Thread: Crazy EFI idea?

Tweet

I mentioned on Dave's FAST EFI thread that I was interested in getting into EFI. The new self tuning ECU's look like they will give someone like me a chance since I don't have access to a dyno anywhere close.

My latest project is going to have a 383. I have a couple of old TPI units laying around, but the general consensus is that without a lot of expensive high flow parts, a TPI will severely choke a 383 above 4000 RPM. It will make a ton of low speed torque, though.

Over the last few years I have managed to get a decent machine shop set up in my shop. I'm also wanting to get into aluminum casting. Hopefully I'll get that done this winter. I can build what I'm about to describe without spending a lot of money, just time.

The other day I was looking at the TPI base manifolds and it dawned on me that they look sort of like a cross ram manifold. I started thinking about converting one by fabricating 4 two barrel throttle bodies to go in place of the TPI runners. The TB's would extend out more or less horizontal over the valve covers with stacks. The stacks could either be capped with long narrow air cleaners or a tubular log that connected to cold air induction forward of the radiator support.

I think the injectors can be turned 180* so that the wire connectors are pointed toward the middle of the engine and the stock fuel rails can be reused. A decorative cover panel would hide the wires and the old EGR valve location(I wouldn't be using the EGR valve). The old EGR passage connects all the runners together and I think that it would serve as a vacuum balance tube.

Another crazy idea is to add an idle air valve to the old EGR location. Theoretically the common EGR passage would allow idle air to bypass the throttle blades. The EGR passage would be blocked off on the exhaust side. A tube would supply the IAC with filtered air from above one of the throttle blades. Sounds like it would work anyhow.

Here is a pic of the TPI base. It sure looks like a cross ram manifold to me.



It would be similar in concept to this.



You could also do something similar to this easier than my original idea using junkyard throttle bodies. This would be a natural for cold air induction and the stock (from whatever?) TB's would already have IAC.



The TPI base can be opened up pretty easy for better flow and the throttle bodies could be fabbed any to any size so it should flow better than a stock TPI. Yeah, it's a lot of work, but it sounds like an interesting project to me. It would look cool too.

OK, so talk me out of it. Any potential problems?

I guess I don't know enough about EFI to say it wouldn't work..... But it would be a ton of fun putting it all together!!! I really like the layout of the EFI in the bottom picture, ready made for twin turbo's and a gazillion hp!!!!

Way back in my road racing days (daze) I crewed on Kemp's IMSA Mustang, it ran a Kinsler crossflow mechanical injection unit---would be fun to convert it to EFI!!!

So with this "EFI" are you looking at the ECU just controlling air/fuel ratio, ala the raft of carb replacement systems out there today or bumping it up to the ECU's that control spark, too? I'm a neophyte with EFI but to me the beauty of the system is having both the air/fuel map and the spark map, where you take all of the guesswork out of what your fuel and your spark timing are doing across the range of rpms.
There's almost a cult following of seat of the pants EFI guys using the likes of the MegaSquirt DIY systems, and I really respect them for their knowledge and ability to adapt pieces & parts to make a great engine management system. There's also a bunch of guys out there who think they know all about EFI, but they quickly taste shoe leather when they start talking specifics. Several of the companies (FAST & Holley for two) sell universal EMS controllers stand alone, laptop interface and capable of interfacing with just about anything you can imagine, like drive by wire throttle input.
No reason why what you're talking about can't work, IMHO, as long as you're prepared to work through some details as you go along and enjoy the challenge of making something from nothing. I'd say "Do It!!", but only if you plan to document your progress here!!

Sorry for the long winded reply, too....

Dave Funny you should mention Kinsler injection and IMSA. Way back when I was a teen, I got to see an IMSA Mustang II in a shop in Jackson, MS. It had a cross ram style injector setup on a 351. May have been a Kinsler. I know I was very impressed with the look of that setup. That's probably the seed for this idea. This engine will be going in a light weight sports car with a low hood.

I'm leaning toward the second idea idea for my first try. Since the all the stuff for both setups would be bolt on for the stock manifold, I could change out to the first idea later down the road. The first setup would look killer in my T bucket, too.

Rspears I do plan on controlling spark with the ECU. The new FAST XFI sportsman has that capability. The Sportsman has self learning mode, but it can be tuned with a laptop if needed. The lower priced EZ2.0 ECU only has self learning mode.

The main problems I've come up with are whether the old EGR passage will serve as a vacuum balance and could it actually be used to supply idle air. Those are fundamental problems. Other than that, it's just straight forward machine and fab work. The throttle bodies will need to be angled slightly up to clear the valve covers.

If the idle air idea doesn't work, I could always go with a simple mechanical idle adjustment like a carb, but I would prefer computer controlled idle. The main fab problem is getting the throttle blades fitted well enough to the bores to seal off at idle. I'll have to have a way of synchronizing the throttle bodies, too.

Of course I would do thread to document the process.

Mike

Originally Posted by Hotrod46

Dave Funny you should mention Kinsler injection and IMSA. Way back when I was a teen, I got to see an IMSA Mustang II in a shop in Jackson, MS. It had a cross ram style injector setup on a 351. May have been a Kinsler. I know I was very impressed with the look of that setup. That's probably the seed for this idea. This engine will be going in a light weight sports car with a low hood.

I'm leaning toward the second idea idea for my first try. Since the all the stuff for both setups would be bolt on for the stock manifold, I could change out to the first idea later down the road. The first setup would look killer in my T bucket, too.

A coincidence indeed!!!!! That was Kemp Motorsports in Jackson----the car I crewed on!

Originally Posted by Hotrod46

....The main problems I've come up with are whether the old EGR passage will serve as a vacuum balance and could it actually be used to supply idle air. Those are fundamental problems. Other than that, it's just straight forward machine and fab work. The throttle bodies will need to be angled slightly up to clear the valve covers.
If the idle air idea doesn't work, I could always go with a simple mechanical idle adjustment like a carb, but I would prefer computer controlled idle. The main fab problem is getting the throttle blades fitted well enough to the bores to seal off at idle. I'll have to have a way of synchronizing the throttle bodies, too.
Of course I would do thread to document the process.
Mike

Mike, just chewing on the EGR passage idea it seems to me that your idea may be a touch of genius! First thought is that it might put more to cylinders closer to the IAC valve, but I don't think it would matter - we're just talking a tiny air bleed so I can't see that the porting really matters. In theory it seems like a good approach to me, but just my opinion.
For the synchronizing of the throttle bodies, I agree that they need to be balanced, but it would seem to me that you would synch full open and closed as precisely as you can, and then you'd have one TPS for ECU/EMS input, is that right? The IAC would kick in to balance idle air, and as you come off idle the A/F map would kick for your transition into fuel control. I haven't looked at the FAST XFI Sportsman, but I'm guessing that you can maybe have individual A/F maps for each cylinder for hole by hole fuel control?

Dave, I stopped by that shop with a friend when I was still in machinist school. The owner said that when we finished school, we should come back. He might could use us. I've often wished I had. Life took me in another direction.

Originally Posted by Hotrod46

Dave, I stopped by that shop with a friend when I was still in machinist school. The owner said that when we finished school, we should come back. He might could use us. I've often wished I had. Life took me in another direction.

That was some great times and a real learning experience for me!!!! The Mustang was a one of a kind, built by Bob Riley, engines were done by Gapp & Roush, Doug Nash 5 speeds modded by Joe Liberty----We were terribly underfunded compared to the factory teams, but we did manage to make a good showing for ourselves most of the time. Had a bunch of off time, the IMSA schedule was a short one then, got to do a lot of prep and repair work on different road race cars, some testing time with Charlie for "other brands", and generally a really good time with tons of hours and a really steep learning curve!!!!!

Originally Posted by Dave Severson

generally a really good time with tons of hours and a really steep learning curve!!!!!

Sounds like just what I was hungry for at that time in my life. If the planets had aligned a little differently, you and I may have worked together.

Mike

I like the way you think Mike. building this yourself is the truest form of hotrodding. so many people listen to the voices saying "you can't" and they never get the pleasure of doing something out of the box.
I guess I never looked at a TPI egr port close enough to know how it's routed in the lower manifold. does it connect with each individual runner? if it does, then I can't think of any reason you couldn't use it to feed air for the iac. you still need a way for the ecm to control that air however. I don't know if you could machine right into the manifold to fit an iac valve directly, but you could thread the opening and plumb it to a block like the one I machined for my Northstar itb system.



The dual TBs seems like an easier way to handle the iac at first blush, but I'm not sure you could use it as-is anyway. trying to control two could be an issue in keeping them in sinc, and you can't just use one side because they are isolated to each bank of the engine.
the dual TBs does make it easy to hook up the TPS control, but that would work the same with any number of TBs. on my Northstar system I used TBs from (2) GSXR600 Suzuki bikes and was able to keep the Suzuki TPS sensor. I think most brands of TPS work in the same ohm range, and any ecm I've used was able to recognize it's parameters without any problems.

regardless of which way you go, have fun with it, and I look forward to watching.

Russ

Roger I think you're correct about just a tiny amount of air required for idle. I'll be checking out the EGR passages a little closer when I get some time.

What you suggested about synchronizing at idle and wide open is right on track with my thoughts. They will all have to be completely closed at the same time or I'm sure there will be idle issues.

I'm not sure if the Sportsman has individual cylinder mapping, but I know the regular XFI does. I'm pretty sure that you would have to be set up for sequential injection for that to be implemented.

Russ The block you have pictured is something along the lines of what I was thinking. I haven't really gotten far enough along to figure out which IAC I would use. When I get a chance I'll make a trip to the UPull It and go shopping. I will search for your Northstar thread. I would like to take a look at it.

I originally did a build thread on another forum, but that was ~7 years ago and most of the photo links are dead now. if you google "ITB Northstar" or "Northstar T" I'm sure you'll find plenty of reading material however. I originally built the injection for a Northstar powered Fiero I had built, and then the injection migrated over to my T bucket for a few years. both cars have since been sold, but I have plenty of photos and am always glad to provide any help I can for those attempting to build their own similar system.

Cheers,
Russ