H22 turbo build from the bottom to the top

[kratemertyl]

finshed the book and fel like i got a good understanding of compressor maps, trouble is im having a hard time finding any precision turbo maps, i emailed one supplier asking for a few so i jus will see if they respond.

as far as turbine sizing goes, it looks likes im in need of the exhaust manifold pressure of the h22. ill google it after this, jus wanna be able to compare.

and,
is split inlet housing the same as v band housing?

[toycar]

finshed the book and fel like i got a good understanding of compressor maps, trouble is im having a hard time finding any precision turbo maps, i emailed one supplier asking for a few so i jus will see if they respond.

as far as turbine sizing goes, it looks likes im in need of the exhaust manifold pressure of the h22. ill google it after this, jus wanna be able to compare.

and,
is split inlet housing the same as v band housing?

V-band is not the same.


V-band is specific to the hardware/flange and how it seals up. Instead of having a | | shaped flange it has a \ / shaped flange. A v-flange. The edges are tapered and it does seal better.


A split inlet, or twin scroll manifold divides the collector for faster response to the exhaust. It keeps higher pressure in the collector/turbine housing and has individual runners/banks and the manifold is designed to have repeated exhaust pulses alternating in either bank for better pressure delivery in the turbine housing.


Again, different applications for different outcomes. All of the above have their pro's and con's


What are you trying to accomplish? So what compression piston are you going to run to best do that?


How much cfm does your engine now produce at what rpm that you plan to spend the most time in? Are your piston speeds acceptable at that engine speed with your rod to piston angle in mind? So what size manifold do you need to fabricate to accomodate that with what length runners? Now that you figured that out, what size turbine housing will best suit that amount of airflow?


With that much airflow, running through that turbine housing, what speed is the compressor thats attached going to spin at? Can your cam's run the valves appropriately to allow that much airflow?


Now you can find injectors to fit that volume of airflow and also, get a map sensor that can read that much pressure.

Then, you'll need some electronics to manage all of this.



God, I should just write a book. I mean, for christ sakes you read some published book, thats highly recommended and from that you gathered a v-band was the same as a split inlet?!!

WOW.

[kratemertyl]

take it easy there hero, no where did i state " hey, i read a book and all the info i came up with was this....." the autor doesnt even mention anything about v band flanges in his book. i appreciate u explaining it, very str8 forward, got it.
he speaks breifly about split inlet housing and its intended purpose, and as im looking up manifold designs i see alot that say "v-band" and i was thinking that if you have a split inlet housing then you should also have a split outlet on the manifold. so was jus makin sure termanolgy didnt change over the years

calculated airflow @4500 rpm on 20psi boost is 366cfm/ stock is 155 cfm
are you saying i can use that number with the compression pressure and burning gas pressure and come up with my manifold pressure?

i was wondering what the stock h22 manifold pressure is so i can calculate preferable turbine and housing size.

im not sure what the compressor speed will be but at that airflow rate of 610cfm @7500 rpm the turbo will be kicking 137% more air into the motor. can the stock cams effectivly manage the valves there, idk. some say yes some say no.

hopefully your book comes out soon though so i can figure it all out.

[cb7 calling]

valve overlap on a turbo car can be a great thing. Not sure what the shit you guys are reading, but, overlap on a turbo car is all about managing cylinder pressure.


No overlap will build boost later, and choke on a higher RPM. Its all in your design/intentions as to what will be best for you. Have fun trying to rev to 8,000 rpm's with no valve overlap and any pressure at all coming out of that compressor. The same engine that rev's to say, 10k rpm's will build boost slower than one that chokes at 7,000. Thing is, cfm's power turbos. Your engine produces cfm's based on a few things and one of the most important is rotational speed. You gain a lot of cfm's by being able to spin the motor faster, and more cfm's equate to more exhaust pressure to power your compressor.


So really, it just depends on what you are doing. For lower end performance, less overlap is ideal. For a dyno queen, 600+whp monster, welp, your going to need valve overlap to flow that much air and thats all there is to it. This is why k series motors are making so much power actually. Being able to adjust the cam timing in addition to the valve timing allows for ideal management of cylinder pressure. Its all about cylinder pressure guys. 100% about cylinder pressure. If you cannot make pressure in the cylinder at whatever rpm you want the engine to perform at, you won't make dick for power.

End of story. I wouldn't run OEM cams for anything turbo charged if you were really doing a "top to bottom" build.

My thoughts.


***Edit***


Pulled from dsm tuners and considered reliable/accurate information;


Advancing Intake and Exhaust : This will provide the car with more bottom end power, and will decrease top end. Advancing both cam gears will move overlap earlier but will not increase it.

Retarding Intake and Exhaust : This will increase the cars top end, but will decrease low end. Retarding both cam gears will move the overlap later and but will not change the amount of overlap.

Advance Exhaust Only : This will help the cars top end, and it reduces overlap.

Retard Exhaust only : This will help the cars mid range power, very useful for cars with big turbos / big cams. By increasing overlap, It decreases lag significantly. Doing this will bring the boost on all at once. Very common DSM modification.

Advance Intake only : This will increase overlap and helps the cars bottom end and mid range power. This mod will bring the turbo on all at once, although isnt a very common mod for DSMS.



Wether you advance the intake or retard the exhaust, increasing valve overlap = power on most turbo cars. Especially dsm's. There is obviously a threshold, but, valve overlap is a good thing on turbo cars. It is.

Thread link;

http://www.dsmtuners.com/forums/cyli...-degreing.html





And just so its out there, most people install 280 cams +1 advanced for intake, -3 retarded for exhaust for the best performance. Valve overlap is your friend in forced induction. You don't want the intake valve choking the compressor. Common missconception about turbo motors. You let the turbo power the motor. The motor doesn't power thet turbo. The motor needs to breath freely under peak boost which is why valve overlap matters so much on a turbo engine. The air coming into the cylinder has already been compressed.


Let the turbo control air flow basically is what I am saying.

I am not the most knowledgable but I thought we should mention when it comes to cams it's all based on the physics of air.

I have never built or tuned a motor. One thing I know is that advancing and retarding exhaust and intake cams designed for turbo is not creating the kind of overlap you would see on a f20c or even an h22.
You say it's all about cylinder pressure, yes it is. The whole need for variable valve timing, vtec all of it is the physics of air. With no air you have no cylinder pressure.Air compresses and stretches in the cylinder which we all know but more importantly when discussing cams it stretches and compresses in the intake/exhaust manifolds. Air does not simply displace the space a piston once occupied it takes time. As piston speeds increase the elasticity of air becomes more of an issue. In an NA motor If there where no valve overlap at 8000 rpm the air would only stretch in the intake manifold and not actually get pulled into the cylinder so as the piston reaches and passes bdc the intake valves stay open and like a rubber band the air still gets sucked in. That same amount of overlap would not even allow the car to run at a slower piston speed like say at idle. That's why if you have big high rpm cams on a non vtec motor it's all blip,blip....blip,blip......blip,blip. Different valve durations create power at different rpms because its all dependent on piston speed no matter which manifolds, head and block you have air still does the same thing it rubber bands it's way into and out of the cylinder.
Now having never built any motor I'm guessing from this point on.
In deeves, and that book he referenced defense on the intake side with forced induction valve overlap is not much of an issue. The turbo pushes the air in as soon as the valve opens and only some of the physics relies on the cylinder pulling the air in. I believe most of the benefit of overlap with a turbo to be on the exhaust side as like with n/a when piston speed increases the gasses compress/stretch more before they actually move so they need more time to exit the cylinder than just opening at bdc and closing at tdc. They will open slightly early and use some of that stretch inertia to suck the air out and close late to allow the compression action of the gasses to relieve and suck some more out of the cylinder. A faster piston causes for more stretch and compression delay time in the head. There is most likely calculations to figure how much based on volume of ports, piston speed and distance traveled fortunately a whole bunch of companies have already done the math (hopefully)and offer various cam profiles for different applications.

[toycar]

I am not the most knowledgable but I thought we should mention when it comes to cams it's all based on the physics of air.

I have never built or tuned a motor. One thing I know is that advancing and retarding exhaust and intake cams designed for turbo is not creating the kind of overlap you would see on a f20c or even an h22.
You say it's all about cylinder pressure, yes it is. The whole need for variable valve timing, vtec all of it is the physics of air. With no air you have no cylinder pressure.Air compresses and stretches in the cylinder which we all know but more importantly when discussing cams it stretches and compresses in the intake/exhaust manifolds. Air does not simply displace the space a piston once occupied it takes time. As piston speeds increase the elasticity of air becomes more of an issue. In an NA motor If there where no valve overlap at 8000 rpm the air would only stretch in the intake manifold and not actually get pulled into the cylinder so as the piston reaches and passes bdc the intake valves stay open and like a rubber band the air still gets sucked in. That same amount of overlap would not even allow the car to run at a slower piston speed like say at idle. That's why if you have big high rpm cams on a non vtec motor it's all blip,blip....blip,blip......blip,blip. Different valve durations create power at different rpms because its all dependent on piston speed no matter which manifolds, head and block you have air still does the same thing it rubber bands it's way into and out of the cylinder.
Now having never built any motor I'm guessing from this point on.
In deeves, and that book he referenced defense on the intake side with forced induction valve overlap is not much of an issue. The turbo pushes the air in as soon as the valve opens and only some of the physics relies on the cylinder pulling the air in. I believe most of the benefit of overlap with a turbo to be on the exhaust side as like with n/a when piston speed increases the gasses compress/stretch more before they actually move so they need more time to exit the cylinder than just opening at bdc and closing at tdc. They will open slightly early and use some of that stretch inertia to suck the air out and close late to allow the compression action of the gasses to relieve and suck some more out of the cylinder. A faster piston causes for more stretch and compression delay time in the head. There is most likely calculations to figure how much based on volume of ports, piston speed and distance traveled fortunately a whole bunch of companies have already done the math (hopefully)and offer various cam profiles for different applications.

I'm gonna dumb it down for you since you have the science down but have 0 hands on experience;




More air flowing requires more time for the valve to be open in order for it to clear the cylinder, right?

At some point, the duration of this event on both ends of the engine require valve overlap to accomodate. How do you think you get a motor to spin to 10k in the first place? Naturally aspirated motors still have scavenging to design around. How would scavenging help anything if the intake valve was closed until all of the exhaust left the cylinder? Scavenging increases intake airflow, but how if there is no valve overlap? See my point? When does scavenging actually happen? Higher rpm's? Yuppers.

This is why big power builds allow the turbo to run the motor, not the other way around like most people believe. Sure, exhaust pressure generates boost. That, however, is not what you are designing the motor to do. You are designing the motor to flow more air. The compressor is what flows the air here, and you need to allow the compressor to dictate the pace. Shorter strokes, open valves-they all allow the turbo to power the motor and not the other way around.

Which is why overlap comes to play at certain thresholds of cfm. Lets stop thinking hp/tq and start thinking about lbs/min of airflow. At some point, the valve event has been extended to a point that overlap has to be created to allow the compressor to flow more air.


You make more power off of advancing ignition timing in these scenarios than you do by upping compression-because you are allowing the turbo to run the motor and not the other way around.



I pulled this off of ls1tech about valve overlap;


I see this subject come up almost daily now and it makes me curious as to why it's such a hot topic. I've decided to run some math and get some hard numbers. I don't have a whole lot of cam experience so I'd like some people who know more to chime in on this topic.

At 3000 rpms it takes 40mS to complete a 4 stroke cycle (720 degrees of crank rotation and 360 of cam rotation). 6000 rpms take 20mS for a complete cycle.

If your cam has 10 degrees of overlap at .050 then here is the time the exhaust valve is open per cycle based off RPM.

3000 rpms 1.11 mS
4000 rpms 0.83 mS
5000 rpms 0.66 mS
6000 rpms 0.55 mS
7000 rpms 0.47 mS

These are very small numbers, for reference and average blink is 300-400 mS, these times are hundreds of times shorter.

There are 3 main stages of exhaust pressure from spool up to full boost of a turbo charged engine.

1. You have what is essentialy very low back pressure when not in boost
and for a very short time (right before making boost) you have equality of pressure in the exhaust.
2. During spool and full boost you have higher than 1:1 back pressure, often 1:2 and higher.
3. As RPM increases under full boost you often have rising back pressure due to the higher mass flow of the engine.

With such short overlap times I don't see how it can make much of a difference by "having boost blow out the exhaust" during spool up and it's impossible under boost.

My very basic understanding of overlap is that it's for used for aiding the evacuation of the exhaust out of the cylinder and also has an effect on scavenging created by the pulses made in the header (if equipped with headers). I can understand how the cam affects spoolup by making more power when not in boost though.

I can see where having no overlap in a high pressure differential (3:1+) of drive pressure to supplied intake pressure (back pressure vs boost) might be beneficial by not allowing the exhaust gas to revert into the combustion chamber polluting the intake charge during overlap, but it's a very short time duration.

http://ls1tech.com/forums/forced-ind...reversion.html

And this trend is common with valve overlap. If you had only 10 degrees of overlap as illustrated in this example(everyone has more than 10 degrees of overlap), you basically have such a short event at a higher rpm that the turbo cannot breath at all. Valve overlap reduces as rpm's increase since the speed of the rotation increases. Make sense? The length of time it happens is reduced greatly. Therefore, to rev an engine to 8500-9500+rpm's, you need valve overlap on a turbo engine that is making any real power. Valve overlap is reduced by more than 50% when comparing 3,000 rpm's to 7,000 rpms. Think about that and how that applies to whats happening at those engine speeds. Turbo is usually making pressure into its efficiency range-why would you want to stop it from flowing air? Every time the valves are closed, the turbo slows down a smidge. The valve closed event is actually a restriction for the compressor, right? Woudn't it flow the most air if the ports were wide open 100% of the time?


The air is already compressed coming into the cylinder, just add fuel and spark and bingo! The motor is just the body for the airflow, the compressor is what is breathing here. For a top to bottom, 500whp turbo build, you'll need low compression pistons and valve overlap to make that happen.



It all has its pro's and cons though. You could make tons of cylinder pressure at 2500 rpms, but the engine wouldn't make any power up top which is where you want to be on a turbo engine. Long stroke, small valve having v8's are torque monsters and don't generally rev much over 5500 rpm's safely. It all depends on what you are trying to do, what rpm the engine will be at when you are trying to do it and how to best accomplish that. This is why vtec is so sweet, even on SOHC engines with intake only vtec. The single valve generates way more pressure in the manifold and the cylinder at lower rpm's, and the second valve activates when the flow is choked on 1 valve operation giving you better cylinder pressure overall across the rpm range.

[toycar]

take it easy there hero, no where did i state " hey, i read a book and all the info i came up with was this....." the autor doesnt even mention anything about v band flanges in his book. i appreciate u explaining it, very str8 forward, got it.
he speaks breifly about split inlet housing and its intended purpose, and as im looking up manifold designs i see alot that say "v-band" and i was thinking that if you have a split inlet housing then you should also have a split outlet on the manifold. so was jus makin sure termanolgy didnt change over the years

calculated airflow @4500 rpm on 20psi boost is 366cfm/ stock is 155 cfm
are you saying i can use that number with the compression pressure and burning gas pressure and come up with my manifold pressure?

i was wondering what the stock h22 manifold pressure is so i can calculate preferable turbine and housing size.

im not sure what the compressor speed will be but at that airflow rate of 610cfm @7500 rpm the turbo will be kicking 137% more air into the motor. can the stock cams effectivly manage the valves there, idk. some say yes some say no.

hopefully your book comes out soon though so i can figure it all out.

Sorry man wasn't knocking your reading comprehension more so as the lack of elaborate details obviously lacking in the outdated book.


Sorry man if I came across like a douche.


For figuring out your exhaust manifold pressure, well, that is pretty elaborate. EGT's have a lot to do with what kind of pressure you are having in the manifold, so, it would be fairly difficult to illustrate and I would have to make a ton of assumptions to do that. Most formulas are considered an estimate.


This is a pretty informative pdf I found just now. Explains it better than I would've/could've;

http://es.delphi.com/pdf/techpapers/2008-01-1004.pdf






What you really need to do is figure the cfm of you engine, and apply that to a turbine housing. The manifold should be designed around this as well. Runner diameter and length should be designed around the volume of air each cylinder moves, and you usually plan for 2 ignition cycles in each runner. FYI.


If you need help on that let me know.

Start with your engines cfm at whatever rpm's you'll be driving at.

Then apply that to your tool box when trying to figure out what turbine housing you want and see what your options are for compressors and go from there.



500'ish whp is going to be in the ballpark of 52-55lbs/min of airflow. So, keep that in mind when you start thinking about what to spend money on. Oversized valves, upgraded valve train and custom cam(s) will be your best friend.

[98vtec]

for a build only using 1000cc injectors (and the kind of HP/rpm use that goes along with that) leave the head alone completely. Have it refreshed with a valve job, seals....etc. stock valvetrain (given the springs dont have too much mileage) is more than enough.

79lbs of seat pressure is completely unnecessary for your application.

however, there is good reason for drag racers using skunk2 Pro1 cams in their engines. they are very versatile for NA and boost and can pick up big power over stock. The type S cams do very well as well, but i would not run them on stock valve train. Roger (fastest turbo prelude) ran type S cams for a very long time with great success.

Engines arent as cut and dry as many think.


special note:

keep it simple. Overthinking unimportant aspects of such a "mild" setup will just lead to wasted money.

a healthy stock ignition, NGK or OEM wires and the proper plug will support more power than your injectors.

[kratemertyl]

No worries bro, I can handle a little constructive criticism
Some interesting reading here, lots to consider

Some of the parts for the motor may be overkill, some of the planning may not be necessary. Thing is, I only wanna do this once. As fun as it is, (which latley its not) I am no good with long term projects or rushing projects. I wanna do it right the first time so I don't have to constantly dick with it and I can jus enjoy driving it again.

This whole bottom to top build thing, yea I may have spoke to soon. I think I'm gonna be over budget here. At first I thought I could build me a solid race motor for 10k (including a jdm h22) well had I not sleeved the block I probably could have stayed in there, but I also had to figure in all the suspension shit to support that motor. And all those lil trips to rock auto for misc shit adds up quick.
I'm not bailing on the build, jus gotta be deliberate , I'm half way through the budget and I haven't even bought any of the stuff that goes In the motor or any of the turbo......

[98vtec]

a very sleeved/built bottom end from golden eagle, stock refreshed head, stock intake manifold (gutted). built bottom from golden eagle will run you roughly 3000.

a new turbo will run you in the 800-1500 dollar area depending on what you go with. the proper manifold and DP setup will run 500-1200 depending on the design you want to use.

then you have intercooler..etc.

check out www.go-autoworks.com

he can make you a full turbo kit and provide you with all the turbo accessories. he usually does package deals so you can save a lot of money getting everything through one person.

i am just saying that you dont have to "go all out" to make a lot of power.

if you map your build out correctly, you can make big power without having to build a "race" engine.

my hatch makes almost 400whp on pump gas with an incredibly basic setup on a stock engine with an intake manifold the internet says isnt good for boost.

[laz93se]

a very sleeved/built bottom end from golden eagle, stock refreshed head, stock intake manifold (gutted). built bottom from golden eagle will run you roughly 3000.

a new turbo will run you in the 800-1500 dollar area depending on what you go with. the proper manifold and DP setup will run 500-1200 depending on the design you want to use.

then you have intercooler..etc.

check out www.go-autoworks.com

he can make you a full turbo kit and provide you with all the turbo accessories. he usually does package deals so you can save a lot of money getting everything through one person.

i am just saying that you dont have to "go all out" to make a lot of power.

if you map your build out correctly, you can make big power without having to build a "race" engine.

my hatch makes almost 400whp on pump gas with an incredibly basic setup on a stock engine with an intake manifold the internet says isnt good for boost.

This is all very good info from Blake. I have purchased a bunch of stuff from go auto works , you can't go wrong with them. You have a basic idea for you build , stick to toycar and 98vtec's advice and you will be fine.

I will start to tare down my motor soon, I have ton's of parts I need to install. Once I get off my lazy ass I can probably help you out as well.

Goldeneagle is top notch, head work at 4piston or endyn is probably your best choice for p&p / angle valve job.

[kratemertyl]

thanks for the link ill lokk into them,
this week has been very productive and costly.
i got all the busnings, ball joints and suspension re installed.
with doing so i noticed my rotors where scared, bearings bad, front hubs stuck on, and pads craked in the middle. i dropped the gas tank for instal of new fuel pump and found about a inch of sand and rust in the tank, pretty much jus a giant tank of rust.
so another 600 bucks in parts i wasnt planning on spending, but hety thats how it goes.
i finshed the driver side wire tuck and got a few freebeis from my boy, like battery wire(for the trunk relocation) boost gauge, oil press and engine temp gauge and a oil catch can. so that helps.
ill post some pics later, having computer issues right now.

[kratemertyl]

got a lil package in yesterday,


boom,



athough that is something to smile about my new(used) crank i got in yesterday too seems to be bent
so im on the hunt for another crank

decided to go with the skunk 2 rather then the supertech

the heads getting decked and a valve job but should be done this week

it feels good to be done with the tear down and clean up and finally ready to reassemble

passanger wire tuck

all gutted and ready for a re wire

[98vtec]

if you are hell bent on using aftermarket valvetrain the skunk2 is pretty good. if they are the PRO series make sure you have the base spring seats for them as well. the pro series spring is simply a b series spring and the spring seat is used to make them adapt to the H22 head. You should have roughly 60lbs of seat pressure. without the spring seats, your seat pressure will drop dramatically and increase potential of valve float.

[kratemertyl]

I do not belive they are the pro series. I picked springs and retainers up for 200 bucks brand new off Craig's list so I couldn't pass it up. I'm jus using my original pads,( spring seat) that shouldn't be a problem huh? They fit nice and snug on the inner spring and the outer sits on it fine.

On another note, iam having little luck finding a crank. Graveyard motor sports offerers one for $320 with a 4 week wait. My guess is its their own design and not oem. Thoughts....

Or their are a few h23 on eBay. I know its a different stroke and I'm not sure how to account for that. Do I jus have to change VTP clearance? Shouldn't TDC height be the same?

[98vtec]

do yourself a favor. find the spring height those springs are supposed to be installed at and have a machine shop check the seat pressure at that height.