Nissan Juke : Juke Forums banner

381 - 400 of 454 Posts

·
Registered
Joined
·
1,485 Posts
Discussion Starter #381
Here's what I'm talking about on the stock Intake manifold design:


The stocker is about 9.0" runner length plus 3" flange/intake valve or 12.0" total length. This gives a peak return pulse on the Crower cams 218* Intake duration right around 6200 rpms. This is pretty much where the Juke spends most of it's time at WOT. If I wanted to peak @ 7000 rpms then I'd shorten the runner to a 7.6" + 3" flange/intake valve. The stocker is plenty darned good where it is actually even for the upgraded cams. Since I will be in CVT mode I could potentially take advantage of the 7000 rpm power band, so an upgrade to 7.6" runner intake manifold would also benefit. Just saying, the intake manifold is the last thing I'd upgrade.
 

·
Super Moderator
Joined
·
10,728 Posts
Yeah but with the upgraded intake runners. Can't the overall throttle response be faster in the lower rpms since you have done all the porting / restriction work. Assuming the turbo can spool fast enough.

Like some day I will get that Chinese Intercooler installed and get a Butt dyno comparison of it compared to my FMIC. Sure it will heat soak sooner but you are eliminating a lot of piping.

BTW after driving today. Put 100 miles on the Juke in 95 deg weather. Came home. Idled to cool her off a little. Oh man. My fenders were HOT. Intake was hot. The whole front end was Hot. Everything was Hot. Hood etc etc.

I always put my hood up when I get home. The heat just rolls out of the engine bay. Especially towards the cowl.
 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #383
Have to try it out. I'd say the cams and headwork will help torque almost everywhere above the boost threshold. An intake manifold and exhaust manifold I'd figure above 5,000-5,500 rpms only, based on my past cars. So different mods, different rpms for power.

You need a vented hood, best invention.
 

·
Super Moderator
Joined
·
10,728 Posts
I might go back to the CAI. I had one but sold it. It was chrome but I painted it black. I dont have to worry about rain so. We shall see.

BUT having the shortest paths possible should have gains in certain areas. It is just fun to experiment.
 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #385
I like having a CAI, I think probably bigger than 3" is needed, more like 4" if you are near 300 w.h.p. My plan is to increase the diameter maybe 12" before the MAF but 4" down to the fender well. I ran into this problem on my other cars, 4" is actually way better.

Shorter is better, true.
 

·
Super Moderator
Joined
·
10,728 Posts
Well the 2J is convertible. SRI and CAI you can set it up as you like plus it doesnt need the Injen little adapter pipe. So you could run that and add 4" in front of the 2J SRi. Just an idea.

 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #387 (Edited)
Yes, 4" as far as you can run it I would think. I ran a 3" DYI CAI before on my 400 whp Eclipse and it just strangled the motor, 4" opened it up nicely.

I just took some measurements: The cylinder head is 4.75" from flange to intake valves, the stock Intake manifold runners are 8.75" C/L more or less. So total length is about 13.5" total from Plenum to Intake valves, roughly. Plenum volume is about 1.2 Liters. So using that calculator the motor is tuned to hit a peak of about 5500 rpms. A power peak of 7000 rpms would require shortening the Intake runners to about 5.85" long. To make the most power plus reduce compressor surge with a bigger turbo, the plenum should be 1.5x the engine displacement, or 2.4L. It's this surge reduction that is a major benefit on the bigger turbo's that spool quick. Pretty standard stuff.
The intake manifold is PA6-GF30 (i.e. 30% Glass filled Nylon) made by Mahle, it's a very nicely made part actually.

So yeah, there is definitely room for improvement on the intake manifold for sure. Just kinda depends on how extreme the setup is.
 

·
Super Moderator
Joined
·
10,728 Posts
Yeah we have not heard of anyone blowing up their plastic intake. Which years ago we would have been. Plastic WHAT ???
 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #389
Yeah, it won't happen with the Mahle Juke Intake manifold. It's insanely strong and light weight, maybe 3 lbs or so. They mainly have to be resistant to backfire pressures of about 75-100 psi. I design stuff similar to this but for water applications, and we can hold about 400-500 psi using GF PPO/PPS filled plastic, though they operate at 30-125 psi nominally. Plastics are absolutely 100% the way to go for an Intake manifold as they have a lot of benefits if made correctly. The stock design is worth copying, just needs those improvements for higher rpms.
 

·
Super Moderator
Joined
·
1,778 Posts
nobody tell Matt about your intake info...

im intrigued how frankenstein your engine is going to be. at first it was CVT rebuild. then CVT rebuild with improved part(s), then CVT complete rebuild of internals and mess with valve body, then add the engine build, valve job, head resurface, cams, exhaust, intake, etc, etc...

It'll still be an MR16DDT, but really only in name and shape.
 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #391
nobody tell Matt about your intake info...

im intrigued how frankenstein your engine is going to be. at first it was CVT rebuild. then CVT rebuild with improved part(s), then CVT complete rebuild of internals and mess with valve body, then add the engine build, valve job, head resurface, cams, exhaust, intake, etc, etc...

It'll still be an MR16DDT, but really only in name and shape.
Hopefully I didn't say anything that contradicts what he's doing, haven't read it.

It's a bit Frankenstein....lol. Hopefully it all comes together. Just need to start building the motor and it'll start coming together. It kinda progressed from one project to another, you know how that goes. I get bored kinda easily.
 

·
Super Moderator
Joined
·
10,728 Posts
There is no perfect way to build / rebuild a motor / CVT. We all have our different goals / styles / knowledge.

I am a big fan of removing the restrictions in the Juke. She was designed for Low RPM torque so the top end is just waiting to be found.
 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #393
True. Designing mostly for high rpm where I can take advantage of the limited torque I can safely put down. Probably I can keep making power to 7000-7200 rpms if my plans unfold correctly. I've run the CVT at 6900 rpms @ WOT in NORMAL mode so I was almost there already, though the power was dropping off a bit with my limited mods. What you've seen on the typical CVT dyno runs is manual mode and shifting at 5900 rpms. There is a huge jump in acceleration running in NORMAL mode, even though it feels slower it's tons tons faster than manual mode. So dyno pulls won't show the true potential. I need the ECUTek tuner to bump the rev limiter up to maybe 7200-7400 rpms or so to lock the CVT near that during WOT pulls.

I mentioned the CVT oil pump being maxxed at 7000 rpms for cavitation, so I'll have to measure CVT oil pump flow/pressure to see where to lock the rev limiter but it'll be close. I'm not really going to be rpm limited on the engine, so much as the transmission. This is where the tuning phase will come in. A custom exhaust manifold and some other stuff I have planned will hopefully make that happen. A 6spd Juke RS wouldn't need to follow the same approach, but with the CVT I have to stretch the limited torque up higher in the rpm band to make the most power. The CVT transmissions introduces so many variables that it's not even comparable to a manual transmission car anymore.

So, very much it's a customized build for my purposes. This build was never for big power and I'm always mentioning that because I can't work miracles on the transmission, though it'll be much better than stock. I'll be very happy if I had 300 w.h.p. and that is by my calculations very doable while still keeping some decent reliability, if only occasionally running at that level. By my estimates the CVT has about a 30% power loss at higher rpms, so I don't see why I can't make 390 h.p. crank with the right turbo. Probably dial that back for daily driving as these cars simply don't need huge power to be fun. My next big project will probably be weight savings.

Still waiting on the darned block.
 
  • Like
Reactions: menshealthadvocate

·
Super Moderator
Joined
·
10,728 Posts
Weight is easy. Just take out the whole rear seat and cargo area. Haha. It is just you and your GF.
 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #395
Lol.....that is true. It's gonna get expensive after the first 100 lbs. We'll see if I'm still motivated at that point. The Juke needs some brakes and suspension, so if I replace that I'll go for weight savings there.
 

·
Super Moderator
Joined
·
10,728 Posts
Yeah wheels. Brakes and Suspension will save corner weight but costs $$$
 

·
Registered
Joined
·
1,306 Posts
Get z1 motorsports 2 piece brembo size rotors wich is -20 lbs total
A wilwood caliper kit for g35 from revolution brakes LLC wich is -10lbs total remove rear seats and spare tire and jack -100lbs
Konig dekagram wheels are about 18lbs for 18x9 so thats about -20lbs total compared to oem
A set of coilovers would be about -30lbs lighter overall
No breakfast -5lbs
No shoes -5lbs
Leave wallet full of cash home just bring id and debit card-5lbs
 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #398
Get z1 motorsports 2 piece brembo size rotors wich is -20 lbs total
A wilwood caliper kit for g35 from revolution brakes LLC wich is -10lbs total remove rear seats and spare tire and jack -100lbs
Konig dekagram wheels are about 18lbs for 18x9 so thats about -20lbs total compared to oem
A set of coilovers would be about -30lbs lighter overall
No breakfast -5lbs
No shoes -5lbs
Leave wallet full of cash home just bring id and debit card-5lbs
I'll have to review this list. I've been pricing stuff out, expensive but if I do it over time I'll get the weight savings too.
 

·
Registered
Joined
·
1,306 Posts
2 piece rotors ar $1k
Wheels are $1k
Caliper kit is $400
All the other stuff is just hard work
 

·
Registered
Joined
·
1,485 Posts
Discussion Starter #400 (Edited)
Ok, another subject here, long but worth the read for entertainment purposes.

Background:
Decided to come up with an Intake manifold design and possible durable prototype for the Juke. Basically I was bored and there wasn't anything else out there that was available that I liked. The entire engine and transmission build was mainly a learning experience and kinda progressed from there. I neither care about being the first to do something, or having the most horsepower, etc., but rather doing it right or the way I prefer it if it's not available. I design products similar in construction/materials to this intake manifold for a living, though in a different field. So it's an interesting challenge if not different. Having to handle airflow, hot engine temps, oil/fuel vapors, vibration, etc. poses it's certain sets of challenges. I have a few other things I'd like to do as well on the Juke but it'll have to wait until the car is running.....one day...lol.

Intake manifold design basic requirements:
Simply, the Juke upgraded intake manifold has to fit in the stock location. All emissions equipment has to bolt-up as stock. That includes all the mounting tabs for wire harnesses, vacuum hoses, etc. etc. have to be there. I have the option to laser scan the stock intake manifold or CMM the mounting points. I may CMM as I only need a reference point and some hard points and a couple of planes. Basically, bolt-hole C/L x,y,z and and zero point. I'm thinking maybe 30-40 points and I can grab the
throttle body bolt-holes, cylinder head flange plane, all the hard points, etc. Those aspects I'm not concerned with right now as the datum planes and mounting points are roughed in for now. The cylinder head flange is sitting at a 36* off vertical plane angle, which was important for designing the runners, this easily measured with a protractor. Basically, the runners are at a 126* bend give or take. I have the cylinder head, block & intake manifold sitting here for test fittment, plus the ability to build a fixture off the stock Intake manifold. I can also potentially point cloud scan the stock manifold and overlay against the CAD design and that'll be the best way to check fitups quickly.

What I did was review the Mitsubishi Intake manifold EVO X design, Cosworth, Blitz, Skunkworks, stock Juke and created my own hybrid design. Non of this stuff is revolutionary. The intake manifold will be designed for my setup. The Mitsubishi EVO X intake manifold is a good example of a deisgn good for at least 7000-7400 rpms on stock cams. Probably it'll spin out to 8,000 rpms when I install my Cosworth MX1 cams. What works on a 2.0L turbo car will work on a 1.6L turbo car, just smaller.

I have a generally good working knowledge of what an Intake manifold does, how it works, and all that good stuff. The car is going to be well supported to utilize it.
The various engineering software(s) will be used for 3D-CAD, FEA, CFD, and pulse tuning. Basically Solidworks and Dynomation-7 will be used. This saves me huge time and money. Analysis first, then functional prototype/engineering samples, then laboratory & field testing.

I'm not doing an aluminum sheetmetal manifold as they don't holdup. I've seen some very clever aluminum sheetmetal designs that looked actually awesome, but ultimately they all fail from cracking. My opinion on stainless steel is that it's strong for this application, but heavy and soaks up heat, so I'm passing. Having said all that: I'll attempt to make the Juke upgraded intake manifold in PLASTIC. Yes, 100% plastic.....lol.

Manufacturing:
The manufacturing process will be an SLS rapid prototyping process. Don't think this is some Makerbot type deal or a junk FDM toy part. This isn't an SLA either. It's basically like a metal SLS, but in plastic instead. SLS metal laser sintering was looked at but the price was astronomical. The SLS process produces are part that isn't porous to air/water and a whole lot stronger than even the best SLA prototypes out there. This basically is as close to an injection molded part as you can get. The surface finish will come in raw at about 6.2 micrometer or 250 Ra finish. Similar to a very rough glass filled molded plastic surface finish, but still acceptable. Post finishing is an option I may look at. I have ZERO experience with this material, this is the first time I try it on anything. I do most of my product development on high strength SLA plastic materials which are nearly as good as injection molded parts. When it comes to GF high strength plastics, it has to be SLS process from what I understand.

I received quotes in the U.S. for Duraform HST and the price was good. The material strength however wasn't good enough for my needs. ULTEM 9500 series or something would also work and is available in the US in SLS process form. I finally selected a material available in Europe called WINFORM SP which is a carbon polyamid plastic with the strength/stiffness near close to PA6-GF30. It's designed specifically for automotive intake manifolds with some ductility and toughness built in to handle vibration, high heat, and shock. Typically used by OEM or racing teams when they don't want to tool something up but need a part to last.

A couple guys have done this on the Porsche aftermarket, FORD have done it too. The Nissan DeltaWing had an SLS intake manifold made for development and even actual race version. So it's been done successfully. The trick is in the design, but the shops that make it have the experience for these applications.

OK, could I fabricate this thing? Yes, that's an option. It would be more of an existing aluminum casting and some selective 5-axis CNC billet machining to make it work.
That is a long term plan perhaps.

CFD Fluid analysis:
On the topic of fluid flow, this is a tricky one. The engine is not constant flow, so modeling this is tricky. I have full access to the CFD sofware to verify the flow. This will require some basic simplifications of fully developed flow and non-transients. Basically, not realistic. But, it does tell you if you have restrictions, dead spots in the flow, recirculation, etc. The key is to maintain equal pressure drops across from one port to the others, thus maintaining equal flows. This CFD software can easily do this. It won't be able to tell you how much reversion and all that going on unless you are willing to completely model an engine. The harmonic engine modeling software does do this, that is later though. I won't be bench marking the stock design as it would require sending it out to flow an airflow bench which I don't consider worth the effort and not totally accurate for a turbo car. For these purposes, we can assume the stock design can use an improvement. Any future dyno runs would be more informational and realistic anyway. Honestly, I almost don't care about base-lining but I'll have one anyway when the car hits the dyno for the engine retune.

FEA pressure vessel design:
This design is primarily built as a pressure vessel. In short, you typically design to 300 psi, though in my field 500 psi is standard for Burst, 2.4x rated pressure for Proof testing @ 15 minutes, then 0-150 psi for 100,000 cycles without leaking. Testing would be done on water, then air dunk tested for leaks. Any air tests would be capped around 30-50 psi due to safety concerns. Pressurized air is dangerous, with plastic much more so. I may run a PRV on the intake manifold to snub any high pressure spikes which would be easy to do, like a high pressure blow off valve.

The main concern is engine backfire in the 75-100 psi range. Operating pressures might be 20 psi boost, plus the intake manifold can easily produce another 7 psi internally due to helmholtz tuning. This is the main concern with sheetmetal intake manifolds. Aluminum is the worst. I try and avoid aluminum sheet metal intercoolers for this very reason, cause the endtanks typically split at the weld line. I spend most of my time here analyzing the stresses. Burst test up to 300 psi is a good place to start. For fatigue life cycling what I do is take the Yield stress of the material & derate it to 40% so that on the SN curve it'll run out to 100,000 cycles or more for instance. This is the SN curve (Stress vs. Cycles). So if the material is good to 11,000 psi, actually it's really only good for 40% or 4,400 psi part stress at high cycle counts. The FEA value should be less than this. You could use safety factors too, but this is more accurate. Anyway, most guys get fixated on CFD and performance and all that stuff, but this is actually where most of my time is spent in the commercial environment. If it's strong, then you can start worrying about performance. The main thing to think about is stiffness. You need a stiff part for it to function properly......otherwise it won't seal the o-rings. All that flexing is bad, weakens joints and/or welds. In some areas I removed ribs to allow flex where it was needed to avoid stress risers. The FEA analysis is where things get critical to making it work.

Helmholtz tuning
I won't go into too much detail as I can just refer a few good books on the subject. But I see info out there suggesting somehow the intake manifold is a restriction. Partly true, partly it's a pulse tuning limitation. By that I mean, the Intake manifold is like a supercharger in-line with the turbocharger. The design affects this. Not all of it is restriction based. It has to be looked at it in terms of reducing restriction, but also increasing the pulse tuning pressures at the right time. Together, these affects will combine to provide the most pressure at the intake valve at the right time to make more power. Obviously the turbocharger has to have the volume flow to support all this, but typically that's easily solved. I already mentioned plenum volume, runner length, runner diameter, runner taper, cam timing, etc. All those factor and the calculators are a start, the engine tuning software will dial it in. The design is modular, meaning I can replace the runner section and optimize it if I chose to.

The intake manifold calculator tends to confirm those findings in terms of runner lengths and what not. The manifold plenum volume on the Juke upgraded design is about 2.2L, roughly 1.4x the engine displacement. This ratio was taken from the Blitz manifold of 2.8L on a 2.0L engine, the software will optimize this but it's a starting point. The stock Juke is maybe 1.2L plenum volume as I've measured with water, way too small for high rpms. The critical thing to remember is the upgraded plenum needs to be "tapered". Not tapering wrecks airflow distribution to the cylinders. Most aftermarket manifolds in the last 20 years.......they taper the plenum. The runners too can be tapered 3-5 degrees but I probably won't as I've run out of room in terms of size/packaging. It's a trick to make the runners look shorter to the engine given space constraints and somewhat extending the reflected pulse wave but lowering it's intensity slightly.

CAD/Solidworks:
This stuff is probably the most fun. You gotta be comfortable with sweeps, blends, surfacing to attempt this stuff. Solidworks isn't the most powerful software, but it is comfortable to use if you stick to the basic techniques. The other thing here is being comfortable designing for injection molding or castings. I could convert this design to CNC machining and fabricating tubing/welding if I had to as well.

Next Steps:
Well, lots of stuff is missing. All the emissions/vacuum mounting bosses need to be added plus the vacuum/pressure ports. I'm using brass tri-serts for the mounting bosses. They are sort of a threaded "heli-coil" for plastic. The plastic is drilled, then these self-tap in, with some loctite to permanently secure them. The thread will be strong as the factory sonic welded/overmolded brass inserts, or close enough.

Stuff is thrown in as place holders for now. Ton's of ribbing needs to be cleaned up and finalized. Wall thickness, rib thickness, port geometry need tweaking. The o-ring runner/plenum crush seal needs adjustment. I have to add steel insert sleeves in the bolt-holes to prevent bolt over-torque, not shown. The port runners start out as 1.625" diameter I.D. round, then transition to ellipse, then I need to transition to the final square slot port shape as I kept them ellipses, gotta fix that but it's close already and looks pretty good. I spent tons and tons of time tweaking the intake runner paths to avoid cheated radii and good flow entry. Was tough to do and probably could be designed with less bend angles, but not that easy to blend a 120* turn in such a short space. If it doesn't flow perfectly, it's still a massive improvement over stock.

The CFD will be used here to adjust and improve the port-to-port flow balances. Anyway, this was mostly a CAD problem, lot's of surfacing was used and some crazy use of LOFTS and a few tricks I had to develop. I like Solidworks but ProE or Creo would have done this stuff easily. It's as good as I can get it given how tight everything is and where it needs to fit. The runner lengths are nearly identical but can be tweaked later as I mentioned. Throttle body flange and gasket sealing gland/groove need adding. Cylinder head flange needs gasket seal gland added, plus ton's of tweaking, extra rib strengthening. That head flange was really tricky to get it strong enough, still not happy with it. I'm working on an aluminum bridge adapter plate that'll mount to the plastic head flange like a sandwich. It'll add (10) mounting bolts to the plastic head flange design with o-ring port seals to adapt it, then use (5) of those same bolts to mount to the head using the OE gasket. There is zero room on the cylinder head to drill/tap for more bolts. Why they thought (5) was enough is beyond me

You can see I use small sized M5 bolts and higher quantities for even sealing on the port o-ring seals, they will get blue-loctite to keep them in place. The plenum needs some major internal flow streamlining but for now it was done to reinforce it to 300 psi burst pressures and it looked good on the computer simulation. Lot's of radiuses need to be added to the keep the ribs from cracking when the manifold flexes. And so on.

Regarding weight, the stock plastic intake manifold is 6.4 lbs fully dressed out. This design is 7.06 lbs without the emissions stuff added. So the weight is close. The wall thickness is coming in at 0.220 to 0.250" to account for the slightly weaker material strength. Strength is much more important but I'll attempt to keep the weight the same.

More later. Maybe I'll get bore and forget about this. We'll see.

188733


188734
 
  • Like
Reactions: menshealthadvocate
381 - 400 of 454 Posts
Top