Relocating turbo
09-25-2005, 02:24 PM
#1
Instructor
Thread Starter
Join Date: Sep 2005
Posts: 131
Likes: 0
Received 0 Likes
on
0 Posts
Relocating turbo
Isn't it terribly inneficient to have the turbo sitting all the way across the engine bay on the other side of the engine?
How *much* of a different does the distance between the turbo and exhaust ports have in impacting spoolup? (I'd settle with a good turbo documenation that explained some math about it...) Also, how does the length of the intake system affect a turbo car? (The length between the turbo and intake ports looks like it would have to increase -- intake could be move to the passenger side...).
(http://www.ststurbo.com/ -- Remote Mounted Turbos, they don't give any useful numbers though ... for all I know it could have a 2 minute spoolup time)
Anyway, if there was a SIGNIFICANT spoolup delay simply due to the length of the exhaust manifold (is that the propper name for the entire system between the engine and the turbo?) then wouldn't one of the ``ultimate'' modifications for a 2.5L 951 be simply the relocation of the turbo itself?
I found a few interesting articles on this -- actually it was over att the 968 forumns and relating to engine swaps and they all seem to conclude that it isn't practical for two reasons
1) If mounted on the passenger side of the engine (front) it makes the belts to hot and they strectch over time.
2) Even more relocation of engine components if mounted on the back of the engine block on the passenger side.
In reguards to point 1 (and the thread recently kicking on this forumn about heat reduction), couldn't a good exhaust heat retention system and turbo shield combat this issue making it a non-issue? It gets hot in the engine bay reguardless so if the temp is kept down to stock temperatures where the belts are then there should be no difference, right?
In reguard to point 2 either system would require alot of relocating and custom fabrication of manifolds so it falls in the same category as anyone trying to make a custome turbo or perform an 'un-kitted' engine swap...
I've searched google but didn't find much information.
Anyone know where to find information on the Callaway configuration as found in the '83 Callaway-turbo'ed 944's?
How *much* of a different does the distance between the turbo and exhaust ports have in impacting spoolup? (I'd settle with a good turbo documenation that explained some math about it...) Also, how does the length of the intake system affect a turbo car? (The length between the turbo and intake ports looks like it would have to increase -- intake could be move to the passenger side...).
(http://www.ststurbo.com/ -- Remote Mounted Turbos, they don't give any useful numbers though ... for all I know it could have a 2 minute spoolup time)
Anyway, if there was a SIGNIFICANT spoolup delay simply due to the length of the exhaust manifold (is that the propper name for the entire system between the engine and the turbo?) then wouldn't one of the ``ultimate'' modifications for a 2.5L 951 be simply the relocation of the turbo itself?
I found a few interesting articles on this -- actually it was over att the 968 forumns and relating to engine swaps and they all seem to conclude that it isn't practical for two reasons
1) If mounted on the passenger side of the engine (front) it makes the belts to hot and they strectch over time.
2) Even more relocation of engine components if mounted on the back of the engine block on the passenger side.
In reguards to point 1 (and the thread recently kicking on this forumn about heat reduction), couldn't a good exhaust heat retention system and turbo shield combat this issue making it a non-issue? It gets hot in the engine bay reguardless so if the temp is kept down to stock temperatures where the belts are then there should be no difference, right?
In reguard to point 2 either system would require alot of relocating and custom fabrication of manifolds so it falls in the same category as anyone trying to make a custome turbo or perform an 'un-kitted' engine swap...
I've searched google but didn't find much information.
Anyone know where to find information on the Callaway configuration as found in the '83 Callaway-turbo'ed 944's?
09-25-2005, 03:28 PM
#4
Pro
Join Date: Mar 2002
Location: S. Florida
Posts: 608
Likes: 0
Received 0 Likes
on
0 Posts
I am with you. Way too complex, the crossover is too long and mounting the turbo under the intake manifold is not great either.
Its obvious that the entire turbo setup was stuffed in to existing space and I cant really see the Porsche engineers endorsing a log type turbo manifold. Since our cars were originally designed for VW with an audi engine, via the 924, I doubt a turbo setup for a 944 engine was in the original design plans. Just conjecture, but I would assume porsche would rather compromise on location, rather than performance, and that just makes these cars a real bitch to work on.
Here is the callaway setup, but the 951 manifolds are definitely better for even gas flow. For normal street driving to about 10 psi it probably does not matter IMO. If you were running high enough rpms, with a lot of boost on a log manifold, you could end up with a lot of back pressure on one cyl and get into trouble. Also log manifolds can expand enough to literally shear off a bolt.
http://members.rennlist.com/tholyoak...ion_parts.html
The 951gt 3L setup above would be the best route to a relocated single turbo, since it has headers, but you will lose the pop up light, probably have to relocate the PS tank or go manual, and the oil filter is under there too so you would need to run a remote for that too.
You could go tiny twins, but thats even more complicated and and more parts to break. Similar to the Jag triple setup on page 123 in the Corky Bell book.
Its obvious that the entire turbo setup was stuffed in to existing space and I cant really see the Porsche engineers endorsing a log type turbo manifold. Since our cars were originally designed for VW with an audi engine, via the 924, I doubt a turbo setup for a 944 engine was in the original design plans. Just conjecture, but I would assume porsche would rather compromise on location, rather than performance, and that just makes these cars a real bitch to work on.
Here is the callaway setup, but the 951 manifolds are definitely better for even gas flow. For normal street driving to about 10 psi it probably does not matter IMO. If you were running high enough rpms, with a lot of boost on a log manifold, you could end up with a lot of back pressure on one cyl and get into trouble. Also log manifolds can expand enough to literally shear off a bolt.
http://members.rennlist.com/tholyoak...ion_parts.html
The 951gt 3L setup above would be the best route to a relocated single turbo, since it has headers, but you will lose the pop up light, probably have to relocate the PS tank or go manual, and the oil filter is under there too so you would need to run a remote for that too.
You could go tiny twins, but thats even more complicated and and more parts to break. Similar to the Jag triple setup on page 123 in the Corky Bell book.
09-25-2005, 04:25 PM
#6
Instructor
Thread Starter
Join Date: Sep 2005
Posts: 131
Likes: 0
Received 0 Likes
on
0 Posts
... and the engine bay wasn't? 
Thanks for the Callaway link. I thought it might have used a different setup.
It doesn't look like to would hard to replace the log manifold in that picture with LR setup (or even stock) -- look at all that empty space! -- so Bob Howard must have a reason...
I still wonder what benefits are realized from the relocation.
The STS people look like they have a rather ingenious setup and it does pretty much eliminate the issue of cooling (I wonder how extra heat transfer fins on the turbo->manifold piping would work
) ... but how does the extra unf feet of piping between the engine and turbo affect lag?
I would think reaction time would be more noticable (not that there isn't 50' of tubing on our cars...) but the dynos they post look, to my very untrained eye, pretty damn good and all the reviews and opinions (of course, selected I'm sure) seem to be quite positive.
This either means that it does add extra spoolup but people don't (choose not to) notice it or any extra spoolup isn't noticeable -- as in, the turbo itself plays a much larger part of the spool cycle counteracting the distance. On this note, the turbo on many airplane engines seems to be located near the back ... but they also server a much different target and purpose.
This however, brings me to two different points:
1) All the applications from STS are for larger engines -- V8's (V6's too?) -- so they start off with SIGNIFICANTLY more power and airflow to begin with. (Poor little 2.5
)
2) They dyno at ~5000ft; how does the this affect the characteristics of a turbo?
On the other hand, some of their customers (not them -- they claim a believable ~30%) claim 50-70%+ gains on 5psi
So, uhm, anyone want to try a rear-mounted turbo setup? Maybe put twins in the boot
Thanks for the Callaway link. I thought it might have used a different setup.
It doesn't look like to would hard to replace the log manifold in that picture with LR setup (or even stock) -- look at all that empty space! -- so Bob Howard must have a reason...
I still wonder what benefits are realized from the relocation.
The STS people look like they have a rather ingenious setup and it does pretty much eliminate the issue of cooling (I wonder how extra heat transfer fins on the turbo->manifold piping would work
) ... but how does the extra unf feet of piping between the engine and turbo affect lag? I would think reaction time would be more noticable (not that there isn't 50' of tubing on our cars...) but the dynos they post look, to my very untrained eye, pretty damn good and all the reviews and opinions (of course, selected I'm sure) seem to be quite positive.
This either means that it does add extra spoolup but people don't (choose not to) notice it or any extra spoolup isn't noticeable -- as in, the turbo itself plays a much larger part of the spool cycle counteracting the distance. On this note, the turbo on many airplane engines seems to be located near the back ... but they also server a much different target and purpose.
This however, brings me to two different points:
1) All the applications from STS are for larger engines -- V8's (V6's too?) -- so they start off with SIGNIFICANTLY more power and airflow to begin with. (Poor little 2.5
) 2) They dyno at ~5000ft; how does the this affect the characteristics of a turbo?
On the other hand, some of their customers (not them -- they claim a believable ~30%) claim 50-70%+ gains on 5psi
So, uhm, anyone want to try a rear-mounted turbo setup? Maybe put twins in the boot
Last edited by pstickne; 09-25-2005 at 04:28 PM. Reason: Typos bugging me
Trending Topics
09-25-2005, 04:54 PM
#9
Banned
Join Date: Feb 2004
Location: In self-imposed exile.
Posts: 14,072
Likes: 0
Received 8 Likes
on
7 Posts
The Callaway setup WILL spool faster. As with most things, it's a tradeoff. The placement of the turbo immediately after the manifold / header assembly results in tremendously fast spool times (a very small volume for the exhaust gas to fill before pressure changes affecting spool yield results, also the charge has a lot of heat energy at that point). It also results in shorter turbo life, especially given that the IHI-6B turbo is not water-cooled. On my setup I'll be running a Mocal external oil cooler, after-oiler and some custom air ducting to assist in turbo cooling, especially if I decide to upgrade the IHI-6B to something else later on.
I actually like the Callaway design about 1000 times better than the 951 design. The crossover pipe design is full of problems - it makes working on the car a humongous pain in the ***, it places a lot of heat right under the rear of the oil pan (which causes the gasket and rear main seals to fail prematurely), and of course it does increase turbo lag since the engine must put out a lot more exhaust gasses in order to raise the pressure in a larger volume chamber (the volume of the x-over pipe) in order to result in a spoolup of the turbine. Also, the charge does not contain as much heat energy as it did closer to the exhaust ports by the time it gets to the turbo, although the double-wall construction does help with this a little. I've been told that simply heat-wrapping the crossover pipe will reduce turbo lag by as much as 200 RPM, so it's something to consider.
The only obvious disadvantage to the Callaway system that I see (other than the turbo being subjected to more heat, a problem potentially addressed by more cooling air and/or water jacketing) is that the manifold is very tight to the block and results in very constricting 90-degree turns for the exhaust gasses. This likely inhibits high-RPM flow somewhat, but the overall power of the system and the greatly diminished turbo lag makes this an acceptable tradeoff in my book.
A long-term goal of mine with the system is to redesign the Callaway manifold slightly to (1) maximize interior diameter of the ports and (2) provide for an increasing-diameter "cone" profile going back and (3), reduce the angle that the exhaust gasses must make when exiting the port. In this way, the profile of the manifold will help to "suck" gasses back towards the turbo turbine and also increase the overall efficiency of the system. It's a "second generation" design, so to speak. I would also like to explore this as an R&D project for the 3.0L engine, since the exhaust ports are situated differently.
Actually, the Callaway manifold I have is a later one that comes in two sections - one for cylinders #1 & 2 and then a second piece for cylinders #3 and 4, with a "sleeve" and lead donut sealing the gap. This makes it a lot easier for the system to bolt on and accommodate thermal expansion - something Reeves Callaway must've learned along the way (I'm guessing a few manifolds came back to him with cracks between the #2 and #3 cylinders). This is one of a list of question I'm putting together for him regarding the 944 and 928 conversions he did in the 1980s.
I actually like the Callaway design about 1000 times better than the 951 design. The crossover pipe design is full of problems - it makes working on the car a humongous pain in the ***, it places a lot of heat right under the rear of the oil pan (which causes the gasket and rear main seals to fail prematurely), and of course it does increase turbo lag since the engine must put out a lot more exhaust gasses in order to raise the pressure in a larger volume chamber (the volume of the x-over pipe) in order to result in a spoolup of the turbine. Also, the charge does not contain as much heat energy as it did closer to the exhaust ports by the time it gets to the turbo, although the double-wall construction does help with this a little. I've been told that simply heat-wrapping the crossover pipe will reduce turbo lag by as much as 200 RPM, so it's something to consider.
The only obvious disadvantage to the Callaway system that I see (other than the turbo being subjected to more heat, a problem potentially addressed by more cooling air and/or water jacketing) is that the manifold is very tight to the block and results in very constricting 90-degree turns for the exhaust gasses. This likely inhibits high-RPM flow somewhat, but the overall power of the system and the greatly diminished turbo lag makes this an acceptable tradeoff in my book.
A long-term goal of mine with the system is to redesign the Callaway manifold slightly to (1) maximize interior diameter of the ports and (2) provide for an increasing-diameter "cone" profile going back and (3), reduce the angle that the exhaust gasses must make when exiting the port. In this way, the profile of the manifold will help to "suck" gasses back towards the turbo turbine and also increase the overall efficiency of the system. It's a "second generation" design, so to speak. I would also like to explore this as an R&D project for the 3.0L engine, since the exhaust ports are situated differently.
Actually, the Callaway manifold I have is a later one that comes in two sections - one for cylinders #1 & 2 and then a second piece for cylinders #3 and 4, with a "sleeve" and lead donut sealing the gap. This makes it a lot easier for the system to bolt on and accommodate thermal expansion - something Reeves Callaway must've learned along the way (I'm guessing a few manifolds came back to him with cracks between the #2 and #3 cylinders). This is one of a list of question I'm putting together for him regarding the 944 and 928 conversions he did in the 1980s.
09-25-2005, 05:10 PM
#10
Instructor
Thread Starter
Join Date: Sep 2005
Posts: 131
Likes: 0
Received 0 Likes
on
0 Posts
No, no, I'm not planning on doing anything that extreme to my car. (Adding heat shielding/wrap is more inline with my level.) However, I *am* really interested in, at least basic, theory and design issues. I like to know why things work.
You mention a much faster spool time ("greatly diminished") -- do you have estimates or observed times?
You mention a much faster spool time ("greatly diminished") -- do you have estimates or observed times?
09-25-2005, 05:17 PM
#12
Instructor
Thread Starter
Join Date: Sep 2005
Posts: 131
Likes: 0
Received 0 Likes
on
0 Posts
I wonder how long the connection between the impellors (or is one an impellor and the other just a compressor?) on a turbo system can be ... with a light enough material that can resist fracturing, twisting and otherwise warping ...
09-25-2005, 05:20 PM
#13
Pro
Join Date: Mar 2002
Location: S. Florida
Posts: 608
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by pstickne
However, I *am* really interested in, at least basic, theory and design issues. I like to know why things work.