Do these pictures need captions? (32V Intake Manifold Study - HP)
04-20-2013, 04:41 PM
#31
Chronic Tool Dropper
Lifetime Rennlist
Member
Lifetime Rennlist
Member
It's gonna need 8 individual flappy valves...
04-20-2013, 04:53 PM
#32
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
Not concerned with the fuel "fog" building up and pooling in the gap between the flanges and the silicone couplers to the runners? One would think having the injector bosses on the other side of that coupler would be ideal. Granted, having them further back, at higher rpm/velocity may yield better atomization pre-valve head. Kind of like stand-off injectors on ITB motors.
This intake system is, certainly, the most difficult thing I've ever attempted, in my entire life.....and I've done more "engineering" than most "professional engineers" have ever done.
This design and application is known as "V17", in the shop, BTW.
Those runners and the lower plenum have been done...and redone...many times. Those runners have less than 1mm gap between the plenum and the runners and the same on the "adaptors" to the head....welding the entire thing into one piece would be very easy. The boots are simply "thermal" separators, designed to keep heat from transfering from piece to piece.
For instance, note that those boots are only silicone on the outside. Silicone is not compatible with fuel and would be a very poor choice, in this application. The boots have an inner liner that is compatible with fuel.
This manifold has been thought about....quite a bit.....as with everything that I design and build.
However, in the end, all that matters is how it actually works.
And it could make less power than the stock manifold....
This is just my own humble, underfunded attempt at improving the 928 engine, for my customers.
__________________
greg brown
714 879 9072
GregBBRD@aol.com
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
greg brown
714 879 9072
GregBBRD@aol.com
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
04-20-2013, 05:13 PM
#34
Addict
Lifetime Rennlist
Member
Lifetime Rennlist
Member
I love this...." For instance, note that those boots are only silicone on the outside. Silicone is not compatible with fuel and would be a very poor choice, in this application. The boots have an inner liner that is compatible with fuel. .."
It may look relatively simple but there is so much more involved.....
It may look relatively simple but there is so much more involved.....
04-20-2013, 05:51 PM
#35
Nordschleife Master
Some random thoughts on manifold design. I know nothing more about this particular manifold than what is shown in the pictures, so this is just general speculation.
It might not be. Even car companies have to do some actual testing on a dyno before being certain. That said, this particular design is very common for fuel injected (D)OCH V8's, and it seems to work. It's common especially in cars with low hoodlines because it uses the otherwise dead space in the valley. And certainly much more common design than the variable geometry single/dual plan stock 928 S4/GT/GTS manifold. Most variable geometry V8 intakes nowadays directly alter the runner length by opening eight butterfly valves.
The challenge in this manifold design is to produce a plenum that feeds all the runners equally. It's easiest to do that if one can slow the air down and the way to slow the air down is to make the plenum large. That's why for example 60 and 62 BMW V8's use the runners as the plenum ceiling, maximizing the volume. The pictures don't tell how the plenum was designed for this manifold.
My personal opinion is that having equal length runners is not an objective on its own. A lot of good manifold designs have intentionally unequal length runners. The unequal length runners can be used to flatten the torque curve. With our old LH computers, however, there's no individual cylinder fuel trim table and therefore equal length runner manifold makes sense for our cars.
The injector placement is an interesting question. I think the consensus is that moving the injectors further upstream produces slightly more peak power, while emissions may suffer slightly. If injectors are moved up, then one has to decide what to do with the port shape. The port shape is hardly ideal once the injectors are removed, so one would ideally like to fill the hole in the port somehow. One option is welding the head, which is really expensive and the project will likely snowball. Another option is to have a tongue sticking out of the manifold flange filling the port. The tongue sticking out of the manifold flange makes it expensive to machine the mating surface and I haven't seen than on a cast manifold. Maybe there's a way to attach the tongue to the manifold after the flange mating surface is machined flat.
Fuel pooling on the walls is not a problem. Fuel will always pool on the walls, and even LH has an algorithm to compensate for that. What is a problem is if the fuel makes it to the cylinder in too large droplets. Since I like turbo cars with port conditions resembling a combination of a tornado and Finnish sauna, I don't have fuel atomization problems. Normally aspirated engines often do. Moving injectors upstream gives fuel more time to mix with air, potentially helping.
The temperature question is interesting. The largest effect of the manifold temperature can be felt when taking off say from a traffic light. The air will measure hotter as it has been heated by the intake manifold. However, once the car is running at full load, the air is traveling so fast that it doesn't really heat up in the manifold - the logged air temperature drops almost instantly. So in my opinion the heating issue is mostly relevant for leaving the traffic light or the line. Of course, cooler is always better if every little bit counts. The modern manifolds are made out of plastic, and the insulation is one reason for that.
There is one additional consideration that relates to manifold construction. One wants to build some sort of flex into to the manifold, because the aluminum engine block and heads expand a lot as they heat up, and the cooler manifold much less. I think that with the stock manifold the flex is built exclusively to the manifold-head interface with a thick flexible gasket, which in theory might cause minor port matching issues. It's probably better to build a little flex in many spots than a lot of flex in one spot.
It might not be. Even car companies have to do some actual testing on a dyno before being certain. That said, this particular design is very common for fuel injected (D)OCH V8's, and it seems to work. It's common especially in cars with low hoodlines because it uses the otherwise dead space in the valley. And certainly much more common design than the variable geometry single/dual plan stock 928 S4/GT/GTS manifold. Most variable geometry V8 intakes nowadays directly alter the runner length by opening eight butterfly valves.
The challenge in this manifold design is to produce a plenum that feeds all the runners equally. It's easiest to do that if one can slow the air down and the way to slow the air down is to make the plenum large. That's why for example 60 and 62 BMW V8's use the runners as the plenum ceiling, maximizing the volume. The pictures don't tell how the plenum was designed for this manifold.
My personal opinion is that having equal length runners is not an objective on its own. A lot of good manifold designs have intentionally unequal length runners. The unequal length runners can be used to flatten the torque curve. With our old LH computers, however, there's no individual cylinder fuel trim table and therefore equal length runner manifold makes sense for our cars.
The injector placement is an interesting question. I think the consensus is that moving the injectors further upstream produces slightly more peak power, while emissions may suffer slightly. If injectors are moved up, then one has to decide what to do with the port shape. The port shape is hardly ideal once the injectors are removed, so one would ideally like to fill the hole in the port somehow. One option is welding the head, which is really expensive and the project will likely snowball. Another option is to have a tongue sticking out of the manifold flange filling the port. The tongue sticking out of the manifold flange makes it expensive to machine the mating surface and I haven't seen than on a cast manifold. Maybe there's a way to attach the tongue to the manifold after the flange mating surface is machined flat.
Fuel pooling on the walls is not a problem. Fuel will always pool on the walls, and even LH has an algorithm to compensate for that. What is a problem is if the fuel makes it to the cylinder in too large droplets. Since I like turbo cars with port conditions resembling a combination of a tornado and Finnish sauna, I don't have fuel atomization problems. Normally aspirated engines often do. Moving injectors upstream gives fuel more time to mix with air, potentially helping.
The temperature question is interesting. The largest effect of the manifold temperature can be felt when taking off say from a traffic light. The air will measure hotter as it has been heated by the intake manifold. However, once the car is running at full load, the air is traveling so fast that it doesn't really heat up in the manifold - the logged air temperature drops almost instantly. So in my opinion the heating issue is mostly relevant for leaving the traffic light or the line. Of course, cooler is always better if every little bit counts. The modern manifolds are made out of plastic, and the insulation is one reason for that.
There is one additional consideration that relates to manifold construction. One wants to build some sort of flex into to the manifold, because the aluminum engine block and heads expand a lot as they heat up, and the cooler manifold much less. I think that with the stock manifold the flex is built exclusively to the manifold-head interface with a thick flexible gasket, which in theory might cause minor port matching issues. It's probably better to build a little flex in many spots than a lot of flex in one spot.
04-20-2013, 09:22 PM
#36
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
Because of the fact that the 928 engine is a fairly "wide" V-8 and the hood sits relatively low, in relationship to the heads, there is "limited space" for a manifold with a "common" plenum....which was one of my design criteria.
One can either put the plenum up high (near the hood), or down low (in the valley.) "Up high" makes for very short runners....which can be detrimental to using/containing the intake pulses....which is one of the things that I'm trying to take advantage of.
"Down low" results in a manifold that looks like this....this is why many current V-8 engines look this way.
One can either put the plenum up high (near the hood), or down low (in the valley.) "Up high" makes for very short runners....which can be detrimental to using/containing the intake pulses....which is one of the things that I'm trying to take advantage of.
"Down low" results in a manifold that looks like this....this is why many current V-8 engines look this way.
04-20-2013, 09:23 PM
#37
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
04-20-2013, 09:29 PM
#38
Race Director
a couple thoughts......1st Doc knows what he is doing......by 1,000%
2nd....airflow is airflow...in 1970 or 2013.....look at high HP "street" V8 engines and what do you see.....manifolds that look like this....look at Porsche manifolds....chebby manifolds....in other words factory made manifolds that offer FLAT torque curves, idle silly low and make emissions standards... they look like this but made of plastic....
2nd....airflow is airflow...in 1970 or 2013.....look at high HP "street" V8 engines and what do you see.....manifolds that look like this....look at Porsche manifolds....chebby manifolds....in other words factory made manifolds that offer FLAT torque curves, idle silly low and make emissions standards... they look like this but made of plastic....
04-20-2013, 09:30 PM
#39
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
Some random thoughts on manifold design. I know nothing more about this particular manifold than what is shown in the pictures, so this is just general speculation.
It might not be. Even car companies have to do some actual testing on a dyno before being certain. That said, this particular design is very common for fuel injected (D)OCH V8's, and it seems to work. It's common especially in cars with low hoodlines because it uses the otherwise dead space in the valley. And certainly much more common design than the variable geometry single/dual plan stock 928 S4/GT/GTS manifold. Most variable geometry V8 intakes nowadays directly alter the runner length by opening eight butterfly valves.
The challenge in this manifold design is to produce a plenum that feeds all the runners equally. It's easiest to do that if one can slow the air down and the way to slow the air down is to make the plenum large. That's why for example 60 and 62 BMW V8's use the runners as the plenum ceiling, maximizing the volume. The pictures don't tell how the plenum was designed for this manifold.
My personal opinion is that having equal length runners is not an objective on its own. A lot of good manifold designs have intentionally unequal length runners. The unequal length runners can be used to flatten the torque curve. With our old LH computers, however, there's no individual cylinder fuel trim table and therefore equal length runner manifold makes sense for our cars.
The injector placement is an interesting question. I think the consensus is that moving the injectors further upstream produces slightly more peak power, while emissions may suffer slightly. If injectors are moved up, then one has to decide what to do with the port shape. The port shape is hardly ideal once the injectors are removed, so one would ideally like to fill the hole in the port somehow. One option is welding the head, which is really expensive and the project will likely snowball. Another option is to have a tongue sticking out of the manifold flange filling the port. The tongue sticking out of the manifold flange makes it expensive to machine the mating surface and I haven't seen than on a cast manifold. Maybe there's a way to attach the tongue to the manifold after the flange mating surface is machined flat.
Fuel pooling on the walls is not a problem. Fuel will always pool on the walls, and even LH has an algorithm to compensate for that. What is a problem is if the fuel makes it to the cylinder in too large droplets. Since I like turbo cars with port conditions resembling a combination of a tornado and Finnish sauna, I don't have fuel atomization problems. Normally aspirated engines often do. Moving injectors upstream gives fuel more time to mix with air, potentially helping.
The temperature question is interesting. The largest effect of the manifold temperature can be felt when taking off say from a traffic light. The air will measure hotter as it has been heated by the intake manifold. However, once the car is running at full load, the air is traveling so fast that it doesn't really heat up in the manifold - the logged air temperature drops almost instantly. So in my opinion the heating issue is mostly relevant for leaving the traffic light or the line. Of course, cooler is always better if every little bit counts. The modern manifolds are made out of plastic, and the insulation is one reason for that.
There is one additional consideration that relates to manifold construction. One wants to build some sort of flex into to the manifold, because the aluminum engine block and heads expand a lot as they heat up, and the cooler manifold much less. I think that with the stock manifold the flex is built exclusively to the manifold-head interface with a thick flexible gasket, which in theory might cause minor port matching issues. It's probably better to build a little flex in many spots than a lot of flex in one spot.
It might not be. Even car companies have to do some actual testing on a dyno before being certain. That said, this particular design is very common for fuel injected (D)OCH V8's, and it seems to work. It's common especially in cars with low hoodlines because it uses the otherwise dead space in the valley. And certainly much more common design than the variable geometry single/dual plan stock 928 S4/GT/GTS manifold. Most variable geometry V8 intakes nowadays directly alter the runner length by opening eight butterfly valves.
The challenge in this manifold design is to produce a plenum that feeds all the runners equally. It's easiest to do that if one can slow the air down and the way to slow the air down is to make the plenum large. That's why for example 60 and 62 BMW V8's use the runners as the plenum ceiling, maximizing the volume. The pictures don't tell how the plenum was designed for this manifold.
My personal opinion is that having equal length runners is not an objective on its own. A lot of good manifold designs have intentionally unequal length runners. The unequal length runners can be used to flatten the torque curve. With our old LH computers, however, there's no individual cylinder fuel trim table and therefore equal length runner manifold makes sense for our cars.
The injector placement is an interesting question. I think the consensus is that moving the injectors further upstream produces slightly more peak power, while emissions may suffer slightly. If injectors are moved up, then one has to decide what to do with the port shape. The port shape is hardly ideal once the injectors are removed, so one would ideally like to fill the hole in the port somehow. One option is welding the head, which is really expensive and the project will likely snowball. Another option is to have a tongue sticking out of the manifold flange filling the port. The tongue sticking out of the manifold flange makes it expensive to machine the mating surface and I haven't seen than on a cast manifold. Maybe there's a way to attach the tongue to the manifold after the flange mating surface is machined flat.
Fuel pooling on the walls is not a problem. Fuel will always pool on the walls, and even LH has an algorithm to compensate for that. What is a problem is if the fuel makes it to the cylinder in too large droplets. Since I like turbo cars with port conditions resembling a combination of a tornado and Finnish sauna, I don't have fuel atomization problems. Normally aspirated engines often do. Moving injectors upstream gives fuel more time to mix with air, potentially helping.
The temperature question is interesting. The largest effect of the manifold temperature can be felt when taking off say from a traffic light. The air will measure hotter as it has been heated by the intake manifold. However, once the car is running at full load, the air is traveling so fast that it doesn't really heat up in the manifold - the logged air temperature drops almost instantly. So in my opinion the heating issue is mostly relevant for leaving the traffic light or the line. Of course, cooler is always better if every little bit counts. The modern manifolds are made out of plastic, and the insulation is one reason for that.
There is one additional consideration that relates to manifold construction. One wants to build some sort of flex into to the manifold, because the aluminum engine block and heads expand a lot as they heat up, and the cooler manifold much less. I think that with the stock manifold the flex is built exclusively to the manifold-head interface with a thick flexible gasket, which in theory might cause minor port matching issues. It's probably better to build a little flex in many spots than a lot of flex in one spot.
04-20-2013, 11:15 PM
#40
Nordschleife Master
04-21-2013, 01:47 AM
#43
Addict
Lifetime Rennlist
Member
Lifetime Rennlist
Member
Greg will not be selling them until he KNOWS exactly how they work !!! Because that is how he does it. Just one of the reasons why I respect the way he does business.
04-21-2013, 02:55 AM
#44
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
Should test in the next 2-3 weeks. Will be using John Speake's Alpha mafless system with Sharktuning, along with Jim Corenman's help. Jim has one of these Alpha systems running on his own car.
Both of these guys are Geniuses....and we are lucky to have them helping us with these cars!
04-21-2013, 05:11 AM
#45
Rennlist Member
FOR SALE: - 1 newly reconditioned & coated stock intake.
PRICE: Offers
AVAILABLE: 1 Week AFTER these new intakes hit the market.
.............................
................................
Myles
PRICE: Offers
AVAILABLE: 1 Week AFTER these new intakes hit the market.
.............................
................................Myles