PDA

View Full Version : Lets talk turbo voodoo



ProStreet R/T
04-25-2006, 11:36 PM
Not any build in particular, but theories and practical application in regards to hot side plumbing. With as many turbo PT cars being built maybe we can all learn a thing or two. If you have any experience on the subject please speak up *cough* Tom Nelson *cough*


As well all know there are two basic directions you can go. Either a nice merge collector, somewhat equal length tubes, a more traditional header type configuration. The other being a simple log, many times just using a stock exhaust manifold.

I've seen both designs make phenominal power in various cars, and often times the same shop using both styles in various builds. W2W for example, built a beautiful 4:1 merge collector setup for the MULE, but used stock manifolds on a 1400hp corvette. As well there are quite a few crazy powerful TT vipers using stock manifolds.



So what do YOU think is the ideal setup? and why? If you have any hard facts please share with the rest of us. I've talked with quite a few well known engine builders and recieved a different answer from every single one.


note: My friend and I are working on a back to back test with a few designs to see exactly what changes so hopefully there will be some 1st hand hard data fairly soon. We have come up with a pretty extensive and educated hypothesis which I will post shortly, but it wouldn't be the first time guesses (even good ones) have been way off.

And yes it is dangerous when a couple engineers are bored and start brainstorming :geek:

nitrorocket
04-26-2006, 08:26 AM
To my knowledge there has been some testing done on 4 poppers with log versus header for a turbo setup. The Header picked up noticable power the higher the boost and rpm's went. For a max power application, the header should make more power, but in a real low horespower car, you might not even see a difference.

Tom Nelson
04-26-2006, 03:41 PM
I use Tubular hedders I've used the banks logs in the past also.I would say the main reason I use a tubular hedder is I only use tangential ex housings.I've found alot of power in that.along with a good reduction in intake temps.if you want to make 700 /800 hp a think a log is good it's surprising how crap you can make a hedder or log and still have it work good.But when you start making big power I've seen good gains from a hedder with the right transition into the turbo and some decent tube length.Also with a hedder you can give some room for all 4 cylinders to run before they get picked up by the wastegate.The logs are pretty tight especially if you have a big motor and alot of turbo.the gate has to work really well.it's hard to control on the log.Also note most logs have cast in hole sizes and bolt patterns for way to small a waste gate.Be careful unless you like variable non controlled boost.I think lastly nothing looks as cool as a well designed tubular stainless hedder.My to cents

ProStreet R/T
04-26-2006, 06:55 PM
That is pretty much what we are expecting, even that the log style may be better for a moderate hp street car. In that we can run larger turbos, keep the air temps down, and not sacrafice any drivability.

note: This is on a 488ci V10 so a little turbo lag really isn't all that bad on a street car.

He just finished one car, bone stock 488ci viper, put 615rwhp and 750ft/lbs at 4 psi. This was with a log style manifold. Granted it's custom fabricated from stainless with a pretty unique design, but same effect.

The same style manifold made did 1188rwhp and 1176rwtq at 15psi on another viper. Granted this one had a built bottom end, standalone pcm, fuel system etc. Still cleaning up the tuning, shooting for 1300/1300... on pump gas. Same turbo kit as the first car.

I think on the 1200hp car it may be giving up 100hp due to the manifold design. A relatively short primary, 4:1 merge collector, header style will likely flow quite a bit better on the top end. But time will tell, soon as we get any hard facts back i'll be sure to share.


And I agree Tom, a well designed header feeding a big fat ass turbo is a thing of beauty. Especially when it's a street car with a pair of GT47-80's hung off the front and twin 5" down pipes :firefire:

Y-TRY
04-26-2006, 11:12 PM
Okay, my .02+++++ FWIW...

I am a real, to-the-point guy. I deal in real-world, average-guy terms, so just think about what I am saying here...

From my own research I've realized a couple things that are not popular with the generally-accepted theories, and I know this will piss some marketing departments off....

No matter what you run, the turbo will always be a restriction. You could have the baddest, most expensive, headers and there will always be a restriction after the exhuast port: it is called a "turbo". This is why turbo headers get so hot. The space between the exhaust port and turbo is always under pressure. The engine will almost always produce more exhaust gas than the turbo(s) can flow. You could spend $$$$ on merge collectors but the fact is that the entire header is under pressure from both ends and that the advantage of the collector is moot (compared to an N/A set-up). The collector could flow GREAT, but the exhaust still hits a dead end at the turbo. Merged collectors just get it there faster and prettier. The trick is determining which turbo compliments what your engine can flow.

Those exhaust pressures, however, actually HELP the turbo spool. How? Thermodynamics. The more pressure in the header = more HEAT in the exhuast = the more pressure against the turbine wheel (this REALLY gets into cam design, but that's another thread). It's kinda a catch 22. Theoretically, if you reduce pressure in the header it would reduce pressure against the turbine wheel, slowing it. Remember: It's not only exhaust gas velocity, but pressure in the header/manifold, that spins the wheel. There is a point where velocity from the engine-side is greater than the turbo can breathe, thus the pressure. We are all consumed with "flow", and rate performace by this term, but we need to consider the balance of header flow and turbo flow to really make a decision.

IMO- Boost is a measure of INTAKE RESTRICTION more than anything else, when it comes to power. Boost is really a measurement of how much MORE the intake tract is pressurized than the engine can breathe. For example: Take a 383 chevy with 195cc intake and 15psi boost...... Bolt on a head with 220cc and boost will drop- with no other changes. Why? Because the engine now consumes a greater ratio of air than with the 195cc, compared to what the tubro provides. The turbo isn't slower, but the heads are now less restrictive. So, even though "boost" drops, the engine is more efficient. Conversely- Take the same 383ci engine and bolt-on some 170cc heads and the boost will SPIKE. The result is LESS power. This is only becuase the turbo flows that much better than the intake path. Get it?

My point is: Spend less time trying to create boost with the turbo and more time trying to reduce it at the engine. When you find that balance you will find POWA!!!!!!!!!!!!!!!!!!!!!!!

Hard facts? I built a 406ci SBC with two 60mm turbos for my '68. So far it's made 925hp on pump gas (787rwhp). We're talking a carb'd small block on pump gas making 900+hp. This is on ported 220cc AFR's and 15psi, no cooler. Dropping to 195cc AFR's it would probably jump to 20psi and make less power. Turbo flow vs. engine VE.

ProStreet R/T
04-27-2006, 12:58 AM
Okay, my .02+++++ FWIW...

I am a real, to-the-point guy. I deal in real-world, average-guy terms, so just think about what I am saying here...

From my own research I've realized a couple things that are not popular with the generally-accepted theories, and I know this will piss some marketing departments off....

No matter what you run, the turbo will always be a restriction. You could have the baddest, most expensive, headers and there will always be a restriction after the exhuast port: it is called a "turbo". This is why turbo headers get so hot. The space between the exhaust port and turbo is always under pressure. The engine will almost always produce more exhaust gas than the turbo(s) can flow. You could spend $$$$ on merge collectors but the fact is that the entire header is under pressure from both ends and that the advantage of the collector is moot (compared to an N/A set-up). The collector could flow GREAT, but the exhaust still hits a dead end at the turbo. Merged collectors just get it there faster and prettier. The trick is determining which turbo compliments what your engine can flow.

I agree on some points, you must match the turbo to the motor. But there is more to the exhaust flow than you're giving credit for. Yes the turbo will be a big restriction, the key to efficiency is to keep it the biggest restriction. The last thing you want to do is choke the motor in the tubing and wasting that heat energy. Remember, just because it's under pressure the laws of airflow still apply. You won't see the scavenging benefit, thats pretty clear, but there is a lot more to it than that.



Those exhaust pressures, however, actually HELP the turbo spool. How? Thermodynamics. The more pressure in the header = more HEAT in the exhuast = the more pressure against the turbine wheel (this REALLY gets into cam design, but that's another thread). It's kinda a catch 22. Theoretically, if you reduce pressure in the header it would reduce pressure against the turbine wheel, slowing it. Remember: It's not only exhaust gas velocity, but pressure in the header/manifold, that spins the wheel. There is a point where velocity from the engine-side is greater than the turbo can breathe, thus the pressure. We are all consumed with "flow", and rate performace by this term, but we need to consider the balance of header flow and turbo flow to really make a decision.

You are correct, they do help spool the turbo. But that brings in a million variables, boost pressure, turbine a/r, turbine size, motor rpm range, etc. etc. I have personally seen first hand that a log style will in fact spool a turbo at a lower rpm than a full length header setup. We didn't do any tests on primary dia, length etc. YET ;) But I strongly feel that once the turbo is spooled the poor airflow of the manifold will hinder performance. How much? That is entirely dependant on the motor. If it's a 5psi street car then it's not even worth debating. But on a high hp car I think there is a real benefit, we'll see how my theories stack up.

One thing with "flow" that is why turbo "flow" is rated in a constant lbs of air ;) You are correct, the pressure in the manifold is just the biggest restriction, but that just brings me to my next point.

What is the pressure ratio on your turbo setup? That is back pressure on the turbine side (inside the manifold) vs boost pressure? You would be shocked how many "good systems" see 3x as much pressure in the turbine as they do on the cold side. If you can accomplish the same thing, but lower the exhaust back pressure to 2:1, or even sub 1:1, (and yes it's quite possible to get under 1:1) isn't is fairly logical to assume you will free up quite a bit of power?



IMO- Boost is a measure of INTAKE RESTRICTION more than anything else, when it comes to power. Boost is really a measurement of how much MORE the intake tract is pressurized than the engine can breathe. For example: Take a 383 chevy with 195cc intake and 15psi boost...... Bolt on a head with 220cc and boost will drop- with no other changes. Why? Because the engine now consumes a greater ratio of air than with the 195cc, compared to what the tubro provides. The turbo isn't slower, but the heads are now less restrictive. So, even though "boost" drops, the engine is more efficient. Conversely- Take the same 383ci engine and bolt-on some 170cc heads and the boost will SPIKE. The result is LESS power. This is only becuase the turbo flows that much better than the intake path. Get it?

My point is: Spend less time trying to create boost with the turbo and more time trying to reduce it at the engine. When you find that balance you will find POWA!!!!!!!!!!!!!!!!!!!!!!!

Hard facts? I built a 406ci SBC with two 60mm turbos for my '68. So far it's made 925hp on pump gas (787rwhp). We're talking a carb'd small block on pump gas making 900+hp. This is on ported 220cc AFR's and 15psi, no cooler. Dropping to 195cc AFR's it would probably jump to 20psi and make less power. Turbo flow vs. engine VE.

Would you mind posting pics of your exhaust setup?

I agree on many motors the difference in power may not be worth the effort, and in some may not even provide a desired result. As I stated the excessive restriction only hurts the motor at higher rpm/HP ranges. On the 4 psi car IMO the logs are perfect, it makes for a fantastic street car, and at his hp level it's incredibly efficient. The gains were near 46hp per 1 psi on a bone stock motor. Though on this motor the "logs" were a custom design and a bit more involved than a stock cast manifold. As well the turbos are a bit larger than one would normally spec for such a tame build, but there is a method to the madness.

But on the reverse side. With a motor seeing 380cfm of intake port flow, custom sheet metal intake manifold, and looking for 1200+ hp there are real gains to be had from a well designed header. And not only from increasing the flow, but attempting to keep the backpressure symetrical for all cylinders. It's clear to see that the cylinders furthest from the exit will see more pressure than those at the front. This alone can cause many tuning headaches and inefficiencies. Is it a 2% cost or 10% is left to be seen.

Don't take my thoughts as stating you're wrong, this is exactly the discussion I wanted to bring up. If you think i'm off my rocker please don't hesitate to speak up, lord knows I wont lol.

One thing to point out. There are many different qualities of "log" type manifolds. One friend of mine has a design that may very well close the gap on the headers for all but the most extreme applications.

Here are a couple tests I found that people have already done. This is a pretty nice fabricated "log" manifold, vs a tubular header.

http://www.full-race.com/articles/Bseriestest_writeup.pdf

This one they used a cast iron manifold, vs a tubular header.

http://www.full-race.com/articles/1.8T_turbo_manifold_test_writeup.pdf


Will we see the same gains (minimum 10% from their findings) on a motor 3x the size as theirs? Hard to say. But it will be fun and informative none the less.



Come on guys, there are a dozen or more people on here with turbo cars being built. Surely you all have something to say, add, a new angle to the whole debacle.

Y-TRY
04-27-2006, 06:53 AM
Thanks for your reply. It's great to air-out my thoughts and have someone agree/disagree with well thought-out responses. These were simply the conclusions made by a guy laying on his garage floor to build his car. I would also like to see what could be tested and proven with all the computer models and flow benches and stuff. But thesee responses do have me re-evaluating some.

Your 1200hp engine example really highlights the balance needed. For all that intake flow, it definately needs an exhaust that can keep up. More in = more out.

Funny thing is- I didn't mean to mislead, I have nice tubular headers with merge collectors. I made a mistake in designing my wastegate so that the flow path is way too perpindicular to the exhaust. But it works.

The one point from above is this: I think that more power can be achieved by increasing the intake flow than focusing on the exhaust. But it's just my opinion. If you ratio Time/effort to resulting power I'd say it's like 1 part effort to 4 parts power achieved for the intake side, while only 1:2 on the exhaust. Merge collector vs. ported heads? I'd vote for the heads.

I have always been curious about how much more power potential there is for short-tube headers (like Banks uses) vs. longer tube ones (like the Mule) Is there more velocity at the turbo when it is that much closer to the port? Hmm. Mine are located alot like the Mule.

ProStreet R/T
04-27-2006, 10:41 AM
I completely agree there are much larger gains to be had from doing head work. As well something like this would really only shine on a well built, unrestricted motor anyway.


The length of the primary tubes is something we have been discussing for quite some time and haven't found a difinitive answer to. My (in theory at least) ideal build would use a relatively short primary (12-16") in an effort to retain as much exhaust heat/velocity as possible. I don't see any major gains coming from using a longer (more traditional 24-30") lengths. BUT, using quality material (good stainless) and having it coated with some type of thermal barrier, the heat/velocity loss would be minimal.

The general consensus is to use the largest radius bends you can, mount the turbo in a way that allows you to run the largest down pipe (within reason) you can, and if the primary tubes end up being 4" longer it's not the end of the world.

Look at the tubes on this thing:
http://www.radrides.com/69cuda.html

The primary pipes must be near 3ft long. And thats feeding a monster turbo for that little 4 banger (as baaadddd as an R5 4 banger is).

In regards to the software I honestly haven't found anything that works reliably with a turbo setup. They can tell you what the motor will do with the boost, but never take into account anything on the turbo side. Either it's just way too many variables to make it accurate, or nobody knows enough about them (from a physics perspective) to write the software. who knows.




Next topic... Turbo Cam selection :hammer: This is one area that seems to cause more confusion than anything, as if cams weren't an art all their own to begin with. I'm going to lunch, but will come back to this shortly.


btw, 85 views and only 4 people talking. Surely (dont' call me shirley :rotfl:) someone else has something to add.

Tom Nelson
04-27-2006, 12:03 PM
Manifolds tend to turn into big heat sinks.look at a tubular hedder in a pull the collector is always what gets red first then it works its way threw out the hedder,versus a manifold it keeps heat throughout the hole deal.I found this kills head gaskets .there are a thousand things going on and what might work well for one won't for another just because of all the different motors/turbos(ar wheel size/housing size and how they all mesh.wholy moly camshaft that changes everything also.It gets crazy when i first started this i read all the maps and listened to the so called experts 90 turbo variations later i've fiqured out one thing .Rather than drive myself nuts with what i think it might do i just do it let the motor eat and see if it's worth a damn.it's always a sucess the hp or tq might be crap but you fiqure out a little everytime.Everybody i ever talked to told me i was running way to big of turbos on my applications the'll never spool only efficent in certain rpm ranges whatever.I listened to the experts for to long and fiqured out they were clueless.I by no means am an expert on this stuff i've just tried enough combo's to know what works for me on my applications.I beleive the hedder is the way to go for power and reliablity as long as it's designed right with the right materials and supports .I learn with each motor i do .we are just at the beginning of knowing what makes these things killer.

slow4dr
04-27-2006, 12:55 PM
It gets crazy when i first started this i read all the maps and listened to the so called experts 90 turbo variations later i've fiqured out one thing .Rather than drive myself nuts with what i think it might do i just do it let the motor eat and see if it's worth a damn.it's always a sucess the hp or tq might be crap but you fiqure out a little everytime.Everybody i ever talked to told me i was running way to big of turbos on my applications the'll never spool only efficent in certain rpm ranges whatever.I listened to the experts for to long and fiqured out they were clueless.I by no means am an expert on this stuff i've just tried enough combo's to know what works for me on my applications.

This is sooooooo true. All to often people try to match turbo applications on paper and it just may not work in the real world. Much of what we do is educated guesses after many years of trials and tribulations.

ProStreet R/T
04-27-2006, 01:42 PM
I agree, for soo long everyone has been saying certain turbos will never work for a street car. But in reality even a pair of Gt47-76's spool wonderfully on the street when behind a near 500" realtively low rpm motor. Even though the maps say they would be terribly oversized and far out of their efficiency range. It looks like the reduction in backpressure greatly outweighs the efficiency drop on the turbo side. You get a setup that spools all of 3-400 rpm later, but not only makes quite a bit more power, but that power due to the smoother application is easier to drive on the street.

Who would have thought a 1200hp car would be easier to drive on street radials than a 900hp car. :3gears:

Fuelie Fan
04-27-2006, 05:59 PM
Depends on which experts you talk to. When I interned at Garrett, one of the engineers would mock Turbonetics non-stop, even though the outsider would probably consider them to be "experts". That being said, paper matching usually got us 90% of the way, and was certainly more than adequate for the street application. It's how he selected his turbos. His comment was that most people either ignored or incorrectly interpreted turbine maps most of the time. But, that was 6 years ago and I haven't been able to do a single thing with turbos since, so I don't have much more useful information to contribute.

Cleatus
04-30-2006, 05:33 PM
I understand the concept of pressure = restriction.....but I'm wondering if it's a good idea to get *too* get carried away with the idea that increased manifold pressure is in fact not always a good trend. It seems that equal pressure relating to compressor discharge and cylinder pressure as the intake valve closes is most desirable........equal is cool, but under certain operating circumstances (cylinder heads, manifold, camshaft variables), striving for the *most* cylinder pressure possible when the intake valve closes is also a good thing.....IOw, whatever it takes!

Intake/exhaust delta P is a whole another thread!

Any way, unless you try it yourself...don't let *anyone/experts* tell you it won't work....people really don't know, too many variables and they just might have a built in mental block.

John

ProStreet R/T
04-30-2006, 09:26 PM
:hammer::Alchy: Sigh... It's been a LOOONG day. For one reason or another we've been chasing our tail with a turbo car and just can't seem to figure out what it's doing, or should I say not doing.

From 16psi, raise boost to 20psi... no hp/tq gains... change timing... nothing. :confused: HP/TQ is still taking a nose dive at around 5400 rpm and it's got plenty of fuel.


Next step, run it with open 4" down pipes, see if there is too much restriction running the full exhaust.

From there we'll have to pull the lifters and see if they are collapsing. (if you haven't changed lifters on a viper it SUCKS. Requires pulling the cylinder heads and such)

:hmm: pretty much sums up my feelings at the moment.

I'll be sure to update the thread with our findings, hopefully it saves someone else a day of headaches.

camcojb
04-30-2006, 09:34 PM
:hammer::Alchy: Sigh... It's been a LOOONG day. For one reason or another we've been chasing our tail with a turbo car and just can't seem to figure out what it's doing, or should I say not doing.

From 16psi, raise boost to 20psi... no hp/tq gains... change timing... nothing. :confused: HP/TQ is still taking a nose dive at around 5400 rpm and it's got plenty of fuel.


Next step, run it with open 4" down pipes, see if there is too much restriction running the full exhaust.

From there we'll have to pull the lifters and see if they are collapsing. (if you haven't changed lifters on a viper it SUCKS. Requires pulling the cylinder heads and such)

:hmm: pretty much sums up my feelings at the moment.

I'll be sure to update the thread with our findings, hopefully it saves someone else a day of headaches.

I've had the same thing happen with centrifugals and both times it was an airflow problem. One was a 180 degree bend at the inlet to the blower and the other was piss-poor exhaust port flow in the heads.

Jody