I'll try to post, but I'm afraid it will be a very busy post...

Numbers at peak TQ - 417.5:
...RPM...4890
...Fuel Pressure...58 (best I can tell, did tests with mechanical gauge, seems to hold)
...Injector Size....45.03 @ 58 psi, 39 @ 43.5 psi
...Duty Cycle.......51.2%
...Air Mass..........297.3 g/sec
...BSFC..............0.474

Numbers at peak HP - spike at 474.3@6320. Real number looks like 474 @6700, so that's what I use.
...RPM...6700
...Air Mass...........382.6 g/sec
...Duty Cycle........65.8%
...BSFC................0.500 (oddly perfect, but that is the number)

Was this what you were looking to see?

 

I found GM's friction number table. It reduced friction to a reduction in TQ output by RPM from 400 to 8000. I ran the data through a statistical least squares polynomial regression, and got:

0.00000182 x RPM^2 - 0.0011 x RPM + 35.62, R-squared of 0.9978.

No idea where GM got these numbers. My assumption is that they are only valid for a 5.7L. It does appear GM considers friction a quadratic function. The computer uses this number to calculated TQ output for its traction control and transmission shift softening modes, based on the HP Tuners description.

Again, I can't validate how GM got the numbers or that they are correct, but It does stand to reason a ton of research went into them

 

Yes, thanks. Pulse width, if you have it handy.

 

You can get pulsewidth from duty cycle and rpm.

 

I know.

I've been busy with work & I figured he had it in the log.

 

I do but I'm traveling for work. I have my spreadsheet and notes but not HPT

 

Ok, I understand. You can PM me if you would like. I think we could have a slightly different conversation that way.

& I don't mind figuring it out. I do always think it's best to hear it from the horses mouth, rather than mine.

 

They might not be related to what you think they are (parasitic loss or friction loss for turning the engine). In fact I am almost certain they are not if the function is parabolic as you suggest, since I can find no reason for friction to be either higher (or lower) at both ends of RPM, and I am surprised you didn't mention this when you found out it was quadratic. Instead I will guess this is a function derived through tests, that software has worked out backwards by eliminating (or discovering) a constant which leads to repeatable/consistent performance benefits when applied this way to some device on the vehicle. The naming of it (friction coefficient or friction table) might refer to something besides the engine for example, like the internal activation of parts within the transmission that have varying "grab-ability" (or friction coefficients) based on the RPM at which they are applied.

 

The only friction variance I could see being related to torque/ HP would be dealing with oil viscosity/ lubricity.

Otherwise, I would bet you are right. & I'm not a betting kind of guy.

Didn't someone mention traction control?

 

Whatever tricks are used to get racecar engines to spin to 10k... all the coated bearings, lose tolerances, perfect oil viscosity, crank scraper, dry sump, vacuum pump, all of it... And did it to the OP's engine with his same cam and intake and tune, would it still stop making power when it did?

 

Certainly there are losses.

I know of a bearing place that designed a very high-efficiency hydrostatic bearing. His number indicate there is power to be had, up to 10% in some types of bearings.

IIRC, on a 700HP engine he had done his bearings on netted about 30HP.

It's still related to oil viscosity/ lubricity unless you have metal on metal surfaces.

 

I don't think it would. I think it would move the peak to the right. I'm also finding that in general, adding air moves the peak to the right. If I take the model and just go +10 cents at all rpms, the peak moves to the right. Even with no other changes. I'm boarding an airplane now, but I'll try to post a few modeled examples later in the week

 

Actually, I think they just named the table friction to call it something. As I said, I don't know what the numbers are for other than the computer calculating TQ output for traction control. But GM got them somehow. Trying to see if they line up with anything.

 

They wanted to call it the flux capacitor table, but that was already taken.

Anyways, back to the friction scenario, I too think that the powerband would shift to the right, with nothing changed but the bearings/tolerances/oil viscosity/whatever racecar friction reduction tricks. It would be very neat to see two engines, identical in every way other than the racecar tricks to reduce friction, but same rotating assembly, heads, cam, intake, fuel, tune, etc tested on the same engine dyno during the same day.

My personal belief is that torque would still peak in the same place/time between both engines, but the horsepower would peak later in the engine with less friction.

It would be very interesting to see if the GM friction table values were different for an engine with the corvette coated bearings. That would allow us to hopefully see if internal engine friction is part of their algorithm at all.

There's just SO much that can effect friction, and I am trying to figure out how we can quantify any of them. Is it even possible to come up with a friction coefficient for our equation? Even if we had access to all the data we wanted, would we be able to calculate 10w-30 is .00047 and 5w-40 is .00039, or whatever numbers they may be? Subtract .002 from the friction multiplier if a vacuum pump is present.

If we ran the two almost identical engines, and ran the two test results through a solver program, we could probably find a correlation, and maybe derive a friction coefficient for that exact combination.

But again, there's so many variables, that any empirical data collected will be specific to that exact combination. You could run the same tests again on a new set of almost identical engines, and solve their results against each other to find a correlation, then solve that correlation against the correlation of the original tests, etc etc, until you have tested every GM factory LSx offering, and solved all the test results into one coefficient for friction... And even then, it would be specific to gm factory LSx motors that have these exact racecar friction reduction tricks, you couldn't even use it as a coefficient of friction for modified motors, or even a gm factory LSx motor with different friction reduction tricks done to it.

Maybe I'm just over thinking this. I should probably drink less coffee...

 

There are certainly coefficients of drag to be calculated.

The main factors being utilized static volume of the column of lubricant, force, lubricity, effective surface area of the bearing & the surface texture of the bearing. Most important is effective surface & texture of the bearing.

You can use effective bearing surface to influence effective static column volume.

That is how I had it explained to me. That is what he worked out on his bearings, & I believe him.

 

Run a junkyard motor on the engine dyno, "test 1", then disassemble it, rough up the bearings with some sandpaper, and run it again, "test 2"... See if test 1 peaks later than test 2.

 

Over polished surface would have a detrimental effect as well.

 

What would the values of the friction coefficient be in as far as units are concerned?

If we did have an overall friction coefficient, or more specifically a bearing coefficient, oil viscosity coefficient, etc, which value would it influence in our equation? Would the friction coefficient be in terms of ft/lb of torque required to overcome this friction? Would it be in terms of heat generated from the friction? Where does such a coefficient factor into our equation?

 

the units in the table are TQ. Either NM or LbFt depending on your unit selection. GM is basically telling the computer "factor this much TQ loss in your calcs".

Friction is a form of negative work, so ultimately the friction units are KW - or horsepower - depending on your units.

 

What are you going to do if you ever figure it out? What changes will it make to your mind? If you understand that a looser engine robs less power as it spins higher, and tighter engines suffer oiling issues if they are too tight at too high of an RPM, you find the middle ground for your application and be done with it. You are not going to want to be tearing engines down and modifying bearings to run back to back tests on which bearing offers 7 extra horsepower because even if it does, only time will tell if it lasts longer which I think is more important than anything if the engine is to have any serious mileage on it.

Thats what I am talking about, oil will vary in a running engine is a great example of why this is a dead end. Its like you said " I want to know how much energy is in a lightning bolt " and I said "well, we need the right equipment to measure that" and you told me that you dont have any equipment, you dont even have a lightning bolt to study- all you have is excel. guys, lightning is pretty dangerous I would just steer clear and measure things you can count with your eyes- OR find reliable resources/references which can be depended upon for their content and share it.