Hi all, I am thinking of getting some inlet manifolds cast for the 2V heads as the spare engine I bought is an injection engine and I'd like to put IDFs or EFI on it. Might be a good option for people struggling to find parts for and tune the Bosch K-Jetronic system. I need 9 other people to reach the MOQ and the price would be $200 each plus post. Let me know if you are interested. I have 2 people who have asked for 4V already and it would be easy to adjust the drawings for the 4V head if there is enough interest. cheers, derek Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login
I've often wondered if it wouldn't be better for an EFI ITB manifold to do a cross ram design like the old Weber small block ford manifolds - here's a Chev small block version EFI setup: Image Unavailable, Please Login That gives you a bit longer runners than the vertical oriented throttle versions, allows more room for desired intake trumpet length, and more room for a proper plenum/airbox and filter setup over the trumpets on each side Image Unavailable, Please Login Those guys who's photos I posted https://www.efihardware.com/ do DCOE and IDF throttle bodies (as do quite a few other shops), a cross-ram manifold for a 308 designed for DCOE /sizing/spacing could do actual carbs or an EFI throttle body on the same manifold. If you're going to do custom manifolds for the 308 - be the first to offer this cross-ram design!
The angled design would be better for EFI but carbs are vertical or horizontal and I don’t think there is space for DCOEs so you have to have vertical. I could have some 45degree versions made but it would be a lot of time to measure, design, model, test etc. and I think I have more chance of getting the numbers with an IDF/IDA manifold shirt enough to allow a decent filter under the deck.
Derek, do you have 4V manifolds for DCNFs? I have a set of 348 heads that I would love to build with my 308 carbs.
Sorry about the slow reply Dan but Brock asked me to look into this again so thought I would give this discussion an update. Casting the manifold as a single piece is going to need very expensive tooling so I will look into having only the tubes cast and then welded onto machined flanges. This opens up the possibility to weld both IDF & DCNF flanges onto the top and 2V or 4V flanges at the bottom. The unit cost will be higher, possibly $250-300 each but we should get more flexibility in terms of head types and carbs. I am quite busy at work these days but will try to get the new drawings done and sent off to the usual suspects for quotes over the next few weeks.
Once you are welding it might be easier/cheaper to just fab the whole thing? I've built a few manifolds using aluminum pipe form mcmaster for the runners, Slight curves get mitered, for tighter cures they sell elbows I miter in, and oval is a round tube crushed in the vise, a tapered, cureved oval is in this case is 3 sections first cut and shrunk to the right perimeter, then crushed to the right ovality, then mitered to the correct curve ...then grind it all smooth and its a "custom cast" intake to put blowers and intercoolers on a QV or ducati TBs on a TR header 400i, its kind of all the same really, cut, weld, grind, repeat. I've also seen both 2v and 4V ITB intakes made on a CNC, but those were for sportbike TBs (which are about $100.set of 4 including injectors and fuel rail on ebay) and don't need a top flange so easier to machine the intake. Image Unavailable, Please Login Image Unavailable, Please Login anything can be made. Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Image Unavailable, Please Login Another slightly off topic item that I'm a big believer in is this guy Image Unavailable, Please Login Image Unavailable, Please Login A good MAP signal is hard to get from ITBs so most people end up using the TPS and running alpha-n (throttle position based load sensing) instead of speed-density(MAP based load sensing) which is really perfered for a lot of reasons on street cars, it just does a much better job at low throttle openings like you are doing 99% of the time on the street....but you need a good MAP signal and without a plenum type intake manifold you don't have a good signal. Many try adding a small plenum and that kind of works but the issue is it's not a real plenum so for every 1 cylinder pulling vacuum you have 3 more leaking vacuum and you end up with a signal about 1/3 as good as you'd get from a real plenum and not in any way linear as the throttle starts to open and the vacuum in the 3 leaking cylinders goes to about 0 meaning the signal all but vanishes....so most just end up going alpha-n and living with it as the best of a bad situation. I found this sketch on the megasqirt forum and had a board made that holds up to 12 map sensors but can be cut to 2/4/6/8/10/12. Its a simple diode circuit that looks at up a bunch of separate MAP sensor and sends the strongest signal to the ECU so the ECU is seeing a MAP reading free of vacuum leaks and you actually get a stronger MAP signal than you would have on a normal plenum intake. The frankenferri with ITBs and big cams (so low vacuum) is idling at about 35kPa which is the strongest MAP reading I've ever seen on anything, my QV intake reading was about 60kPa at idle iirc. Its also very linear as the engine comes off idle and starts making power. I was very honestly amazed how well this thing worked and completely solved one of the biggest tuning issues with ITBs.
I forgot this, because each cylinder has its own MAP sensor I was also able to send the separate signals to the ECU so I can display them for syncing the TB. Mine are from a motorcyle so everyone needs to be adjusted, this makes it quick and easy. Once I had this working I also added a cylinder trim setup so the ECU can automatically adjust fuel cylinder by cylinder to compensate as the TBs go out of adjustment and light a warning light when they are X% out so I know its time to retune them. Obviously not every ECU can do stuff like this but I can share the model I wrote with anyone who runs AEM/Enginelab ECU hardware (about $900-$1200 on ebay). I think you can do stuff like this with the motecM1 ecus....but the ECU is $4k then you need a $2500 software license and you'll be at $7-8k before it runs, or $1000ish for a setup like mine Anyway, when you are running EFI, the ECU and good signals to the ECU matter is the message. Image Unavailable, Please Login
Truth! I am amazed at how flexible these small Italian companies are at such challenges too. Ferrari whipped out a set of 4v Dino V6 heads in no time. Maserati did the same with their SM/Merak V6 when it was raced for Ligier at Lemans in 74, just casually whipped out a set of 4v heads on their 3L v6 revving to 9300rpm with 430bhp, in no time flat, with mechanical injection system. Can't be more than a dozen sets made, all crunched out seemingly at the drop of a hat. Love the custom fabrication side of things and the creativity it births
Why not cnc them? I 3D printed a set for my QV and I’m sure I can machine them on my four axis without much trouble. If they were taller and more complex I’d prefer a five axis but they are pretty simple.
It's a question of having both the time and the equipment/skill to make several sets. My plan is to get them assembled in China and let them decide whether to cnc the tubes or cast them, for 10 sets that's 80 tubes so having them fit well and require less welding, filling, grinding etc. is a big plus.
I found this sketch on the megasqirt forum, it 12V not, I was starting to mark it up to tell the guy doing the PCB what I wanted. All the files are opensource license and found here. There is a problem with the opamp (it doesn't pull low enough iirc) so I have it bypassed on my board https://github.com/jharvey/MultiMAP I'm not sure how easy it would be with programming vs HW.....i guess you could just scan in AN inputs then if/else your way through them looking for the lowest and do a running average on that result. i have a running average filter on my input from the multiMAP HW that I may adaptive, I change the coefficient based on MAP reading and rpm so try to make it very steady at idle/cruise but less laggy elsewhere. Image Unavailable, Please Login
