Switched Hot/Cold Air Intake - Gas Savers

rh77 · 01-21-2007, 06:14 PM

OK -- here's the premise. After messing around with hot-air intakes, warm-air intakes, and cold-air intakes, it turns out that none of the above will work for my application. A combination is needed with the radical changes in ambient temperature experienced in this region.

Background Testing: Using the 1998 Acura Integra LS Automatic test vehicle, seperate tests have shown that at the same speed, road and wind conditions, 50F air at 20% TPS results in 25 mpg, 90F at 20% TPS yields 33 mpg, and 120F+ at 20% attains 28 mpg and reduced performance with pre-ignition/detonation unless a higher octane fuel is used. Air colder than 50F reduces FE linearly. It is apparent that a constant intake air temp of 80-90F should increase available power and highest engine efficiency from this variable.

Hypothesis: Based on fuel maps for my application, a switched intake to draw cool outside air and warm/hot air from 2 separate sources will hold the intake air at a pre-determined temperature range of 80-90F, using either a manual switching mechanism, or a fully-automatic control module using the IAT sensor.

Implementation: This is where I need help.

Cold Air -- I found a conveniently removable plasctic notch in the wiper tray area where the hood seals the engine compartment near the battery, that exposes a 1-inch tall slot by 12-inches long. This could be the "cold air source" although this is usually a "dead air space" at speed. Otherwise, a notch could be cut into the right turn signal to draw outside air directly from the front of the vehicle.

Hot Air -- bit of a specialty of mine. I've been able to draw up to 220F actual IAT temps off of the exhaust manifold. This is the target location.

Control System -- I have no idea. MetroMPG mentioned that a Honda Insight owner developed a completely automatic system, but I did a deep search on the site mentioned and through search engines I wasn't able to find anything. There are 2 ways to approach:

Manual: The driver controls a knob or to direct the intake air based on ScanGauge values
Automatic: The IAT signal is tapped, and wired into a circuit board that interprets the signal to control the valves.

Valves, Piping, and Fail-Safe -- So, what kind of valves and piping are available for this application? I've performed some DIY plumbing repairs and know PVC fairly well, but it all comes down the valve type and how it works. Any suggestions? Lastly, a fail-safe mechanism should be implemented so intake air isn't inadvertently choked-off partially or completely: in either automatic or manual mode. Each valve's closure should be offset by the other valve's opening (for example if the hot-air valve is open at 80%, the cold air valve is open at 20% to ensure 100% air flow).

Climate -- Kansas City is known for its extreme temps. In the last 365 days, outside temps have ranged from -1*F to 107*F; thus enhancing the importance of a more controlled environment for this OBD-II system.

Personal Statement -- This has been on my mind as a project for quite some time. With the cold weather, I need hot air on startup to prevent running rich or in open loop for extended periods, and once up to temp, a slight cooling of the air is required to keep it out of the 100F+ range (which has been happening in temps as low as 30F ambient outside). When in that sweet spot of 90F IAT, I've achieved superior in-town FE as indicated by the ScanGauge.

Plea for Help -- Any engineers, folks with this experience, or brainstormers that can help figure this out? All piping will terminate to the airbox where a K&N cone filter can allow any cutting or compromising of the box itself.

Thanks in advance for the advancement of Science and FE!

RH77

lovemysan · 01-21-2007, 08:27 PM

Quote:

Originally Posted by rh77

OK -- here's the premise. After messing around with hot-air intakes, warm-air intakes, and cold-air intakes, it turns out that none of the above will work for my application. A combination is needed with the radical changes in ambient temperature experienced in this region.

Background Testing: Using the 1998 Acura Integra LS Automatic test vehicle, seperate tests have shown that at the same speed, road and wind conditions, 50F air at 20% TPS results in 25 mpg, 90F at 20% TPS yields 33 mpg, and 120F+ at 20% attains 28 mpg and reduced performance with pre-ignition/detonation unless a higher octane fuel is used. Air colder than 50F reduces FE linearly. It is apparent that a constant intake air temp of 80-90F should increase available power and highest engine efficiency from this variable.

Hypothesis: Based on fuel maps for my application, a switched intake to draw cool outside air and warm/hot air from 2 separate sources will hold the intake air at a pre-determined temperature range of 80-90F, using either a manual switching mechanism, or a fully-automatic control module using the IAT sensor.

Implementation: This is where I need help.

Cold Air -- I found a conveniently removable plasctic notch in the wiper tray area where the hood seals the engine compartment near the battery, that exposes a 1-inch tall slot by 12-inches long. This could be the "cold air source" although this is usually a "dead air space" at speed. Otherwise, a notch could be cut into the right turn signal to draw outside air directly from the front of the vehicle.

Hot Air -- bit of a specialty of mine. I've been able to draw up to 220F actual IAT temps off of the exhaust manifold. This is the target location.

Control System -- I have no idea. MetroMPG mentioned that a Honda Insight owner developed a completely automatic system, but I did a deep search on the site mentioned and through search engines I wasn't able to find anything. There are 2 ways to approach:

Manual: The driver controls a knob or to direct the intake air based on ScanGauge values
Automatic: The IAT signal is tapped, and wired into a circuit board that interprets the signal to control the valves.

Valves, Piping, and Fail-Safe -- So, what kind of valves and piping are available for this application? I've performed some DIY plumbing repairs and know PVC fairly well, but it all comes down the valve type and how it works. Any suggestions? Lastly, a fail-safe mechanism should be implemented so intake air isn't inadvertently choked-off partially or completely: in either automatic or manual mode. Each valve's closure should be offset by the other valve's opening (for example if the hot-air valve is open at 80%, the cold air valve is open at 20% to ensure 100% air flow).

Climate -- Kansas City is known for its extreme temps. In the last 365 days, outside temps have ranged from -1*F to 107*F; thus enhancing the importance of a more controlled environment for this OBD-II system.

Personal Statement -- This has been on my mind as a project for quite some time. With the cold weather, I need hot air on startup to prevent running rich or in open loop for extended periods, and once up to temp, a slight cooling of the air is required to keep it out of the 100F+ range (which has been happening in temps as low as 30F ambient outside). When in that sweet spot of 90F IAT, I've achieved superior in-town FE as indicated by the ScanGauge.

Plea for Help -- Any engineers, folks with this experience, or brainstormers that can help figure this out? All piping will terminate to the airbox where a K&N cone filter can allow any cutting or compromising of the box itself.

Thanks in advance for the advancement of Science and FE!

RH77

I've thought about this some. For switching between ambient and heated air you could use 2 large used throttle bodies. One for each type of air. they could open and close together thus creating a mixing device of sorts. I would install these upstream of the airbox to facilitate even distribution of the heated air. It would be farely easy to fab linkages if you got the hardware thats with them at a pic and pull junkyard. I'd look for something 65mm or larger. You could also install a vacuum opened door to allow unrestricted air for full throttle use.

I don't know how you would control the mixing system but I'll give some thought. Do you know any electro mechanical engineers?

lovemysan · 01-21-2007, 08:31 PM

Rh your in Kansas city right? I'd be glad to help in the fab work. I have plenty of tools including a mig welder. That is if our schedules could align.

CLint

Red · 01-21-2007, 08:36 PM

Depending how handy you are with electronics and some programming you could build an intake controller using a basic stampII

http://www.parallax.com/

lovemysan · 01-21-2007, 08:39 PM

It will have to be economically feasable or its a waste of time. I would a say a total cost under $100 would be a good starting figure.

Red · 01-21-2007, 08:48 PM

Cheapest would be pulling the intake setup off an 80s GM truck with either a carb or TBI. That will give you the plumbing for a dual feed vacuum controlled intake.

rh77 · 01-21-2007, 09:32 PM

Clint -- Yup I live in Lee's Summit (commute through the Grandview triangle and into the Sprint area in Overland Park to work -- Quintiles new location). I take 150-highway on the return leg due to clogging up at I-470 and 50-highway -- that is when I'm not driving 52 miles each way to the airport for work.

If we go with the throttle-body design, I can probably use some a strapping or angle iron to secure the pieces. If I need some welding, I'll definitely let you know, and thanks! I can imagine maybe 2 throttle cables attached like old chokes. But the ultimate would be to have it be fully automatic (and yeah, less than $100 is a good target project cost). I don't know an electrical engineer but...

Red -- I build computers, networks, and HAM radios. I have a basic understanding of schematic interp, programming, and soldering. I'd love to make it fully automatic using a bread board and programmable IC, but I may know just enough to get me in trouble. I'll review your link.

Ironically, my parent's '92 Chevy K2500 Extended Cab Pickup blew the engine (350ci, carbed V-8) 2 weeks ago. Is this too new? They're in Ohio, and I'm in Missouri, so I can't swing-by to see if it fits the bill. If it does, they can pull it off and send it to me. IIRC, I don't think it has the dual feed setup, but I'll have to research the topic.

Keep the ideas coming folks -- I think we're getting somewhere. Another goal is to perhaps adapt it so others can easily customize their own applications as well.

Thanks guys!

-Rick / RH77

Ryland · 01-21-2007, 09:50 PM

I would say the easiest rout to take is to use off the shelf parts from almost any carburated vehicle, as the carburators are tuned to run with 100F air, so they have a very simple system with one or two air temp sensors that are simply a bi-metal tab that releases the vaccum that controle the warm air intake air motor, it is able to maintane a very stedy 100F with a total of 3 simple parts, and a few feet of vaccum hose (the other part is a restrictor valve, to prevent total vaccum loss), it would work alot better then reinventing the wheel.

Red · 01-21-2007, 10:22 PM

The 92 should have at least some form of intake heating due to the carb, so it could be a viable donor.

cfg83 · 01-21-2007, 11:21 PM

RH77 -

I had a similar but different idea awhile back. I didn't want to create a "closed system" hot air intake because I am in LA and wasn't sure if an HAI wouldn't get out of control. Instead I wanted to have the best of both worlds. My idea was that a fan powered HAI would work for me. I also wanted to have control of the fan, so I wanted a fan controller. Here is what I did :

1 - I put a 12V/5Amp Yellowtail In-line 3" Marine Blower into my Hot Air intake tubing ($20 on the net but I paid $32 retail) :

http://www.defender.com/product.jsp?...2133&id=570315

2 - I built a 12V DC motor controller kit that I velcroed to my dashboard ($35) . Power came from the cigarette lighter :

http://www.ramseyelectronics.com/cgi...tion&key=MSC1C

I wanted a reversible DC controller to "shutdown" the HAI completely, but that would have been more $$ and I think it turned out to be unnecessary.

3 - I ran the HAI tubing into the airbox, in the hole that *used* to connect to the airbox resonator (to lower engine noise at high RPM, but otherwise useless).

4 - I moved the IAT sensor into a mid point position in the airbox between the CAI and HAI inputs.

5 - I used the ScanGauge to look at the IAT temperature and I would change the DC fan speed accordingly.

6 - NOT DONE : Automate the fan controller to be temperature controlled. This would have been an inverse temperature control system. Instead of "fan on until temp goes down to X", it would have been "fan on until temp goes up to Y".

Results : When the fan was off, the fact that it was inline would limit the "opening" from the HAI tubing to the airbox, so while the IAT temp was up, I think it was maybe only 20-30 degrees F over ambient. I always had a self-imposed limitation of 140 degrees F because this is a plastic fan that I *think* is only designed to work in temps up to 160 degrees F (that's what a different manufacturer rated theirs at, couldn't get the specs on this one). The implication here is that too high a temp will break and/or melt the fan.

I abandoned the idea when a machine screw came loose in the blower fan housing and caused a nasty noise

. I switched to the normal "passive" HAI and I think it works better for me. I think this is because even though I have CAI, the hot to cold convection allows the heat to increase as the RPMs go up.

Here is what it looks like right now :

Better Solution (?) : I think a stepper motor controlling a low cost butterfly valve would be better. By limiting the size of the opening, you should be able to control the rate of flow of HAI input air, and thereby control the temps.

CarloSW2

markweatherill · 01-22-2007, 12:27 AM

Is there anything inherently *wrong* with the wax pellet or bimetal vacuum-door air mixing system referred to earlier?

lunarhighway · 01-22-2007, 04:36 AM

my car's carburator air intake is equipted with this system ( bimetal vacuum valve controling air temperature by mixing hot and cold air with a butterfly valve). as far as i can see it's working nicely.even after 20 years. i don't know it it's opperationtemperature can be adjusted though or if it changed over time but it's still opening and closing. if i can find a suitable thermometer that i could place in the airflow but read from inside the car some it would be interested to see what the temperature does

the only downside i can see is that this system will take time to adust to temperature changes... electronics with a thermometer will work much faster.

also i think this system is linked to the trottle ... letting more cold air in at higher trottle pressure...though i'd have to look up how eccactly... i'm still amazed at how clever some things are engineerd with 20 year old technology.

rh77 · 01-23-2007, 10:48 PM

Quote:

Originally Posted by markweatherill

Is there anything inherently *wrong* with the wax pellet or bimetal vacuum-door air mixing system referred to earlier?

The problem is precision and speed. It is a simple system, but here's a problem I foresee (please correct me if my assumption is incorrect):

15F, Cold Start at work. 100% HAI input off of the exhaust manifold / 0% Cold Air. The engine warms up and intake temps hit 100F after 10 minutes of driving in the city. 110F is reached and the machanical system opens the cold air tube and closes the hot: a rush of cold is introduced. IATs are now down significantly. The mechanical system takes a while to change back to a warmer mix. The result is a wider swing of temperatures and a potential for a low temperature average.

An electronic system would recover more quickly and precisely control actuators to instantly keep the temperature range in check. A threshold of 85 low and 95 high would be programmed to keep those servos busy, especially during higher engine speeds. With luck, 80-100 maximum range would be realized.

Basically, here's what I've come up with so far:

2 throttle bodies from a "U-Wrench it"-type place, and rubber tubing from perhaps a used vehicle as well (for the cold air especially and to fit the throttle body; hot air will utilize a metallic end source into a plastic or rubber tube. Both outputs will terminate into a 'Y' directly to the airbox. I'd like to run it farther up the flow, but the airbox contains the filter and is the only way to provide clean air.

Now the tricky part: control. At first, I'd like to try wiring a knob that, for example, ranges in motion from 9-o'clock to 3-o'clock. The "Noon" position will allow 50/50 mix where 7-o'clock gives 70 hot/30 cold and 9 = 100 hot/ 0 cold (and the same on the Noon to 3 on the cold side) -- or if that's too complicated, 2 knobs at first for each source.

To control the throttle bodies, servos or actuators will move the valves based on current from the knobs. Someone mentioned using actuators from an old fax machine. If that doesn't work, I could find a (recommended

) company that sells servos cheaply. Then badda-bing controlled intake temps.

The final stage: full automation. If this proves to be work, I'd like to use Red's example using products similar to those from Parallax to program a range from the IAT to move those servos without thinking about it

Carlos-

That is quite some system -- good work! Unfortunately for my setup, if temps go higher than 120F, then detonation, power loss, and an overall loss of FE will occur. I hear Saturns love high temps, though. For the Integra, this may be impossible in the summer when the CAI is in full ops and it'll get what it can, but for the remaining parts of the year, the mix should help.

Keep it coming folks -- good stuff. I can't go to sleep without brainstorming about this first. Call it what you will, but it's a version of cabin fever: not easily being able to modify the car in Winter!

RH77

Red · 01-24-2007, 12:00 AM

Random idea from a sleep deprived brain, some throttle bodies on newer engines have an IAC servo strong enough to move the butterfly of the throttle body. I don't know the range of these servos but something like that would at least simplify the opening of the throttle body

larryg · 01-24-2007, 08:43 PM

Don't a lot of the luxury cars have a climate control system that adjusts the heated and cooled air to maintain the exact temperature you set to keep your car toasty in the winter? seems like the bulk of what you need could be found in the dash of a junk yard caddy- you could could even adjust the temp as you drive and use the mpg guage to get the correct setting-

rh77 · 01-24-2007, 10:22 PM

Quote:

Originally Posted by larryg

Don't a lot of the luxury cars have a climate control system that adjusts the heated and cooled air to maintain the exact temperature you set to keep your car toasty in the winter? seems like the bulk of what you need could be found in the dash of a junk yard caddy- you could could even adjust the temp as you drive and use the mpg guage to get the correct setting-

Hmmm...you may be onto something there. First, a vacuum-controlled ventilation system that directs air to the floor, vents, defroster, etc. could be the mechanism that directs the air. Eventually, I'd like to fully automate the system thru electronics if possible -- but to start, this auto-luxo system could be hacked. That's a good idea.

I need to find a Pic-n-Pull place and go "shopping"

RH77

jeff88 · 03-11-2013, 12:28 AM

I was thinking about this idea and found it in a search. I just wanted to know if this went anywhere. At this point I have the "Y" intersection with a cold and hot air side and a "wastegate" of sorts to switch between them. At this point, the only part I don't know how to do is attaching a "computer" which will take inputs like engine temp and engine revs and use the data to determine which intake side to open/close.

I would like the HAI open during engine startup/cold op until the engine is at proper engine operating temp and the CAI open during high revs/accelerating. I want the computer to determine based on temp and revs which to do.

Any help would be greatly appreciated and I just wanted to know what came about of this.

Thanks!