07 coldair intake, is it safe? Options

[mtlelantra]

I tried a 3 inch CAI, but it threw a lean code, but a smaller diameter tube may not lean the mixture that much.

Yes, the airbox air does cool the ECU. Hyundai designed it in that way for a reason.

So...... when you installed the CAI, 1) how did you route it? and 2) what did you do with the stock box/ECU??? ??? ???

[cclngthr]

So...... when you installed the CAI, 1) how did you route it? and 2) what did you do with the stock box/ECU??? ??? ???

The ECU can't be moved due to the wiring harness being there and it is tough to reroute without undoing a shitload of stuff. The CAI I tried was an older version that had the filter in between the radiator and the transmission, like the older LC Accent CAI's. The stock airbox can be removed, but the bracket remains in place.

I reinstalled the stock airbox.

*SIGH* . . .what happened to all of our Pioneers? THe ones that took the chances to find out what worked? It seems people just want it on a silver platter but don't understand the sacrifices people like Jay, Ford and the like did to make it happen so long ago.

Basically, until some one tries it out, we won't know. Period. Colin and his doom&gloom theories about how cars are to never be touched are way to the left, while Joe's carefree "it'll be fine" attitude is to the right. The real answer is probably inbetween.

Whether you like it or not, cars are programmed differently than older cars, mainly due to the tightening emission controls. In 2005 and 2007, cars have to emit less HC and CO/CO2. The way manufacturers did this is to reduce the adaptability parameters the ECU because the emission systems (mainly due to cost) are not equipped to handle much exhaust emissions beyond factory.

Are you willing to risk damaging the ECU with something stupid as that? Hyundai designed the airbox to cool the ECU for a reason. They did not do it for shits and giggles.

It is not doom and gloom. It is basic common sense.

[mtlelantra]

The ECU can't be moved due to the wiring harness being there and it is tough to reroute without undoing a shitload of stuff. The CAI I tried was an older version that had the filter in between the radiator and the transmission, like the older LC Accent CAI's. The stock airbox can be removed, but the bracket remains in place.

So presumably if you had some narrower tubing that fit that way and just leave the stock airbox in place along with some of the stock tubing to force air through the box to cool the ECU then you might find a winner setup...

[popeye]

Presumably you could cool it with anything that did the job... Peltier device, a rad Thermaltake CPU cooler, but it will require some serious consideration and how much is one to gain in this adventure for the cost?

[cclngthr]

So presumably if you had some narrower tubing that fit that way and just leave the stock airbox in place along with some of the stock tubing to force air through the box to cool the ECU then you might find a winner setup...

Actually, the stock box has close to the same amount of airflow as a cone filter if the box contains a K&N filter. The HD has a redesigned intake system where air comes in quicker and isn't hampered by a resonator that restricts airflow. The XD has a resonator that air must go through before hitting the stock airbox. Since it is a ram air system, airflow is not restricted like a typical standard factory intake. It already is a cold air intake system; and the only thing that you would gain is a bunch of underhood noise.

The 3 inch CAI does work on the XD though.

Presumably you could cool it with anything that did the job... Peltier device, a rad Thermaltake CPU cooler, but it will require some serious consideration and how much is one to gain in this adventure for the cost?

It would take a larger shroud and enough cool air to be effective, and with underhood temps reaching 190+ degrees, cooling the ECU would be harder without some kind of tubing so the air temperature would be cool enough actually cool the computer. Most computer fans would not have the volume of air to sufficiently cool the larger computer used in a car. Most PC CPU's are about the size of a 2 inch speaker, and the ECU is 5x7 inches by 1.5 inches deep.

[CRF450Jim]

I have had the Apollo CAI installed on my 08Elantra since Dec07 and for 7,000km without any issues. Its been taken off and put back on twice as I have taken it to the dealer for oil changes (45min to take off) … dealer provides oil changes for life.

The sound is similar to stock except at WOT where it sounds great even though it is a lot louder. I have noticed about a 4% milage benefit. My seat of the pants dyno show better torque.

WRT ECU overheating … manufacturers build to worst case scenario … that's why we are able to do mods that increase performance but really do not impact reliability in the environment we use the car. There would have to be really high ambient temperature for the ECU to fail with this mod. The ECU box heat sink is very large and the ECU does not generate that much heat.

I intend to keep using this CAI … love the sound and performance … not a big risk.

[mtlelantra]

Awesome- way to go!

[cclngthr]

I have had the Apollo CAI installed on my 08Elantra since Dec07 and for 7,000km without any issues. Its been taken off and put back on twice as I have taken it to the dealer for oil changes (45min to take off) … dealer provides oil changes for life.

The sound is similar to stock except at WOT where it sounds great even though it is a lot louder. I have noticed about a 4% milage benefit. My seat of the pants dyno show better torque.

WRT ECU overheating … manufacturers build to worst case scenario … that's why we are able to do mods that increase performance but really do not impact reliability in the environment we use the car. There would have to be really high ambient temperature for the ECU to fail with this mod. The ECU box heat sink is very large and the ECU does not generate that much heat.

I intend to keep using this CAI … love the sound and performance … not a big risk.

Without an actual dyno, there is no telling what the increase would be.

A flow test on the aftermarket setup vs the OEM setup should show if it is worth the money for the aftermarket setup. If there is little difference in flow rate I would not bother changing it. I do think the HD stock system is designed much better than the older Elantras and the new setup should be equal to the aftermarket if the panel filter was used.

Manufacturers design the system for a reason. Hyundai designed the OE air to cool the computer not just because of ambient temps, but also underhood temps, which can get 200 degrees or better. Computer equipment does not do well with that kind of temperature, even external temperature. The heatsink needs to be cooler than the internal temperature to be effective. 195-200 degrees is not cool enough to keep the computer happy. It actually heats it up instead.

[CRF450Jim]

Without an actual dyno, there is no telling what the increase would be.

A flow test on the aftermarket setup vs the OEM setup should show if it is worth the money for the aftermarket setup. If there is little difference in flow rate I would not bother changing it. I do think the HD stock system is designed much better than the older Elantras and the new setup should be equal to the aftermarket if the panel filter was used.

Manufacturers design the system for a reason. Hyundai designed the OE air to cool the computer not just because of ambient temps, but also underhood temps, which can get 200 degrees or better. Computer equipment does not do well with that kind of temperature, even external temperature. The heatsink needs to be cooler than the internal temperature to be effective. 195-200 degrees is not cool enough to keep the computer happy. It actually heats it up instead.

I would agree that without an actual dyno there is no telling what the increase would be. In a similar fashion, without a measure of actual flow rate, there is no telling what the flow rate would be compared to the OEM setup or the OEM setup with a K&N panel ... even if you think that the HD stock system is designed much better. Its kind of like my "seat of the pants dyno" guesstimate ... both are guesstimates.
That being said, here are many good "dyno measured" examples (with a variety of motors) about the proven performance benefits of CAI vs stock intake. I have not seen any that show the flow rate of the stock intake is similar to that of aftermarket CAI's. The other benefit with this CAI is that it will not be the weak link in delivering performance as other mods are installed.
WRT ambiant temp ... the steady state under the hood temp will be inflenced by the heat output from the motor/xmission as well as the ambiant air. In addition, hotter ambiant air tends to make the motor/xmission create even more heat.
ECU's have protection via an internal temp thermal ... this means that the ECU with shut down to protect itself should its internal temp be out of range. That being said, ECU's can handle a significant amount of heat. The following has been taken from a generic ECU spec sheet that shows just how wide of a temp range is acceptable ...

A.21 Engine ECU Temperature — Internal air temperature of the engine ECU.

Parameter Data Length:

1 Character

Data Type:

Signed Short Integer

Bit Resolution:

2.5 °F

Maximum Range:

–320.0 to 317.5 °F

ECU's do not contain state of the art electronics such as todays hi speed processors that can only handle 70 degrees F of ambiant temp. Instead, the ECU's contain IC's that are more heavy duty in nature. In fact, the number of IC's used to measure very hot spots around motors have increased dramatically ... this is due to adoption of the SI2C sensor interface standard. These sensors (read IC chips) can play in the exhaust system where the temp is in the very hot 400C or 750F range.

However, most under hood temp seldom get beyond the 150F mark. Chk out this link for under hood temp with a Honda Civic Sir that has a modifed from preformance motor from Australia with max underhood temp of only 60C or 140F in their summer ...
http://www.teirney.net/civic/TemperatureReadings.htm
So, seeing that;
1) There is a lot of data out there that confirms performance benefits in a variety of motors using a CAI over stock and none that confirms air flow from stock air intakes come close to that of aftermarket
2) knowing that the ECU can handle a lot of heat and will thermal out first before it becomes throwaway
3) knowing that the under the hood temp of my low performance elantra, in most ambiant temp, is not going to overheat the ECU
... it does make sense for some of us to spend $120 to add some performance and sound while getting better gas milage ... if only we could keep our foot out of it!
... +7,000km and still going with my ECU surviving under the hood temperatures

[cclngthr]

WRT ambiant temp ... the steady state under the hood temp will be inflenced by the heat output from the motor/xmission as well as the ambiant air. In addition, hotter ambiant air tends to make the motor/xmission create even more heat.
ECU's have protection via an internal temp thermal ... this means that the ECU with shut down to protect itself should its internal temp be out of range. That being said, ECU's can handle a significant amount of heat. The following has been taken from a generic ECU spec sheet that shows just how wide of a temp range is acceptable ...

A.21 Engine ECU Temperature — Internal air temperature of the engine ECU.

Parameter Data Length:

1 Character

Data Type:

Signed Short Integer

Bit Resolution:

2.5 °F

Maximum Range:

–320.0 to 317.5 °F

ECU's do not contain state of the art electronics such as todays hi speed processors that can only handle 70 degrees F of ambiant temp. Instead, the ECU's contain IC's that are more heavy duty in nature. In fact, the number of IC's used to measure very hot spots around motors have increased dramatically ... this is due to adoption of the SI2C sensor interface standard. These sensors (read IC chips) can play in the exhaust system where the temp is in the very hot 400C or 750F range.

However, most under hood temp seldom get beyond the 150F mark. Chk out this link for under hood temp with a Honda Civic Sir that has a modifed from preformance motor from Australia with max underhood temp of only 60C or 140F in their summer ...
http://www.teirney.net/civic/TemperatureReadings.htm
So, seeing that;
1) There is a lot of data out there that confirms performance benefits in a variety of motors using a CAI over stock and none that confirms air flow from stock air intakes come close to that of aftermarket
2) knowing that the ECU can handle a lot of heat and will thermal out first before it becomes throwaway
3) knowing that the under the hood temp of my low performance elantra, in most ambiant temp, is not going to overheat the ECU
... it does make sense for some of us to spend $120 to add some performance and sound while getting better gas milage ... if only we could keep our foot out of it!
... +7,000km and still going with my ECU surviving under the hood temperatures

This is not a Honda. The Hyundai ECU on the 07+ car is a 32 bit processor and is the same as a regular IC. Most ECU's are 16 bit processors. Only difference is the casing acts like the heat sink. They do not necessarily shut down when they overheat; I have replaced ECU's in cars that did overheat and the IC chipset did show signs of overheating, which is the main reason why they fail. Any electronic devise is made in the same manner whether it is for a home or car. Equus can give more info because that is his area (he used to program the ECUs). Your link only shows intake charge temp differences, but not IC temp differences. Don't confuse the two. I'm talking ECU/computer tech here, not intake air temps.

As for a flow rate reading, that can easily be done with the right stuff. I made a flow gauge for the oil filters and tested several of those and making one up for the intake is similar to what I did for the filters (I will try that this weekend). Having the same/similar air volume as an aftermarket setup causes me to rethink the purpose of doing the mod. Measuring the amount of restriction in the intake setup also gives you a baseline which one to use.

Colder air on the stock HD intake is colder than what is seen on the XD because that design has air coming in from a colder area. Stock Honda intakes are not designed like the HD setup as well. Most of those are restrictive. I found that the HD setup when I had mine off is much better because the resonator is not within the airflow path, which on most cars the airflow passes through the resonator (which is restrictive) before hitting the filter. On the HD, the resonator is on the side of the intake tubing and airflow passes by it freely, although the resonator can slow the air down slightly, but not as when the air goes through the baffles in most resonators. On a stock Honda intake, the incoming air is within the underhood area. The HD setup is getting air out of that area because the intake port is outside of that underhood proximity.

Look at my dyno graph in the dyno area. I had the engine bone stock at that point. Later on this month, I'm redoing the dyno with the intake and exhaust modifications I've done since then. That will show you the difference in where the power ended up at.

[CRF450Jim]

This is not a Honda. The Hyundai ECU on the 07+ car is a 32 bit processor and is the same as a regular IC. Most ECU's are 16 bit processors. Only difference is the casing acts like the heat sink. They do not necessarily shut down when they overheat; I have replaced ECU's in cars that did overheat and the IC chipset did show signs of overheating, which is the main reason why they fail. Any electronic devise is made in the same manner whether it is for a home or car. Equus can give more info because that is his area (he used to program the ECUs). Your link only shows intake charge temp differences, but not IC temp differences. Don't confuse the two. I'm talking ECU/computer tech here, not intake air temps.

As for a flow rate reading, that can easily be done with the right stuff. I made a flow gauge for the oil filters and tested several of those and making one up for the intake is similar to what I did for the filters (I will try that this weekend). Having the same/similar air volume as an aftermarket setup causes me to rethink the purpose of doing the mod. Measuring the amount of restriction in the intake setup also gives you a baseline which one to use.

Colder air on the stock HD intake is colder than what is seen on the XD because that design has air coming in from a colder area. Stock Honda intakes are not designed like the HD setup as well. Most of those are restrictive. I found that the HD setup when I had mine off is much better because the resonator is not within the airflow path, which on most cars the airflow passes through the resonator (which is restrictive) before hitting the filter. On the HD, the resonator is on the side of the intake tubing and airflow passes by it freely, although the resonator can slow the air down slightly, but not as when the air goes through the baffles in most resonators. On a stock Honda intake, the incoming air is within the underhood area. The HD setup is getting air out of that area because the intake port is outside of that underhood proximity.

Look at my dyno graph in the dyno area. I had the engine bone stock at that point. Later on this month, I'm redoing the dyno with the intake and exhaust modifications I've done since then. That will show you the difference in where the power ended up at.

I do appreciate the info as there are some real truths in it ... however, I am compelled to respond to the FUD portions.
1) I understand that an Elantra is not a Honda. Don't be confused with the Honda link example ,,, as I stated in my initial post, the only reason that I added this link was to show that underhood temp are typically less than 150F as opposed to the rather high +190F as suggested. This had nothing to do about ECU temps or here the intake air comes from. In the next 30days I will have put in place monitoring of under hood temp on my Elantra to minimize the confusion. As a side note, the under hood temp clearly show the potential benefit with the denser charge of a CAI. I have added nitrous to a number of motors and found that half the performance benefit comes from the nitrous "supercooling" the intake as opposed to just adding the extra oxygen molecule with more fuel.
2) WRT ECU temps ... interesting factoid of 16 bit vs 32bit ... but not very relevant to the topic. The main contributer of heat with uprocessors is its clock cycle and how much the processor is used ... typically both are significantly less in ECU's than what we see in todays PC's (where 64bit is the standard).
3) WRT ECU overheating ... it would be more factual if you could state in confidence that the ECU's you relaced that "looked like the processor overheated" were actually tested to prove that overheating caused an issue ... rather than just your thougts ... its kind of like a "guesstimate" or "SOTP Dyno". It would also be good if you could state that the Elantra ECU "does not" have an ECU thermal out. Most newer technology ECU's do but the Elantra may not have this. BTW, an easy way to resolve the potential of overheating is is to use some of the stock intake to flow air over the ECU heat sink.
4) IC manufacturing ... all IC's ARE NOT created equal ... some are made for more hostile enviroments ... please take some time to investigate the SI2C technology that I mentioned in my initial post ... a technology that is typically used in a car but not in a home . I did not see any response to the ECU temp of 317.5 °F from another non-Elantra ECU.
5) why ECU's fail ... the main reason is external issues as opposed to primarially heat. The external issues are due to things such as static, shorts in external wire harness causing failure of internal ECU comoponents, water, just bad manufacturing/design ,,, etc. A transistor junction is about 7microns. A static charge tears off about 3microns. The resulting 4microns allows the junction to continue working but causes it to significantly overheat and typically fail quite rapidly ... 30days or more depending on use. Note, this may be the "signs of heating" that you observed in other ECU's that were dead for some reason. It is interesting to note that many failures with ECUs under the dash are casued by leaks in the car's heating sys. So, maybe the engine bay is a less hostile enviroment?
6) Flow rate ... you keep stating that "Having the same/similar air volume as an aftermarket setup causes me to rethink the purpose of doing the mod". Lets discuss this once you have proven that the flow rate is the same. BTW, go back an look at the pics I posted of stock vs current intake ... do you really think that flow rate would be the same? That being said, its more than just flow rate, its also the benefit of cooler charge of air with a CAI. If you trully believe that the under the hood temps are >190F then this is a case by itself for a CAI. Those underhood temp are really heating up the stock air box that corresponds to heated intake air. In my discussions with K&N people, the Apollo can deliver more air flow than my motor can demand in a normally asperated mode.
7) It would be great if you could dyno the CAI as a comparison ... this would be closest to the "real world" performance test. Although, the CAI would flow even colder air and deliver a denser charge with the car actually moving.

I do not want to turn this into a "back and forth" discussion on this topic. I did want to follow your lead on stating the facts as opposed to conjecture ... from your intial post "Without an actual dyno, there is no telling what the increase would be." Your posts seemed to contain a some decisions/suggestions that were backed-up up by conjecture as opposed to facts.

This thread was really all about is it safe to CAI the 07Elantra. Its truly up to each individual to decide. Based on what I know, both hard facts and experience, I have decided to do it and I am enjoying the results. Others may not chose to do it. I have tried to lay out the facts that I know as well as the "intuitive" details that are based on experience. Its truly important not to dwell on Fear Uncertanty and Doubt (FUD). These should not be part of the decision as much as a healthy respect for knowledge and risk ... there is a significant difference between the two.

I hope that I have been able to arm readers with more factual information to allow them to make an informed decision ... to CAI or not to CAI.

[cclngthr]

1. The ECU on the Hyundai is under the hood and exposed to temps as high as 200 degrees, including the 850 degree F temperature of the first catalyst. HMC programmed the ECU to make the catalyst operate at that temperature, and with it under the hood as well as the engine and transmission, the casing on the ECU needs to be cooler than what the temperature of the ECU is putting out. This is not CAI temperatures. You seem to confuse a CAI temperature with underhood temperature. 2 different issues here. I am specifically talking about underhood temps as related to temps of the engine and trans, which are higher than CAI temps. CAI temps are lower by nature because the air is cold as it enters the filter and flows at a high rate thus warming up the air is not as easy if plastic is used. Since most CAI's are metal, underhood temperatures heat up the metal faster because metal transfers heat better than plastic, what temperature of air is entering the metal tubing quickly warms up once it reaches the intake manifold. With a plastic tube/intake system, heat transfer is not as easy due to the material.

2. The higher processor produces more heat, as with normal IC's. The HD ECU speed is operating faster due to the design of what it is supposed to do. Its decisions make the engine run better, but also decisions regarding the whole electrical system is integrated with that same system. Honda's started using this around 5-6 years ago, where the ECU does more tasks than engine and transmission management. However, Honda/Acura has had multiple electrical gremlins that posed problems regarding the ECU management. HMC decided in 06 with the NF and TG that increasing the processor speed as well as other revisions can improve the reliability of the electronic systems. Apparently, that worked for Hyundai.

3. ECU overheating causes certain functions to intermittently fail and sometimes all functions fail. Voltage is usually passed through chips, and if the voltage does not go through the chip at a certain rate (time or voltage) the chipset is bad or the board is bad. On a 99 Mazda 626 I worked on, there was a bank 2 o2 sensor inefficency code at times and the sensor and catalyst tested out OK. With both the cat and sensor testing fine, the wiring was checked and that also tested fine. On the ECU, testing voltage at the chipset found voltage was entering parts of it, but not exiting where it should be. Discoloration of the IC also shows signs of overheating. Some chipsets have a microcircuit that is visible and under magnification, that chipset shows breakage of the circuitry. Usually, under normal conditions the ECU stays stable, but if it gets hot, instability is seen, which can result in a variety of things, from codes to drivability issues to loss of drivability.

6. Flow rate measures restriction and volume of air through the system that is bench tested. When I bench tested oil filters, I used a stable liquid pressure (psi) and had a gauge before the filter measuring the psi going into the filter and another at the exit measuring the psi level of the liquid exiting the filter. A similar devise can be used to test the stock airbox on the bench where a stable pressure and volume is entering the intake and the second set of gauges reads the volume and pressure of air exiting the system.

You keep saying cooler air. On the HD, the incoming air is not entering the system from the engine bay area. It is entering the system just outside that area ahead of the radiator and AC condenser. Most cars have intake systems that the incoming air enters the system from behind the radiator, which the air is hotter than ahead of the radiator.

You also are neglecting that the intake air has to be at a certain temperature for fuel to burn efficiently and cleanly. Most cars do have intake manifold TB's that have coolant lines running to them. The Elantra HD and XD both have this not only due to emission requirements, but also having the throttle body operating correctly. If the incoming air is too cold, the burn is not as efficient thus higher emissions are seen. The same can be said if the air is too hot.

7. When I tried the CAI I temporarily used (a 3 inch long tube) I ended up getting a CEL with a lean code. There was too much air for the ECU to handle. Although the same intake works fine with the XD, the ECU adaptability is narrower on the HD to solve several issues. 1. The XD suffered from a 3800-4200 rpm bog, which is not present on the HD. Here, the fuel ratio was changed to prevent this bog from occuring on the HD. This was noticed when I changed my plugs, which were a medium brown rather than a light tan to gray color that is normally seen. The XD has a gray color on the plugs when they are removed. Here, a leaner mixture is seen. Since the darker color is evident on the HD plugs, the mixture is richer to offset any bog in an rpm range. You don't want to run a engine too lean or the piston will give out on you.

Here is my dyno sheet with the stock HD.


After this run was made, I modified the intake by removing the resonator and capping the hole and putting a K&N drop in filter in and removing the stock 1.8 inch exhaust pipe and resonator and replacing that with a 2.25 inch pipe from the 2nd cat to the muffler and retaining the stock 1.8 inch tailpipe from the muffler back to the bumper.