Crazy Custom EFI Idea

[Chained_360]

This idea assumes a lot of things. First of all, this would take a lot of time to build, and a lot of research as well. That's why I'm posting this, is to see if anyone has any input or objections to this idea. It also assumes that the engine computer is going to be a MegaSquirt unit, seemingly the ultimate budget/DIY computer for fuel injection conversions. Finally, for ultimate cheapskate-ness, it's assuming a fuel-only system; the ignition timing would still be fully vacuum/mechanically controlled.

For the sake of reliability (and lack of patience), I've been considering converting my '68 Notch to fuel injection. Since it's my daily driver, I can't always afford to dick around with tuning the carburetor (and I also have about zero experience). And more importantly to a cheapskate high schooler, fussing around with the electric choke and fast-idle cams uses precious gasoline. More than anything, this idea just seems like an interesting thing to kick around.

A while back, Holley introduced the new Terminator Stealth EFI, which is basically a throttle body injection kit that disguises the unsightly throttle body as a traditional carburetor -- presumably to make people like me feel better about wussing out on carburetion.

-- And this got me thinking... why can't you just build one yourself? Rather than spending $2,000 on the kit, of course.

I got down to thinking about it, and as far as fuel and air control are concerned, an EFI throttle body really only needs provisions for a few things:
- Injectors
- Throttle plates
- Throttle position sensor
- Manifold Air Pressure (MAP) sensor
- Idle Air Control Valve

So here's where I'm either spot on or dead wrong; I think that all of these items could be retrofitted to a dusty old Holley 650 I have sitting in my garage. Now, of course, this will take some cutting and drilling on the old unit to make things work, but I think its plausible.
Fuel Injectors: Get four fuel injectors, remove the boosters from the barrels (?), and maybe drill them out to remove as much of the venturi shape as possible.
Throttle Plates: I'm not sure how nicely vacuum secondaries would play with a MegaSquirt ECU, so I figure it would just be easier to permanently connect the primary and secondary blades to each other and let the computer take care of the rest of the adjustments.
Throttle Position Sensor: I'm not too sure on this one, but I think that it would be somewhat not impossible to retrofit either a junkyard or a universal throttle position sensor to the factory throttle plate on the carburetor.
MAP Sensor: After a previous escapade with my '68 Notch's 360, I discovered that Holley carburetors (at least some) have a 3/8" vacuum port behind the carburetor underneath the secondary bowls. I would imagine that this would allow a MAP sensor to be tucked away and connected underneath the carburetor in a nook in the manifold.
Idle Air Control: Finally, and this is the biggest "if" (asides from drilling holes and mounting injectors in the main carburetor body), I had the mad-scientist idea of using the existing idle speed screw and a stepper motor/servo to control the idle speed. The IAC valve was always my biggest issue when trying to figure out this idea, and I think this is the best solution available. Since Holley 4 barrels have a screw that slowly opens the throttle blades to adjust the idle speed/idle air flow, I was thinking that a small servo could be mounted on or connected to the idle speed screw and programmed to control the idle RPM. This would probably take additional coding and would not be as great as a real IAC valve, but I don't know how else you could easily put an IAC valve in a standard carburetor body.

This post may seem like a pile of garble, but I think that this idea is something that would be insanely cool if it was pulled off. Again, I'm posting this here to unite the experience and expertise of some of the best mechanics and people around with that of a lunkhead teenager. Any words of widsom or skepticism are greatly appreciated. Cheers, and have a good one!

 

[ScampMike]

an interesting read - let me go back up to what looks like the "issue" here though - what is up with your current carburetor?

 

[Chained_360]

an interesting read - let me go back up to what looks like the "issue" here though - what is up with your current carburetor?

Well, the primary issue is operator error; I'm still learning how to tune carburetors. Aside from that, I am running into issues with cold starts -- the electric choke and fast idle cams provide some challenges for me. Understandably it runs rough until the choke opens, but the fast idle cams are just absolutely ridiculous. If I let it sit for more than five minutes off, it has to run through the fast idle settings all over even though the engine is warm.

I was planning on addressing those issues in a separate thread, and they're not the complete reason why I would go to EFI. Other reasons would be gas mileage and such. This thread is more exploring if I were to go EFI, this would be a hell of a cool way to do it.

 

[KitCarlson]

Sounds like you need to learn how to set up a carb, ideas about EFI need much more thinking and experience.

 

[Johnny Mac]

Sounds like thousands of man hours with HUGE margin of error, and $600 worth of hard parts, to avoid paying $795 for something that already exists.
FTH-30003 - FiTech Go Street 4 Injector 400 HP Carb Swap EFI System with Cast Finish

 

[ScampMike]

Well, the primary issue is operator error; I'm still learning how to tune carburetors. Aside from that, I am running into issues with cold starts -- the electric choke and fast idle cams provide some challenges for me. Understandably it runs rough until the choke opens, but the fast idle cams are just absolutely ridiculous. If I let it sit for more than five minutes off, it has to run through the fast idle settings all over even though the engine is warm.

I was planning on addressing those issues in a separate thread, and they're not the complete reason why I would go to EFI. Other reasons would be gas mileage and such. This thread is more exploring if I were to go EFI, this would be a hell of a cool way to do it.

well, you have some interesting thoughts about retro fitting a traditional carb with EFI - but in all practicality, it's not going to happen. If it could be done with any kind of ease, someone would have done it by now - just for haha's I did a Google search for "convert carburetor to fuel injection" - EVERY single hit explains how to take the carb off and install an EFI system.

 

[dukeboy440]

This idea assumes a lot of things. First of all, this would take a lot of time to build, and a lot of research as well. That's why I'm posting this, is to see if anyone has any input or objections to this idea. It also assumes that the engine computer is going to be a MegaSquirt unit, seemingly the ultimate budget/DIY computer for fuel injection conversions. Finally, for ultimate cheapskate-ness, it's assuming a fuel-only system; the ignition timing would still be fully vacuum/mechanically controlled.

For the sake of reliability (and lack of patience), I've been considering converting my '68 Notch to fuel injection. Since it's my daily driver, I can't always afford to dick around with tuning the carburetor (and I also have about zero experience). And more importantly to a cheapskate high schooler, fussing around with the electric choke and fast-idle cams uses precious gasoline. More than anything, this idea just seems like an interesting thing to kick around.

A while back, Holley introduced the new Terminator Stealth EFI, which is basically a throttle body injection kit that disguises the unsightly throttle body as a traditional carburetor -- presumably to make people like me feel better about wussing out on carburetion. View attachment 1715059543
-- And this got me thinking... why can't you just build one yourself? Rather than spending $2,000 on the kit, of course.

I got down to thinking about it, and as far as fuel and air control are concerned, an EFI throttle body really only needs provisions for a few things:
- Injectors
- Throttle plates
- Throttle position sensor
- Manifold Air Pressure (MAP) sensor
- Idle Air Control Valve

So here's where I'm either spot on or dead wrong; I think that all of these items could be retrofitted to a dusty old Holley 650 I have sitting in my garage. Now, of course, this will take some cutting and drilling on the old unit to make things work, but I think its plausible.
Fuel Injectors: Get four fuel injectors, remove the boosters from the barrels (?), and maybe drill them out to remove as much of the venturi shape as possible.
Throttle Plates: I'm not sure how nicely vacuum secondaries would play with a MegaSquirt ECU, so I figure it would just be easier to permanently connect the primary and secondary blades to each other and let the computer take care of the rest of the adjustments.
Throttle Position Sensor: I'm not too sure on this one, but I think that it would be somewhat not impossible to retrofit either a junkyard or a universal throttle position sensor to the factory throttle plate on the carburetor.
MAP Sensor: After a previous escapade with my '68 Notch's 360, I discovered that Holley carburetors (at least some) have a 3/8" vacuum port behind the carburetor underneath the secondary bowls. I would imagine that this would allow a MAP sensor to be tucked away and connected underneath the carburetor in a nook in the manifold.
Idle Air Control: Finally, and this is the biggest "if" (asides from drilling holes and mounting injectors in the main carburetor body), I had the mad-scientist idea of using the existing idle speed screw and a stepper motor/servo to control the idle speed. The IAC valve was always my biggest issue when trying to figure out this idea, and I think this is the best solution available. Since Holley 4 barrels have a screw that slowly opens the throttle blades to adjust the idle speed/idle air flow, I was thinking that a small servo could be mounted on or connected to the idle speed screw and programmed to control the idle RPM. This would probably take additional coding and would not be as great as a real IAC valve, but I don't know how else you could easily put an IAC valve in a standard carburetor body.

This post may seem like a pile of garble, but I think that this idea is something that would be insanely cool if it was pulled off. Again, I'm posting this here to unite the experience and expertise of some of the best mechanics and people around with that of a lunkhead teenager. Any words of widsom or skepticism are greatly appreciated. Cheers, and have a good one!

I like the idea and the sound of it. But I think you're gonna end up spending several hundred bucks in parts and who knows how much time involved trouble shooting. It would for sure be neat to see though and awesome if you got it to work.

 

[dukeboy440]

Sounds like thousands of man hours with HUGE margin of error, and $600 worth of hard parts, to avoid paying $795 for something that already exists.
FTH-30003 - FiTech Go Street 4 Injector 400 HP Carb Swap EFI System with Cast Finish

i think his whole goal was to make it look like an old school holley carb but function like EFI.

 

[ScampMike]

I like the idea and the sound of it. But I think you're gonna end up spending several hundred bucks in parts and who knows how much time involved trouble shooting. It would for sure be neat to see though and awesome if you got it to work.

assuming you get it right on the first try... which is highly unlikely - so how many carbs do you end up destroying trying to get all this to happen? 2, 3, 4 - - 100's of dollars?? more like thousands..

 

[Johnny Mac]

i think his whole goal was to make it look like an old school holley carb but function like EFI.

If only holley would have thought of that.lol (kidding) Yes i get the whole, lets save some money, and get a cool idea to work. It's just too much BS as scampmike is pointing out. These holley and fitech type units aren't just 4 injectors bored into a carb base and sticking out the bottom...they have annular discharge, pulse width modulation, integrated TPS, IAC, etc etc... I don't deal with megasquirt, but the ECM is what? at least $500 right? even adding some junkyard injectors and such, you're out of the ballpark on perspective costs vs Labor hours to make anything even close to this work.

He'd have better luck installing injectors MPFI style into the runners, and trying to use the carb as a throttle valve, but then you're into fuel rails, hold downs, on and on. To quote a famous British Auto Journalist..."Ambitious but rubbish"

 

[dukeboy440]

If only holley would have thought of that.lol (kidding) Yes i get the whole, lets save some money, and get a cool idea to work. It's just too much BS as scampmike is pointing out. These holley and fitech type units aren't just 4 injectors bored into a carb base and sticking out the bottom...they have annular discharge, pulse width modulation, integrated TPS, IAC, etc etc... I don't deal with megasquirt, but the ECM is what? at least $500 right? even adding some junkyard injectors and such, you're out of the ballpark on perspective costs vs Labor hours to make anything even close to this work.

He'd have better luck installing injectors MPFI style into the runners, and trying to use the carb as a throttle valve, but then you're into fuel rails, hold downs, on and on. To quote a famous British Auto Journalist..."Ambitious but rubbish"
View attachment 1715059561

im not disagreeing. I think the cost/savings ratio is gonna be slim to none, especially if you count the man hours and custom work needed. Still, if he could pull it off, and wants too try, I'll encourage him. After all, thats what hot rodding is all about, figuring out what does and doesn't work. Sometimes you win, sometimes you never should have tried in the first place. Wont know unless you try

 

[roccodart440]

Well, the primary issue is operator error; I'm still learning how to tune carburetors. Aside from that, I am running into issues with cold starts -- the electric choke and fast idle cams provide some challenges for me. Understandably it runs rough until the choke opens, but the fast idle cams are just absolutely ridiculous. If I let it sit for more than five minutes off, it has to run through the fast idle settings all over even though the engine is warm.

I was planning on addressing those issues in a separate thread, and they're not the complete reason why I would go to EFI. Other reasons would be gas mileage and such. This thread is more exploring if I were to go EFI, this would be a hell of a cool way to do it.

Many of us don't even run a choke and aren't having a cold start issue.

Your mileage isn't going to go up that much vs. a properly tuned carb and ignition. Combine that with cost vs. end money saved. lets say you get 2MPG more. OK, so 30 more miles per tank or so. This isn't big savings especially considering gas isn't outrageous currently.

It takes about 10mpg more to feel a difference in your pocket. Like when my cummins went from 12-13 to 20MPG and I drive 55 miles per day at least. Yeah, felt that. When my wife went from 30mpg to 50mpg. Felt that.

Fuel injection is great, truly it is. Don't think that it never has issues. When it does, you'd best be really knowledgeable about what to do when chasing down those gremlins.

Personally, I'd get that carb straightened out, even if it meant having a professional do it for you. OR I'd set your car up to run E85 if it's in your area since it is dirt cheap.

 

[tanis4457]

If you want to do efi down the road you'll need a 02 sensor anyway, id get a wideband 02 sensor and tune the carb or save the r and d on building an efi system and just buy a fitech or holley sniper setup, you will be way ahead. Depends if you want mpfi or tbi efi, tbi has come a long way. From my research mpfi has benefits at idle and part throttle over tbi but in wot driving both systems are the same.

 

[cannucky]

Wow you guys are brutal and so wrong . You're assuming his spare time is worth something when in fact this is his hobby and tinkering in the garage beats the hell out of spending money going for other forms of recreation . Second you are assuming he doesn't have access to some of what he needs .He isn't on the clock and may very well have access to used parts on the cheap . This is how new products are born ,yes holley is doing it for 2k , I would point out you couldn't get into Fi for under 4k before a couple of guys pieced together a cheap throttle body setup and started selling complete master kits for $1300 now everyone in the game has amazingly found a way to knock 75 % off their price too .
I say give er bud if I come across anything you can use I'll send it to ya on my dime , this is how our hobby is supposed to work -rock on .

 

[dukeboy440]

Wow you guys are brutal and so wrong . You're assuming his spare time is worth something when in fact this is his hobby and tinkering in the garage beats the hell out of spending money going for other forms of recreation . Second you are assuming he doesn't have access to some of what he needs .He isn't on the clock and may very well have access to used parts on the cheap . This is how new products are born ,yes holley is doing it for 2k , I would point out you couldn't get into Fi for under 4k before a couple of guys pieced together a cheap throttle body setup and started selling complete master kits for $1300 now everyone in the game has amazingly found a way to knock 75 % off their price too .
I say give er bud if I come across anything you can use I'll send it to ya on my dime , this is how our hobby is supposed to work -rock on .

Very good points. tell you what, i'm gonna take it one further. @Chained_360 I have a disassembled Holley 600, i got it given to me to tinker and learn how to rebuild holley's on. All the pieces should be there, nothing special but would work for you to experiment with. Yours free if you want it to play around on. Just PM me your address

 

[Chained_360]

Wow, this blew up overnight! I understand lots of the concerns with drilling a bunch of holes in an old carburetor, there certainly is a lot of room to mess things up -- I do have drill presses and conventional homegamer shop tools, unfortunately I don't have a fancy schmancy CNC or oldschool bridgeport to be as precise.

I also understand that I wouldn't save money by any means, I could drive for an entire summer (or more) on the same amount spent on building this system. Regardless, the cool factor is still pretty high for me, and you don't see people bailing on EFI just because it's cheaper not to.

The concerns and doubts that a lunkhead teenager could cobble this together are very valid, and I certainly do appreciate the offer for free parts as well! That's very kind of you @dukeboy_318, I really appreciate it. I guess the only thing standing in my way is that if I'm gonna have others help me out with parts and advice, I should probably do the whole shebang. Which isn't a problem, it just means that I'm gonna be even more broke if I decide to pull it off.

 

[KitCarlson]

Being a teenager makes a difference, you have determination, that is a good thing.

Engine controllers is my hobby. I have done about a dozen cars. My website may give you ideas, I can also help get you on the right track with MAP sensor, port injection, and temperature sensing that I not included in plan.
KitCarlsonEMS

 

[Chained_360]

Awesome! That would be good to test the unit with should a working one come out of this. I get paid tomorrow, so I'll see how much I have saved up and then we'll go from there.

 

[Chained_360]

Quick late-night update. I found these super tiny 8mm X 9.2mm stepper motors on Amazon for ridiculously cheap: Amazon.com : 10PCS 4 Wire 2 Phase Mini stepper motor micro stepper motor D8xH9.2mm with cable and small output Copper gear : Camera & Photo

After doing a quick search, it turns out that MegaSquirt controllers can use stepper motors in either a pre-set (closed loop) or variable (open loop) configuration in order to control the idle speed.

 

[KitCarlson]

I use a solenoid idle air control valve from a Neon. It is much simpler than a stepper motor. It takes only one transistor to drive a single wire. PWM is used, where just setting a duty cycle in a register controls idle air. I make a plate to add two hose hose ports, one hooks to filtered air, the other central position in manifold.

Save your money, use for education if you desire. Take classes to meet requirements of enrolling in engineering. I started at age 12, working at a car wash, and mowing yards, electronics repair on the side.

 

[Chained_360]

Save your money, use for education if you desire. Take classes to meet requirements of enrolling in engineering. I started at age 12, working at a car wash, and mowing yards, electronics repair on the side.

I'm already working on the classes for engineering part, and I'm currently working at a classic car restoration shop to save up some money. This project wouldn't be full splurge all at once, rather small expenditures along the way as I figure out what the hell I'm doing.

I'm currently sitting in front of my disassembled Holley 650 4 barrel, and I think I might give this a shot. I can sit and wonder all day, but I'll never know for sure until I start getting my hands dirty. I think the first place to start is to start taking measurements and see what parts I'll need. From the looks of it, I'll either need some very short fuel injectors or I'll have to make some sort of spacer that will place the fuel bowls farther away from the main body to allow room for the injectors and the fuel rails themselves.

After doing some research, it looks like most fuel injectors are around 2 inches long. Chevrolet injectors for the LS3 and LS7 are super tiny at only 1.5 inches long, but these suckers are expensive.

In regards to actually mounting the injectors, I think the best way to go would be to find or make some proper injector bungs and then drill the carburetor body out to mount the bungs, rather than trying to machine the body to directly fit the injectors.

 

[KitCarlson]

Adding injectors to a carb is a bad idea. TBI injectors are much larger than the LS injectors. TBI injection was a temporary step in progress, MPI is much better, fuel does not drop out as much. Also fuel proceeds the air, so when throttle is applied, the fuel enrichment responds much better. Picture the manifold filled with lean cruise mixture, then open throttle, the TBI will inject fuel, but the engine has to consume air fuel charge, already in the manifold first, so TBI has transient fuel issue. Yeah on a dyno, similar WOT HP, but real driving, not nearly as responsive, smooth, nor fuel efficient.

I still think you should get to the carb to work first, learn from that, nothing lost in learning. The system you describe is 80's technology with TBI, and mechanical igniton the 70's. It seems like much effort, for minimal satisfaction.

Thinking beyond todays tech, researching ideas while in school, taking additional classes that assist needs, may be good planning.

 

[AJ/FormS]

While I can see choke issues in Alaska for most of the year, choke issues are not that hard to overcome.Or maybe I should say that if you understand the choke system, then issues are not that hard to overcome. So as in most things knowledge is power.

 

[rumblefish360]

@Chained_360

I say have a go at it!

I do remember a long while back Steve Dulich (sp?) of Mopar Muscle (back in the day, now doing engine masters with Dave Freiburger) used a TQ as a throttle body for a F.I. Set up. He installed some F.I. parts on the outside of the carb. I can't remember the whole thing, but he did get it to work but I don't think he did what you want to try. Retro fit an entire carb.

 

[cannucky]

Since Holley is selling exactly what you are trying to build why not try to get some pics / details on how they are doing it , someone must have been "more money than brains " enough to drop 2k on one somewhere . even a call to Holley tech might get you the part #s for whatever injectors they are using to fit in a 4150 replica .