If air restriction was lessened, and that is, more air was able to enter the engine at a given pressure, then boost would be reached more slowly, right? That is, for a given amount of airflow, with the lower restriction air intake, boost should be lower. I would think that the less the restriction, the lower the boost, so that boost might build slightly slower.

The best way that I can put it is that, if the engine is getting more air for the same throttle level, wouldn't boost then be lower? Since boost is really only a measure of how much air isn't getting into the engine. Of course, it would make more power at less boost and thus feel like boost was building faster. Or am I missing something?

 

^ not true at all. the reason is that any air that enters the engine also exits the exhaust, thereby spinning the turbine. the more air you shove in there, the faster the turbine spins. in fact there is a certain amount of CFM for any given engine/ turbo combination at which the turbine begins to push enough air that it sort of causes itsself to spool, which is why you have lag, then a sort spool time, then full boost. spooling is sort of an exponential curve, because the more air the turbo forces into the engine, the more exhaust there is to spin it faster.

it really comes down to the fact that restriction is ALWAYS bad for WOT acceleration. any restriction or flow resistance you must overcome is just wasted power, because the system must displace the air, and it takes energy to do that.

not a really good explanation, but hopefully it makes sense

pat

 

That's not the only reason. The center rotor also gets more heat from both sides of the engine making it naturally run more lean.

 

I still don't quite agree with you here. IIRC, RotaryGod has said the same thing about intercooler piping. (That it doesn't do you much good to run larger intercooler piping than your turbo outlet)

Let me bring up an example here...I have a GT40R on my 13B-rew...For those of you that don't know, the outlet is 2". Therefore, if I ever max out the 2" outlet on the turbo, that's it...Period. What will 3" intercooler piping do for me? There's going to be a restriction at the compressor outlet, and if you choose to enlarge the piping after the turbo, I'm pretty sure it's not going to increase the power. If you can provide some sort of proof via a dyno chart, and prove that only the intercooler piping has changed, then I'll be more than happy to take back my statement.

So now, in the case of a 90mm TB, and a GT40R, it's just pointless IMO. You have a 3.5" TB, with a 2" outlet on your turbo. My turbo will never be able to flow what the 3.5" TB can.

Sure 3" intercooler piping would be less restrictive than 2", but it'll also decrease velocity, and increase the amount of time it takes to build up pressure. (boost) You just have to find a happy medium, which I feel is the compressor outlet.

Something else to think about here...If Garrett felt that larger intercooler piping is better, then why did they choose to make the GT40R with a 2" outlet, and not something like a 2.5"-3" outlet? Probably because the GT40R will max out close to the same time 2" intercooler piping will, therefore anything larger than 2" intercooler piping is useless!

As for your statement, that in general, bigger is always better, is complete BS. Just because I can run 4" intercooler piping, doesn't make it any better. Just because I can run an intercooler core size of 4.5" x 22" x 12" doesn't make it any better than a core size of 4.5" x 16" x 12". I've spoken with Gerhard Schruf, over at Bell Engineering, about this in great detail. I just hope your statement wasn't ment to be as broad as you made it!

As i've stated before, it's all about your setup. There's no single way to do things between diffrent setups. Some setups may call out a larger intercooler, or larger intercooler piping, but that doesn't mean that another setup will benefit from them.

-Alex

 

That's pretty much it Alex. I've gotten into the same type of debates when it comes to exhaust system size. People will say that a larger pipe can flow more air than a smaller pipe. This is true but if the smaller pipe can flow all that your engine needs, anything larger is a complete waste of time. The key is to go only as large as you need to and no larger as anything larger than what you need is bad. When someone asks if a pipe CAN flow more than another one, ask them if it actually WILL. The responses are usually pretty entertaining. You only need larger if what you have is too small for your needs. That's it.

If you have a 2" outlet on the turbo, you do need a larger throttlebody as the actual throttle plate(s) do restrict flow a little bit when fully open. This doesn't mean that you need a honking big 90mm unit. How much total area is in all the intake runners combined? Remember that air pulses through them as well with each pair in a 2 rotor having alternative pulses. Once your throttle plate flows all the the turbo outlet can flow, you go no higher. Larger does not help one bit. What you will do in the case of intercooler piping is to create more area for the turbo to pressurize before it does your engine. I know that some people say that the small amount of extra air isn't much more when it comes to lag time but any amount more than I need, right down to the millisecond, is unacceptable as far as I'm concerned.

Another thing that many overlook is the fact that speeding up and slowing down air takes energy. This even means changes in piping size as well as going through the intercooler and into each intake runner. It all uses a little bit of energy. The best rule to go by is to keep it as simple as possible. This means that you should try to keep the total area within the airflow path as close across the whole distance as possible. This is obviously not going to happen when you get to an intercooler but also remember that an intercooler does have a pressure drop. Fortunately they more make up for this in charge cooling.

Air is funny stuff.

 

Well, after the last statement from RotaryGod, among others I've spoken to, I feel we can start to sort out some facts to help guide people when they're sizing a TB, and intercooler piping:

1.) Try and run the same size intercooler piping as on your compressor outlet. Running anything larger won't give you any sort of benefit.

2.) When sizing a TB, choose one that's one size larger than your intercooler piping to compensate for the butter fly valve and shaft, unless you're using a slide throttle setup!

-Alex

 

This discussion is really shaping up! Re. IC piping size, when I spoke to Bell about selecting a size for my project (600rwhp<) they encouraged me to use 3" rather than 2.75. They said 2.75 would proably be fine for a straight pipe, but is less than ideal when the pipe has bends in it.

 

I thought the GT42R had a 3" outlet?!

-Alex

 

FWIW, I laid all the irons out on the CMM to figure out the surface area of the ports on a 2 rotor 13B-REW:

Secondary ports - 2.1025 square inches each

Primary ports - 1.1266 square inches each*

* - Square inches slighty off due to gouges/damage around the port

If anyone has ever looked at the ports on our engines, they sort of "flare" outward right before they get to the surface that the intake manifold bolts to. I'm willing to bet that if I took measurements before the flare, and on a primary port iron without damage, that the primary ports would actually be half the surface area as the secondarys. Kind of interesting IMO!

Anywho, back to what I was getting to...The primary TB, on the stock 2 rotor 13B-REW TB, is 45mm (1.7716"), which is 2.4637 square inches. From the measurements I got above, which are a bit off, the surface area of the primary ports combined is 2.5384 square inches. IMO, if you take out the "flare" on the primary ports, and get a nice reading off a better conditioned iron, the stock primary TB will be just a bit larger. I'm also willing to bet the combined area of the primary ports, and the area of the primary TB, would almost be exact if you compensated for the area the butter fly valve takes up!

Now, on to the secondary ports...On the 2 rotor 13B-REW TB, each secondary TB is 50mm. The area of 50mm (1.9685") is 3.0418 square inches. Now, each primary port measured 2.1025 sqaure inches. If you took the measurements before the "flare," it would come out to less. Then if you compensated for the area the butter fly valve takes up, they would come out pretty close to the same. However, I feel the secondary TB's would still be a bit larger than the secondary ports.

I'm not sure what the sizes of each TB on the 3 rotor are, but if they're the same, then I would say the stock TB is to small. That's why people see an increase in power with a larger TB on the 3 rotor. I know most of you go ahead and port the engines as well, so your ports are even larger than stock, therefore creating even more of a restriction at the TB. IMO, I would size the TB to the area of the ports on your engine. Then, I would run intercooler piping the same dia. as your compressor outlet on the turbo side of the intercooler. On the intake side of the intercooler, I would run the dia. of the TB, or a dia. that has the same area as your ITB's combined. This would keep you from having to make a transition.

I would love to hear your opinion on this RotaryGod!

-Alex

Edit: Here's some pictures showing the irons being laid out on the CMM, and the damage on the one iron. The last picture also shows the "flare" i've been talking about. The "flare" is found on all of the irons, but it's more significant on the primary port iron.

 

i know i am not much good at relating what i mean in type, but you guys should consider reading up on fluid flow a little bit. Youre right, my statement was way too broad, thats true. I assumed that it would be taken with a little bit of common sense to mean that if you are deciding between 2 sizes, bigger is better, not that everyone should use 6" piping for everything.

as far as IC piping, you guys can argue whatever you want, but think about it this way: the single factor that makes some intercooler setups much better than others is pressure drop through the system. Pressure drop is caused by fluid drag on the walls of the piping and IC tubes. The less pressure drop you have, the better. period.

Now, as far as what RotaryGod said about bigger tubing flowing more air than is needed; this is not the issue here. The flowrate into the engine is going to be within a general range, no matter what size pipe is used, so youre right, the capability to flow more air isnt necessary.

what you are overlooking is that a bigger tube has the capability to flow a set amount of air WITH LESS RESTRICTION than a smaller tube. That is why you use bigger pipe, becuase the less restriction you have, the less pressure drop there will be, which makes the system more efficent, and nets you more power in the end.

there is of course a limit above which it makes negligible difference whether you increase the diameter any more. 12" pipe is not goign to net you any more power than 4" pipe on an engine that flows 650 cfm. but 4" pipe will net you a very significant increase in power over 2"

also, to everyone in general, PLEASE stop using the argument "if it should have been different, the manufacturer would have designed it that way" look at what this thread is about, and then look at that statement. if you dont at least crack a smile i'll buy you a coke.
the idea that OEM designs are infallible and perfect in every way is the stupidest thing ive ever heard. if that was true, why would you modify anything? why would there be recalls? why would there be a whole multimilloin dollar industry in aftermarket parts?

you guys arent far wrong, but i really do think that if the blanket statement of this thread is:
1.) Try and run the same size intercooler piping as on your compressor outlet. Running anything larger won't give you any sort of benefit.

2.) When sizing a TB, choose one that's one size larger than your intercooler piping to compensate for the butter fly valve and shaft, unless you're using a slide throttle setup!

then it should be changed to use one size larger IC piping as the compressor outlet. I can prove that to you mathematically, I can get you statements from engineering professors, race car designers, you name it. It is just a simple fact that I cant believe anyone would argue with.

pat

 

I agree patman, just to piggy back on it. The FD throttle body is fairly large, and I doubt that its anywhere near the combined size of the two teeny tiny outlets on fd turbines? Im running a T88 with a 2.5 inch ID outlet. ill run 3inch piping on the hot side of the I/C and then step it up to 3.5 for the cold side to mate up with the TB.

 

.

 

Sorry for the above post. I had one wrote out, and decided I wanted to change it all, but I had to get in the shower.

I never said anything about running 6" piping, but that's really besides the point...

The two cores I listed in my example above were the two I was debating on. After speaking with Gerhard Schruf at Bell Engineering, he assured me that with a side to side core with that depth, and that height, going any wider than 18" wouldn't do any good. Pull out all the math you want, because he already did, and it clearly showed no point in going larger. Therefore, this disproves your theory that when debating between two sizes, go with the larger of the two. Period.

Edit: I just noticed I said 16" in my previous post, and it should have been 18".

It may work sometimes, but not always. There's more to a setup than simply going with the largest...

Sure, less pressure drop is better, your 100% correct. However, this being the single factor that makes one intercooler setup better than another is false. There's much more to factor in than just pressure drop, and there's other things you can do to increase performance from your intercooler, even with a little more pressure drop. There's intercooler size, intercooler type, end tanks, ducting, etc...

Less pressure drop will always be more efficient. Atleast we can agree on something here!

However, there's draw backs to larger intercooler piping...The larger the intercooler piping, the more time it's going to take pressurize the intercooler piping, therefore increasing the time it takes to start building power. Some say it doesn't make much of a difference, some say it does. IMO, it does.

So now your increasing efficiency, while decreasing responce. What you have to look for is something in the middle, not one or the other.

Edit: Depending on what the car is used for will actually depend which one to choose as well. In other words, a drag car doesn't worry about lag, where as a road racing (road course, autox, etc.) does. So If your car is designed for one specific application, you should/can pick one or the other. If you have a street car, I'd find something right in the middle.

Now, as I've stated before, it all depends on your setup. For example, in my setup, my turbo literally sits a 90 degree bend away from in intercooler. Trying to enlarge my piping would be retarded, and have zero benefits.

Now if it was a bend, and a few inches, or even a few feet or so of straight piping, I'd still say to stay with the same dia. as the outlet on the turbo.

However, most 3 rotors seem to run with a FMIC. This increases the amount of bends, and length of piping. Here I would say it would be a benefit to increase the piping to the next size up that's available.

Bingo!

Here's the point I'm trying to bring across. (Can't really speak for RotaryGod)

Lets use my example here again...The outlet on the GT40R is 2". Lets say I'm running it at a power level that would create a flowrate of ~half of what 2" can atually flow, then there's no point in upgrading to larger intercooler piping. That 2" piping isn't breaking a sweat, even with a FMIC, and several bends.

Now, lets use the same turbo and outlet size again. Now I'm running a power level that would create a flowrate of ~80% of what the 2" can flow, and I have alot of bends, or long piping, then you would benefit from larger intercooler piping.

If you've noticed, I've always been trying to say that there's not much of a point in running larger intercooler piping than the size of the outlet on the turbo. If I've slipped up, and said there's no point in doing so, then I'm sorry, and that's incorrect. The point behind wording it the way I did, is to bring across that it all depends on the setup. I would assume RotaryGod worded it this way for the same reason.

I agree with you 100%, but I think my statement was taken a bit out of context.

You can't really compare an OEM design, lets say one of Mazdas, to a Garrett design. Mazda has restrictions, and costs to think about. Within reason, Garrett doesn't. I've never heard of a recall on an intake manifold, or turbo, because the outlet/inlet isn't the correct dia. Therefore, IMO, if Garrett was making a turbo that can handle 700fwhp (GT40R) they wouldn't use a 2" outlet if it couldn't flow the amount it takes to reach 700fwhp. I don't really see how you could argue with this...

However, Garrett can't compensate for how much piping your going to have, or bends, so AGAIN, you just have to look at your setup, and apply the information in this thread to it.

Excuse me, I jumped the gun, and went against what I stated previously in other posts...Each setup is different, requiring different things.

I beleive the TB sizing is a pretty close guideline to use. However, if you want to go more in depth with what size TB to use, you can use the information I posted about the surface area of the ports on the engine and so on.

As for the intercooler piping, I think statement that if you don't have many bends, or an extreme length of intercooler piping, stick with the dia. of the compressor output on the turbo, is correct.

If you have alot of bends, and/or an extreme length, you may want to consider going with the next size up. To determine wether or not to go with the next size up, figure out how much your turbo will be flowing for your hp goal. Then figure out how much the dia. of the compressor output will flow. If your close to maxing it out, and have alot of bends, and/or an extreme length of piping, go a size up. If your close to maxing it out, but only have a bend or two, and not an extreme length of piping, then keep the dia. the same.

As for engineers, race car designers, and whatever else goes, I'm sure most everyone would agree with what has just been stated. In case you don't already know, Corky Bell and Gerhard Schruf are both engineers, and race car designers, that have worked with and on race car teams. I already know they back up what I've stated, as well as a Garrett engineer, and Spearco. (Turbonetics Inc.)

-Alex

Edit: I hope we can start to agree on this so we can get back on the main focus here...The intake manifold!

 

Well, the only real way to know is to measure them, and find the surface area. IMO, the FD uses 2.75" piping to compensate for all the bends, length, and how the crossover pipe goes flat. I bet the outputs on the factory turbos is ~ equal to the surface area of a 2.5" pipe dia. This would support what's stated above, and further support my findings on the ports surface area compared to the TB surface area, and how well the intercooler piping dia. matches both!

The problem with most people is they just assume, and take the words of others. I don't expect anyone to just simply read what I post, and beleive it. If you want the perfect setup, take the time to do the math. After you do that, use what you come up with to find out which opinion you agree with, and apply all of it to your setup!

-Alex

 

w3rd

 

-Alex

 

i still dont really agree, but lets just agree to disagree about it.

as far as the manifold, i just think the stock one is not well suited to higher than stock power levels, and my one question is is there anyone on this forum that can get us a small scale production run of a custom manifold for a reasonable price? that seems like the best way to go. i am planning to just weld one together myself, but if someone was producing a nice one, i'd take it just to save the time and trouble.

i have access to a waterjet, and could have all the parts CNC cut if anyone wanted to make a sheet metal manifold..... I just dont have my own TIG, so welding more than one together isnt really an option for me on borrowed equipment...

i could probably get a decent price from the shop to make them, if someone wanted to help me out on the design (i can do the CAD, and a portion of the flow analysis, but if i was going to produce them i'd want to know it was pretty damn good...) and enough people wanted one to make it worthwhile...

pat

 

Fair enough! At this point it may just be easier to provide the math for people to come to their own conclusion. There's no single answer to this problem.

In case you dont know, I started the machining on Rich's (AutoIllusions) first intake manifold, and will be machining his new design as soon as I find some free time. I have all the bolt locations for the 20B, but not the size of the stock ports.

I've been speaking with CCarlisi about doing a small run on the UIM, but would want 5-10 people. However, it wont be fabricated, nor machined runners. The best way to do this is with a casting. The intake flange, runners, plenum, and TB flange would all be casted in one. The TB flange would be casted so that later on someone can enlarge the opening, and I can provide a TB adapter plate.

There's several issues with casting the UIM/LIM. If you'd like to discuss them, shoot me a PM with your e-mail addy, and we can talk about them...I'll post them later on for everyone else to see as well, I just don't have the time to at this exact moment!

As for cost...I would probably require enough money up front to cover the machining of the mold, then the rest would be required once they're casted. As for the actual cost of the intake manifold, it depends on wether it's the UIM, LIM, or both. Also depends on how much it'll cost to cast them and so on. Personally, I'd rather do both, and at the same time. This rids of most of the problems in which I was speaking about in the paragraph above. However, this requires more money up front from individuals.

I personally own 4 3 axis cnc machines, 2 4th axis cnc lathes with live tooling, and 1 cnc wire EDM. I think I can provide any and everything needed to do the project. However, right now I'm over loaded with military work, and I have a project for CCarlisi to finish up as well.

I can do the drawing, programming, machining, and maybe the flow analysis. I'd have to check to see if I renewed that option with SolidWorks. There's always a way around that if not, and I'm sure you know what I mean!

-Alex

 

well hell sounds like youre all set!

i agree that a casting is best, i just figured it would be harder for someone on here to find a place that would do it on a small quantity basis. sounds like you probably know some people tho...

my personal idea of the ideal manifold would be a single piece that replaces both the LIM and UIM. I'd like to see 25-40% larger runners (ID) and about 2-3 inches lower profile for hood clearance (i already cut mine, but others shouldnt have to, and also i'd like the room for a strut brace right there). I'd also like to see it still have a decently sized pressure chamber, which ive noticed a lot of custom manifolds dont have.

I realize that the one-piece idea is quite a bit harder to cast, but it would be sweet. 2 piece wouldnt be much worse tho. also im guessing you probably already have all that pretty much figured out... as long as i wasnt broke at the time, i'd definately be in for one when you get to it. if theres any way i can help out with the design or anything, shoot me a pm, i'd be glad to contribute.

pat

ps, i'd rather not put a bunch of math on here, every time i do everyone just skips over it. i know i do a lot when other people post it too.... if you'd like to talk math via pm, that would be fine tho

 

wow I definetly agree with you pat. You seem full of useful knowledge.
Regarding the diameter of the IC pipe. The same ideas should apply to electricty. since the fuse block uses a 10amp fuse, probally like 20 or 22 gauge wire, you should be able to run a 22 gauge wire all the way back the tail lights to power 10amps. Doesnt work this way right. There is a "bottle neck", but resistance is length and diameter. Im sure theres a formula for electricity. Same with hydrodynamics. I assume, well it makes sense to me, that they all are relevant to each other. Im using 3" pipe, and 4" exhaust. BTW corky bell says that a 4" exhaust flows about equal to 2x3"s. Confused me, but thats corky bell. Didnt think it be that much of a difference.
About the TB, Im wondering how the q45 is driveable, its a stock part, yet were all having problems figuring out how its tb cant be streetable on 20bs. If a supra (3.0l) can run a monster like those 90s and 100s we should. Not that Im gonna take this on anytime soon.
Im quite happy with the stock one for now. I think its going to the recyclers when Im ready to mod the intake. Anyone seen that 20b with the log type manifold? It looks like a supras. Super clean. I think thats one of the most practical designs yet. Anyone with a tig and a saw can make it. those manifold are $$$ though, even for ebay knock offs.

 

Although I still don't think this will change your opinion, I decided to post this for others to see...

I've spent the past 2 hours finding every photo I could of Garrett sponsored drag cars, along with manufactured turbo kits from Turbonetics, Greddy, HKS, APS, Rev Hard, Drag, and a few others that I can't even remember. There was several photos that I couldn't make out, but everyone I found that was worth a damn showed the intercooler piping, between the turbo and intercooler, being the same dia. as the output on the turbo. These drag cars/turbo kits ranged from short length 1-2 bend intercooler piping setups to extreme length, several bend setups.

However, there was something I found quite interesting...Alot of the Garrett sponsored drag cars had larger intercooler piping between the intercooler and TB. The piping between the intercooler and TB always matched the dia. of the TB. The TB was almost always larger than the turbo output as well. This does follow what RotaryGod said though...Use a TB that's one size larger than the intercooler piping to compensate for the butter fly valve. Since most of those drag cars are running turbos around the size of the GT42R and larger, the outputs on those things are like 3.5". The next size up TB would be a 4", therefore they run 4" intercooler piping between the intercooler and TB so they don't have to make a transition.

I noticed a few of the turbo kits did the same thing, but most still use the same dia. intercooler piping after the intercooler, and used a transition coupler at the TB to attach the intercooler piping. Some of the smaller displacement engines didn't need a transition coupler, as the TB was already the same size as the intercooler piping.

I did find something I thought was kind of funny though...Just about every custom turbo system I found for street cars, and weekend drag racers, had as large of intercooler piping as they could fit.

-Alex

 

A 4" dia. has a surface area of 12.56 square inches. A 3" dia. has a surface area of 7.065 square inches. Therefore, a 4" dia. pipe has just about as much surface area as two 3" dia. pipes. (Two 3" dia. pipes has a combined surface area of 14.13 square inches.)

Want to know something really funny though? A 6" dia. has a surface area of 28.26 square inches.

-Alex

 

^ yeah, and also flow through a pipe is related to diameter to the 5th power. so a 4" pipe has 1.8 times the cross sectional area, but flows much more than that. In any given situation, a single pipe with the same cross sectional area will flow Much more than 2 pipes.

what you say about the hot side being small, and the cold side big makes sense, because a diameter change cuases turbulence, so what they are doing is combining that with the turbulence already caused by going thru the intercooler, so that you get better flow on the cold side, and less overall turbulence in the system. In fact, that turbulence is the only reason you wouldnt want to adapt to a bigger size right at the turbo, i was thinking about doing the math to see if the reduction in flow caused by the turbulence might actually be greater than the increase caused by the larger diameter pipe if it was a short tube length. if it was long, then it wouldnt be, but for a 18" pipe, its possible.

 

As with anything that is explained over the internet, many things would need to come with a disclaimer that was longer than the actual post. For the most part the numbers I give are really guidelines to get you close. You won't go wrong following them but it's not to say that they can't be tweaked a bit for a little better number. As has already been mentioned, number of bends and how sharp they are can really affect things. The sharper the bend the larger I want the pipe to be (to a point). Going up or down 1 pipe size isn't that big of a stretch and nothing I consider significant. As Patman said, it's not like we are talking about going from a 2" to a 12". That would be ludicrous. For the most part I like to keep everything as close as possible in area throughout the entire system. Yes there are times where you want to deviate from this a small amount but for the most part it's safest to just follow that rule as you can't go wrong that way. Just remember that all the math and all the theory in the world only mean so much. Many things happen in practice that weren't accounted for on paper. That's why many things should be looked at as guidelines rather than absolutes. You need to start somewhere. You just tweak from there. A big mistake that people make is to assume that all their design work is going to come out perfect in practice. That's usually not quite the case but hopefully you got close.

 

^ i agree.

judging by some of the designs I have had to fix to make them work, some people are lucky to even get close. I have worked on projects where the cost of modification of a unit is more than the original cost of manufacture. that dosent mean that you shouldnt do the math tho...