You don't have to use your fist, I believe you! But really what I wanted to say is greetings, I met you and your son at Spring Mountain in August 2011.

Best,
Tom

 

Fair enough as far as the volume. But considering that the motor supposedly won't change for the forced induction model, I'm sure the injectors should handle a very good amount of power.

I wasn't saying that it'd hit 1000HP on E85. I was stating that if the injectors are good to 1000HP on 91, they should be good to switch over to E85 in totally stock for with no problem at all.

As for the measurable forces comment, when I read that I was thinking the same thing. You can't get energy from nowhere. The pump pressurizes fuel to a high pressure. That requires a lot of energy. If there were hardly measurable forces, you could run that pump and pressurize liquid to churn a turbine that drives the pump. Not only would you have perpetual motion, but you'd be gaining energy from energy sources that don't exist. Obviously that's a load of it.

What he may have meant is that they somehow did something along the same lines as what they did with the AFM; creating air springs and re-using as much energy as possible.

HAHAHAHAHAHAHAHA!!!!

You know nothing.

Does this take into considering that the injector can't stay open as long as a standard injector?

 

I've been building various forms of racing engines since I was a pup..nearly fifty years and I have an intimate knowledge of the small block Chevrolet both the original and current formats. There is no question that the normally aspirated push rod engine has been a successful format. My concern is a cam driven fuel pump that is buried in the lifter valley developing more than 2000 psi. Any fluid that is put under high pressure develops heat..simple thermodynamics....the pressure has to be developed through mechanical action with the CAM...cam lobe degradation has to be a concern...particularly over time...variable valve timing...acting on both the intake and exhaust valves is particularly inefficient...with a DOHC format...the availability to make cam adjustments independently on both the intake and exhaust valves provides more versatility...and better control of the cylinder pressures. I can go on but as you point out....who am I to comment. I am sure the new engine will be a marvel...at first....the truth will be known when the engineering marvel is 10 years old and have 100,000 miles of abuse...on the street...not in a lab environment.

 

So the Ford Powerstroke diesel has a cam driven pressure pump, they regularly run 400,000 miles without a rebuild (the engine not the pump). Tell me why you don't think GM has this figured out. Why do you think you are smarter than they are? Do you know the alloy and hardening process they use on the cam in the LT1 that hasn't been released?

GM has found they get 92% of the benefit of independent exhaust and intake timimg with their system. What have your engineering studies found? If you have a study please publish them.

On the supercharged 4.5L DOHC engine being lighter and less complex, please point to an engine in a production car that demonstrates that.

 

In theory you are correct as any pressure developes heat, but in the diesels we see similar go 500k miles. Also, the GM LLT & LFX DI's are driven from a 3 lobe on the rear of the drivers side exhaust cam and we have seen zero failures or noticable wear on the lobe.

Some of the DI's are 4000 PSI, but the GM ones we work with are around 2000-2300.

 

Following the Bernoulli equation for pressure vs flow, we see that this 50% increase in pressure yields a smaller flow increase than you think. It takes the injector from 125.7#/hr to a maximum of 153.95#/hr. Considering that these can not, I repeat CAN NOT run anywhere near 100% duty cycle in the cylinders, we are left with only about 1/4 of this flow rate as a comparison to typical port fuel injector sizing: approximately 38.5#/hr. Someone please show me how to make 1000hp with only eight of these.

More flawed math here. Your ASSumption of a BSFC of 0.45#/hp-hr is a bit "ambitious" shall we say. Most 1000hp engines I have worked with came in closer to about 0.6 #/hp-hr. Redoing your math now requires something closer to just under 370lph, which is well outside the capability of the OE pump here.

See below...

No, he didn't.

Surely I'm not the only engineer reviewing these numbers, right?

 

Your math is correct Greg by what I can determine.

 

Hey Greg,
Not an engineer but when I first tripped over this thread, I was thought "No Way". Don't have to be Cal engineer to get that.
I just know after playing with lots of injectors 45 60 80 95 120 160 210
that there is little reason to have OEM injector oversized by 2x.

 

Exactly how many engineers do you think are on here?

 

How do you arrive at your "¼" number? A 4-stroke engine has a cycle of 720° of crankshaft rotation. Do you know when the DI starts? Even if injection doesn't start until the intake lobe is at .050" lift (let's say 20° ATDC), there are 200° the injector can flow. But the injector can continue to flow after the valve has closed and even after combustion has been initiated, possibly as much as 20° ATDC. This gives flow for 360° of crankshaft rotation which is ½ of the Otto cycle. Even using your 153.95 lb/hr flow rate, that gives ~77 lb/hr...with 8 injectors, that's 8*77*3.785/6≈388 L/hr. The question you should have asked yourself is why would GM size the fuel pump way over what you think the capacity of the fuel injectors...maybe then you wouldn't make bad ASSumptions.


My math is perfect, what you're disputing is my BSFC number. My .45 lb/HP-hr may even be high...check this link:

http://www.land-and-sea.com/dyno-tec...using_bsfc.htm

For a high compression 4-stroke with fuel injection, it lists a BSFC of .42 lb/HP-hr. Now you're really making an ASSumption of yourself.

If you'll review my post, it's very clear I was talking about the HP capacity of the fuel pump only and the calculation is completely independent of the fuel injectors...comprehension is everything.

No, and you certainly aren't the only engineer on this forum. I have a mechanical engineering degree from Georgia Tech with a minor in internal combustion engines. My senior year I was the teaching assistant to the ICE professor and became quite familiar with the extensive engine research facilities which also had a CFR engine...I think I know my way around these calculations and can make some reasonable assumptions.

However, a quick search of your work turned up this fine little gem:

http://calibratedsuccess.com/AboutUs.htm

Which contains this work of art:

Let me clue you in, all engine/chassis dynos are "load bearing"...they all present a load to the engine. With chassis dynos, there are three types: 1) pure inertia, 2) pure absorption, and 3) a combination inertia/absorption. A pure inertia dyno will spin up relative to the HP of the engine. A pure absorption and combination absorbtion/inertia dyno can control the rate of RPM change within the capacity of the absorption medium (eddy current or hydraulic). Most of all, an inertia only dyno does not "freely accelerate".

Seriously, if you wrote that drivel, are we expected to believe you're an engineer? Did you really not know all dynos present a load to the engine? I mean this is very basic stuff for an engineer.

People who know a lot more about this engine than you or me have said the fuel system will support 1000 HP. I'll freely admit I'm not up to speed on DI and qualified my calculation by pointing out my assumptions. At this point, this is just a discussion throwing out "possible" scenarios...if you have absolute facts, please share them with us. If not, please humor those of us who want to make assumptions.

 

My ASSumptions come directly from having been the lead calibration engineer on a turbocharged V8 DI engine program back in 2007 where I lead the dynamometer and in-vehicle calibration activities on a similar engine with similar limitations. Dare I ask how many programs like this you have worked on?

It was our experience that we could really only reliably inject fuel during the intake stroke, as it must have time to mix with fresh incoming air charge, evaporate, and homogenize prior to ignition if you want clean combustion. Too early means injection while the exhaust valve is open, leading to EXTERNAL combustion. Too late means inadequate evaporation and locally rich combustion within the chamber with high CO and HC emissions and poor torque. Putting these limitations into practical play yields an effective injection window that really is approximately 1/4 of the previous (port fuel) window.

Let's take a closer look, shall we? 0.42 lb/hp-hr is their estimate for "4 stroke, high compression, closed loop". While this might be ok for the typical small block at stock power levels, it is NOT typical for 1000hp engines. In order to make 1000hp, I would suspect that some form of supercharging would be employed. That is, of course, unless you're ASSuming that we're more likely to build 1000hp naturally aspirated engines within the typical enthusiast's budget. I calculated my fuel needs based on how we typically make 1000hp in this industry. I stand by my math.

My intent in that article was to explain the difference between static and transient test conditions. (Did they teach that at Georgia Tech?) What type of load that may be available determines whether static high power testing is even an option. Simply put, an inertial load can NOT by definition support static testing at high engine speed and load. Further, transient sweep rates will also be limited by the rotational moment of inertia of the dynamometer, rather than by the combination of RMOI and load cell retarder forces (which are often finitely controllable via computer) which may be added to properly simulate real world vehicle conditions. Do you really want to parse words about whether an engine on an inertial dynamometer is "free to accelerate when power is applied" versus "freely accelerates" when attempting to explain the difference to a lay person? If you want to get the college level explanation, feel free to come to one of my live classes.

Do you really want to have the nerd fight over this? If you read so much of my website, how did you miss the facts that I graduated from GMI with a BSME (auto specialty) and have worked in the industry as an OEM calibrator for quite some time? I do this for a living, and even did it at GM for a while.

 

Does octane still play a role with DI engines? I suspect that we could use 87 or 91 with no difference in the new LT1. Compression can be high because it works more like a diesel engine than previous gas engines. (Diesel with a spark plug, for the imagination)

 

Like any new technology people try to make it fix into what they know. My wife drives a 2007 GTI with direct injection which has been trouble free for 60,000 miles. The direct injection allows the fuel spray pattern to be even. Thus there are no hot spots or conditions that would cause pre ignition. High test gasolines burn slower for a reason. I worked for Sunoco in the race fuel department for many years. If there is one thing I understand it is gasoline there are many misconceptions about gasoline and one is high test gasoline has more power or BTU's. in truth they do not but they do, by flame speed etc, allow an engine to be tuned for more power through compression etc. way to much to go into on a thread.
Bill

 

the deposits occuring in DI engines are kinda scary (based on the pics)... not making me a big fan of this....how does the diesel guys handle this?

 

I always liked to pump Ethyl. She was a real party girl. A slow burner.

 

lol....Joe

 

JoeC5... Post of the week!

 

I guess when GM cuts us a car early we can figure it out...

Great pissing match for sure!

HT

 

What a fantastic thread. Thanks to those who contributed.

 

Greg, cool that you weighed in. We've been doing work with piston dome designs on the LNF and 2.3 disi mazdas that intercept the spray pattern for an increased dwelltime when the piston is around tdc. This goal being a longer pulsewidth before and after tdc and give the tuner a improved mathematical window to inject fuel . The burn efficiency may suffer a bit because of the increased dome rise and bowl design, but we're able to fuel the flame a little longer for extra horses before hosing down the cylinder wall.

I've also heard that over years, Mitsubishi, Mazda, Chevy Ford, and others have all tried slighty different injection schemes and your comment on the intake stroke is interesting. Because cylinder pressure is quite a bit lower than fuel pressure, I believe some have done a secondary injecting event btdc at the end of the compression stroke and a small portion atdc into the powerstroke before combustion pressure overcomes fuel pressure. I don't know if this does any damage to the injector itself, but considering what diesel injectors go through....probably not. I've also thought cooling on the piston crown from the fuel should be beneficial mechanically.

I've also seen some people talking about modding the pumps. We have some customers blueprinting the factory mechanical pumps and apparently extending the window. Because of the sheer volume of programmiing by the o.e. under different conditions, I'm wondering if there are any "constants" that can be changed in the GM software that rewrites all the pages at once rather than making one page right and 200 wrong. Testing a revised aftermarket fuel pump flow curve as accurately as the o.e. isn't an easy feat.

Any thoughts on these things?



It was our experience that we could really only reliably inject fuel during the intake stroke, as it must have time to mix with fresh incoming air charge, evaporate, and homogenize prior to ignition if you want clean combustion. Too early means injection while the exhaust valve is open, leading to EXTERNAL combustion. Too late means inadequate evaporation and locally rich combustion within the chamber with high CO and HC emissions and poor torque. Putting these limitations into practical play yields an effective injection window that really is approximately 1/4 of the previous (port fuel) window.