skirkland1980

iTrader: (2)

I came up with this idea last night to convert any square flange, or spread bore carb setup to an EFI. The material can be had easily from ebay or metal supply store. You just need some minimum 1/2" aluminum plate to make the flanges, 3" hole saw to cut the holes in the 2 flanges, 5/16 drill bit, jig saw, and a drill (to name a few)
For the bent pipe, that, you can also get from ebay. ($84 +shipping) Just get a 3" intercooler pipe kit. Use the 90* for the manifold and use the rest of the kit to plumb the air filter, or turbo/supercharger.
The tough part will be the injector bungs. You may be able to purchase them already made or use some thick aluminum bar stock. I would use at least 4, 2 on each side. Then, once you've done the mock up, take it to a welding shop and have it welded. A good welder can have it done in a few hours I think. And at $60-80 per hour thats not too bad.
For injectors, I would go with the 440cc 1980-1985 ford cfi injectors minimum for stock/na applications. For higher HP and/or forced induction applications, I would use the mazda RX7 injectors. These style injectors dont require a fancy fuel rail so you can save on cost. You can just use hose and clamps on the inlets. Although custom rails would still be fairly inexpensive and look great.
A salvage yard throttle body will also help save on cost. I like the nissan KA24 throttle body. It's 60mm and easy to find. For fuel management, just a simple, inexpensive, easy megasquirt I v2.2 will work great. You can use an external electric fuel pump in place of a mechanical pump, like an MSD 2225. Feel free to use this idea. I'm not looking to get rich off the patent. Thanks for looking

Pocket

iTrader: (24)

Check this out. Theres a ton of different types
http://www.summitracing.com/parts/EDL-38493/

Then read over these
https://www.thirdgen.org/forums/alte...-manifold.html
https://www.thirdgen.org/forums/alte...ake-parts.html

83 Crossfire TA

iTrader: (2)

This seems sort of badly thought out, at best.

a stock TPI throttle body, assorted mustang, 65mm explorer, LSx... would all be easier to get, flow more and cost less.

If you have to pay someone to do the welding that setup will cost more than buying one, and the injector bungs are just a PITA like that with no real advantage... If you're making it from scratch, use a 3/4" or 1" plate and drill the sides to mount the injectors (for most standard injectors it's a 17/32" or 14.5mm hole, you'll probably have to order one, you can get them fairly cheap on ebay, but check around, I got a silver and demming 1/2-1" set for less than I could get the single bit.

injectors... If you're going with something that can handle peak and hold injectors like a megasquirt you could go with the 160lb/hr ford injectors, 4 of them will feed 1200hp, if you're going with something that doesn't 63 or 83lb/hr seimens/dekka injectors will still feed a lot of HP and work with _any_ injector driver. Fuel rails can be made using that same drill bit and fuel rail stock, or 1/2" copper pipe actually has the perfect ID for it and is cheap, easy to get and will handle plenty of pressure (if you can't sweat copper joints, don't blame me if you end up with leaks, find a friend that has a clue, don't burn down your car).

Why is this better then grabbing a TBI setup for $50 and slapping it on any carb intake with an adapter plate? I don't know, you tell me...

83 Crossfire TA

iTrader: (2)

for that matter, if you have the ability to do a good job doing this kind of thing, I don't see why you wouldn't just add injector bungs to the runners of the intake unless you're making an auxiliary setup

skirkland1980

iTrader: (2)

Badly thought out? Thanks. Trying to tame a wild mountain lion and keeping it as a pet is a badly thought out idea, not fabbing your own custom intake. The idea here is to have a custom, one of piece that doesn't require much on car work. You can build the piece, while still driving your car, and have the conversion done in a day. I figure it would be cheaper to use 4 larger injectors than 8 smaller ones. And the welding part isn't too bad. I had a friend tig weld my fuel cell for me. I gave him $20 even though he said don't worry about and he was happy. Also it can have custom features built in that off the shelf part won't. Like you can use a plate to try different size throttle bodies. And I know this unit will flow better than any stock unit. I am going to build this eventually and report back with my success

White'89

iTrader: (8)

83 Crossfire TA is shooting you down for many reasons. First off is that The injector idea is of the same basic design as a TBI and will be incredibly inefficient. And second not everyone has a buddy that can tig weld.

To sum it all up it would be ten times cheaper to just go find a fuel injected car in the junk yard and take all the parts you would need. I can guarantee it will cost more to do what you are saying and will have crappy power and fuel mileage. Want to prove us all wrong? Do it.

89_RS

I concur with 83 Crossfire as well. I've made custom one off intake parts (from TB's to the Manifold itself) for FSAE cars before, and the level of CFD that goes into making sure that a particular intake will work goes well beyond 3" pipe + 4 monster injectors = FAST.

That elbow is going to be a monster bottelneck to making power. You don't have any throttle blades in there, so how is the engine going to idle? Even if you bolt on a standard TB, you still have to make brackets & cams to correct for the improper TB cable geometry & length. Feeding the injectors is another issue as well. The large amount of fuel each injector will need means fuel rails are required. Another thing to consider is fuel atomization. Those large injectors plus a very unstable airflow means that you will have a lot of fuel falling out of solution and making the manifold very wet. Lets not even start in on the lack of basic EFI electronics either.

See where this is going. I can TIG weld, CNC machine, and fab just about anything on a car with a few exceptions. But if its a poorly designed and integrated system to begin with, no amount of awesome fab is going to make it perform any better than poorly. Having machined & fabbed components to race with in FSAE, I can tell you that you are way in over your head.

skirkland1980

iTrader: (2)

yes, it is designed to use the throttle body of your choice. Hmmmm. 4 monster injectors? I have the same injectors on my tiny little datsun L28(6 of them) and it runs great, thats why i suggested those. They are stock injectors for ford 5.0L throttle body injecton from 1980-1985. I am certain this set up will out perform any stock unit. And custom fuel rails and custom throttle linkage...yes that is what makes this a custom setup. I got most of these ideas from customizations that i have already done to my datsun. I guess I will start gathering the parts to make this work. Thanks for the inspiration.

White'89

iTrader: (8)

Do you have pictures of this monstrosity?

You put 6 of these injectors on a 6cylinder but wanna put 4 on a V8

skirkland1980

iTrader: (2)

no pics. I have not started building yet. wanted to get opinions first but it seems like the only people who have opinions are negative. These injectors flow 440cc/min. each. It only takes 2 to run a ford 5.0L. You can step up to the mazda RX7 injectors which are about 670cc (if i recall correctly) for higher HP applications. The trick to running massive injectors in a tiny engine is the aftermarket FMU, like megasquirt, haltech, wolf, etc...

89_RS

Stock TBI units on the Thirdgen use a pair of 55pph injectors @ 9-13psi. In order to justify the large injectors, you need to move a ton of air. Holley & FAST use 88pph injectors in their systems, but the ECM also tunes to a specific fuel pressure & pulse width which is why you can slap those systems on any engine that makes HP close to or under the advertised sticker.

If EFI was as simple as you're making it out to be, there wouldn't be an overwhelming rush towards the new self tuning systems from Holley & FAST.

It doesn't matter if its the best fabbed system on Earth ever. What does matter is that its properly integrated & setup.

skirkland1980

iTrader: (2)

Any fabrication project involves math, geometry, and mechanics, pretty much everything you would learn in school and even some college or trade school. I've custom fabbed a ton of parts for my cars and built a megasquirt 1 and megasquirt 2 from the ground up and have gotten pretty good at tuning. For me this would be very simple. Is amazing that someone can say that someone else is "in over your head" when they have no clue about the skills that they have. For that matter noone should ever lift the hood of their car. There is way too much stuff under there.

89_RS

CFD: Computational Fluid Dynamics isn't something you learn in trade school or an Associate program. Try 4yrs of school in Mechanical or Aeronautical Engineering before you even have a clue how fluids (both liquid and gaseous) work. Then tack on another 5yrs experience doing CFD analysis & design before you can go on your own. You need to know how the air & fuel will both flow and interact with each other in the intake before you can begin to guestimate about performance. Things like RPM resonance, pressure waves, injector pulse, fuel injector cone pattern, and many more details that are well beyond either your or my skill sets need to be taken into account.

Like I said, making an intake work great goes far beyond 3" pipe + 4 monster injectors + awesome fab = FAST. Until you've seen what it takes to design an intake for a car that makes only 80FWHP, you really don't know what you're getting yourself into. The more power you make, the more these little details matter.

We had 100 variant models in the PC to run CFD on before we ever got to the RP stage. At RP we made full scales of the top 5 designs, dynoed them, and then honed in on a final desing before creating it in on the car materials (CF, Delrin, and some stainless & aluminum). All in all it was a 5 month process. Creating an intake isn't as simple as slapping some OTS parts together and saying it works. Sure, it may work, but how well?

skirkland1980

iTrader: (2)

You're making out to be more complicated than it really is. We're not building something to run at 24hrs @ lemans, really, or break land speed records. Are you going to shoot down someone's plans to build a custom tunnel ram. My buddy built one and his car has run in the 4s in the 1/8. He has no degree. I think he even dropped out of high school, he is just a good welder/ fabricator. I think you missed the point I tried to make in post#5. This unit works much like a tbi without 2 big injectors to block airflow and will look and function much better in my opinion. And really, how well was the tbi thought out? Does that mean everyone who has one is an idiot? All the tbi does is mix air and fuel and send it into the manifold, and that's all we're doing here.

89_RS

Care to point out where I said TBI was good? I must be an idiot since I drive a car that has TBI on it.

Custom Tunnel Rams aren't what you're talking about. Tunnel Ram was made in the late 40's and it exploits a very basic principle of manifold design: longer runners make more power. Now that only works to a point and is an over simplification of the design of a tunnel ram. Point is the system has been around forever and its exploiting some very fundamental power principles which means that pretty much anyone who can weld can make a very crappy knock off and still get some decent power out of it.

A TBI with 4 injectors in it and a 90* turn right after the TB. Its horrid on so many levels. That elbow is going have terrible flow period. I've seen enough CFD data analysis to know that. The area along the centerline of the elbow will have good flow, but the inner corner is going to have high pressure turbulence and the outer corner is going to have something resembling an eddy pool with a low pressure in it. Most of the air & fuel will get stuck in the elbow and then fall violently into the manifold, creating horrible pressure waves that will disrupt manifold flow.

Simplicity is often the key to DIY EFI. Tunnel ram with custom length runners, trumpets with there own injector & butterfly blade, etc. are all very "simple" systems in terms of air & fuel flow compared to what you are thinking about. Heck, stock TBI isn't too bad in the regard that there is a low pressure region directly below the injector. To make TBI better, you either raise the injectors up & port the TB, or you put them under the blades like jets in a carb (look at Holley & FAST systems and you will see this).

Getting the injector out of the path the air takes is good, but making that air take a path with huge turbulence is doing yourself no favors.

skirkland1980

iTrader: (2)

ok. say i use one of these and weld injector bungs to it? it eliminates some of the cool custom fab idea.

ericjon262

want a throttle body with multi port injectors? then buy one of these, and wa-la you got it. alot simpler, and bolts to any square 4 bbl intake.

http://www.summitracing.com/parts/FS...-KIT/?rtype=10

89_RS

No dice. Its not designed to be wet. Its an air intake and the injectors are deep down in the manifold.

http://www.edelbrock.com/automotive_...e_elbows.shtml

Start drilling holes in it, welding injector bungs to it, and putting injectors in it and you will induce turbulence and those numbers will drop. As I have stated before, this isn't as simple as 3" pipe + 4 big injectors = FAST. Theres a ton of CFD (be it in a PC or by hand) done to make sure the airflow is proper in any intake & manifold for the required power.

skirkland1980

iTrader: (2)

let go of the "3" pipe + 4 big injectors = FAST" for some reason you're the only one saying it and it's lame. Maybe thats why it takes you several months to design an intake, because what a computer says should work and what really does are not the same.

89_RS

I never designed the intake, I fabricated it off the computer file. A group of people far smarter than me designed the intake. CFD is processor intensive. You setup a simulation for a single design and it runs for 6hrs before you have data to analyze. Do that 100 times and then a couple times more and you've easily spent 3 months before you are ready to move to the RP & dyno stage. Its a hell of a lot easier to spend the time to get it right the first time than it is to just slap some OTS parts together, call it good enough, then race and loose due to poor performance & economy.

What a computer says will work and what does work are exactly the same. They all follow the same laws of physics & fluid dynamics. You seem to be the only one who doesn't understand that. That Edelbrock elbow used a baffle designed through extensive use of CFD to make sure that the turbulence I talked about earlier doesn't happen. You go putting something in that CFD verified air path and you're now introducing turbulence that kills airflow. You would be shocked if you saw how much turbulence the throttle blades alone introduce into the air flow.

The engine will run as long as it gets fuel & air, no one in this thread has disputed that fact. What is in dispute is your belief that you will have better performance than a carb from a TBI style setup. I guarantee that a stock carb or TBI unit has better economy & performance than the setup your thinking about making. The turbulence alone will make it perform horrid, the large amount of fuel and other issues such as cable length & geometry for both gas & trans will just keep driving any performance you get from the dyno stand down further.

I've said it before and I'll say it again: its more complicated than 3" pipe + 4 big injectors = FAST.

White'89

iTrader: (8)

^^All of that is truth. Although I understand after rereading the original post that your not trying to make it faster, your just trying to convert it to EFI as cheap as possible(by using designs which have been thought of before). We are just trying to tell you that Air/fuel ratios are going to be so horrible that tuning is going to be awful, and the trade offs may cause power to suffer.

83 Crossfire TA

iTrader: (2)

You know, I have a hard time even agreeing with the people trying to agree with me in this thread...

No, your idea how to do it is badly thought out. Just because you disagree with me doesn't make it any different, for that matter, just because you keep changing it without much of a pattern reinforces that assessment...

Sure, make something that you can bolt on when it's ready with minimal downtime is great... your plan how to do it isn't

As far as the rest of the comments...

- TBI has it's merits, that is not what he's building here
- CFD... yea, that's great... but that doesn't mean it will work or not work... there are plenty of examples of stuff that works great on paper/computer that is totally worthless in practice, and stuff that you can't seem to model that works well IRL. People joke that I make stuff that really shouldn't work work, where I would argue that you sort of have to have an instinct for these things. Edelbrock is a great example of this... they have all the equipment and tools to model and test everything, and I would argue that they make good products, but in almost every category there is something that works better that has much less modeling (if any at all) out there.
- Just because there is a curve/elbow doesn't mean that there will necessary be turbulence and pooling because of it. Just because it wasn't designed for wet flow doesn't mean that it won't work. Likely, maybe, to be honest, I don't get why so many of the elbows are designed like they are, not so much from a turbulence standpoint but from a/f distribution perspective. For that matter, turbulence is not always bad.
- I don't see a problem with 4 of those injectors on a mild small block... AAMOF we ran 4 of those (2 80's ford CFI setups on a dual quad manifold) on a quick 302, you'll have about the same fuel flow as a stock TPI/LT1/LSx, but they are rated at lower pressure so you can gain a little flow running them at higher pressures. They're pretty fast with a proper driver and they're very similar to standard port injectors. If you're looking to run 13's, maybe high 12's they'll be fine...

Andrew91GT

Did this thing ever get built?

anesthes

iTrader: (12)

Thanks for bringing this back to the top. I missed it the first time and should have locked it.

-- Joe