Custom EFI, 14mm Port and other Questions
 

Closed the other thread and moved it into this section.

First off, I have to thank deatschwerks and diyautotune for answering my questions.

This is most likely a one time crazy person attempt, doesn't deter me from trying though.

Back-story: The 465 Holley setup on my streetport 12a leaves a lot to be desired. Decided to design my own fuel injection manifold, rail, intake manifold etc. This will use an FC ignition system and modified FC throttle body. Fuel and spark will be taken care of with an MS3. I'll be reusing the Holley intake manifold, spacing it out from the engine with the aluminum chunk below. It is the shape it is for ease of machining since I'll be doing it all on a manual mill. Connecting the existing holley square bore flange to the FC throttle body I can do a couple things. The easiest would be to use this:
https://cimg1.ibsrv.net/gimg/www.rx7...e5d8c8ef85.jpg


...and build an adapter plate to hold the throttle linkage, omp linkage and the body itself. That one would act more like a plenum and air speed through the tubes would only really be independent once it entered the holley casting.

I could also build my own adapter between the flange and throttle body. While I am at this, I could extend the throttle body over towards the driver’s side around 20” or so. That would give me an intake runner length of nearly 30-35”. This one would also be individual aluminum tubes from the port all the way up to the throttle body. My goal is to have mid rpm range power by sacrificing higher rpm (+7000) breathing. I am now trying to sort out if it would make sense to offset the injector port through holes on the primaries so they are off of the center line of the port. I have enough space to machine the primary ports either way. The secondaries, due to packaging constraints, force the use of an offset. This got me thinking, that offset might end up helping with flow by inducing slight vortices in the air stream due to the injector pulses into the air stream. That vortex (not much of one) should help in getting the air fuel mix around the corner into the chamber a little.

Here’s what I am talking about:
https://cimg5.ibsrv.net/gimg/www.rx7...4598eec224.jpg


OR
https://cimg2.ibsrv.net/gimg/www.rx7...907c97d48b.jpg


Turns out the offset primaries end up with a spacing to the secondaries similar to the 2006 miata fuel rails. I could cut one of those in half, weld it back together in the middle and get something that could work for little money. But, the purple injectors run at 60 psi and only give around 280-290cc/min. I am thinking I want more head room available for tuning. I also plan on reducing the engine side port size of the secondaries to help with reversion. It might not matter at higher rpms but I want the secondaries to be clean even at lower rpms when they are opened.

Some background per PeeJay
https://www.rx7club.com/1st-generati...nifold-115465/

Power goals of this are mild. With the 465 setup, I made 141 whp and 117 or so torque. I have that somewhere on another computer. I would most likely be able to broaden out that torque profile and gain more low end torque with the intake design and the timing control.

Estimating 175 hp at the crank, a BSFC of 0.45, 4 injectors and a duty cycle of 0.85 gives an injector flow rate of around 23 lb/hr or 240 cc/min.

If those numbers are wonky for a rotary, please let me know. Using all sorts of disparate sources and sometimes the info has to be made up until I know better.

After talking with anyone that would respond; deatschwerks and diyautotune have been the most helpful. DW were able to send along the 2d drawing of a 14mm injector for me to design around. The one in these pictures is a placeholder that I got from grabcad.

https://www.summitracing.com/parts/d...3504/overview/

http://www.deatschwerks.com/sites/ww...terization.pdf

That injector is from a 2014 MAZDA MX-5 MIATA 4Cyl 2.0L and the 18U-00-0350-4 injector set would work well (I think) since they are flow matched and could be used with kurt staging if I am having trouble keeping up with fueling. Those also have nice linearity down to small dwell times. I plan to run staged injection with only the primaries being used for low load and cruise. I plan to start with Alpha N control and see how that goes before even looking at speed density. If I go with individual runners, that makes things tricky and the vacuum signal on my streetport is not the best either.

The fuel system is also getting upgraded with one of these and FC fuel tank parts… another thread at a later date.
Walbro 255 LPH Fuel Pump Kit - 86- 88 Mazda RX-7 [GSS341-400850] - $109.98 : Himni Racing, Turbocharger, Turbo, Garrett, Turbo Kit, GReddy, Mazda RX-7, HKS, Apexi, TiAL, TurboXS, TurboSmart, Flange, Turbonetics, Exhaust, Intercooler, ACT, Intake

Back to chunks of aluminum. Here’s a few detail shots of this thing so far. The underside of the block has to be edited to block off the ACV port and allow clearance to the exhaust flange.
https://cimg6.ibsrv.net/gimg/www.rx7...b77afd6173.jpg
https://cimg7.ibsrv.net/gimg/www.rx7...a8a7f47659.jpg
https://cimg8.ibsrv.net/gimg/www.rx7...cef713bb4d.jpg
https://cimg9.ibsrv.net/gimg/www.rx7...18cfed9da1.jpg
https://cimg9.ibsrv.net/gimg/www.rx7....png" alt
https://cimg0.ibsrv.net/gimg/www.rx7...3e335b152c.png
https://cimg1.ibsrv.net/gimg/www.rx7...5aeecb9f62.png

Rummaging through sites with any mention of drilling your own injector port seats got me to those last dimensions. Checking with my machinist handbook today though gave me a whole other set of numbers. Those are dealing with pressures of 100psi through 5000psi though. Completely different world than 43-60 psi. Side note: Viton seems to be the preferred o-ring of choice for fuel injectors.

https://cimg0.ibsrv.net/gimg/www.rx7...b7c41b2b0e.jpg


So, all that to get to the part where I’m asking you for help…

- Can anyone see a reason why using the offset primary ports would help or hurt?

- Are the injector cushions always used? On a 14mm oring setup, can I get away with just having the injector body held down in direct contact with the aluminum? (assuming I cut the port counterbore correctly)

- From various sources, it seems 11mm and 14mm orings are interchangeable?! What the? Is this injector specific or are they all the same gland size and the 14mm o-rings just have a higher durometer and more gap to deal with?

- 14mm is about 0.551”, if I cut the bore at 0.531” (17/32) is that considered too much, enough, not enough squeeze? I can drop down to the next 64th size and get 0.515” (33/64) and end up much closer to the handbook’s recommended tolerance with the injector body… that seems too tight though. Either way, hand polishing those bores to get around a 16-32 micro inch finish will probably be done, partly for sizing and partly just to get the best sealing surface and to smooth out the transition from the lead in chamfer to the bore.

- Does anyone have a 14mm port tool they wouldn’t mind loaning for a short time?
The obvious thing to get would be the Ross Machining one, I'm trying to figure this out with tooling I already have on hand.

- Does anyone know if a 14 or 11mm o-ring port is governed by any SAE, ANSI, ISO spec? More curious than anything, found these port tools and thought they were neat...
Manufacturer of Metal Cutting Carbide & Porting Tools-Davis Junction, Illinois-Form Relief Tool Co., Inc.

This fuel block would be able to turn any 12a engine into EFI. As long as the intake manifold doesn’t use the coolant passages any longer, it looks like it will work… just a whole lot of details to sort through before I start making chips.

If you made it this far, congratulations, please add any suggestions, comments; ideas are always welcome

No one brought their ten foot pole?

I should have time later this week to clean up the underside to get it closer to the finished shape

This sounds like a lot of effort, I didn't quite make it through the end, but you do realize they make IDA and DCOE style EFI throttle bodies that can mount onto any available intake manifold, right? There are some disadvantages to having the injectors further away and not being able to control the primaries and secondaries separately, but it's a quick way to get EFI without fabrication and reports are that they are very streetable, the same way a DCOE is streetable at least.

Building a car from nothing sounds to be a whole bunch of effort as well... but building these things is half the enjoyment, plus fabrication is some of the best parts of that fun. What am I going to do when I run out of stuff to build or weld? Probably build another one, but better since I'll know what not to do.

If you are referring to the holley efi or fitech efi's, then yes, I have looked into those and wankel=awesome makes the best point about their limitations with tuning on a rotary. There's also a 6 port thread of a guy in Italy that had all sorts of trouble trying to get his fitech to run well. The megasquirt has enough history and users to sort out most issues.
https://www.rx7club.com/build-thread...ngine-1075767/

Goals of this are great throttle response, which is why I want the primaries as close to the 13b location as is possible

Here's a few questions from another member that I figured would aid with anyone else's...

"....
A. Will the injectors be too close to the engine and possible melt or have problems?
B. How well can you bolt on an intake that will cause the right amount of air flow at the high RPMs?
C. Can we still use the stock, or in my cause Direct spark system, for spark?
D. Would it just be easier to run different injectors due to part availability, example, LS injectors?
E. Would a hand built intake make placement of the injectors easier?
12A flange, https://www.mazdatrix.com/e3.htm

and my reply...
Looking to use miata injectors from an NC that flow about 300-350 cc/min. Using a megasquirt 3, although I guess an FC ecu could be tricked into working. Fuel pump will be an upgraded walbro for an FC.

A: Injectors and heat soak should not be an issue. 13b's have the primaries in the center iron, the aluminum dissipates heat quickly, my rb exhaust has heat wrap and i'll be making a heat shield to go between the exhaust and intake.

B: I'll be using an FC throttle body, should flow enough air for what I am trying to do power wise.

C: Yes, depends on how you want to control fuel too. Can either to fuel or fuel and spark. I'm going with both so I can control timing directly. The mechanical advance curve from the 12a distributor cannot be tuned in the same way.

D: Any injector can be used with my setup. I chose 14mm ones due to flow rates and ease of machining the ports needed. As long as you can find a flow rate that works, it doesn't really matter which injector body style you select as long as they are long enough for a fuel rail to fit on the injector. These are at a 40 degree incident angle with the airstream, if i go further packaging becomes more difficult.

E: They (mazdatrix) only have exhaust flanges, a 12a intake flange to make one with tubing is not available (from what I have been able to find in my searching uk, oz, us...). The primaries being partially overlapping on the correct size tube lead me down the road of machining the aluminum block to make the holley intake work.

I can model up some tubing to get an idea of what will be needed to carry the holley square bore flange to the FC throttle body. That model shouldn't take much effort. I found a model of the 4150 holley flange, just have to make one of the FC throttle body and work the blends in between.

and yes, I write a lot, could be accused of being verbose or highfalutin, it's better than the alternative when dealing with engineering problems.

OK, there's no real reason this won't work, it's just weird. There's Weber style throttle bodies like so: 3002 Series - IDA Style Throttle Bodies which bolt to available Weber style intake manifolds.

Personally I was sort of working on casting a manifold to take an Arctic Cat throttle body setup (cheap off ebay!) but I'm not confident it'll fit into anything other than a Locost.

It is a very different and interesting take on EFI. It'd be much easier to just get an IDA/DCOE manifold and some Jenvey throttle bodies to match. Or get an intake manifold welded up and use some of their singular throttle bodies.

https://store.jenvey.co.uk/throttle-...hrottle-bodies

As I was saying above, and others have confirmed, you are trying to reinvent the wheel here. There are off-the-shelf intakes that can be mated to throttle bodies that come with injector bungs and fuel rails.

People haven't had much luck with the self-tuning Holley style kits that bolt to the Racing Beat Holley manifold, but that's just because the self-tuning is looking for an open plenum. There are non-self-tuning systems that can be pieced together.

Atkins sells a bolt on kit that uses a twin 45mm TWM throttle body in a short side-draft style manifold.

EFIHardware sells downdraft kits for street or Track than can dump a ton of fuel depending on your needs.

Me personally, I'm going to use a wrap around side draft manifold with a Jenvey twin 45mm throttle body. Depending on how it behaves, I may tap the manifold for extra injectors closer to the intake ports. I'm not going to bother until I get a feel for it first.

Interesting links for the ida's. The thing that gets me is that I already have everything needed minus the fuel block shown above, the intermediate tubing (call it an upper intake manifold) and a fuel pump (needed either way). If I went with the idas, I would sell everything I have and still end up spending more money than if I just make these two custom pieces. Sure it's weird, doesn't mean it won't work well though. The car this is planned for is a shell right now, it literally has one brake line and two fuel lines. Everything else had been removed for refurbishing or the trash bin... Oh, the accelerator pedal is still attached for some reason... Working on finding the motivation to finish the tri-link and panhard bar. Only then can I get it to be a roller and start mocking up the engine... Point is, that is going to take a while and I don't mind building everything from scratch in the meantime. It forces me to research more so all the different aspects of the build are understood completely, which should make troubleshooting slightly less convoluted on the future... Plus, this setup will lend itself well to intake length tuning and forced induction. Not saying you don't have a point, this idea has been in my head for close to 5 years and the whole car build closer to 15. It is going to be different and that is kind of the point.

Huh, forum ate my response... I'd just be careful you don't run out of room between the engine and the strut tower. I'm not sure offhand how much room the RB Holley manifold would take up and it would be annoying to build this and then not be able to fit it.

As another minor thing I'd change it so that you can get to the nuts for the adapter-to-rotorhousing nuts from some side other than the bottom, otherwise it'll make everything a bear to work on - it suuuuuucks designing something from scratch and missing something that'll make it annoying to work on for the rest of the part's existence. Maybe omit the helicoiled holes where the Holley manifold bolts up so you can get a socket through there. Food for thought.

Edit: Actually, looking at it, bolts for the other holes are going to be impossible. Best to figure that out now when it's nice and cheap electrons to change.

This place does like to eat entire posts if you take too long to type, there's no warning of when you automatically get signed out... makes things interesting and frustrating sometimes. I've resorted to typing longer posts offline then copy pasting.

Almost all of the geometry is preliminary right now. The plan is to design the nut pockets so I can use a ratcheting box wrench with enough clearance to get the wrench out when it is tight. The inner under ones will also hold the heat shield and will most likely require the use of a number of wobble extensions to get in there. The underside needs to be cleared out for the exhaust flange anyways.

Distance to the strut tower is about 5" right now from the nearest point on the holley flange. I'll double check as the snow around that car melts. Once I have the engine out and the intake in hand, I'll double check all the measurements so I know it works.

The stud locking inserts are there so I don't have to worry about threading into aluminum and those failing over time... problem is I can only find stainless ones at mcmaster, thread locking compound will be needed to reduce the chances of galling.

The problem is that the stock intake manifold to engine stuff is, other than the aluminum rotor housings, bolts and not nuts on studs. And I meant more, leaving a hole where the nuts for the rotor housings would go (IE, just using what would be the outer 4 bolts, not the 2 studs for the rotor housings) for the Holley to adapter interface so that you can use a socket to tighten the nuts for the adapter to block interface.

I'm not sure of your background - but if you design stuff as part of your career, please think of the poor bastards who have to work on it after it's made. ;) Seriously, I'm an engineer, but I feel like a lot of my stuff has been improved by the side stuff I do where I'm the poor bastard wrenching on the stuff that I design, which makes it much easier to find the person responsible and have a word with them.

Oh, as to the 11 vs 14mm injector o-ring stuff, just a different thickness o-ring usually. Like, 7mm ID in either case (or whatever, eyeballing ID drawings) and o-ring either 2mm thick or 3.5mm thick. There's at least a few applications where the injectors are constrained just by the port and fuel rail o-rings without injector "cushions". All of this stuff seems to be specced around "nominal" sizes - so for the "14mm" injector o-rings you make... a 14mm hole! Or given as how you're looking at the manifold side which is only sealing vacuum, 9/16" is probably close enough, though going a bit undersize is probably a good idea.

MechE background, actually waiting for creo to come back from being locked up in flexible modeler right now... Ahh, good to know. It's been so long since I have seen my engine together that I had forgotten if they were all studs or a mix of studs and bolts.

Once I can get the shell on the ground I can turn it around in the garage and really start looking into the details of all the engine attachment points. I think it is feasible for studs and bolts to work. There is enough material in the aluminum block to allow for tool clearances. Since I'm the poor bastard doing the wrenching on this, I will be making a concerted effort to think ahead for ease of maintenance and assembly.

I kinda figured MechE background somehow - we know our own. ;)

I think this is the way to go if you want to make your own injector block. The block you have CADed up looks like a bear to install/remove if the engine is in the car.

How was the FC throttle body modified? The 2 smaller butterflies open a little after the primary one, and those fed the secondary ports with the stock manifold. It would be interesting to see how it works with a plenum.

Like others have already said I think the simplest way to is piece your own EFI set up with a 4 bbl throttle body since you already have the manifold. Like this one.
This will also save you the trouble and room for the air filter.

This block weighs in just under 8 lbs, unless I royally messed up my material property library. Here's some updates, throttling the secondary port down for reversion, cleaning up the underside, adding in the holes for weigh savings and accessibility from the top to get to those nuts. I plan on changing out all of the mounting hardware of this to studs and using flanged nuts to mount this. The holley intake on the other side would use all bolts... or really, could just use studs and nuts too since I'll have more than enough from one box from mcmaster. Enough thread engagement in the aluminum with loctite red would be good enough for me to compress the gaskets. The secondary ports can be cleaned up by hand to get that surface smooth.

https://cimg1.ibsrv.net/gimg/www.rx7...be487f1b29.jpg
https://cimg2.ibsrv.net/gimg/www.rx7...d589315895.jpg
https://cimg3.ibsrv.net/gimg/www.rx7...b9ae2401c2.jpg

Next up are the tubing sweeps to mate the holley flange to the fc throttle body. The FC throttle body is simplified with the mechanical logic butterflies on the secondaries removed. The only thing it really does is flow air, connect to the omp, throttle cable and I'll be adding a mounting block for a GM TPS.
On this project, this is the simple route.

But it's $3,700 and it doesn't allow you to substitute the Microtech ECU it comes with. Isn't EFI Hardware's kit significantly cheaper?

Keep scrolling on their site and you'll find you can buy a manifold for $250 (shorty) or $450 (wrap around), the throttle body for $1150, or a package with the two for $1350. It comes with the fuel rail and FPR.

You can also just get the manifold and combine it with a throttle body from Jenvey which ends up costing around the same once you piece together their parts for fuel rail, FPR, throttle linkage, TPS, etc.

Of course, that's just the start, you still need a fuel pump, fittings, lines, injectors, and some sort of EMS. I'm not saying it's cheap, but it doesn't require fabrication.

Compare that to the cost of the Holley manifold the OP is using, plus the Edelbrock adapter, plus an FC throttle body, plus fabricating this injector plate, plus whatever has to be fabricated to connect the FC throttle body to the Edlebrock adapter. Then a fuel rail has to be fabricated.

Even if he completes this, there are two issues. Injectors close to the intake port is good for low speed, but further away is better for high speed as it gives more time for the air and fuel to mix. That's why there are two different injector locations on the later intake systems. The second issue is runner length. The ports between the front and rear rotor should remain separate for longer than the Holley manifold. Numerous reports are that an open plenum on that intake sucks on the street. It may work OK at very high RPMs but there is a reason the later manifolds have long separate runners. Torque.

I'm building this for torque and not any power higher than 5000-6000. The beauty of this setup is I can make my intake any length I want. My space constraints are the driver side inner fender/driver side shock tower. With this I can get upwards of 40 inches of individual intake runner length, maybe more. Find me any setup that comes even close to this level of adjustability for runner length. The plenum at the top of the thread is only there to show a stop gap and not the end goal. Here's an extremely simplified model of the FC throttle body and the simple tube bends that will be needed to carry the holley flange to the FC throttle body. A support post would be added around where the rats nest used to be and it should be solid.

I will be using much better injectors than anything the FCs ever had. I am either looking at RX8 disc injectors or deatschwerk pintle style ones for NC miata's. I would love to use Injector Dynamics ID725's... http://injectordynamics.com/injectors/id725/
But that would be overkill even if their linearity goes down to the dwell times I need for my meager power output.

Point is: Atomization will be better with these injectors, flow rates will not be expected to be much more than 7000 rpm (I'm expecting a power drop off around 6500) so flow mixing with the better spray patterns should make up for the less distance traveled.

https://cimg1.ibsrv.net/gimg/www.rx7...724da6a5d1.jpg

https://cimg3.ibsrv.net/gimg/www.rx7...be86ada79a.jpg
Creating things from scratch is more enjoyable for me than buying something and bolting it on... even if it takes years instead of days.