Page 1 of 1

Extractors and turbos

Posted: Sun Mar 15, 2009 4:40 pm
by TheOtherLeft
I know the reason for extractors on NA engines but what about on turboed engines? Are they needed since the turbo itself acts as a restriction and increases back pressure etc?

Posted: Sun Mar 15, 2009 5:11 pm
by KYSI
you dont have extractors when you are running a turbo :?

you have a exhaust manifold, which the turbo bolts to and a dump pipe coming out of the back of the turbo which is the start of the exhaust so to speak

Posted: Sun Mar 15, 2009 5:19 pm
by tweak'e
you can have them. its just a custom manifold thats tuned more than a stock one. not worth bothering with unless your doing an extreme engine build.

Posted: Sun Mar 15, 2009 5:22 pm
by KYSI
Image

Posted: Sun Mar 15, 2009 5:29 pm
by KYSI
tweak'e wrote:you can have them. its just a custom manifold thats tuned more than a stock one. not worth bothering with unless your doing an extreme engine build.
is this what your talking about? coz i would still call it a exhaust manifold and not extractors

Posted: Sun Mar 15, 2009 5:36 pm
by -Scott-
KYSI wrote:Image
That looks like what I would call a "split-pulse" manifold, intended to make better use at low revs of the pulsatile nature of exhaust flow, helping the turbo to spool up faster.

The length of the individual runners is important; get it wrong and you may not be much better off than a simple log manifold.

Whether you're talking about NA or turbo, not all exhaust manifolds are the same; proper design makes a significant contribution to performance. If you take the view that "extractors" are about taking advantage of exhaust pulses then it's not unreasonable to call a well designed turbo manifold "extractors".

Posted: Sun Mar 15, 2009 5:45 pm
by KiwiBacon
-Scott- wrote: The length of the individual runners is important; get it wrong and you may not be much better off than a simple log manifold.
The length itself doesn't matter other than shorter is better. You can't pulse tune when the exhaust gas changes temperature soo much.

Equal length headers like that will be better at high end, but for transient response the smallest dead volume wins. Notice all the modern diesels have the most compact headers they can, helps to keep the turbo spooled to meet emissions.

Posted: Sun Mar 15, 2009 5:52 pm
by -Scott-
KiwiBacon wrote:
-Scott- wrote: The length of the individual runners is important; get it wrong and you may not be much better off than a simple log manifold.
The length itself doesn't matter other than shorter is better. You can't pulse tune when the exhaust gas changes temperature soo much.
True; absolute length is probably not important, but to take advantage of the "split pulse" concept all runners should be as close to the same length as possible. Otherwise, what's the point of a split housing, such as that manifold is designed for?

Posted: Sun Mar 15, 2009 9:10 pm
by smccask
as the RPM increases the flow more closely approximates a continuous one and like any pipework it will flow more for a given pressure if it is designed with non-restrictive bends etc. These designs are to give lower pressure drop at high flows exactly the same as extractors, not really suited to a 4wd due to high volume and its effect on lag. Also, all that welding on formed tube is bound to crack given some rough and hard 4wding, not to mention excessive use of underbonnet space.

Posted: Sun Mar 15, 2009 9:28 pm
by Dirty
Have applied the "extractors" theory in the pass to making manifolds for turbos, but to avoid the tight turns and getting the required matching lengths takes ALOT of time and a bit of space. We had both on a sports sedan, but we also didn't have any proven dyno evidence that the extra work produced a reduced spool-up time.

For a 4WD, would be a good project, but I don't think there is any benefit in the bush. For a comp truck it might be a different manner though.

- David.

Posted: Mon Mar 16, 2009 6:35 am
by KiwiBacon
-Scott- wrote:True; absolute length is probably not important, but to take advantage of the "split pulse" concept all runners should be as close to the same length as possible. Otherwise, what's the point of a split housing, such as that manifold is designed for?
Yep.

Posted: Mon Mar 16, 2009 6:45 am
by tweak'e
KYSI wrote:
tweak'e wrote:you can have them. its just a custom manifold thats tuned more than a stock one. not worth bothering with unless your doing an extreme engine build.
is this what your talking about? coz i would still call it a exhaust manifold and not extractors
yeah thas one example. i've seen other deigns where they pair up certain ports, kinda like a short 6-3-1 or 6-2-1 extractors. bit more common with petrol motors as you have some RPM to play with.
complicated, exspencive and gives small gains. but if your chasing every last drop of power and have unlimited budget........

Posted: Mon Mar 16, 2009 3:01 pm
by KiwiBacon
tweak'e wrote:
KYSI wrote:
tweak'e wrote:you can have them. its just a custom manifold thats tuned more than a stock one. not worth bothering with unless your doing an extreme engine build.
is this what your talking about? coz i would still call it a exhaust manifold and not extractors
yeah thas one example. i've seen other deigns where they pair up certain ports, kinda like a short 6-3-1 or 6-2-1 extractors. bit more common with petrol motors as you have some RPM to play with.
complicated, exspencive and gives small gains. but if your chasing every last drop of power and have unlimited budget........
It is supposed to work best on 6 cyl engines and the result if done right is earlier spool.
On a 4 cyl the cylinder pairing doesn't look good and isn't easily acheived so it's not as common.

Posted: Mon Mar 16, 2009 8:16 pm
by berad
KiwiBacon wrote:
-Scott- wrote: The length of the individual runners is important; get it wrong and you may not be much better off than a simple log manifold.
The length itself doesn't matter other than shorter is better. You can't pulse tune when the exhaust gas changes temperature soo much.

Equal length headers like that will be better at high end, but for transient response the smallest dead volume wins. Notice all the modern diesels have the most compact headers they can, helps to keep the turbo spooled to meet emissions.
With my petrol fun toys, i was under the impression, longer runners = torque curve change, shorter = power curve.

Posted: Mon Mar 16, 2009 9:31 pm
by -Scott-
berad wrote:With my petrol fun toys, i was under the impression, longer runners = torque curve change, shorter = power curve.
Are you talking intake or exhaust, na or forced? Tuned length runners work best at a given rev range - the shorter the runner, the higher the rev range. And, as others have mentioned, I believe forced induction systems really prefer shorter, more for reduced "dead space" to reduce lag.

Posted: Mon Mar 16, 2009 10:18 pm
by brad-chevlux
berad wrote:
KiwiBacon wrote:
-Scott- wrote: The length of the individual runners is important; get it wrong and you may not be much better off than a simple log manifold.
The length itself doesn't matter other than shorter is better. You can't pulse tune when the exhaust gas changes temperature soo much.

Equal length headers like that will be better at high end, but for transient response the smallest dead volume wins. Notice all the modern diesels have the most compact headers they can, helps to keep the turbo spooled to meet emissions.
With my petrol fun toys, i was under the impression, longer runners = torque curve change, shorter = power curve.
considering power is calculated from torque and rpm, if the torque changes the power changes at the same point.

Posted: Tue Mar 17, 2009 7:01 am
by KiwiBacon
berad wrote: With my petrol fun toys, i was under the impression, longer runners = torque curve change, shorter = power curve.
That only works on non-turbo engines. With a turbo your gas temp changes constantly, which changes the speed that pulses travel through the gas. The result means pulse tuning is limited to keeping them evenly spaced (which is where the equal length comes into play).

Posted: Tue Mar 17, 2009 10:13 am
by coxy321
KiwiBacon wrote:
berad wrote: With my petrol fun toys, i was under the impression, longer runners = torque curve change, shorter = power curve.
That only works on non-turbo engines. With a turbo your gas temp changes constantly, which changes the speed that pulses travel through the gas. The result means pulse tuning is limited to keeping them evenly spaced (which is where the equal length comes into play).
Thats right. Thats why the variable intake length mechanisms seen in the N/A falcon 4.0L's are locked in the turbo variants.

Tuned length turbo manifolds are probably more important on 4 cylinder engines more than anything else because the turbine wheel is only receiving four pulses per revolution (from only four cylinders firing per revolution). To set up a tuned length turbo manifold (4 cyl), ideally you would have the pipes tuned to a length that would allow the turbine wheel to recieve four individual pulses per revolution, which would inturn give a more constant turbo RPM and also give marginally better spool-up times.

At the end of the day, i dont see ANY benefit in doing something like this unless you are racing and need that extra 2HP, or that 0.2 second faster turbo response.

Most people call them turbo manifolds, some people call them turbo extractors (possibly due to their appearance). Its a bit like "rim locks" and bead locks....

Posted: Tue Mar 17, 2009 1:26 pm
by MightyMouse
coxy321 wrote: (from only four cylinders firing per revolution).
IIRC a four stroke cylinder fires ever second revolution ? making it two firings / rev ?

Of course the underlying issue is as you described - less firings, greater need to harness pulse energy.

Posted: Tue Mar 17, 2009 1:46 pm
by coxy321
MightyMouse wrote:
coxy321 wrote: (from only four cylinders firing per revolution).
IIRC a four stroke cylinder fires ever second revolution ? making it two firings / rev ?

Of course the underlying issue is as you described - less firings, greater need to harness pulse energy.
My bad - you are spot on the money there. Always half the firing per revolution as there is cylinders.

Posted: Tue Mar 17, 2009 5:15 pm
by BEU77y
As most have said, you can do it, but it's a fair amount of work for what could be little gain. I watched a mate build his own manifold for his GQ turbo set up. He didn't go as far as tuned lengths because there is almost no space under a GQ bonnet. Granted it took him about 3 weeks of arvos and weekends with a welder and a die grinder. But I think he did make some gains when it comes to all out efficiency as compared to the off the shelf rail style manfolds that you get with a $3500 kit. They are nasty when you look at simple flow design.
His GQ is running a Garrett TB3404 and puts out 15psi at 1900rpm. Pretty bloody good for a home made system... and it flys!

I bought the same turbo with the same ideas, but never got around to it. I've since used the tubo on a Safari system that was on our 105. It runs a lot better but I don't think it spools up as quick as my mates GQ. There are other issues, it's not just the manifold...

Posted: Tue Mar 17, 2009 7:34 pm
by DamTriton
I would have thought that with turbos (petrol and diesel) that keeping all the turbo to engine volumes at a minimum would have been more important than any equal length issues.

The minimal volume b/t the turbo and engine will reduce the amount of gas exhaust that is required to spin up the turbo (if you like, "primary lag") as well as reduce the time it takes to pressurise the inlet manifold ("secondary lag").

In the exhaust to turbo example it should also reduce the heat loss if the manifold pipes are shorter and thicker (volume to surface area ratio greater). The turbo will have a natural damping effect ("elastic" effective volume effect) on any exhaust pulses that may scavenge gas from the cylinder.

From what little I know, tuned pipes of any description are only of value in high or contant RPM engines where you are trying to scavenge as much exhaust gas as you can out of the cylinder.

Posted: Tue Mar 17, 2009 8:09 pm
by coxy321
True, but i think that with an extractor style manifold, there is only so much you can do to make it all smaller/shorter (hence the octopus-like appearance of some). As an example, log style manifolds have less volume than an extractor type manifold, however they are less effective at distributing the pulses in an equal fashion. However, i think the cast units would transfer/hold the heat better than extractor type unit. Some people dont realise that the exhaust gases are only the half of it - its also the heat that helps spin that little wheel up.

Posted: Wed Mar 18, 2009 12:59 am
by zagan
isn't extractor just the nick for exhaust manifold?

I never really heard of anyone making out an extractor is sonething different.

Bit like Torque converter is the same as stall converter or stally.

That pic would be a single bank, as the exhast all ends up at the 1 point, I that it the other piping is a bypass for excess exhuast gas, unless it's for a twin turbo setup.

Posted: Wed Mar 18, 2009 5:12 am
by KiwiBacon
DAMKIA wrote:I would have thought that with turbos (petrol and diesel) that keeping all the turbo to engine volumes at a minimum would have been more important than any equal length issues.

The minimal volume b/t the turbo and engine will reduce the amount of gas exhaust that is required to spin up the turbo (if you like, "primary lag") as well as reduce the time it takes to pressurise the inlet manifold ("secondary lag").

In the exhaust to turbo example it should also reduce the heat loss if the manifold pipes are shorter and thicker (volume to surface area ratio greater). The turbo will have a natural damping effect ("elastic" effective volume effect) on any exhaust pulses that may scavenge gas from the cylinder.
Yes, that's why all the factory diesels have the smallest volume exhaust manifold they can use. Keeps lag to a minimum so they can meet euro 5 emissions and still wind 280hp from a 3 litre diesel.
DAMKIA wrote: From what little I know, tuned pipes of any description are only of value in high or contant RPM engines where you are trying to scavenge as much exhaust gas as you can out of the cylinder.
The other point to mention, most turbo engines run around no valve overlap, where the typical NA setup with extracters can run heaps.

Posted: Wed Mar 18, 2009 7:54 am
by coxy321
zagan wrote:isn't extractor just the nick for exhaust manifold?

I never really heard of anyone making out an extractor is sonething different.

Bit like Torque converter is the same as stall converter or stally.

That pic would be a single bank, as the exhast all ends up at the 1 point, I that it the other piping is a bypass for excess exhuast gas, unless it's for a twin turbo setup.
IMO, an exhaust manifold is exactly that. An aftermarket manifold designed to improve the path/flow of exhaust gasses is call a set of extractors. Also, turbo manifolds are just turbo manifolds - however i think its fair that people call the custom units extractors.

Any of you northerners want a potato cake?? :D

Posted: Wed Mar 18, 2009 4:56 pm
by cooki_monsta
coxy321 wrote:
zagan wrote:isn't extractor just the nick for exhaust manifold?

I never really heard of anyone making out an extractor is sonething different.

Bit like Torque converter is the same as stall converter or stally.

That pic would be a single bank, as the exhast all ends up at the 1 point, I that it the other piping is a bypass for excess exhuast gas, unless it's for a twin turbo setup.
IMO, an exhaust manifold is exactly that. An aftermarket manifold designed to improve the path/flow of exhaust gasses is call a set of extractors. Also, turbo manifolds are just turbo manifolds - however i think its fair that people call the custom units extractors.

Any of you northerners want a potato cake?? :D
LOL potato cake