After you complete this engine, how long do you think it will be before you supercharge it???
I've been on a 3 year development cycle with this car I think......and a supercharger would be cheaper then the custom crank.....
Maybe I missed something here but is it really necessary to stroke the crank?. There are plenty of very successful over-square engines out there. I would be inclined to go with a good standard crank, spend the extra dough on better rods and pistons and BLOW IT. It should give you more room to tweak and develop the character and performance curves of the engine whereas the static configuration of a naturally aspirated engine will limit your choices.
There is absolutely no good reason to stroke it other than I got it in my head the it would be pretty cool to get to 6 liter .double stock. It serves no other purpose really. It would give more torque, which I like, and if I could have done it for $1000 I would have but I think thats dead at this point. All joking aside, I see no reason to supercharge this engine. Im pretty sure Ill be able to ring about 600 hp out if it as it is and that is about 200 hp more then plenty and I dont want to mess with what should be a magnificent sound. There is no supercharger in the future that I see.
I just gave the vems site a good read....it appears to be an improved version of a megasquirt and my initial feeling is that I'm not in love. I'll talk to the guy and see what he is willing to correct, but I have the feeling it won't be enough to sell me even a free system. It uses the same 12x12 ignition and fuel table set-up as megasquirt and I just don't think that is enough to get a good tune on a street engine. haltech is 32x32 and motec is 22x40 I think and I think it’s for a reason. My first haltech was 16x 16 and that really didn't cut it, there were spots that weren't quite right that I could never quite fix. My second haltch that had 32x32 tables never had any such problem. It might have been processor speed related too, and this is a naturally aspirated engine so it won't need quite as many points but.... This system will do sequential injection and waste spark on a v8 and it also appears to also be able to do twin distributor non of which the megasquirt will do. But it looks like no way to run sequential injection on a 12 cylinder because it only appears to have a total of 12 drivers which leave none for ignition. It is cheap enough to just use 2 of them, but then there’s that 12x12 table problem We'll see....I would like to save money....so I'll listen before I decide.
Are you sure you need 12cyls sequential? I've spoken to a number of people and the consensus is that all you might lose is a little bit of smoothness at tickover, otherwise it makes no difference. I'm planning to use a DTA S80 unit on my 575M engine.
Great project, my hat's of to you! Not to confuse the issues, but you seem to be very focussed on low end torque. I'm sure that the trans can handle the 600 or so planned HP, but can it handle the torque the powerplant will crank out at low revs? I'm a novice to much of the specific Ferrari stuff, but I'd be worried about the drop gears, for instance. I've been tossing this whole thing around in my mind for the past few days, and thought I would go a totally different route, and rather de-stroke the engine a little to, say, 4 litres or so, and make it a real screamer. With the standard diff ratio the 308 runs out of revs with more power, but if you can gain some top end revs it immediately adds dramatically to the top end of the vehicle. It would also create a more than glorious symphony! Happy motoring, Jack V.
Im a bit confused with your term "de-stroke". This generally refers to making the stroke shorter, and would by reference lower engine displacement. Also you refer to 4 liters. The 308 is 3 liters, and attempts to get it beyond 3.4 liters havnt yet materialised and probably wont. Perhaps your suggesting the V-12 be reduced to 4 liters? Unless it was seriously destroked, getting any significantly higher speed out of the engine and keeping it reliable will be periless. As Mark has observed, the heads are not going to flow well enough to make much more power without drastic head work. Even at 4 liters power would probably still fall off before 8000. And as to the gearbox, I dont think anyone has yet found its limits. Norwood has probably done more projects with 308's that have pushed high horsepower, and as far as I have heard they never blew one up. They even had a blown Chevy sitting on top of one if I recall. In any event, there have been more than enough turbo kits and blowers on these cars, and the gearbox has never been cited as being weak. I believe it was James at Norwood who told me the general thinking was it would handle over 800 HP, and possibly 1000. So I dont think Marks measely little 600 HP will do much harm The gearbox would be the last of my concerns.
That is a good question and I'm really not sure. I'm sure it makes no difference at all for hp. But the higher the output per liter, the more it matters at idle I think because injectors big enough to feed the max power end up at a pretty short pulse at idle. On the supercharged engine there was a marked improvement in idle/low power smoothness with sequential vs multipoint, but that had nearly double the hp/liter this engine will...so I'm just not sure. I would probably be willing to yield on sequential for this engine as not all that important. Im not willing to yield on the 12x12 maps being useless for this engine though. I went over it last night again and I come up atleast 4 points shy of being able to get a good fit the expected shape of the torque curve and even 16 would leave nothing for an unexpected shape variation. Im leaning toward twin distributors simple because I prefer to see distributors. That simplifies the computing needs and would let a $1400 haltech E11 handle things pretty well. It has 14 outputs, so 12 for sequential injection plus 2 for the distributors I need to double check that the software will do that, but I think it will. A motec m800 will do the same thing for more money, and for even more money could also do coil on plug is I change my mind and go that way. So much to think about ..
My focus on low end is because I've learned over the years that cars with more torque are just better drivers around town. The supercharged 308 engine was an absolute joy to drive and it made something like 250 ft-lb at idle and peaked at 420. With those kinds of numbers in a 3000 lb car you shift when you please. You want to just leave it in 3rd all the time including redlights? Go ahead, it drives fine. If you feel like a hearing it a bit, downshift and let it spin, that's fine too. If you have the torque you have a choice I guess is the best way to putit. If you don't have the torque, you don't ahve a choice, you have to shift. What engine and heads are you talking about Jack? A 400/412? With the stock heads and a little porting they pull to about 6000 rpm with stock displacement, so at 4.0, I guess it would pull to about 7300 and the optimum shift point would probably be around 7800. For any more rpm you'll need more air....maybe you can get it with the stock heads, I'm not sure. revs do sound good though and even at 4.0 it will still have 30$ more torque than the 3.0 you pull out.
Yes, I'm refering to the 400/412 block and the TR heads as you are configuring now. I recon that the mods to the head for airflow may be less than for the type of engine you are building, as your displacement would be 20% less. That would, with thr right cams, create some room for more revs. At 4 litres you would also get shorter rods and theoretically a lighter crank, so you gain there. Where is the RPM limit of the heads from a mechanical aspect? What limits it? As far as reliability and longevity are concerned, it shouldn't be a problem to get to 9.000rpm safely (even Honda can do it), and you'd be flowing as much air as with the 6ltr @ 6.000rpm, at least in theory. Nice not having to worry about the transmission The motivation for effortless in-town driving is one thing, I'm just thinking about the all important fun factor! Happy motoring, Jack V. .
I can't believe any one reading this thread would still be contemplating building a copy of this engine...I sure wouldn't be I think if you start with an early 365 block the studs will work out much better and you’ll have a 71mm crank vs the 78mm and save a lot of money on the de-stroking. Maybe. On the trans, I was putting 420 ft-bl to it with the supercharged engine, right about where this engine will be. And the norwood guys build a couple 800hp turbo cars that have to be pushing 500 ft-lb that seem to live. I'm told the shafts and gear sets look just like thr TR gear in dimensions. edit - the shorter stroke will mean longer connecting rods becasue you can't low the deck height on the block and the piston has to get to the top.
Didn't consider the deck-height to be a problem, my bad.. A 365 block would be quite a bit more expensive to start with, no doubt. Oh well, it was just an idea Happy motoring! Jack V.
You might be right. I'm going to end up with about 30 hours into moving the studs I think. For me that cost about $20 in argon, $10 in electric, another $10 for welding rod and 3 or 4 scoldings from Lana. At the local welding shop it would cost $80 per hour I think. I'm thinking I see about another 30-50 hours in the head install portion of the job. Then the actual engine install begins and I'm thinking there's 100-150 hours there. Oh then then make headers/exhaust for another 40-50 hour. This is way less an easy swap than I had hoped it would be....I can hear Charles laughing all the way from atlanta
Charles isn't the only one laughing, I busted a gut when I saw those ground out stud locations. Wanted to post my dig about hacking a v12, but can't find it.
It was in a PM you sent me a few weeks ago I believe I was busting your chops about the second QV at the time
Mark, Its been over 24 hours since the last post. I'm getting the withdrawal shakes from lack of info. Plus, I needed to set this to instant email notification. Gene P.S. You rock man....
LOL The latest breaking news then! Yesterday I got a new 5lb box of welding rod and some new gas cups for the tig torch in the mail. I got the argon today and also a new O2 tank incase I want to warm things up faster. Now I just need to fit the leak in the tig torch and Im ready to get back to welding Saturday morning. On a more interesting note Ive exchanged a couple emails with the guy from VEMS about what they can do for a control system for me. The unit is actually more capable than my read of their somewhat lacking web site lead me to believe. They are now capable of 16x14 fuel and ignition map tables and I think that would be acceptable. The 12x12 tables were my biggest concern really. Its funny that the computer guys think no engine needs more than 8x8 and they only went to even 12x12 as marketing gimmick .but engine guys (me) wouldnt even consider an 8x8 set-up because its absolutely useless. It doesnt change the peak hp any, but boy does the drivability S*CK when you dont have enough points. I think 16 rpm point is the minimum as long as they are user defined rpms points and not just evenly distributed (waiting for an answer). This unit also has up to 20 high power drivers, not 12 like I thought. For you V8 guys, that is enough for sequential injection and COP ignition. It also appears (Im confirming) that the coil charging control is built in to the ECU, no external ignition boxes to buy and install which can save you several hundred dollars. For my 12, I can run sequential injection and then have my choice of distributor or a waste-spark ignition of some kind. They suggested 2 3x2 bosch coil packs. Im waiting for an answer if I could connect 2 of the cute motorcycle COP coils to 1 driver and get rid of the ignition wires. The unit will also run 2 wide-band O2 sensors among other things so its not a bad unit and its like $700 with the various options. The real problem I see left is there documentation is just plain bad, they need to spend some time on that. They guy is offering to spec everything and then coming down and help install it and get it working though so I guess in that case it doesnt matter about the documentation. Ive got a few more questions to resolve, but it might be an offer that is to good to pass up.
Mark, If you get time can you explain the 12x12 vs XXxXX? My uneducated guess is that this is the number of "modes" that the computer has, i.e. if these sensors are doing XYZ then change spark and fuel ABC. Thanks! Gene
The fuel and ignition maps are 3 dimensional. There is a 2D table that is rpm v load that is then filled with value for fuel injector pulse time on the fuel map and ignition advance on the ignition map. A 12x12 table is simply 12 rpm points on one axis and 12 load points on the other. I think the minimum points requiredfor an 8000 rpm engine looks something like: 0 750 1000 (idle speed) 1250 1500 1750 2000 2500 3000 3500 4000 4500 5000 6000 7000 8000 That gives good control of idle with points just above and just below, a good smooth ignition map, and a good fit to changes on the torque curve (which indicate a change in air flow and therefore fuel requirements). Less than 16 rpm points and I think you start making compromises to the driveability, and thats no good.
hi Mark, just a simplistic metallurgy observation, but grinding and filling and then tapping new holes for the stud positions must surely change the metal characterisitcs of the block in those locations that have been worked on, no ? Will you have to do any baking to normalise the block after all this reconstruction ? I take it you will have the whole thing crack tested but what about grain structure or is this not a consideration ?
I wondered how I'd gone this long without anyone asking this. I was really hoping not to have to weld because the metal does change but I couldn't see any way around it. The block is heat treated and when the metal gets up to around 300 degree, you start to pull the hardness out. The weld zones will be dead soft. My book's at work, but off the top of my head it loses about 40% of it's strength I think, which sounds pretty bad, but it's really not. Aluminum is a pretty flexly metal. Steel has a modulus of elasticity of 30Mpsi but aluminum is only 10Mpsi. Something like an engine block needs to be rigid so you end up adding metal or making the making the cross section bigger until the appropriate stiffness is achieved. All the things you do to make the part stiffer also make it stronger, generally much much stronger than required. The other thing working in my favor is that the casting process requires some minimum wall thickness and the walls have to contain the cooling water so they have to go all the way around, again adding a lot more metal than actually required. I had thought about re-heat treating the whole thing but was wisely cautioned against it. The reason being is that would cause every machined surface to move and require re-machining but there isnt any extra metal, so every machined surface would need to be welded up that would be a mess. There is some risk, but Im pretty sure it will be just fine.
I have been wondering about the relative strength of the weld-fill areas (particularly their ability to hold threads under some significant tension). I figured you didn't need some observer second guessing what you are already committed to testing first-hand. I've got my fingers crossed for ya. I can't wait to see how it all turns out.
