Do these pictures need captions? (32V Intake Manifold Study - HP)
10-20-2013, 12:51 AM
#166
Instructor
...the entire top piece, with the multiple entry angles and planes....digitally mastered and then cut from one piece of aluminum on a CNC. Expensive in small quantities....but this thing is going to be a real bitch to remake, by hand....even with this prototype sitting in front of us.
Many moons ago I did a project vaguely like this on an OTS (Other-Than-Shark), albeit on a 4-cylinder car, and I was floored at how many dependent variables there were. And we didn't have the space constraints for the plenum that you do.
The CNC approach is probably the right answer but I do not envy you the rapid prototyping costs to validate the code. You will, at the very least, know exactly what qty 1 will cost (including a non-amortized setup cost of course), and can simply supply them on demand rather than stocking inventory. If customers push back, offer to let them pay the machine shop directly and simply require them to pay a licensing fee from you
10-20-2013, 01:12 AM
#167
I was one of the passengers who time travelled in Rob's Millennium Falcon. All I can say is, hide your daughters away. This GTS is a menace to society (and any unwary AMG.)
The new John Holmes has entered the room.
The new John Holmes has entered the room.
10-20-2013, 01:21 AM
#168
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
With a working prototype...... additional CNC pieces makes sense, but what about injection molded plastic like all the LS intakes?
Like everything else it takes VOLUME to offset the startup costs & I would imagine there won't be enough to sell to make it worthwhile...
Kibort could use one, same for Fan & others
Like everything else it takes VOLUME to offset the startup costs & I would imagine there won't be enough to sell to make it worthwhile...
Kibort could use one, same for Fan & others
Huge investment.
__________________
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!
10-20-2013, 01:28 AM
#169
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
That would, in theory at least, also simplify the creation of a 5.0 version. However with the reduction of runner diameter you'll also probably have to change the radius of every runner's primary curve to keep everything centered and positioned correctly.
Many moons ago I did a project vaguely like this on an OTS (Other-Than-Shark), albeit on a 4-cylinder car, and I was floored at how many dependent variables there were. And we didn't have the space constraints for the plenum that you do.
The CNC approach is probably the right answer but I do not envy you the rapid prototyping costs to validate the code. You will, at the very least, know exactly what qty 1 will cost (including a non-amortized setup cost of course), and can simply supply them on demand rather than stocking inventory. If customers push back, offer to let them pay the machine shop directly and simply require them to pay a licensing fee from you
Many moons ago I did a project vaguely like this on an OTS (Other-Than-Shark), albeit on a 4-cylinder car, and I was floored at how many dependent variables there were. And we didn't have the space constraints for the plenum that you do.
The CNC approach is probably the right answer but I do not envy you the rapid prototyping costs to validate the code. You will, at the very least, know exactly what qty 1 will cost (including a non-amortized setup cost of course), and can simply supply them on demand rather than stocking inventory. If customers push back, offer to let them pay the machine shop directly and simply require them to pay a licensing fee from you
A 5.0 version will be somewhat easier, but certainly not simple.
We will probably start on this, soon, funding permitting.
The results from the 6.5 version are far beyond anything I expected.....
10-20-2013, 01:38 AM
#170
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
With a working prototype...... additional CNC pieces makes sense, but what about injection molded plastic like all the LS intakes?
Like everything else it takes VOLUME to offset the startup costs & I would imagine there won't be enough to sell to make it worthwhile...
Kibort could use one, same for Fan & others
Like everything else it takes VOLUME to offset the startup costs & I would imagine there won't be enough to sell to make it worthwhile...
Kibort could use one, same for Fan & others
Pretty sure Rob's car is faster than Joseph's race car.....which is what I was shooting for.
The "lack" of a central plenum on that Threshie intake doesn't do much for inertial supercharging.....and the "update" on the "Alpha N" injection is also a great improvement.
10-20-2013, 02:09 AM
#171
Archive Gatekeeper
Rennlist Member
Rennlist Member
Rob not only exceeded the speed on the GB test track, he blew it into oblivion. Previous best was 95mph. Now 140mph
10-20-2013, 11:04 AM
#173
Nordschleife Master
Looking forward to seeing the dyno graph from that engine.
I don't see much peak power benefit from an aftermarket ECU with an equal length runner, single plenum manifold. Each cylinder is getting about the same amount of air at each rpm. The duty cycle of the injectors is high. The gains from either individual cylinder trim or sequential injection are going to be modest at best. People are making a lot power with carburetors, after all. I think that blanket recommendations to switch the ECU are bad advice, unless there's a specific problem that one is trying to solve with the ECU. My opinion anyway.
The geometry of fitting long runners on a 928 engine gets more difficult as the runner diameter and taper increase. The heads are quite far apart from each other, and even with the 122 mm bore spacing the ports are pretty close to each other. This makes any manifold in which the runners cross between banks pretty challenging, especially if the ideal degree of taper is used.
I think that a cross-ram manifold could be made to work with a large enough cross-over pipe that makes the two plenums look like a single plenum from pulse tuning perspective.
Here's a question about this kind manifold with a single plenum, like the one on Rob Edwards's car: Why does the plenum volume matter? It's a single plenum V8 manifold. The plenum sees pulses that are garbled/cancel each other. The plenum volume doesn't really do anything in terms of tuning. I believe that one just needs a plenum that is large enough to slow down the air and distribute air equally to the runners, and the volume beyond that is unimportant. In this manifold, I see the plenum challenge being creating a plenum that fits in the space and distributes air equally to the runners. That's my reaction, obviously without having tried to fabricate one.
I don't see much peak power benefit from an aftermarket ECU with an equal length runner, single plenum manifold. Each cylinder is getting about the same amount of air at each rpm. The duty cycle of the injectors is high. The gains from either individual cylinder trim or sequential injection are going to be modest at best. People are making a lot power with carburetors, after all. I think that blanket recommendations to switch the ECU are bad advice, unless there's a specific problem that one is trying to solve with the ECU. My opinion anyway.
The geometry of fitting long runners on a 928 engine gets more difficult as the runner diameter and taper increase. The heads are quite far apart from each other, and even with the 122 mm bore spacing the ports are pretty close to each other. This makes any manifold in which the runners cross between banks pretty challenging, especially if the ideal degree of taper is used.
I think that a cross-ram manifold could be made to work with a large enough cross-over pipe that makes the two plenums look like a single plenum from pulse tuning perspective.
Here's a question about this kind manifold with a single plenum, like the one on Rob Edwards's car: Why does the plenum volume matter? It's a single plenum V8 manifold. The plenum sees pulses that are garbled/cancel each other. The plenum volume doesn't really do anything in terms of tuning. I believe that one just needs a plenum that is large enough to slow down the air and distribute air equally to the runners, and the volume beyond that is unimportant. In this manifold, I see the plenum challenge being creating a plenum that fits in the space and distributes air equally to the runners. That's my reaction, obviously without having tried to fabricate one.
10-20-2013, 01:35 PM
#175
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
Looking forward to seeing the dyno graph from that engine.
I find rear wheel dyno charts interesting, at best. A quick look through dyno charts provided by just the people that are active on Rennlist reveals the obvious problem.....there is a severe issue with accuracy and reality. I believe that this information is only useful as a tuning tool specific only to the same dyno, the same vehicle, with the same "preparations" to the target vehicle, prior to testing.
I don't see much peak power benefit from an aftermarket ECU with an equal length runner, single plenum manifold. Each cylinder is getting about the same amount of air at each rpm. The duty cycle of the injectors is high. The gains from either individual cylinder trim or sequential injection are going to be modest at best. People are making a lot power with carburetors, after all. I think that blanket recommendations to switch the ECU are bad advice, unless there's a specific problem that one is trying to solve with the ECU. My opinion anyway.
I totally agree that an aftermarket ECU offers very little benefit.....and could be a negative thing. I believe that knock sensing and correction is very important. I consider the Bosch knock sensor system to be very good.....and from my experience the aftermarket knock sensing systems are crude (keep in mind that I do not have experience with all of the aftermarket systems.)
And anyone that has converted from carburetors to fuel injection on a well tuned hot rod engine will tell you the same thing......carburetors make more power.
The geometry of fitting long runners on a 928 engine gets more difficult as the runner diameter and taper increase. The heads are quite far apart from each other, and even with the 122 mm bore spacing the ports are pretty close to each other. This makes any manifold in which the runners cross between banks pretty challenging, especially if the ideal degree of taper is used.
Agreed. Very quick measurements in the 928 engine bay point out the obvious. Eight 2.125" runners equal 17"...if they are all touching each other perfectly in one straight line. Take out your tape measure and measure from the suspension cross bar towards the rear of the engine and see where that 17" ends up.....
I think that a cross-ram manifold could be made to work with a large enough cross-over pipe that makes the two plenums look like a single plenum from pulse tuning perspective.
I studied this, in the "design phase", at length. For the pulses to "look like a single plenum", the connections from plenum to plenum have to be fairly large....and connect in several different places, for all cylinders to get the same results. Connecting those two separate plenums, from one side of the engine bay to the other is a very difficult task. in the available space of the 928 engine bay. It would not be an issue, if one wanted to put on a large bulge in the hood, but virtually impossible with the stock hood. Very few of my 928 customers would "accept" a huge bulge in the hood.
Here's a question about this kind manifold with a single plenum, like the one on Rob Edwards's car: Why does the plenum volume matter? It's a single plenum V8 manifold. The plenum sees pulses that are garbled/cancel each other. The plenum volume doesn't really do anything in terms of tuning. I believe that one just needs a plenum that is large enough to slow down the air and distribute air equally to the runners, and the volume beyond that is unimportant. In this manifold, I see the plenum challenge being creating a plenum that fits in the space and distributes air equally to the runners. That's my reaction, obviously without having tried to fabricate one.
The size of the plenum has been a big question, for me, also. Quickly looking at the plenum volumes of what Ford and Chevy use, in their various displacement engines, makes me think that it must be important. However, what effect it has on the running of an engine is a mystery. We are completely prepared to change the volume of the plenum we are currently running....as it is a bit larger than what either Chevy or Ford would use for an engine of this displacement (and making it smaller is much easier than making it bigger....however the costs of testing all of these individual "potential" changes is very large, for an effort like this. Unless we see some significant and obvious "holes" in the dyno results, I doubt we will worry about this.
I find rear wheel dyno charts interesting, at best. A quick look through dyno charts provided by just the people that are active on Rennlist reveals the obvious problem.....there is a severe issue with accuracy and reality. I believe that this information is only useful as a tuning tool specific only to the same dyno, the same vehicle, with the same "preparations" to the target vehicle, prior to testing.
I don't see much peak power benefit from an aftermarket ECU with an equal length runner, single plenum manifold. Each cylinder is getting about the same amount of air at each rpm. The duty cycle of the injectors is high. The gains from either individual cylinder trim or sequential injection are going to be modest at best. People are making a lot power with carburetors, after all. I think that blanket recommendations to switch the ECU are bad advice, unless there's a specific problem that one is trying to solve with the ECU. My opinion anyway.
I totally agree that an aftermarket ECU offers very little benefit.....and could be a negative thing. I believe that knock sensing and correction is very important. I consider the Bosch knock sensor system to be very good.....and from my experience the aftermarket knock sensing systems are crude (keep in mind that I do not have experience with all of the aftermarket systems.)
And anyone that has converted from carburetors to fuel injection on a well tuned hot rod engine will tell you the same thing......carburetors make more power.
The geometry of fitting long runners on a 928 engine gets more difficult as the runner diameter and taper increase. The heads are quite far apart from each other, and even with the 122 mm bore spacing the ports are pretty close to each other. This makes any manifold in which the runners cross between banks pretty challenging, especially if the ideal degree of taper is used.
Agreed. Very quick measurements in the 928 engine bay point out the obvious. Eight 2.125" runners equal 17"...if they are all touching each other perfectly in one straight line. Take out your tape measure and measure from the suspension cross bar towards the rear of the engine and see where that 17" ends up.....
I think that a cross-ram manifold could be made to work with a large enough cross-over pipe that makes the two plenums look like a single plenum from pulse tuning perspective.
I studied this, in the "design phase", at length. For the pulses to "look like a single plenum", the connections from plenum to plenum have to be fairly large....and connect in several different places, for all cylinders to get the same results. Connecting those two separate plenums, from one side of the engine bay to the other is a very difficult task. in the available space of the 928 engine bay. It would not be an issue, if one wanted to put on a large bulge in the hood, but virtually impossible with the stock hood. Very few of my 928 customers would "accept" a huge bulge in the hood.
Here's a question about this kind manifold with a single plenum, like the one on Rob Edwards's car: Why does the plenum volume matter? It's a single plenum V8 manifold. The plenum sees pulses that are garbled/cancel each other. The plenum volume doesn't really do anything in terms of tuning. I believe that one just needs a plenum that is large enough to slow down the air and distribute air equally to the runners, and the volume beyond that is unimportant. In this manifold, I see the plenum challenge being creating a plenum that fits in the space and distributes air equally to the runners. That's my reaction, obviously without having tried to fabricate one.
The size of the plenum has been a big question, for me, also. Quickly looking at the plenum volumes of what Ford and Chevy use, in their various displacement engines, makes me think that it must be important. However, what effect it has on the running of an engine is a mystery. We are completely prepared to change the volume of the plenum we are currently running....as it is a bit larger than what either Chevy or Ford would use for an engine of this displacement (and making it smaller is much easier than making it bigger....however the costs of testing all of these individual "potential" changes is very large, for an effort like this. Unless we see some significant and obvious "holes" in the dyno results, I doubt we will worry about this.
10-20-2013, 06:52 PM
#176
Instructor
Join Date: Feb 2005
Posts: 209
Likes: 0
Received 0 Likes
on
0 Posts
I believe cross rams are only useful on pushrod engines with low profile hoods. I think they do look cool though. Has there ever been a oem cross ram style application? I think the style of intake Greg is building suits the 928 perfectly as that's what Porsche and Audi are currently using. In regards to throttle response, I thought it was universally agreed that all manifold designs bow down at the feet of the All mighty ITB? They sure are a pain to build an air box for though. Making one that appears almost factory stock like Gregs manifold? Good luck with that if you live behind the green curtain.
10-20-2013, 07:29 PM
#177
Three Wheelin'
This early v8 Aston has a mechanical bosch system with the throttle butterflies under the air filters at the end of the intake tubes. Pretty close to the length of a cross ram.
http://www.boldride.com/ride/1970/as...-v8#gallery/11
http://www.boldride.com/ride/1970/as...-v8#gallery/11
10-20-2013, 08:14 PM
#178
Rennlist Member
I was there, riding shotgun and helping Rob tune the Alpha system fuel maps. I don't recall much, however.
Trying to focus on AFR numbers became near impossible when everything turned plaid about halfway through fourth gear... Thank goodness for ST data logs!
Trying to focus on AFR numbers became near impossible when everything turned plaid about halfway through fourth gear... Thank goodness for ST data logs!
10-20-2013, 08:20 PM
#179
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
I believe cross rams are only useful on pushrod engines with low profile hoods. I think they do look cool though. Has there ever been a oem cross ram style application? I think the style of intake Greg is building suits the 928 perfectly as that's what Porsche and Audi are currently using. In regards to throttle response, I thought it was universally agreed that all manifold designs bow down at the feet of the All mighty ITB? They sure are a pain to build an air box for though. Making one that appears almost factory stock like Gregs manifold? Good luck with that if you live behind the green curtain.
Mostly "show"....not much "go".
ITB's certainly allow unrestricted flow though their given butterfly, but unless they have a runner above them, leading to a central plenum, there's never going to be any cylinder filling above 100%......inertial supercharging is completely "out the window".
10-20-2013, 08:44 PM
#180
Rennlist Member
On a serious note - as 3D printer and material technology advances, I wonder if these runners could be printed?