Turbo Theory and Selection...why the 14b ain't bad

[mtnickel]

The 16g's do have 7cm^2 housings vs the 14b's 6cm^2 housing...
I was under the assumption that the 6cm housing flows enough to support about 300whp...but perhaps it plays a small role in power loss in our engines...meaning, past 5500rpm it chokes the engine noticably more than the 16g. That said, the 14b as a compressor wouldn't be "blowing hot air" but just causing a power loss from the small turbine housing (6cm = about .41 garret).

Ideally, i should do some real world testing...dyno with the 14b on...measure power, and intake temp...how much/close to detonation we are...

then swap directly out for a 16g and do the same.

[F22Turbo702]

No "real world" testing needs to be done, it's simple fact that a smaller A/R exhaust housing will choke down an engine in comparison to a larger A/R housing. Which attributes to the quicker spool time vs. loss in upper rpm horsepower. That's the reason we have the T3/T4 turbo, larger compressor housing for better power output, smaller exhaust housing for a quicker spool time. The peak power isn't as great as a full T4 turbo would produce, but the rpm range is much more usable on a motor that doesn't rev to 10k.
Price, availability, and condition are merely a decision making tool, they have nothing to do with the physics behind the turbo itself.

-Kyle

[ludeude1992]

I have a 14b for sale! hahaha

[mtnickel]

the only thing holding me back from getting one is the fact that the numbers dont' seem to add up in other cases...

Take straightedge for example... 270whp @ 9psi.

270whp = about 31lbs/min = 455cfm quite outside the efficiency map...and moreoever, doesn't even make sense with a 2.2L engine at that pressure ratio...
134cu in X 6000rpm x 100%VE (Optimistic...he has a cam) * PR (1.61) /1728/2 = 375cfm...

So A) we have a way better BSFC, or B) the cam is really doing wonders, or C) the math is not really THAT important.

The jury is sitll out for me and my 220-240whp...but if i could do it at 9psi instead of 11-12...i'd be a happier camper.

[ludeude1992]

get a 16g

[owequitit]

Just a couple comments, thoughts and observations here.

1) You are dealing with thermodynamics, so to some degree, the math is hypothetical, especially if it is just a convenient little plug and play formula.

Large companies like Boeing and Honda pay millions of dollars for computers that can't calculate this stuff to 100% accuracy.

It would be like a rule of thumb that passes the TLAR (That Looks About Right) check.

You won't be able to quantify for certain until you actually get the stuff on the car and tuned.

This is partly because of design differences (a head like the F22 might respond much better at a given pressure ratio than a 1950 OHV design), and also because of variables within each individual engine.

I personally have seen some pretty hellacious low boost gains from Hondas because of their flow capabilities. In some cases, they far exceed the 10HP/PSI rule.

Here is a good example. A new Accord V6 6 speed typically lays down around 215 WHP stock.

There is one kid that supercharged it with a Comptech kit (no intercooling, and less efficiency than a turbo), and at 4.5lbs of boost, he layed down 308WHP with a mild tune.

That is a 21 HP gain per lb of boost, or more than double that rule of thumb...

His buddy had the same car, only with an auto tranny. He went from around 200WHP to about 270WHP with no tune at all.

That is a 15HP gain per lb of boost... So even untuned, he saw 150% of that 10HP "rule of thumb."

So you can see that there are 2 real quick easy real world examples where that math doesn't work. There are many more, but I happen to like those two because of the car and engine.

It is designed to give you a general idea of what to expect.

2) With regard to the F22 powerband. VE does play a role in how the engine will respond, because that is simply a measure of how easily if flows air at a given pressure, in this case (NA) ambient or 1 atmosphere.

If you increase the given pressure, you will make more power by virtue of the fact you are moving more air, but you will still expect to see a mirrored power curve if everything else remains the same. So at 2 atmospheres (14.7lbs of boost) the engine's ability to flow air will still taper off after 5200 RPM, so you expect power to do the same. There are some variables there, like if the turbo can maintain a higher flow rate passed 5200, and we continue to increase boost slightly, then we can get more power longer, but with everything remaining constant, it should mirror the original power curve.

Now, I also know that you could take an H23 intake manifold and TB, and a stock F22A6 cam and bolt them onto an F22A1 and now you will have an engine that peaks closer to 6K, while pulling to 6800 NA.

How does that effect the use of a 14B?

3) I am not saying the 14B is a bad turbo. The DSM guys have proven that. They have extracted boucoup power from them. I have seen that too. But one variable that you might be neglecting is engine construction.

The 4G63 is stout to begin with, and it was designed to be boosted from the factory. The F22 was not.

That means that design wise, the 4G63 may be tolerant of things that the F22 is not. The ring landings are a good example. There are many things that cause stress on the ring landings. One of them is temperature.

The effect of a 9 degree increase in temp on a 4G63 may not be the same as on an F22. Plus, 9 degrees isn't as little as it seems for intake temps.

That would be an additional 10% increase over ambient if you are driving the car on a 100 degree day.

Temperature is almost always the enemy. Would that 9 degrees necessarily be bad? I don't know. But neither do you until you actually get some seat time on it.

Also, running the turbo down the efficiency map is silly if there is another affordable alternative, that will work better. What is the sense at forcing a 14B down to 50% efficiency, if you could get the 16G to do the same thing at 60% efficiency? Especially since there aren't a lot of other tradeoffs, in terms of power delivery.

It has been a long time, but I remember when I was going to boost (I may still), and I was looking at the 14B map, the F22 DID fall within the map, but it wasn't necessarily the best part of the map.

Just my .02

[mtnickel]

Just a couple comments, thoughts and observations here.

1) You are dealing with thermodynamics, so to some degree, the math is hypothetical, especially if it is just a convenient little plug and play formula.

Large companies like Boeing and Honda pay millions of dollars for computers that can't calculate this stuff to 100% accuracy.

It would be like a rule of thumb that passes the TLAR (That Looks About Right) check.

You won't be able to quantify for certain until you actually get the stuff on the car and tuned.

This is partly because of design differences (a head like the F22 might respond much better at a given pressure ratio than a 1950 OHV design), and also because of variables within each individual engine.

I personally have seen some pretty hellacious low boost gains from Hondas because of their flow capabilities. In some cases, they far exceed the 10HP/PSI rule.

Here is a good example. A new Accord V6 6 speed typically lays down around 215 WHP stock.

There is one kid that supercharged it with a Comptech kit (no intercooling, and less efficiency than a turbo), and at 4.5lbs of boost, he layed down 308WHP with a mild tune.

That is a 21 HP gain per lb of boost, or more than double that rule of thumb...

His buddy had the same car, only with an auto tranny. He went from around 200WHP to about 270WHP with no tune at all.

That is a 15HP gain per lb of boost... So even untuned, he saw 150% of that 10HP "rule of thumb."

So you can see that there are 2 real quick easy real world examples where that math doesn't work. There are many more, but I happen to like those two because of the car and engine.

It is designed to give you a general idea of what to expect.

2) With regard to the F22 powerband. VE does play a role in how the engine will respond, because that is simply a measure of how easily if flows air at a given pressure, in this case (NA) ambient or 1 atmosphere.

If you increase the given pressure, you will make more power by virtue of the fact you are moving more air, but you will still expect to see a mirrored power curve if everything else remains the same. So at 2 atmospheres (14.7lbs of boost) the engine's ability to flow air will still taper off after 5200 RPM, so you expect power to do the same. There are some variables there, like if the turbo can maintain a higher flow rate passed 5200, and we continue to increase boost slightly, then we can get more power longer, but with everything remaining constant, it should mirror the original power curve.

Now, I also know that you could take an H23 intake manifold and TB, and a stock F22A6 cam and bolt them onto an F22A1 and now you will have an engine that peaks closer to 6K, while pulling to 6800 NA.

How does that effect the use of a 14B?

3) I am not saying the 14B is a bad turbo. The DSM guys have proven that. They have extracted boucoup power from them. I have seen that too. But one variable that you might be neglecting is engine construction.

The 4G63 is stout to begin with, and it was designed to be boosted from the factory. The F22 was not.

That means that design wise, the 4G63 may be tolerant of things that the F22 is not. The ring landings are a good example. There are many things that cause stress on the ring landings. One of them is temperature.

The effect of a 9 degree increase in temp on a 4G63 may not be the same as on an F22. Plus, 9 degrees isn't as little as it seems for intake temps.

That would be an additional 10% increase over ambient if you are driving the car on a 100 degree day.

Temperature is almost always the enemy. Would that 9 degrees necessarily be bad? I don't know. But neither do you until you actually get some seat time on it.

Also, running the turbo down the efficiency map is silly if there is another affordable alternative, that will work better. What is the sense at forcing a 14B down to 50% efficiency, if you could get the 16G to do the same thing at 60% efficiency? Especially since there aren't a lot of other tradeoffs, in terms of power delivery.

It has been a long time, but I remember when I was going to boost (I may still), and I was looking at the 14B map, the F22 DID fall within the map, but it wasn't necessarily the best part of the map.

Just my .02

Pretty well said. The whole 10hp/psi is hocus though...eg...you think 10psi on a v8 is only gonna make 100hp more :P not likely. It's a pressure ratio...

So given we are at 1.0atm...the above 5psi is 14.7+5 / 14.7 = 1.35

215whp x 1.35 = 290whp = 75whp on 5psi...Again, strange that they don't add up to 308. my only logic is, NA the manifold sees some mild vacuum from frictional losses...etc.

consider it this way. if the NA engine was indeed at 1atm, then it should really make as much HP as the Absolute best setup intake. That means, intake manifold, Throttle body, runners...everything should yield around a 1.0atm at the valves...this rarely happens though.

what i'm trying to say is that the turbo can just work a little harder to compensate for the frictional losses, and guarantee an 8psi in the manifold.

i like what you said about the costs though.

16g's really aren't that much in the world of turbo prices. The evo3 16g is like $500us shipped...that's a hell of a deal for a turbo optimally suited for this engine (brand new too).
but for those on a budget setup and stock boost...i'd say the 14b is just fine as well.

[owequitit]

Pretty well said. The whole 10hp/psi is hocus though...eg...you think 10psi on a v8 is only gonna make 100hp more :P not likely. It's a pressure ratio...

So given we are at 1.0atm...the above 5psi is 14.7+5 / 14.7 = 1.35

215whp x 1.35 = 290whp = 75whp on 5psi...Again, strange that they don't add up to 308. my only logic is, NA the manifold sees some mild vacuum from frictional losses...etc.

consider it this way. if the NA engine was indeed at 1atm, then it should really make as much HP as the Absolute best setup intake. That means, intake manifold, Throttle body, runners...everything should yield around a 1.0atm at the valves...this rarely happens though.

what i'm trying to say is that the turbo can just work a little harder to compensate for the frictional losses, and guarantee an 8psi in the manifold.

i like what you said about the costs though.

16g's really aren't that much in the world of turbo prices. The evo3 16g is like $500us shipped...that's a hell of a deal for a turbo optimally suited for this engine (brand new too).
but for those on a budget setup and stock boost...i'd say the 14b is just fine as well.

I know the 10PSI rule is hocus, which is EXACTLY why I posted those examples. But still, it is commonly used as a general rule of thumb, and many people use it as gospel.

Also, you are approaching it as though I don't know how to do the math. And I do.

And yes you are right, that an engine is not 100% efficient, and you will never see quite 100% VE. However, in the typical Honda, you will be well north of 90%, and in some cases see very very close to 100%, so the difference can be negligible. Will you see this on a stock F22? Probably not.

You are also neglecting that the definition of VE is the % of total capacity being moved by the engine at 1 ATM.

That means that under boost, most engines see in excess of 100% VE.

Your claim of spinning the turbo faster to compensate, is exactly what I mentioned, but that is only effective to a certain point, and certain flow rate.

I would rather invest a little bit in a better turbo, than to put myself somewhere on the efficiency map that I don't have to be.

And yes, the 16g's are affordable, which is why I was going to opt for it instead. In many cases, I found one comparable to a used 14B for not much more.

[ludeude1992]

i got a 16g for 250$

[mtnickel]

You misunderstood my point.

the VE is the engines ability to get air into the engine.

I wasn't refering to that...more so the intake systems ability to add frictional losses and a pressure drop such that the pressure in the manifold is lower than 1atm say 1" mercury. The engines VE remains the same, but with the air being less dense at 1" mercury, a power loss ensues. This is where intakes add power to engines...they reduce resistance and allow manifold pressures to be closer to 1atm. They do not however affect the VE at all...that is left up to cams and the head.

if you notice my last post...i made no reference to VE at all. I'm fully aware of it's definition.


Mark

[owequitit]

You misunderstood my point.

the VE is the engines ability to get air into the engine.

I wasn't refering to that...more so the intake systems ability to add frictional losses and a pressure drop such that the pressure in the manifold is lower than 1atm say 1" mercury. The engines VE remains the same, but with the air being less dense at 1" mercury, a power loss ensues. This is where intakes add power to engines...they reduce resistance and allow manifold pressures to be closer to 1atm. They do not however affect the VE at all...that is left up to cams and the head.

if you notice my last post...i made no reference to VE at all. I'm fully aware of it's definition.


Mark

As am I. And yes, something like an intake reduces "pumping losses", but it WILL have an effect on VE.

There are many many things that effect VE, not just head and cams.

An engine is a complete system, and must be considered as such. I think the disagreement is your consideration of VE is the theoretical capability of the engine, whereas mine is the total VE including all losses associated with plumbing.

Since you know what VE is, you know that increasing the total amount of air entering the chamber even slightly will have an effect on VE. It doesn't matter how we do it, it just matters that it happens.

Really it is a moot point. I don't know why I thought you mentioned it a second time, but you are right. It never entered into your response.

You are trying to define terms that we both know, which really have nothing to do with the efficiency mapping of a turbo.

If an intake manifold creates a pressure drop for flow reasons, that will still most likely be there, even in the presence of boost, which makes arguing about it silly. Also, the 1" or smaller pressure drop in the IM is mostly negligible.

I am not sure exactly what your point is apparently.


Yes, you can spin the 14B a little faster. You can also put a 16G on it.

One will result in slightly quicker spool up with less head room, and one will result in keeping the turbo in a slightly better area of its compressor map.

Having looked at both maps and compared them to the flow rates of the F22, I would choose a 16G, but that is just me.

[mtnickel]

ya...nuff said on that matter.


if someone was to hold out the 2 turbos, i'd take the 16g in a heartbeat...it's a cost/performance thing.

My only point was to be that, "theoretically" (and this is a BIG theoretic) the 14b is nearly usable out to around 250whp on our engines while still being in the efficiency range of the MAP.

soo, if you're only running 220 or under, i would see very little reason to spend 200 more and get the 16g (if you don't plan on upgrading, etc).

After all my engine work and whatnot (hypereutectic pistons..and just finished head port and polish...also have h23 intake and whatnot)...and my goal or 240-250whp...it's borderline to run the 14b as the OPTIMAL turbo...meaning, there are better suited turbos (mainly the 16g).

So, i'll probably just rebiuld the evo3 16g i've got here. Damn thing has some scoring on the compressor housing. wheel was rubbing a bit (on the backside). Might need a new wheel, and will definately repair the housing...some sort of filler.

This brings me to another interesting topic.
My genius car friend of mine who knows everything and everything about cars thought about adding some relatively soft compound (relative to aluminum) to the compressor housing...then spin up the turbo to hone/cut the excess material off...The resulting wheel fitment would have very good wheel to housing clearances...and the better this clearance gets, the better the efficiency gets.

I thought it was quite interesting, so i googled it and aparantly there are patents on it. Some porsche guys were having it done, etc.
Patents pertained to the idea as well as what compound to use, etc.

So...considering my housing is a little scraped up, i thought it'd be cool to do.

haven't decided on the material yet. I was thinking a fibreglass bondo or something like that. Some sort of semi-hard epoxy paint. i donno.

Just has to be softer than aluminum.

Carving would be acomplished by attaching 10mm 12point socket on turbine wheel and attach to a drill and then lower the housing more and more...it should fit on nearly all the way, so wobbling shouldnt' be a huge deal.

If i do it, i'll be sure to DIY it for your viewing pleasure.

[ludeude1992]

the truth of it is!!! Reguardless of what anyone says the 14b and the 16g are both good turbos!

NOW i would choose a 16g because it spools up a bit slower than a 14b. I want this because of traction.

I was rocking a t25 and i had NO problems...BUT i had instant boost and it was rediculous..it also didnt pull to redline.

My ported small 16g pulls right up to redline and have a very nice power/tq curve. Ive never used a 14b..but its bigger than a t25 and thats all that matters. I would assume it would make power quicker..which would be nice at the track if you had slicks...

[SN2BH22CB7]

thank you all....from observing this thread..Hmm..I know which Turbo to go with after I do my H23/F22 hybrid!

NOT TELLING...yall would just have to wait and see!

WEEEEEE... ( Gathering MORE parts )

Its a SECRET..not telling what I ALREADY HAVE