Jump to content

180 Degree L28ET Intake


Recommended Posts

Their classes seem pretty decently prices too. Too bad they're a solid hour and a half drive away. I wonder if there's anyone like them up in the sac area.


I've actually considered trying to get in for classes at a school like Davis just for machine shop access, though that's a mighty expensive route to go.

Link to comment
Share on other sites

  • 2 years later...

This thread shall rise again!



So we all knew this was a bit of a pipe dream and as much of an engineer exercise, but I'm back at it. The plan is definitely still going to be running E85, as it's practically my religion now. Local station this week was sub $2 a gallon, while crappy CA 91 octane is hovering around $3.75-3.90. Obviously "saving money" and "car hobby" go together about as well as windex and eye drops, but E85 is so much cheaper than race gas that it just plain makes sense. The caveat here is that it's going to be a dual-fuel setup most likely, only running E85 under boost.


And once again, my design has changed again... The more I gain experience with E85, and the more I see other people's builds, the more I'm convinced I don't need an intercooler for my goals. In fact, I'm pretty positive that with the standard head cooling mods I should have very little heat issue. The crux to achieving that of course is to generate as little heat as possible to begin with, so that the E85 can do it's job of soaking up what heat is there to keep temps manageable. This of course means a BIG turbo, and I don't expect it to produce full boost at 3,000rpm, but I'm okay with that. 


So this means it won't be a T3 flange. This also means I don't need to run a plenum on the opposite side of the engine, and it also means I don't need to run the compressor outlet around the engine... In theory the standard J pipe would work fine, as seen on 510six's build (easily googled by YOU). Except if I'm going to go through the work of doing a legit dual injection intake manifold, I might as well fab an exhaust manifold too right?


So the plan is to build a manifold that might somehow magically avoid the steering shaft (will just relocate it if it just becomes a problem), and basically allows the turbo compressor to exit with a straight shot to the throttle body, which will be center mounted. I plan to run JUST enough gap between the turbo and throttle to install a BOV. I hope to run a reverse rotation turbo to make packaging easier, as this will put the compressor outlet on the "outside" against the strut tower, instead of the block.


And of course, any update without pics is worthless right? I worked on this all morning, with my newborn boy snoozing in my arm. Gotta keep productive right?




You can kinda of ignore the turbo. I modeled it real quick to get an idea of packaging, and how well my measurements all match up as far as flange for/aft position compared to the intake centerline.


The header is a simple design, and you can tell I'm not concerned about equal length, or even perfect flow. What I care about is packaging and easy of fabrication. I do have access to a TIG, and even a decent operator if I don't like my stainless work, but I don't have a good bender, so I'll be buying bends to make it all work. So the simpler the design the better in that regard.




The flange isn't perfect yet, as it needs more clearance for the intake on top, but it's a start. Obviously the pipes will mate to the flange CORRECTLY.... Duh. But that flange is exactly as spec'ed on Garrett's website for the GT3788, though that's not a reverse spin turbo, it's just roughly the size I'm looking at.


Where the primaries merge I'll probably also be going up in size to reduce possible restrictions.


Overall I'm liking the idea of separating 123, and 456 on a twin scroll setup like that, but I'm worried that unless I give them a good 3" of merged pipe at the end that a quick spool valve is out of the question. Thoughts are welcome.


This intake is also one I made just for flow analysts, and I'll probably have to run longer runners to make space for the second set of injectors. I do plan to run a good thick plate at the beginning of the runners/end of the plenum, which will allow some significant radius to encourage flow without going insane with having bell mouths inside the plenum.




And for those curious, I got better flow on this design than almost any other box plenums I came up with. The CRAZY thing though... is how much my flow analysis changed once I added DYNAMIC pressure inputs to create realistic pulsing from only having valves open for a certain time and such. I went with data I found elsewhere as far as pressure difference created, and I'm still not sure how "accurate" my calculations were, but it was a night and day difference going from a static flow. If you just assume all valves are open equally during flow testing, then the OEM manifold would leave certain runners DANGEROUSLY starved compared to others, and we know by looking at our plugs that this isn't the case. The fact that not every valve is ever open, and that adjacent valves are never demanding air at the same time helps smooth out runner to runner inconsistencies, helping balance out even the poorer designs.


But all that really should be in a different topic, just thought it might interest those that have followed this thread in the past.


So yea, I now have access to quality fabrication tools, as well as a few people to hold my hand learning to do some of this stuff, and already my welding skills have gone from naught, to not terrible quite swiftly. Now I just need the time and money, which won't be overnight, but I can sort of see it on the horizon. These manifold won't get made until MS3 is installed and well tuned on pump gas, then I'm going to sort out some other basics before I get to building these. As I'm building these I'll also be building up a new head on the side, and maybe starting a collection of junkyard short blocks. So this obviously isn't starting tomorrow, but unless I sell my Z (which I don't think the wife would let me do) I'll be building something related to this in the future.

Link to comment
Share on other sites

Not that I've ever had a disrespect for people who design and fabricate manifolds, but damn, it's even tough to design them 3D modeling! My hat is off to all those that did it old school, by measuring and cutting, all while anticipating what obstacles they'd be putting in front of them.


For the sake of seeing how hard it would be to fit, I made a manifold with each runner making it all the way to a collector, so as to not limit flow as much... not an easy task. This iteration wouldn't really work perfectly, but it's close to being "plausible".




Edit: Oh yea, all that above is 2.5" radius bends. I'm trying to stick to a single radius and going larger than that makes things QUITE difficult to package.


Next step, in order to make my next version more realistic, is to take detailed measurements of my constraints (block, steering, etc).


Will post measurements soon.

Edited by Gollum
Link to comment
Share on other sites

Well I'm close enough to the ballpark that my rough measurements tell me I need to take more accurate measurements to go much further on the exhaust manifold. My mock up turbo is now woefully not detailed enough, but all in time, as the nature of modeling and prototyping. Here's some pictures of the manifolds + turbo in relation to the steering.




Note the fore-aft position of the compressor outlet versus the intake inlet... The goal would be perfect alignment, but I worry the compressor might have to go back a bit, or at least up, which will make plumbing fun to say the least.




I'm not absolutely sure what block clearance looks like yet, but I tried to make sure I gave ample room there. More measurements to follow will allow me to make a mock-up block, head, and valve cover to get a better idea of the package.




This gives you an idea of where the real-world clearance lies. That hot side wouldn't be quite that close, due to taper, but the compressor side is a larger OD, so it WILL be tight. Right where the rod representing the steering shaft ends is where the coupler is, which means another obstacle to consider as I get close to it.


Overall it's certainly looking possible, and I think I can raise the turbo up a taste if I change the collector around a bit, and there's also the option to just step up pipe sizes as they merge and merge them further up to simplify the design back to be more similar to the earlier renderings.


Oh, and as far as the intake goes, all looks to clear nicely, and that's a 7" runner from head surface to outer plenum surface. TOTAL length from gasket to outside is 13.68", more than short enough to not hit the shock tower. The only concern is hood clearance considering the angle the engine sits at, which will get measured soon as well. Worse case scenario I just add some radius to the runners. Not ideal from a fabrication standpoint, but an easy way to keep it all under the hood.

Link to comment
Share on other sites

  • 2 weeks later...
  • 4 years later...

Since we're still reeling from the Great photobucket crash of 2017 I'll upload some NEW work:




Yes, I'm still planning on making manifolds eventually, but I'm pretty focused on just having a running car at the moment. The long term plan is to pick up a spare block to build and mock up manifolds on once life settles down a bit, then I can have a full motor to drop in and tune with existing MS3x setup.


Also, this was done with onshape, which is free for public use (basically follows the github service model). Sharable link here:



I've also got a dimensioned Garrett G25-660 in my files if you want to copy that as well. (pictured is a turbo dimension slug based upon the BW EFR 7163). Both turbos I've mocked up are intended for a vband clamp on the turbine inlet. This is intentional as it would allow me to focus on getting the flange "flat"(relative to ground)  and then rotate the turbo as necessary to make the downpipe as straight as possible while clearing the steering shaft.


Disclaimer: I warn you that I can't guarantee the accuracy of port locations, ESPECIALLY THE EXHAUST MANIFOLD. The intake should be close, as it's based off Ron Tyler's flange dimensions which put my port floor even with the oem port floor, requiring porting to bring top of port up to flange dimensions. The exhaust side is largely a guess of the MN47 round ports with their liners and all. If/when I get to building this, I suspect I'll make the flanges "blank" with locating dowels, then drill roughly located ports and then grind them up to match the head port. This would ensure the flange fits MY car, and the onshape parts are simply for mock up estimates only. The steering shaft should be +/- 1" (or less imo) but like me you should plan to place your own turbo in your own bay, weld up stay rods to match the flange to turbo locations, then transfer that to your welding bench so you know it'll clear once made. I take no liability for ASSumptions you make by viewing these documents :-)


Link to comment
Share on other sites

As much as some rave about the EFR turbos, even if the new Garrett G series isn't as fast to spool, reverse rotation just makes packaging so much easier. I'd likely end up needing to push manifold outlet centerline about 3" back from center (being between cylinders 3 and 4 intake runners) unless I want to modify the engine mount, which is how much this manifold is pushed back. Note that all runners going into the collector "easily" merge in at straight angles, no bends going past the collector weld point. The other nice thing about pushing the centerline 3" back is that with the slight kick out of the compressor towards the shock tower the outlet starts to look preeeetty nice in relation to the intake throttle location. Going back another 2" I can straighten the turbo out some, but then merging at the collector starts introducing some harsh angles that'd be difficult to work with.


Also for the record, the exhaust is 1.5" stainless pipe 16 guage (.065 wall thickness, giving 1.37" ID) all 2.25" radius bends. Basically modeled after this: 




Some might say to go with a thicker gauge, which I wouldn't exactly disagree with, but I'd be more likely to opt for an extra plane of support for the manifold to help support the weight. Wouldn't be hard to tie in a support at the engine mount, possibly even with some form of spring steel to allow for a bit of growth at temp.



Link to comment
Share on other sites

Yeah, steering shaft is what's in the model. The EFR turbo is quite a bit longer (almost 20%) plus normal rotation only, so the turbo HAS to go further forward, as you can only get the header so close to the block.


Getting block dimensions to make a block mock up slug is the next step. I know the block is slightly narrower than the head in some areas but also has the PCV bulge to deal with.


My steering rack dimensions likely aren't perfect, but probably +/- half inch, which I could easily adjust for at the manifold output plane come time to fabricate.


Also, it helps that the G25 is TINY. If Garrett makes good on it's claim the g25-660 will be the smallest turbo capable of breaching the 600hp range (that I know of at least). The G25 turbos are SMALLER than the stock L28ET turbo...

Link to comment
Share on other sites

I took some time to get rough measurements of the block and then also refined the compressor dimensions (radial diameter was short a few mm). Keep in mind there's no oil pan modeled, but the clearance looks better than oem in several directions. Also for the record, the downpipe pictured is about 51 degrees on each bend. Not quite down to 45 where I'd like, but there's some wiggle room there.




Edited by Gollum
Link to comment
Share on other sites

More updates on the turbo model. Turbine side is now much bulkier to match the drawings exactly, and the compressor outlet extends further out, which matches pictures as well as I could (no dimensioned drawings from the front face I can find), and also extended it towards the inlet a tad. I also added the dimension for the back plate ring for the compressor, though the front edge of that isn't "realistic" as it doesn't match that diameter in a full 360-degree ring, so if I can fit what I drew then I'm on the safe side. The turbo "slug" as I call it, now matches the dimension drawing from Garrett as identically as feasible, other than the cartridge area, which is of little concern to me. Also for those curious, centerline of exhaust outlet is 2.5" back from the manifold centerline, and the turbo is clocked 15 degrees. Alignment looks great, though I also know I need to go get some shock tower measurements... But hey, what's a hammer for anyways?


Also, go figure, it's MUCH easier to run the downpipe when you toss in prefabbed dimensioned pieces and just rotate to fit right (kinda like real life, knock on wood). I used two 45 degree bends and getting it to fit is trivial, though it looks like the straight between the two bends could be swapped out for a 3" length (I used a 1, 2, and 3 inch straight as seen).




Link to comment
Share on other sites

Interesting concept on the intake manifold!

I'm curious about the flow characteristics with the inlet coming into the bottom center of the plenum. I could see the flow becoming pretty well mixed up after it hits the top of the plenum.

If you'd like to send me the model, I could put it in Solidworks' flow simulation to see what it'll do.

Link to comment
Share on other sites

Yeah, I did flow analysis of a similar design in SW before but I've given up on maintaining a SW installation. One of the nice things about onshape is that it's public, so if you follow my links and create an onshape account you can export the design yourself :wink: Could even copy it and make adjustments to your copy :-D


In my previous flow calcs it was really hard to get what I felt was accurate results. The problem is that flows are dynamic in any engine so you need to model a plethora of pressure differentials to really see how the intake performs in transient areas as well as raw mass WOT flow. 


In ANY case, the flow should be better distributed than OEM as well as most non tapered front entry designs.


Judging from some of the other flow simulations I've seen, the best addition I could add would likely be a tapered entry point coming down to the throttle body mount. That would help prevent massive low pressure areas right outside the throttle inlet at that entry surface. The far side being flat is likely not as much of an issue as one might assume since there's almost always negative pressure from valve to throttle. The HIGH pressure areas of concern stem from a relationship between valve closing and runner length. When the valve closes pressure builds up and can start disrupting flow on adjacent cylinders. This is really mostly an issue with short runners at lower rpm (which is like, duh) but it's a design consideration that should be attended to when looking at drawbacks/pros/cons.

Link to comment
Share on other sites

The plenum was measuring mightly close to the shock tower, plus my throttle body was based off a 65mm unit, which seemed too big imo, so I swapped out for smaller diameter pipe for the outside curves, which also use a three-inch radius bend instead of 4. I also cut the bends short, so instead of rotating 90 degrees they come out 70 degrees and then terminate, giving a more squat plenum. It might look small, but calcs are still showing it's a 3+ liter plenum. Plenty enough for a turbocharged L series imo. I might be worried if it were an all out NA build, but I don't think top end would suffer much, especially since if there are lean/rich cylinders I can adjust accordingly per cylinder as long as I know which ones they are...


Also, swapped the throttle body sketches out to adhere to this generic 60mm unit:



Link to comment
Share on other sites

Woops, I lied. I changed the diameter of the pipe, but not the radius. Fixed it. Interior volume is down to about 2.4 liters, put the bends back to 90 degrees since the depth is well under control now. Distance in the axis the runners aim in the intake has a max length of 13". If you tape a tape from the gasket surface to shock tower it sits at about 15" as straight of a line as you can go, so this is certainly in the ballpark of "just fine" for prototype dimensions.


Also, aluminum pipe dimensions are 2.25"OD with 3" radius bends for those corners.


Edited by Gollum
Link to comment
Share on other sites

7 hours ago, seattlejester said:

That angle almost looks like you plan on running intercooler less?

E85 or die bro!


But really, MN47 and no cooler and I'm not even remotely phased. All my research on the topic in the last 5 years only confirms my feelings on the topic. There's plenty of big blocks running e85 with giant roots blowers pushing 14+psi and e85 on 12:1... I'm targeting 14-20psi range and lower compression. Should be fine.


In fact, there's plenty of 6 second alcohol class drag cars (meth not eth) running without intercoolers...


Now, the six second SR20 car is running an intercooler THIS season, but didn't last season... It's all in the tune. I might be giving up some power, and definitely giving up some fuel economy under boost (run richer to keep cooler since alcohol tolerates rich mixtures with much less power drop and spark fuss compared to gasoline) but the trade off is worth it in my opinion. Quicker boost, less plumbing, weight, and really it's about the response on the street. Should make transients much nicer.


Link to comment
Share on other sites

Don't those big block engine usually run carburetors though?


My understanding is that with a carbeurator the fuel vapor actually pulls heat out of the air essentially super cooling the air before pulling it into the blower to be heated up. 


I just watched the "engine masters" on the intercooler pretty interesting their results, granted with a boost controller and an efficient turbo you can net whatever result within your efficiency island so maybe a moot point.


Are you going to run a blow off valve? There was this really interesting blow off valve that was basically a clamp/sleeve that would be super sleek.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...