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.
Turbo/Supercharged LS straightedge torque curve build?
Hi everyone, I'm currently in the research phase for a project of mine in the next few years (once I complete my mechanical engineering degree). I don't know as much about actually building or tuning EFI engines than I do about suspension and chassis setup except some theoretical aspects, so I'm asking for some insight
The build is for a highly modified BMW E34 wagon, with 100% custom suspension all around. I'm currently thinking up a full tubular chassis/hidden roll cage with double A-arm suspension up front and De Dion axle in the back, which is essentially a hybrid between live axle and IRS. It allows for a very light setup with low unsprung weight due to the differential being mounted to the chassis, while keeping both wheels connected with a lightweight tube separate from the axles. A properly setup De Dion tube would have no wheelhop, like a solid axle, while the lower weight should provide better ride quality and handling. But mostly I like the engineering challenge Oh yeah, I'll be making carbon fiber panels, too as I'm trying to get the weight down from 3500lbs to >3000lbs while keeping amenities and some sound reduction.
But enough of that or I'll never stop typing, lol. My question to you guys is: how do modern car manufacturers achieve such flat torque curves? Specifically some turbo engines by Audi and Mercedes-Benz, which literally look like the dyno graph was drawn with a straightedge.
Example:
Could an LS engine ever have a torque curve like that? Maybe a boost controller or some sort of extreme tuning?
Right now I'm making up a plan for the engine's shortblock, and my goal is OEM-like drivability, reliability and the best fuel economy I can muster. The transmission will be a 2.66/0.50 T56/TR6060 or maybe even a TR6070 if a front-engine variant is released in the next couple years, with 3.42-3.55 gears out back (shorter if the TR6070 becomes an option, obviously). I'm looking for around 500RWHP/RWTQ with some room to grow, while maintaining a flat-*** torque curve, so the engine will be overbuilt like crazy. I want to hit peak torque as close to 2000RPM as possible or lower, and hold it flat as long as possible with a redline of 6500RPM (maybe a soft rev limiter after 6250?). I'm thinking an LS2 (4.000" bore) or LS3 (4.065" bore) block with a 3.85" stroke, 6.200" rods and 1.115" CH JE FSR pistons for a displacement of 6.34L or 6.55L, respectively. With those numbers the pistons will be flush with the deck, so the headgasket and head selection shouldn't be too difficult when the time comes, although I'll admit I know nothing on head/cam selection for a build like this, so I'll probably go custom cam and have the cam guy select the proper heads.
I'll definitely take advantage of forced induction to retain drivability and fuel economy off-boost, but I'm wondering whether a remote mounted and intercooled roots blower (I currently have a spare M122 from a GT500, although packaging will be hard) or small twin-turbo setup would be better.
So what are your thoughts? Should I go with a smaller displacement? Turbo/blower? Am I insane?!?! () This post got a little long, congrats to anyone who actually makes it here!
Another way to get 6.30L would be an LS3 block with a 3.70" stroke, 6.100" or 6.098" rods and a 1.304" CH piston, which would leave the pistons 0.012"-0.014" out of the block. However, I'm wondering whether or not the extra piston weight would make a noticeable difference when revving, and whether strength under boost would be noticeably better.
The scale on that dyno curve is making it look much flatter than it actually is. Here's a real dyno pull from a c63 amg black and it's not really that far off my 418 LS3 torque curve (which I seem to have lost my copy of, DOH!). From what I've seen LS strokers have much flatter torque curves because they build so much more low end than the smaller displacement motors do.
The scale on that dyno curve is making it look much flatter than it actually is. Here's a real dyno pull from a c63 amg black and it's not really that far off my 418 LS3 torque curve (which I seem to have lost my copy of, DOH!). From what I've seen LS strokers have much flatter torque curves because they build so much more low end than the smaller displacement motors do.
Thanks for the graph! Wow, the C63 drivetrain has a lot of loss it seems (or maybe Mercedes-Benz overrates the engine), the graph I uploaded shows 800Nm which is 590 pound feet! How is your fuel economy with the 418?
I guess this still begs the question of whether or not it's possible to adjust output of the engine to achieve a flat torque curve à la electric motor! More research for me, lol.
EDIT: I just realized the graph you sent is from a different engine. The one I showed is from the new CLS-class (M157), I did find a graph for it that shows basically the same thing as yours, but this shows actual output.
Last year with the 4l60 and 3.73 gear I was getting 21-22mpg on the highway. This year I switched to a 4l80 with a 3.08 gear so it should be slightly better with any luck. Not bad for 630HP/575ft-lb at the crank and the ability to cruise at 1800-2000 RPM.
Thanks for the graph! Wow, the C63 drivetrain has a lot of loss it seems (or maybe Mercedes-Benz overrates the engine), the graph I uploaded shows 800Nm which is 590 pound feet!
The DynoDynamics is showing TRACTIVE EFFORT at the rollers, this not the same as Torque measured by DynoJet.
FWIW, the C63 is a NA 6.2L engine, and the CLS63 whose flat dyno graph you covet is a twin turbo 5.5L. Also that's not a real dyno graph.
If you want torque at 2000 rpm that doesn't go away, get a twin screw or TVS blower but watch what you wish for. Monstrous torque off idle looks sick on a graph but can be difficult to control in real life.
Here is a graph from a roots blown 408 LS, This is more of a prostreet/strip style cam so a different cam could flatten the torque curve even more. An engine utilizing variable valve timing could make that line nearly flat.
If you want a flat torque curve then either a positive displacement supercharger or small *** turbos. That will the be only way really to build an engine that starts peaking torque values at 2000rpm and holds them to redline. A naturally aspirated engine will have a very very hard time doing that, especially when it's trying to accomplish some of the other things you mentioned (reliability, fuel efficiency, driveability etc.)
If you do go ahead with the non forced induction engine, then your best bet is to get a setup with a long stroke. The longer stroke builds much more torque off idle than a shorter one, and with the proper top end can carry that power up top (relatively). Something in the 416-427 range with a 4" stroke and small high velocity heads will get you the type of power and reliability you desire.
LSA block/crank (CTS-v), good rods/pistons, stock LSA heads/LSA supercharger, stock injectors can meet your goals and then some. 550rwhp/550rwtq is a cakewalk on 91 octane.
If you need more power simply swap pulleys for more boost/retune/larger injectors.
As a stock CTS-V is 4300lbs and the manual version has 3.73's.... your car could get away with less gearing out back and get better fuel economy as a result. Less weight also equates to better economy.
That top graph you posted looks like the marketing team got a hold of it, unless they are supplementing the gas engine with electric controls but that would be very fancy since the electric would have to constantly increase and decrease power to maintain a fixed total output as the gas engine revs.
You dont want torque to peak that early, especially in that light of a car. It is very difficult to drive since you will basically spin tapping the pedal. A roots/screw blower would be best suited for it, but will run out of breath up top. Turbo(s) would be more ideal for higher revving, but any designed to peak early will also run out of breath up top.
What you are looking at in the first post is a function of quick spooling small turbos (hence the low RPM peak) and electronic boost control, throttle management, and torque management need to create drivetrain longevity.
You've already been told how to achieve your goals - I would argue that a variation of twin charging (smaller PD blower and bigger twin turbos) would be hilarious.
Thanks for the responses guys, I'm definitely going to look at fitting my M122 remotely, but we'll see how that pans out in the E34 engine bay. I'm thinking I could run a blow-off valve with a boost controller to limit power in first gear. Maybe a centrifugal blower would be better? The progressive boost could help with traction.
A centri is not going to deliver the type of torque curve you want.
It will be more than enough for that car, but it will not be flat as a pancake from 2000 to 6000rpm. It builds boost progressively with engine speed, so you get a peakier curve compared to a roots blower.
A centri is not going to deliver the type of torque curve you want.
It will be more than enough for that car, but it will not be flat as a pancake from 2000 to 6000rpm. It builds boost progressively with engine speed, so you get a peakier curve compared to a roots blower.
True but you can overdrive the centrifugal using a smaller pulley, and use a wastegate to bleed off pressure. This will bring the power in sooner.
True but you can overdrive the centrifugal using a smaller pulley, and use a wastegate to bleed off pressure. This will bring the power in sooner.
This is actually what I've been reading up on the past couple of days! There's also the Procharger i-1 which is interesting but it needs an OBDII connection and is only sold in complete kits for late-model cars. If I found a used one for sale by itself I'd be tempted to try and rig something up with megasquirt to manage it and the engine.
Build a 4 inch stroke ls2 or ls3, 10.5:1 compression, use ported 243 heads and strap on a set of gt2871rs's or for quicker response gtx2860r's. Fast 92 or 102LSXRT and keep the piping short for throttle response. If you use E85 or spray meth you can get a rude flat torque curve with an 800+hp top end. Twin gtx28r's will spool almost instantly under load with a fairly high compression 400+ci motor. None of the turbos I mentioned is cheap but you do get what you pay for.
I worked on and drove a C5 with a low compression 427 LS that had ported 241's, an LS6 intake and tb and twin gt2871rs's. Power was a little over 740rwhp but it felt much stronger. It had an extremely violent transition into boost around 2400rpm in 1st and 2nd gear, 4th gear hit just over 2k rpm. Even though it was only 740rwhp or so it felt stronger than cars with 4 digit power. It was well beyond the limits of sticky street tires and would violently spin them in any of the lower 4 gears as soon as you rolled into the throttle. A 400+ci engine with decent heads and 2 small, high quality turbos will do what you want easily.
This is my n/a 6.0 with a healthy cam 234/250 .621/.629 and ls3 heads thru a th400. The torque would have been more flat but i tapered the timing off where i figured it made peak torque and I didn't do it right. I think with a wastegated centri it would hold the same curve just woth added power. The operator shut the dyno down early due to a tire vibration. My truck was pig rich when we dynod it. Last year i shifted at 7200 at the track then bumped it up to 7800 to keep from shiftong before the traps but it didn't change the mph just my et dropped some.