When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Okay. Crazy question of the day. Yes, I searched. Not just here. The entire internet. Literally. Before you ask, I have built several engines myself, so I am not exactly a novice here. I even own a few books on TPI. The information I am looking for seems to be a tad obscure. My question is this: can TPI run an engine with a cylinder count different than 8 if the displacement is fairly similar? Say, a 305 cubic inch V6? Or, just for grins, potentially 2 smaller engines whose combined displacement is around 5.0 or 5.7 liters? If I understand correctly, TPI is batch fire and essentially just a slightly smarter version of TBI. I would be using an MAF setup, which might make things a little easier. If it fires injectors in banks, then injector sequence would be irrelevant, but is there a spark timing issue? At this point I am theorizing that I can modify an intake manifold to accept 8 injectors and modify a distributor for the appropriate number of cylinders and the computer won't know or care how many cylinders are actually firing. Am I wrong? I can't seem to find any information or knowledge on the subject. Maybe I'm asking the wrong question when I search. Any help or advice or direction toward the knowledge I seek is greatly appreciated. Just trying to use up all my TPI parts and do something crazy at the same time. I do realize that if the answer is yes that it will require fabrication and tuning. Thanks in advance. I look forward to the replies.
What top secret motor are we putting TPI on here? You are correct that batch fire MPFI doesn't care what it's actually running but I'm mostly wondering what you're doing for a lower manifold if it's not going on a sbc.
Thank you for the reply. I am likely going to build a plenum and machine injector bungs into the sides and modify (hack) a set of aftermarket fuel rails that are already a little modified (hacked). That will get attached to a lower manifold modified to accept said plenum. I should still get good atomization; no worse than TBI at least. My only real concern is putting this all together and the computer insists on firing 8 cylinders. That's why I ask about the distributor. I am hoping TPI isn't smart enough to time spark other than initial base and whatever is written or tuned into the spark tables. Then I can just build a distributor for the desired number of plug wires and it fires every time the rotor makes a contact. Optimistic, but I think still a reasonable expectation. I will spill the secret motor build if I can get a definitive answer on spark distribution, or at least be pointed in the right direction.
My only real concern is putting this all together and the computer insists on firing 8 cylinders. That's why I ask about the distributor. I am hoping TPI isn't smart enough to time spark other than initial base and whatever is written or tuned into the spark tables. Then I can just build a distributor for the desired number of plug wires and it fires every time the rotor makes a contact.
The computer doesn't know how many cylinders it's firing. It's got two injector drivers. It fires them when it gets a "now, now, now" signal from the pick up coil in the distributor. That p/u could be 8 pulse/revolution, 6, 4...5. It wouldn't know or care, it'd just take that signal and combine it w/the MAF, CTS TPS, etc. and fire the injectors. I don't see why you couldn't make it work. I think that you could use any GM (or ford, etc) EFI system from the era and accomplish a similar outcome, too.
Thank you, Tom. I feel certain you are right about pretty much any Ford or GM EFI from that era working. I know EEC IV can be switched to batch fire, and I have a lot of EEC IV stuff lying around, and it would probably be a lot easier to make that work, but I'm just trying to find a good use for my TPI parts stash. EEC might be what I end up working with though. What you said makes sense about combining inputs. I'm probably overthinking all of this, but I would hate to get it all put together only to have some surprise that can't be worked around. So, I guess I'll spill some details here. The project is a Jaguar V12. Tough engine with lots of potential, but zero support or development for the existing engine management. I know I can go Megasquirt, but I'm not looking to build 1,000 hp, and I don't want to build maps and tables and scalars and strategies from the ground up. If I can adapt something simple but tunable, like TPI or EEC, pretty much all the reliability issues the Jag V12 is notorious for should just fade away. Nothing against good old American iron, but the reason people replace Jag V12s with American V8s is not because the engine itself is weak or incapable of producing power. The electric bits and electronics are just terrible. Injector placement will be an issue. Plumbing in sensors and building accessory brackets will be necessary. Building a distributor is going to be fun too; already got a few ideas. If I end up going EEC I can go distributorless and just run an extra coil pack. I think? I can weld and machine and run a tape measure. Clearly I have plenty of imagination. Now all I need is an accurate knowledge of some TPI specifics, which I think yourself and Komet have helped me with. So, I thank you. If I get far enough into this project in the near future, I will update here, or just start a build thread. Thank you both, again. If you have any further information or advice, it is always welcome. If I can return the favor, please don't hesitate to ask.
I've always thought that a Jag V12 was a great engine (mechanically), hobbled with a BS fuel "management" system. Like you, I've always thought that w/decent controls, why wouldn't it be a great running....and great looking, engine?? You're doing it. Please post updates on this project. I'm ALL about it.
The Jag has a distributor, (2, I think) doesn't it? What kind of p/u coil did they use? I wonder if you could run two GM HEI 6 cylinder p/u could inside the Jag dist(s)?
I don't want to build maps and tables and scalars and strategies from the ground up.
Well I hate to break it to you, but that's exactly what you're going to need to be doing.
Originally Posted by clyde427
If I can adapt something simple but tunable, like TPI or EEC, pretty much all the reliability issues the Jag V12 is notorious for should just fade away.
Oh no, your issues will be multiple, unique, and poorly understood. TPI is an antique fuel injection system practiced by the esotericists here but none of them have it running on a Jag v12. Take note of the extreme dedication some members have taken to recalibrate their tunes to accommodate after (comparatively) simple changes in runner length and plenum volume only. I'm not saying it can't be done, but it sounds like you're going to need a lot of help in the tuning department and your custom design would likely be better suited to a modern aftermarket engine management system where you can simply tell it your number of cylinders, firing order, displacement, etc, and get a base tune that has somewhat of a chance of popping off when you turn the key.
I'd also look into abandoning the distributor entirely in favor of a coil near plug arrangement controlled by the same aftermarket ecm mentioned earlier.
It's a 5.3L, ~300hp engine that peaks at ~5000 RPM. Sound familiar? I don't think it will be "perfect", running on an L98 control system....but I think w/some dinking around w/injector sizing and fuel pressure, it could work better than good enough.
I think that it's obvious that the OP can't just "bolt on" the injectors/harness and ECM of the L98 to the Jag. I hope that wasn't implied in my earlier post. I think it CAN be done, and I think it CAN work pretty good....just like we can...and have.... cranked the power way, WAY up on an L98 (and other '80s' controlled cars) w/o building maps and scalars and strategies (tuning) from the ground up. It's a similar sized engine w/similar power and a similar RPM range....I think it can work decent and be a totally cool effort. I'd drool over that, at Park City Car & Coffee. It's not another LS and it's not one of 40 other 911 GT3 RS's. It's different and interesting.
I'm with you Tom. Not to be argumentative or even just contrary, but simply for the sake of keeping the discussion going and hoping others may chime in with knowledge or even a helpful opinion, I'm not trying to build the next International Space Station. I realize that TPI isn't the end all of anything. It really wasn't great when it was in production, and the ever progressing engine management technology has certainly left it behind. Buuuuut, a lot of people have made it work quite well, and some have even used it to make a tremendous amount of power. Reliably and consistently. As I said before: I know it will require tuning. I am not under the impression that GM made TPI that wasn't ideal for the engine it was applied to but somehow is coincidentally perfect for an engine with a different displacement and cylinder count and engineering dynamics and made by a completely different auto manufacturer. Its simple. That's a huge plus here. I have played with EEC IV enough to know you can run a 408 on the stock 302 computer. It doesn't run very well, but it will stay running and you can drive it. Admittedly, a lot of my motivation to make TPI work here is from a stubborn desire to use what I already have. I sincerely doubt it will require more effort to tune the TPI than to start from scratch on a standalone, but I also know I could be entirely wrong. The idea here is to try to find out as close to a certainty if it just will not work; not if it will require effort on my part to sort out issues and make it run like stock. Like Tom said: displacement is close enough and with some testing and tuning I feel like it could be made to work reasonably well. I'm just looking for a nice, smooth running V12. Maybe this isn't the way to accomplish that, but I haven't found a reason that it isn't. Yet. Feel free to give me one.
Oh, yeah. The distributor. I have looked at building a side by side like they used to build for the big hemis in dragsters and funny cars, and also stacking them. Both present some challenges, but neither are unheard of. I can go distributorless with EEC IV, but that may entail fabricating some sort of magical reluctor wheel. I need to look at that option a little more thoroughly. I might even be able to build a set of distributors that mount horizontally and run of the front of the cam. Sort of like a Jesel belt drive type thing. As always, open to suggestions...
I have played with EEC IV enough to know you can run a 408 on the stock 302 computer. It doesn't run very well, but it will stay running and you can drive it.
I hear that. I ran a SBC 400 on a Trans Am 305 ECM...no tune at all; on the stock tune. With some fiddling, it eventually did run well. At the time, it surprised me but looking back, I hadn't changed the shape of the tq curve at all...I'd just made it "more"....so fueling it appropriately on the stock tune was easy.
Originally Posted by clyde427
I might even be able to build a set of distributors that mount horizontally and run of the front of the cam. Sort of like a Jesel belt drive type thing. As always, open to suggestions...
Do you have a pic of the inside of the stock Jag distributor?
I have driven the factory long runner multi point injected version of the old 250 I6 that was built until ~2000. Long runner manifold with port injection and the factory dual split manifold exhaust really woke that engine up compared to the smogged up varajet 2bbl version when it stopped being made here. The long runner manifold looks like TPI runners-attached to a Ford 300 I6 upper plenum.
Sorry about the absence. Sometimes the rest of my life just doesn't understand that car projects are all that really matter.
Tom: I do not have any pics, but as soon as I do I will see if I am smart enough to post them on here. May be a good gauge to determine if I should attempt this TPI adaptation madness.
Fast: was that a Chevy 250 I6? I would love to know more about that. Like, did it have a modern crossflow head as well? Straight sixes are just cool. Smooth power and all way down low. I am looking at using all my spare runners to fab up a long runner manifold.
In the mean time I have been pondering solutions and probably just coming up with ways to make this more complicated, but I wonder if one could splice in more injectors into the engine harness. More signal would be required, but that may be accomplished with a simple resistor swap. I'll have to take a look. More injectors would greatly improve fuel delivery and distribution and atomization. Batch fire would eliminate the need for injector timing. Pulse bandwidth and location would be the biggest issues. Feel free to shoot me down here or just have a better idea.
Komet: I definitely agree with you that my issues here will be multiple and somewhat unique, but I disagree that they will be poorly understood. Its still a 4 stroke engine, and TPI has been around long enough that all of its issues have been addressed, if not solved more or less. Either way, I appreciate you as the voice of sanity, even if I tune you out in favor of the voice of cool factor or thrift. After all, car projects are about learning and a lot of time innovating. What is learned here may help someone else build something crazy that no one would have thought would ever work. Or maybe just talk them out of wasting time and money.
Thanks all, just the same. Looking forward to going broke and insane. That kind of rhymes.
I prefer aftermarket ecms for builds that deviate substantially from stock because they have streamlined the iteration cycle that you need when dialing in your garage engineered deathtrap. If you were running a small block chevy, I'd tell you to look at FiTech and https://www.dynamicefi.com/EBL_Flash.php as well, but I'm really not sure about driving the extra 4 injectors with those ecms.
Ha ha ha...the best kind of death trap, b/c it's YOUR death trap!
I've built myself one of those, but for some reason, I'm still alive.
Originally Posted by Komet
I'm really not sure about driving the extra 4 injectors with those ecms.
I'm not a smart guy...but I don't know what the problem is with this, other than driver failure from too much current? Battery power is supplied to the injectors via the injector fuses. Simple. The drivers in in the ECM ground that power and the injectors fire. Why can't someone splice in more in the same-bank circuit? If the issue is too much current for the ECM's drivers, could you run a higher ohm injector?
I'd try it b/c it's free, interesting to do....and if I couldn't make it work, then I'd move on to my next idea, and try that. That's, "hod rodding".
Ha ha ha...the best kind of death trap, b/c it's YOUR death trap!
I've built myself one of those, but for some reason, I'm still alive.
Agreed, I have one as well and I'd do it again.
Originally Posted by Tom 400 CFI
I'm not a smart guy...but I don't know what the problem is with this, other than driver failure from too much current? Battery power is supplied to the injectors via the injector fuses. Simple. The drivers in in the ECM ground that power and the injectors fire. Why can't someone splice in more in the same-bank circuit? If the issue is too much current for the ECM's drivers, could you run a higher ohm injector?
I'd try it b/c it's free, interesting to do....and if I couldn't make it work, then I'd move on to my next idea, and try that. That's, "hod rodding".
Driver failure would be my concern. The scenario I envisioned was one where it works all through the short test phase until summer when you're 45+ minutes on your first extended cruise and everything finally gets to critical temp and lets the factory smoke out. Or worse, drivers that don't fail but act in unexpected ways from the increased load (thus the 'poorly understood' aspect I mentioned earlier). This is all above my pay grade, perhaps Rbob could elucidate on how viable adding 4 injectors to the mix would be.
I guess if I really did have an ecm and harness kicking around, the first thing I'd do would be hook it up and see what she'll do. OP is giving off first-timer vibes to me and I think it's important to at least conceptually understand a few things:
There is no thrift in building a custom MPFI system. You will absolutely spend at minimum a four digit sum. I think mine clocked in at around $1.5k and over half of my parts were free. That being said, it's pretty easy to do that with a carburetor these days as well. Depends on your definition of thrift I guess.
It's important to have a basic understanding of how every sensor affects the algorithm used to calculate your pulse width. 'There's guys who have figured all this stuff out' is a recipe for disappointment because you're left grasping at straws when it's troubleshooting time.
TIG welding is as much art as it is science. The learning curve is steep and used antique aluminum is the definition of impurities.
Know the process for collecting a datalog and modifying your tune from start to finish.
That would solve the injector driver issue, and it would make MAF's easy, but you'd still need to figure a way to run two "cam sensors" (pick up coils). Which I think is totally doable, but you need 12 pulses/cam revolution, one way or another.
Even the OEM's do this kind of "creative" stuff. Early Viper used one V6 coil pack on one bank, and one Neon 4 cyl coil pack on the other bank to fire 10 cylinders. Kind of cheesy...but it worked and made 400hp in 1992.
That would solve the injector driver issue, and it would make MAF's easy, but you'd still need to figure a way to run two "cam sensors" (pick up coils). Which I think is totally doable, but you need 12 pulses/cam revolution, one way or another.
Even the OEM's do this kind of "creative" stuff. Early Viper used one V6 coil pack on one bank, and one Neon 4 cyl coil pack on the other bank to fire 10 cylinders. Kind of cheesy...but it worked and made 400hp in 1992.
My 1994 Ram had the same dual coil pack approach. Some of the European V12s actually had 2 of everything incuding ECMs and treated the engine as 2 seperate I6 engines for fuel and spark.
I remember the car wizard put GM TBI on a Jag motor. I wonder if it would easier to adapt V6 electronics and run 2 ECUs
Ford also did that with their test build V10 or maybe V12 Mustang. I forget if it was BMW or Jaguar but one of those also came from the factory with 2 ECMs.
The BMW M70 v12 used two ECMs, fuel pumps, distributors, mafs, practically everything was doubled up. I think the Porsche v8 in the 928 did as well. Good 'ol Bosch squaring the failure rate. You can tell how good of an idea this is by how many manufacturers still use this system today.
It wasn't a good idea b/c it was expensive. Fundamentally, it was a reasonable idea/solution. Also, a major factor in this thread is: Criteria. In the first case (dual Bosch), some criteria are; any duff can get in the car, turn the key, drive away. A second criteria is: do ^that^ for 100,000+ miles/decades. It turns out that it wasn't great at the latter. And it was expensive. I don't think that the OP has the same criteria in mind.
There is no reason why a 2-V6 ECM solution should be any less reliable than two, GM V6's fuel delivery systems.
There is psychology at play here. I put an '03, 4.0L Jeep engine in a 1960's Mercedes Unimog, replacing the 2.2L Mercedes gas, inline 6. To make it work I fabricated an intake manifold out of mild steel, used a 2 bbl Holley carb,
Carb hat from a late '90's Durango/Dakota 5.2L
.....and ran a late '70's AMC 6cyl electronic ignition system....then wired the whole thing 12v starter/ignition/fans & 24v the rest of the vehicle, using one Jeep alternator, rejiggered to put out 24v. Also made a flywheel, two plate adapter, to use the Unimog clutch on the Jeep engine, w/the jeep starter.
Could have been a thread called, "Use of Jeep motor in something other than a Jeep". You know how many on the 'Mog forums said "That'll never work!"? A lot. Most. "Intake won't work! -Intakes are very complex devices!!" "Can't run a 70's ignition on an '03 engine!" "Can't do that to an alternator!" "Clutch will never hold!" On and on and on. People often point out what "won't work" -and they do that, b/c THEY don't know how to make it work. Inconceivable. Putting down an unproven idea is easy. What I find impressive, and what I LOVE....is when people show us what CAN be done.
I want to see this Jag project go.....even if, "It'll never work!!"
Grab vortec heads, this intake and paint silver, done.
Except those are not Vortec heads. Those are CPI heads. Totally different intake port design and intake bolt pattern. The fuel spiders were also pretty much junk and no longer available anywhere. For 4.3L a better port fuel system would be the 4.3L Marine MPFI intake and that one will bolt to the 4.3L vortec heads.
Except those are not Vortec heads. Those are CPI heads. Totally different intake port design and intake bolt pattern. The fuel spiders were also pretty much junk and no longer available anywhere. For 4.3L a better port fuel system would be the 4.3L Marine MPFI intake and that one will bolt to the 4.3L vortec heads.
01-24-2024, 07:32 PM
