Is cold air intake necessary on a turboed car?
 

After refitting my charge pipes, I no longer have room to run a cold air intake.

Now, my intake hangs off the turbo sandwiched in-between the exhaust manifold and the radiator. It just looks wrong...

I could maybe get some flex hose routed from the headlight area, but there's really not much room. Certainly not enough room to run a hard line to relocate the filter to the wheel well area.

What have others done?

Is cold intake that helpful? Or, does the turbo just heat soak the intake air stream anyway? ... and the intercooler has the biggest effect?

colder air inlet sure as hell doesn't hurt. Look up flyinmiata kit. They use a 90 silicone elbow and locate the filter behind the headlight area with some heat shielding to isolate the heat from the engine bay as much as possible.

I moved this to the DIY area because it is not race prep.

I wouldn't bother with a CAI if you're already turbo. (Though placement is still important)

Turbo will heatsoak either way, and also consider drag if you have lengthy pipe sourcing the cold air.

Why not do some IAT tests of your current setup? :dunno:

Would you be able to post a photo?

it definitely matters, but it's not a dealbreaker

It will look better in my opinion

My box is also cover with insulation on the inside the keep the heat outside of the box as much as possible

https://scontent-ams.xx.fbcdn.net/hp...03&oe=55763499

Next week im beginning my mild wiretuck:facepalm:

I've done both with my old GT3271 setup. Hot air intake, pulling in exhaust heat/radiator fan air, and then a custom CAI. I posted all teh data here when I made it, though it would probably take a while to find. It was worth probably 40HP on my car. It made A LOT more power with the CAI, I had to retune the car cause it went lean! I don't remember how much, but it was very noticable and power went up a good bit, very much worth it. Also AIT's dropped as I wan't dumping as much heat into the intercooler.

IMO it's a must, but a lot of people don't build them. It can be a pain to build, but it does add HP. I gotta build one for my new setup, and I know it's gonna suck, but I gotta do it for the power!

I'd say CAI is power found.

Yes. Pre turbo air temp and pre-turbo pressure drop are the biggest hidden power suckers you can find in a turbo setup. Like pat said, especially on a big power setup, they can kill you. With the subaru EFR stuff people are finding huge power gains going from a 2.5" intake tube to a 3" because of the pressure drop, even a couple tenths of a psi pressure drop in the intake tube puts a lot more heat into the intake charge and drops the max possible boost. Thats more drastic because of how convoluted their intake is, but do not discount the importance of the turbo intake.

I'd worry about a large intake and a quality intercooler as steps 1 and 2, then if you still have issues consider a cold air intake of sorts.

But sacrificing pipe size for cooler air will do more harm than good. Volume is the most important factor with a turbo IMO.

Not always true. The law of diminishing returns applies here. The tubes become a very small percentage of intake system pressure drop when compare to the intercooler once you reach a certain size. For our 250 WHP turbo cars, the difference betwwen a 2.25, 2.5, and 3 in is very small. However, go bellow 2 in and you intake tubes pressure drop will increase greatly as you start to choke the system at your top end CFM.

I guess it's case-by-case, but in the mazdaspeed3 world going from a 2.5" intake to 3" with the exact same routing will net ~10hp with matching tune adjustments. The g/s increase is significant. That's with a k04

yes, case by case. It's a function of max CFM through the pipe which is a function of MAX whp. A 400 WHP speed3 will benefit more from a larger intake pipe than a 250 WHP turbo miata.

Let boost be at 6 psi. The temp gain with turbo is about 14/15F per psi, or about 90°F.

On a 90°F ambient day, ambient absolute would be 90 + 460 = 550F.

The charge temp absolute would be about 550F + 90F from the turbo, which comes out to 640°F

Density would degrade by 550/640 = .86

The pressure ratio at 6 psi is 1.41.

With the heat from the turbo, then the 141% more density from compression will be degraded by .86 x 1.41 to about 1.21. or a 21% performance increase.

With a 90% efficient IC, .90 of the 90°F temp rise will be taken out, or about 81°F removed, thus leaving a net gain of 9°F.

So 550F + 9F = 559F

550/559 = .984

This would leave a density loss of only 2% from the heat, or a net gain of 1.41 x .98 = 1.38. or a 38% performance increase.

----

Now let's say we aren't using a cold-air intake and now we have 180°F temps under the hood (double the ambient!).

460A + 9F (from turbo minus IC removal) + 90F (intake delta increase) + 90F (ambient) = 649F

550/649 = .85

This would leave a density loss of 15% from the heat, or a net gain of 1.41 x .98 = 1.2. or only a 20% performance increase.


Or we can say, if you expect 10rwhp from every 1psi. that maybe the difference between your 100rwhp car going from 148rwhp from a warm air intake to 158.8rwhp with a cold air intake.


something like that. i probably messed up the math, but i dont tend to let facts stand in my way.

not talking about intercooler pipes, we're talking about pre-turbo intake pipes. Even tiny fractions of psi have a noticeable effect here. For example, assuming you're at sea level and your ambient pressure is 14.7psi and your target boost is 20 psi. That results in a 2.36 pressure ratio with no intake restriction, now put an air filter right on the turbo inlet (lets call it a .2psi pressure drop) and you're already up to a 2.39 pressure ratio. Add some fairly restrictive bends in and you're at a 2.45 pressure ratio. To put numbers to this, lets assume that you're getting 100°F air into the compressor with all of them and lets use 70% efficency for the compressor. In the open turbo inlet example you get 320°F air exiting the turbo, with the air filter its 322°F, with the somewhat restrictive intake its 331°F. And at higher mass flows that nearly 10°F temperature difference is a whole lot more heat the intercooler needs to reject to get you down to 20°F above ambient. Now for kicks if that restrictive intake got you to some ambient air of 70°F your outlet temperature is now 288°. Yeah thats a lot, and thats why this post took so long to write I had to re do that math out like 4 times to make sure I didnt fuck up.

What this doesnt account for is if your turbo efficiency goes to shit due to the higher pressure ratio, or if the higher PR forces the turbo to run out of steam. IE, you're running ALLOFIT but, like most compressors, the highest flow area is at a lower pressure ratio.

bigger intake pipes is power found from compressor efficiency losses.

https://www.miataturbo.net/attachmen...ine=1426263792

When I started putting this together, it seemed like I had enough room. But, I found that things were tighter than I thought.

As you can see, I'm using a MAF which takes up some space. Also, my crankcase catch can is also in the way.

Vincentmiata: That cold air intake box looks great! I wonder if I could fab up a container for the filter like yours and then duct cold air into it?

Solution: K&N RC-5052AB, K&N Apollo Unique Universal Air Filter | K&N

I really wonder how much that setup adds in restriction. The walls of the cone being so close to the filter itself add a lot of restriction. It would be interesting to compare pressure drops with a manometer of that filter vs a normal cone filter in a diy sealed box thats also fed cold air from the same type of hose.

From the CFD I've run the spacing between filter and housing wall looks adequate. Typically about 1.5-2 in is ideal. It should taper out some to create a constant press delta past the filter surface. This would load the filter media equally throughout the filter length rather at just close to the filter outlet like most radial filter airboxes. I have some pretty CFD pictures I wish I could share but I probably shouldn't unless we release marketing data on the study for work.

The best part of this system to me is the bell mouth outlet of the filter. This feature can not be created anywhere near a generous of a radius as your typical filter.

I'm considering on picking one up and can run it on the flowbench and post the results.

It would be cool if K&N actually followed that same principal in the design. I doubt they did, not many automotive aftermarket companies do.

Looks a little restrictive to me.

Look how big of an airbox they use in WRC

https://www.miataturbo.net/attachmen...ine=1426268176

You routed your hot side IC pipe through your cold air source. Reroute that hot side pipe and find power with a cold air box.

I messed up math. Intake plus turbo equal 270f. 270 x .9 = 27f

550/577 = .953. Vs .984

Somewhat trivial gains.

Yes, I agree. I might just have to reclock the turbo and route it down.

But, it wasn't me!... It is a BEIG kit for an MSM. Apparently, they tried to follow the OEM hotside route in their kit.

In it's past life, I solved this problem by using the hollow frame member to route cold air from a filter stashed just behind the bumper cover. (You can see the old route the from the picture). Since then, I ended up plating this opening and used the structure to mount a nerf bar.

http://https://www.miataturbo.net/me...2151018a-4052/

Really interesting read as I'm just now thinking about this. :D

Flame suit on.. :lowdown::rofl:

Soooo.. The calculations were on 180 degree engine bay temps v 90degree outside temps yes? And so at 6psi, it would be about a 10hp difference?

Does this take into account how others are talking about the size of the charge pipe making a difference? If not, is there an even smaller difference with a filter right on top of the turbo v a 1 or 2 foot long charge pipe?

Does colder air move more slowly enough to make a difference in this situation?

Not 10rwhp.

If you assume 10hp per psi.

Then 9.5 x 6 = 57hp. Vs. 9.8 x 6 = 58.8hp from the turbo at 6psi.

Measurable difference but very trivial. Intercooler is key.

What is missing from your calculation is the fact that if you're running the same "boost" in each scenario, the turbo has to work harder (more energy into compressor) to get the same boost from the hot air, as it's less dense. So that means more energy out of the turbine, so to do that the wastegate clamps down to raise turbine pressure to get the extra energy.

So two identical engines, one with cold air intake, one with hot air, the cold one will actually flow more air since the turbine pressure goes down, less likely to detonate (turbine pressure goes down). It's more than just boost and intercooler temps.

I have a back to back of this exact filter, with and without the end piece that you can put a hose on.
No hose either run, both runs had the enclosure in the power steering reservoir location.

GT2560, well flowing cast log, 2.5" exhaust. 20ish psi tapering on.

This is 285rwhp, and the difference where you can see it averages 5rwhp.

https://fbcdn-photos-g-a.akamaihd.ne...338c4c84fdda1d

Adding to what brain is saying, the math shows its trivial, and that's before what no one has yet mentioned.. Delta T.

If the intercooler is fed hotter air, the difference in intercooler surface temp compared to ambient air temp cooling it is greater, (greater Delta T). The greater the difference in temperature the faster the heat transfer.


Ie, when the air into the cooler gets hotter the cooler gets more efficient.

I really don't bother with cai ever on low boost setups, I just add one click on the MBC.

Dann

Does the length of the charge pipe also have resonance effects like the intake?

Imagine if you had a huge fan in the engine bay, and all it did was suck up engine bay heat, and blow it into the mouth of the car at the heat exchangers.

This is exactly what your turbo is doing when you have a hot air intake. Makes all the heat exchangers from the intercooler back less effective since they are getting warmer air.

There is no question that a CAI is the better way.

To try to put numbers to how much it helps, to do it somewhat close, you'd need a map sensor before and after the turbo, temperature before and after turbo, temp after intercooler, turbine inlet pressure, and preferably a MAF sensor too. And if you want to put numbers to how it affects say, radiator effectiveness, it would get even more complex as then you'd need water temp in/out of radiator and air temp before/after radiator for both setups. To measure heat soak, you'd have to datalog all of this and graph it, and measure time between cooldown after a set of runs or something to accurately know how much it matters.

Or you could just accept the physics that cold will make more power and make the engine less likely to detonate and build a CAI accordingly.

Sorry, I edited my question while you were answering it.

His math is wrong, it's not trivial! See my post right above this. Leafy touched on it too, the pressure drop upstream of the turbo is a big deal too! Having a good flowing CAI is much better than sticking a little filter on the inlet of the turbo.

my math on the inlet has been corrected. if you're going to bring up the turbine efficiency, show us a math formula that factors it in.

a 90°f intake delta, plus the 90°f turbo delta, after intercooler would be an extra 27°f on a day 90°f day assuming a 90% effective IC.

you're only 18°f degrees more likely to detonate from intake temps. in the grand scheme of things that nothing and the compressor/turbine efficiency difference between the two is also going to be negligible.

I'm not saying CAI isn't better, but im saying it's not a make or break big deal, assuming you have a good IC. I made efforts to source ambient temps myself. But I'd be more worried about living in the rocky mountains vs. the dead sea really...

Brain there are other math that uses the turbine work to determine the temperature increase across the compressor. But the math going on with the turbine, the efficiency and rho is a whole lot more difficult to do. You're into the case where you have to solve a non linear equation. Either using matlab or brute force in excel like a child.

The equation for the turbine is the change in Enthalpy across it. Enthalpy is denoted H, and H = U + P*V, where U is internal energy, P is pressure, V is volume. Thus you need all the stuff i posted above (temps, pressures, MAF sensor) to measure it.

Like leafy said it's several things changing, not just one. When you change one, it changes all the other things. Kinda like how if I put a turbo on my car, more air goes in, but then air gets hotter, and then cooler from IC, and then denser, and then need more fuel, and then have to change timing now, and so forth. One thing changes other things in the system.

I'm not going to makeup 14 assumptions for temp/pressure and try to solve it, I don't have a software that will do it worth a crap.

But I'll at least show you this to help everyone understand. Say it's an 80*F day. With a CAI, temps going into the turbo are 80*F, with a hot air intake, lets just say 100*F HIGHER, so 80 + 100 = 180*F into the turbo.

I'll assume we're running the same boost before/after, and that the boost reference is at the intake manifold.

And let's just say we're at sea level, 40% relative humidity.

Case 1: Density Air at 80*F, 0. 0727 lb/ft^3
Case 2: Density Air at 180*F, .0569

So if you have a 300whp miata, you're flowing about 30 lb/min through the compressor.

At 300whp, airflow is 30lb/min, so with a 3" inlet pipe hooked to the inlet of the turbo, air speed is

Case 1: 30 lb/min / .0727 lb/ft^3, = 412.7 ft^3/min, or CFM
Case 2: 30 lb/min / .0569 lb/ft^3 = 527.2 ft^3/min.

So at this point, we can see that with a hot air intake, the CFM going through the compressor inlet had to go up by 527.2 / 412.7 = 27.7% just to match the same mass flow rate of air.

This is not without consequence. I said we have a 3" inlet pipe, so at 3" ID, Area is A = pi*r^2 = 7.0868 in^2. One foot of pipe would be 12", so one foot has 7.0868 *12 = 85.04 cubic inches. There are 1728 cubic inches in a foot (12x12x12) so 85.04 cubic inches is 0.04921296296 cubic feet in 1 foot of pipe.

So case A, 412.7/ .04921296296 = 8386 ft/min, 5280 ft in a mile, so 8386/5280 = 1.58826 miles/min, times 60 is 95.3 mph. Speed of sound is 761.2 mph, so 95.3/761.2 = .12 mach

Case B, 527.2/.04921296296 = 10713 ft/min, 10713/5280= 2.0289061833 miles/min, times 60 is 121.7 mph, so 121.7/761.2 = 0.16 mach, or 0.16/0.12 = 33% faster air.

Pressure drop vs speed is not linear, it's squared!

I'm not keeping units anymore, but lets say you have 10 units of crickets worth of pressure drop in Case A, Case B is going to have 16.31 crickets worth of pressure drop, or 63.1% higher pressure drop!!!!!

If inlet temps are more than 100*F higher (say your turbo is bolted to a hot exhaust manifold, for example) the numbers get worse. Or if you have anything smaller than 3" pipe for your inlet, again, it gets worse.

I would like to point out that I ONLY did some of the calculations for pressure drop upstream of the compressor. I did not touch anything else!

And just in case it's not obvious, asking your turbo to flow 27.7% more pushes you to the right of the compressor map, and most people here are already right of the "sweet spot", so then the compressor is becoming less efficient (more heat in charge, more turbine inlet pressure to get energy to drive compressor).

There is a reason that every turbo engine in production in the world has a cold air intake from the factory.

that doesn't really touch on anything I said.

It's my opinion that you use the largest intake pipe you can fit to reduce the pressure drop between the filter and turbo.

one example of a warm air intake on a turbo car: the mazdaspeed miata.

You said it's trivial, I showed you why it's not. You were only looking at one aspect of it, not all of it. Even I didn't look at all of it, just another aspect that is a lot more significant than the one you considered when you stated CAI is trivial, intercooler is key. No disrespect, but you can't draw a conclusion looking at just one part of the picture. Even my math is only one part, I didn't consider the turbine, compressor efficiency, etc.

You also asked for the equation for the turbine so I posted that too.

The point is that CAI does in fact make a difference, it is significant, and there is a lot more benefit than just 3* colder AITs as you posted. I want people to realize that and not assume that is the ONLY consequence of a hot air intake. There are in fact many reasons to run a CAI, and 3* or whatever cooler AITs is not the only reason.

you're ignoring a lot more than i am. and yes, you really lost track of your units, conversions, and relevance there.

you're forgetting how much a good IC equalizes the density ratio of the air that the motor ingests on each gulp, and that's really the most important factor when talking WAI vs CAI. Yes, the turbo will have to work a little bit harder to compress that air, but I don't you're going about determining that correctly.



here, this one factors in most:

Not2Fast: Turbo Calculator

it agrees that the change from 26.6*C to 82*C intake temps doesn't change the HP significantly once you input the rest.

I said I was ignoring everything, and did an example to show flow and pressure drop on the inlet side of a compressor.

What where did I loose track of units, conversions or relevance? I didn't check my math, but since you say it's wrong, what is wrong?

A lot harder, it now has to flow 27% MORE in my example to produce the same mass flow of air.

If I'm wrong, and I may very well be wrong, please explain/demonstrate. I am not perfect! All I did was lookup the density online and work some maths with a calculator. Do you think I did it wrong, or do you know I did it wrong? If you know, what did I miss?

Leafy, does my math look ok, or is there a units error or something else off?

Maybe tomorrow am I'll reread everything and think about it more. but I really don't think you're looking at it correctly.

its time to cook dinner and get drunk on margaritas.

Sounds like fun, I'm dissecting the factory wiring harness as needed, which surprisingly isn't going as bad as I thought it would. Keep having to look up connectors/wire colors which is always fun.

Pat I think you're math is good. But I really really fucking hate doing fluids or thermodynamics work in imperial units. My step 1 when doing it with them on paper is to convert to metric because I've been burned by a weird unit conversion before. Its why I like mathcad so much, you can give your inputs units and it'll handle all the units conversion for you and tell you if theres a units mismatch.

The losses at the engine are fairly low in my scenario--you're ignoring the DR completely and focusing at airspeed at the intake.

remove the intake altogether.

I'm suggesting on same day, with just different intake temps you might only lose 3% power from doubling the intake temps. You're still taking all that less dense air, then packing it back together after the intercooler, just not QUITE as well [3% less effective] because it's slightly warmer and more energized.

https://www.miataturbo.net/attachmen...1&d=1426433585

https://www.miataturbo.net/attachmen...1&d=1426433585

Losses of turbo efficiency might skew the numbers slightly, but that's going to be seen as slower boost threshold. Like how you can overboost on a cold day and underboost on a warm one, if using a MBC. Sure, you'll make more power on that colder day because the temps are lower and you're packing more air per volume. But the DR represents how effective your PR is. In a perfect world PR = DR.

To make up for the loses when it's hot, you spin the turbo a little more. The airflow on the turbo compressor map doesn't change but the speed/PR at the compressor outlet will increase to make up for it. Basically it just takes a little longer to make the boost pressure.

https://www.miataturbo.net/attachmen...1&d=1426433585

I may be oversimplifying it, but I think you're over-complicating it.

the intake temps may have doubled, but the IC being so effective brings the gains of 38% more oxygen to burn per volume down to 34%.

If it was a 40°F day, and still 180°F intake temps, then yeah the loss will be more significant for sure, would be nice to know what intake temps on an n/a miata typically reach using a WAI. But even here that's 63°F intake temps from a WAI after IC vs. 49°F on the CAI--that's still 1.385DR vs. 1.348DR--so it doesn't really appear so.

You are only looking at temps and PR's after the compressor, I was only looking at temps and PR's before the compressor. We're both "right" on those numbers. Your numbers show that the intercooler works to fix the density ratio, my numbers show that a restriction upstream of the turbo is very bad.

The point I was trying to make was that only looking at one number (temp, pressure, doesn't matter) doesn't give you the whole picture.

So making a statement that inlet temps are trivial, intercooler is key is wrong. There is more to it than that. We both know boost is a measure of restriction, that is important here! Your statement about "Losses of turbo efficiency might skew the numbers slightly" should be "will skew the numbers". It will. Anyone wanting to test this, go put a restrictor on the inlet of your turbo, or in the intercooler pipe, or in the downpipe, and report back. Air flows through the entire system, changing anything in the system will affect HP.

We're actually both oversimplifying it to be honest. We have not touched on what's happening at the turbine.

The whole point of my post was to show
1. As a system, changing one thing does change several things.
2. Some of the things that change are more important regarding HP than others.

I believe my post explains these two points, and that's all it was supposed to do. As I said, I don't have the software to model EVERYTHING that would be required to accurately calculate how much it would change hp.

But I do in fact have real world experience on this exact topic, and have done same day testing on this (hot air intake vs cold air intake) on a high boost turbo miata and confirmed that the SYSTEM flows more air through the engine with a cold air intake at THE SAME BOOST PRESSURE. So much so that I had to retune my fuel as the engine went lean with a CAI.

And to add to my last post, the REASON the car had to be retuned/flowed more air is because the compressor on the turbo was more efficient (required less HP) to supply the boost pressure commanded with the CAI. So since it needs less HP to drive the compressor, the turbine doesn't need to supply as much power to it, so the wastegate can open up a bit more (let more exhaust go around the turbine to reduce power to the turbine). This drops turbine inlet pressure, which reduces the backpressure on motor and flow goes up. Thus more air, more HP, thus why I had to retune fuel when I installed my CAI on my turbo setup.

I was only running around 17 PSI back then too, so it would have been an even bigger difference on a high boost setup.

I didn't ignore it, and a CAI would be more of a restriction... :P

lets take the intake tube completely out of the picture and only focus on the inlet temps.

its too bad Garrett and others don't provide turbo compressor efficiency maps at different intake temp values to actually see the effect on compressor efficiency; Garrett maps are done at 30°C intake temps.

Garret publish their test data at the temperatures they are designed to operate at? No OEM designs their turbo car to pull in hot air, they ALL have a CAI. Feeding it hot air is going to reduce its performance and lifespan.

You can take anything you like out of the picture. I'm looking at the engine as a system. As a SYSTEM, a CAI on a turbo car is important for the reasons I have stated.

yes. that's the math formula on their plots.

take the compressor map i was using:

https://www.miataturbo.net/attachmen...ine=1426453553

inlet temp = t1c = 545° 545-460=85°F/30°C
ambient pressure = p1c = 28.4 in.Hg. ~14psi
discharge pressure = p2c as seen in the pressure ratio equation on the x axis.

I already stated the MSM does not have a CAI (i dont think my 626 turbo did either), it sources air from inside the engine bay behind the headlight. granted there's a little hole to allow fresh air in, but it's not a cold air intake.

I don't disagree that having the lowest possible intake temps before and after turbo is a good thing, but I really think you're overstating how drastically important/beneficial it is. When you have a good working IC on your setup it eliminates most power losses between the extremes. Obviously you want to run the coldest air possible, but I don't think it's a make or break decision.

Fixed

https://www.miataturbo.net/attachmen...ine=1426458436

https://www.miataturbo.net/attachmen...ine=1426458436

Dann

What did you fix?

Increase intercooler efficiency with increased Delta T

Is intercooler efficiency usually in the 90's%?

Depends on a lot of different things.

yeah. that's probably a really good ic. just the number i picked, a good IC is probably somewhere between 80-90%. I wouldn't ever see very much over ambient boosting around over the years.

why would it do that?

IC Efficiency = ( compressor outlet - ambient ) / ( compressor outlet - air after IC )

I can see the efficiency dropping, or reaching diminishing returns, but not efficiency increasing as you add more heat to it. It can either exact the heat or it cant.

....Fuck dude.

Heat transfer coefficient - Wikipedia, the free encyclopedia

http://physics.tutorvista.com/heat/heat-transfer.html

Rates of Heat Transfer

First 3 links on google, and literally high school physics.


TLDR-
Heat transfer rate depends on the materials, the surface area of the materials, and the difference in temperature between the materials.



Between this and the torque thread, I seriously think you and Vlad need to stop acting so authoratative.

Dann

I'm not throwing insults at anyone, that doesn't help. I'm here to learn and help others. But I do agree that if you don't understand something fairly well, you shouldn't make statements as though you do. Because as you say, people read it and see who said it and just assume it's true. I know there's some other engineers on this forum, I wish a few of them would chime in and offer there experience/understanding on this.

You are correct about the heat transfer improving with higher delta T improving the effectiveness, but in your "corrected" pics you kept constant turbo efficiency which is a pretty large error, and what I kept trying to explain above. Do you understand why assuming it's constant is wrong?