valve to connect the intake heater to the supply heater hose. When disengaged, the heater supply will go directly to the heater core as normal. When engaged, the heater supply will be re-routed to go through the intake heater first, then go to the heater core as before. The heater valve is controlled by engine vacuum - I will use a vacuum solenoid, controlled from the cabin, to switch the intake heater on or off.

The above setup is for demonstration purposes; once I have seen a satisfactory A-B-A result set, I will probably do a Y branch instead, in order to have full heat capacity to both the heater core and the intake heater.

I plan to have the hoses and control circuit hooked up tomorrow, prior to my work commute.

This is really creative, but are there any advantages over pulling in warm air from exhaust side area? I guess if nothing else ur bay will look more clean and stock.

It's a lot easier to route a couple of heater hoses, than it would be to fabricate something that could take usable heat off my ceramic coated headers. Besides, with the way my stock induction system is set up, I think it'd be next to impossible to route hot exhaust-heated air from the left bank and still have it be kept warm enough to do anything meaningful. The filter sits on the front of the right fender.

I also consider coolant to be a more stable source of heat.

"stable source"

best I can tell, the constant temp is the key to further tuning.

If you want to further control the airbox heat, you can use a thermal switch and an electronic valve. Place the thermal switch somewhere in the intake tubing to monitor air temps, wire it to a controller on the valve you already have in place. In this way, the valve will shut off at X temp, and open back up below X temp, keeping a close range of temps in the intake tract. Different switches will allow you to fine tune for the best IAT for FE.

I had a similar idea quite awhile ago, wherein I'd use a small heater core mounted in the airbox. I never got around to doing it, but at least someone is... if it works for you, I might also give it a shot.

I'm a bit puzzled why you would want hot air in the intake.

The adiabatic expansion implies the colder the air you can use, the more delta T you get.

The greater the delta T, the more effiicent the 4 stroke engine?

I guess I'm curious why you would want a hot air intake.

Here's my solution:

http://i12.photobucket.com/albums/a2...n/IMG_0883.jpg

The stock filter sits off to the driver's side more, in front of the battery. It also pulls cool air from the fender well and is blocked off from the exhaust heat. To build the one in the pic, I simply purchased a cheap "racing" intake, cut the pipe and rotated it 180 deg.

Next step is using a exhaust cutout, an rpm window switch, and a temp sensor with a Y that goes to cold air as well. Above a certain temp, cold air will be let in, and also above a certain RPM. For now, this seems to work just fine, though the loss of power is quite noticeable.

Nice setup!

Do you have temperature measurements from inside your intake manifold?

No, only temp sensor is the one right behind the filter. I don't even have any instrumentation.

There are some well-documented setups on Saturnfans.com that are much closer to the exhaust manifold and produce higher temp air. Apparently the point at which the stock pcm starts to retard timing is quite high for most operation (around 150+ F). I may borrow my roommates scangauge/OBD2 reader/whatever to verify I'm not in the crazy temp range, but I'm in no rush.

Unfortunately, there is really no reliable way to A-B-A test anything around here since the nearest traffic free section of road is probably over an hour away, unless I feel like doing my testing at 3:00AM. I'm happy with just logging my mileage as usual for now - after all, its gas usage that I'm really after. Even if its a placebo that makes me subconsciously drive better... I'll take it :)

Your post did make me think about the intake manifold a bit... the stock manifold does have a coolant line coming into/out of it. It would be quite easy to add another heat exchanger in-line to extract more of this energy if I so desired. Although, as I pointed out above, the only intake air temp sensor is near the filter, so the computer compensation wouldn't be as accurate.

The intake coolant line is there to provide warmth to the intake air and prevent sonic freeze at high air velocities. It doesn't do much, but removing it creates a traceable increase in power when tuning. At least it did on my Civic, 3hp avg over 5 runs on a chassis dyno at 3k rpm. I never had a problem with sonic freeze, either.

I am lucky in this regard. My nightly return commute is usually pretty stable. The highway's pretty empty, and it's really easy to just set the cruise to 72 and do a good 25 mile stretch without even having to change lanes. That takes out a good number of variables, right there. Of course, outside temperature and humidity still factor in...

When I get this thing working tomorrow, I am going to monitor throttle position and manifold air pressure, in addition to intake air temperature.

I wonder if the engine will make enough heat to get to proper running temp during cold weather if you are using the intake heater and the cabin heater.

The heat used by the intake heater goes right back into the engine, so should have no effect on warmup time. Theoretically.

On the other hand, an intake heater that uses exhaust manifold heat should theoretically warm up faster.

Somebody want to run some tests of warmup time?

It'll warm up faster with a core in the intake, too. Using warmer air changes the ratio of heat dispersion from the cylinder into the cold air that's taken in. If the air is already warm, no heat is lost from the engine to warm up the air.

Thereby, decreasing warmup times, albeit slightly.

Coolant, vacuum, and electrical work were completed this afternoon. I temporarily re-wired my foglamp switch to control the vacuum solenoid to open and close the 4-way heater valve.

Initial observations:
Outside temperature: 78 F
Intake air temp with no heat: 90 F
Intake air temp with heat: 115 F

When switched on, the air temperature takes about 30 seconds to stabilize at the higher value. When switched off, the air temperature takes several minutes to stabilize at the lower value.

I think that's a good start, considering that the intake heater only covers about 70% of the air filter area. Air is free to flow around the remaining 30% uncovered area, without getting heated. Fuel economy observations will be conducted during the next run.

Looking forward to results.

Warm air is less dense and has less oxygen molecules therefore less oxygen in the exhaust stream for the 02 sensor to detect and the IAT sensor sends a warmer temperature reading to the ECU, which fools the ECU into thinking the outdoor temperature is hotter than it actually is and leans out the mixture. All cars run richer when cold air is being introduced into the engine. Many have found that warmer IAT's increase their mileage.

They don't run richer, they just add more fuel to compensate for the presence of more available air. "Richer" suggests an AFR different than desirable, as in "too much fuel for available air mass".

ummmmm,,,,, well, no.

it takes a certain amount of oxygen molecules to make a certain amount of power. If the molecules are warmer, they take up more space, so you have to give more throttle position to get the same amount of power.

But when done, it still takes the same number of molecules to make a certain amount of power.

If the molecules were colder to begin, expansion ratio would be higher, making the engine more "efficient" = more power with identical amounts of fuel and air.

Ehh??
CAI or WAI? Which one!
(The age old topic that never gets any solid data for or against).

Okay, the results are in...

Distance travelled: 58.7 miles (94.5 km)

Yesterday:
Outside temperature: 58 F (14.4 C)
Intake temperature: (not recorded)
Fuel consumed: 3.18 gallons (12.02 L)
FE: 18.5 MPG (12.7 L/100km)

Today:
Outside temperature: 55 F (12.8 C)
Intake temperature: 98 F (36.7 C)
Delta T: 43 F (23.9 C)
Fuel consumed: 3.04 gallons (11.49 L)
FE: 19.3 MPG (12.2 L/100km)

So, about a 4.4% improvement in fuel economy.

Keep in mind that this is not a strict A-B-A test. For one, the intake heater was not connected to the cooling system last night. For two, there were two stretches of highway that were under late-night construction last night, causing me to slow down for the better part of a mile.

Tomorrow, I am going to block off the uncovered parts of the airbox roof (where the intake heater is mounted), so as to force all of the intake air through the intake heater. Hopefully, this will be able to cause intake temperature to reach 130 F during the nighttime (return) commute. In the slightly longer term, I am going to find a more efficient transmission cooler (possibly one of them plate-and-fin types) and use that, instead of the tube-and-fin cooler I currently have installed.

No. The mixture stays at 14.7 (or however close your factory ECU can hold it to this number). It takes a mass X of air to make Y power. The warmer the air, the higher the volume. The higher the volume the wider you need to open the throttle. Higher throttle position = less pumping losses.

So the throttle plate is the 'largest' restriction?

which means it is all mouse nuts.

(too small to matter unless you are really into examining mouse nuts).

statistically, this is about as valid as the results when measuring fuel magnets.

one test. not aba, and you don't even know the temperatures involved.

Pretty much, yah. The intake poppet valves are a secondary restriction, but while the engine is heavily throttled, their restrictions are quite small as compared to the throttle plate.

Drove the family up to Niagara Falls, Ontario (!). Along the way, gathered a few tank fills of data. Again, not strictly A-B-A, but the results are nonetheless telling.

May 11 - 363.8 miles - 19.113 gallons - 55.4 MPH average speed - 19.0 MPG (12.360 L/100km)
May 12 - 410.4 miles - 20.517 gallons - 57.0 MPH average speed - 20.0 MPG (11.759 L/100km)

That's a 4.9% increase.

Unfortunately, I did not have the time to install the blocking plates that I had wanted to make.

The intake heater, as installed, appears to give a reliable 45 F delta T between the outside air temperature (measured by the overhead display) and the intake air temperature (measured by ScanGauge II), when the truck is moving level at 72 MPH. Inclines and changes in vehicle speed change this delta T.

Ummmmmm, yes. You answer the question yourself.

More throttle = lower pumping loss = greater efficiency.

Is it a huge change?

No, but, apparently it is measurable and eventually enough "mouse nuts" add up to a substantial difference.

Part Load Pumping Losses in an SI Engine

This website page gives a pretty concise explanation about pumping losses in a gasoline engine. Well worth a read.

Changes of Low Load Engine Parameters by Temperature of Mixture

This technical paper goes over the effects of installing a hot air intake onto a test engine driving a small load. For purposes of discussion here, I have considered only the testing temperature of 90 C (194 F). I have not considered the testing temperatures of 140 C (284 F) and 190 C (374 F) because they are a bit higher than is reasonable. I want to save fuel, not bake things.

For 90 C, the paper mentions that pressure inside the intake manifold rises by a small but measurable amount, which supports the pumping loss reduction idea.

However, the pressure rise (about 2%) is a bit less than the 7% reduction in BSFC seen in the paper. The paper further shows a negligible increase in throttle position. I have seen this effect as well on my ScanGauge - whether going up or down inclines, or even on straight and level roadways, the truck's manifold pressure and throttle position do not appear to have noticeably changed. I used to think that HAI gains were solely from lessening pumping losses. The smaller-than-expected pressure rise and the negligible throttle opening do not really support this idea.

The paper also shows warmer air leading to faster combustion rates, and the effect is certainly seen in the graphs at the end of the paper. Combustion peak pressure appears to move leftward from baseline, which is equivalent to advancing ignition timing by a few degress. This makes sense. In fact, the paper goes on to infer that faster combustion rates are the primary factor in HAI gains.

I think that this paper is showing that the resulting gain in fuel economy from installing a HAI is a combination of the two factors. In any case, the gains I've seen are real, and they merit further investigation.

Edit: Hey! That might could explain why MAF-based cars don't really see any sort of improvement using a HAI. Their engine computers might be overcompensating for higher intake temperatures.

Pictures?

Intake heater element
http://www.tom-viki.com/spgm/gal/Car...1205130002.jpg

Hose connections to air cleaner roof
http://www.tom-viki.com/spgm/gal/Car...1205130001.jpg

4-way heater valve and vacuum solenoid
http://www.tom-viki.com/spgm/gal/Car...1205130000.jpg

I have made a vehicle entry in the garage here, "Dakota A-B-A HAI", specifically to capture trip data related to my hot air intake. I have written down the rules for trip acquisition, which happen to be based on how I would normally drive home. These rules further restrict trip acquisition based on events that would affect fuel consumption (rainy or windy weather, delays due to road construction, elephant races, etc.) The rules below should minimize most of the variables that could significantly affect trip data.

-------------------------------------------------

Each fill-up entry represents one commute from work (or co-worker's house) to home. Three different paths are considered.

• Route A - Work to I-480E to I-77S to home (58.7 miles)
• Route B - Work to I-480E to I-271S to SR 8 to co-worker's house to SR 8 to I-77S to home (61.8 miles)
• Route C - Co-worker's house to SR 8 to I-77S to home (26.5 miles)

Temperature snapshot will be taken at I-77S exit 109, where there is a large sign present that alternates between outside air temperature and the time. The display temperature has been verified to read within one degree F of the vehicle's outside air temperature reading on it's overhead display. The installed ScanGauge II shall be used to record the vehicle's intake air temperature.

Fuel consumption will be recorded with installed ScanGauge II.

All entries are generated between 12:30 AM and 2:00 AM, where there is little traffic present. Each trip begins at work (or co-worker's house), and ends at home.

Vehicle will have sat at work (or co-worker's house) for at least 8 hours prior to trip.

Vehicle does not have any aerodynamic mods, save for a chin spoiler.

The stock cruise control is used, to the maximum extent possible. Speeds to be kept to 7 MPH over the posted speed limit while the vehicle is on the highway, and shall be kept at the speed limit when the vehicle is off the highway. Lane changes to be kept to a minimum. Any attempts to pass slower vehicles to be made only with cruise control set at speed, with no additional acceleration, and by using lane changes as necessary.

Vehicle air conditioning will not be used.

Vehicle to make interim stops only at traffic lights, toll booths, the gas station, and at co-worker's house.

The following will invalidate a trip from being recorded: Road construction along route that spans over 0.5 mile. Delays that reduce vehicle speed to below 50 MPH on the highway. Rainy weather. Wet highway roads. Noticeable winds with windspeeds over 5 MPH, as verified by weather website. Stops made for any purpose other than described above. Driving vehicle for any reason during 8-hour work period. Use of vehicle air conditioning.

YES!!
Finally, an ABABABABABAB test of HAI under controlled conditions.

Done, a few years ago. :)

http://ecomodder.com/forum/showthrea...-up-11471.html

Okay, some good news and some bad news:

First, the good news: I could do a run with the HAI turned off.

Route C Data Set (26.5 mile distance)

Label	Date/Time Stamp	Consumed	FE	Outside Temp	Intake Temp	HAI Status	%Change
B	2012-05-16 1:40 AM	1.40 gal	18.93 MPG	62 F	104 F	Active
A	2012-05-17 1:47 AM	1.49 gal	17.79 MPG	42 F	58 F	Off	-6.04%

This B-A test set shows a decrease of 6.04% from B to A. (and apparently we can use table bbcodes... cool!)

Now, the bad news: The heater valve, used to control the intake heater, allows a little bit of hot engine coolant to the intake heater when the valve is switched off. The supply line was warm to the touch at the end of the test route, but the return line was cool. This caused the warmer-than-expected intake air temperature observed during this night's run. If pure results are desired (HAI turned absolutely off), then I might have to invest in some sort of globe valve to install in-line with the heater supply.

Route A Data Set (58.7 mile distance)

Label	Date/Time Stamp	Consumed	FE	Outside Temp	Intake Temp	HAI Status	%Change
B	2012-05-12 1:40 AM	3.04 gal	19.31 MPG	55 F	98 F	Active
A	2012-05-18 1:27 AM	3.12 gal	18.81 MPG	49 F	68 F	Off	-2.56%

Another B-A set, though I don't think I will be doing any more for a bit.

As can be seen, there's a small difference between the 58.7 mile run with the HAI active, and with the HAI turned "off." I put "off" in quotes because there's still a 19 F difference between outside air temp and intake temp. Apparently, that heater valve leaks more than I had thought. I must put in a heater valve that doesn't leak when closed (preferably one that can be electrically controlled from the cab).

It would be great if you could get a few more runs in with the same ambient temperature.

Yes, yes, it would. If only this weirdo weather will cooperate... One night it's sultry, the very next night it's practically near freezing.

In the meantime, I will try to find a heater valve that will actually, you know, shut off flow. I think that will also help.

Route A Data Set (58.7 mile distance)

Label	Date/Time Stamp	Consumed	FE	Outside Temp	Intake Temp	HAI Status
B	2012-05-12 1:40 AM	3.04 gal	19.31 MPG	55 F	98 F	Active
A	2012-05-18 1:27 AM	3.12 gal	18.81 MPG	49 F	68 F	Off
B	2012-05-19 1:33 AM	2.99 gal	19.63 MPG	58 F	104 F	Active

Hopefully, this weekend I'll be able to install a heater valve that doesn't leak. If so, I'll throw out the "A" data gathered so far, and recollect more "A" trips.

This sounds like something ill have to try.

More good/bad/indifferent news...

Good news: I redid the heater hose plumbing, replacing the heater bypass valves that I was using, with heater shutoff valves that are specified for a late model Chrysler K-car. I tested them before installation, and they do appear to completely shut off coolant flow when they close. So, perhaps I can get better A-B-A data.

http://www.tom-viki.com/spgm/gal/Car...1205190000.jpg

Along with that, I replaced the tube-and-fin heater element with a plate-and-fin element. Instead of a 45 F differential between outside and intake air temperatures, there is now about a 60 F differential. It reflects favorably on the test drive, too. On an all-city driving test run of about 8 miles (between my home and my parent's house), I am estimating a 15% improvement in fuel economy. (That's all in-town miles, there. My commutes will doubtlessly show a smaller increase than 15%.)

The indifferent news: I did not test the heater shutoff valve, because I wanted to do a leak test of the reworked heater hose plumbing.

Now, the bad news: All that A-B-A test data above? Say buh-bye to it... I'm starting over from scratch.