03-07 Calculations and Models (Solidworks, MATLAB, EXCEL) cont...

[Cummin-a-long]

This is just the break-off and continuation of the Brian beat down....

From the other thread....

For those that have never actually calculated some of this stuff, here is just a small example of how it gets done.

This is just a VERY simple calculation that gives you the TOTAL amount of time that you have at a given RPM, with a given amount of timing BTDC, and whatever you set as your latest injector cutout ATDC.

This IS NOT the actual fuel quantity you'll need, this is just giving you the actual time you have available and can help you spec your injectors so you actually get full power at the RPM's you want to turn.

Here is an example of my truck at 5200 RPM's with 35*'s of advance and a max of 20* ATDC before injector cutout.

Now, in order for me to stretch my fuel system out and drop pressure, based on what I know my pumps can flow, I'd need to be exceeding 2400 uS up in those RPM's and in order to do that I'd need to be injecting WAY beyond that 20* ATDC...40*, which is pretty useless at those piston speeds, is actually where this would occur. This would be a hot, smokey mess.

Now, just to show that this function DOES NOT tell you what your fuel quantity ACTUALLY needs to be, here is the total time you'd have at idle (800 RPM's) with 0*'s of advance and 20* of max timing ATDC. You'll see the available time is HUGE and is why you see some of the really ridiculous values that are in the OEM fuel mapping.

This would look like an old coal fired train idling in your driveway if you used this value.

I have others that I made for timing, ACTUAL duration, injection velocity, cylinder pressures, etc., but those I keep pretty close hold.

But, you get the point here...this stuff is not hard to calculate if you know what to look at.

 

[Cummin-a-long]

Sorry, that was a little hard to see.

Here is a combined blow up of that stuff that is easier to read.

When you run the calculation in the RPM's where I originally spec'd these injectors, you'll see it's almost dead on, including electrical to hydraulic delay, to the what I actually need to produce the power I wanted.

 

[Cummin-a-long]

Now, if you go back to post #377, where I talked about the exit velocity at the nozzle, piston speed, spray angle, etc....this is all crap you must know at each instance to make sure what you're spraying is actually going in the bowl and you have the air to use it as well as exactly how long you have to get it there.

With the basic fluid properties of diesel and the parameters that you are specifying in your tune, this is all stuff that can be calculated for any given instant.

Just knowing the time per degree of crank rotation is also not, in and of itself, enough to get total time as the piston speed is almost zero at TDC vs at it's max speed at +/- 90 degrees. Those positions, velocities, and to a lesser extent the accelerations of the piston and it's relative position to the injector tip, along with the velocity and angle of the spray pattern is what gives your true time available while staying in the bowl.

 

[Cummin-a-long]

Just to demonstrate that graphically, I'm doing a REALLY QUICK stick and block model in solidworks and I'll run the simulation to show the vertical piston speed, crank angle, and the relative velocities associated at each point.

I still need to stick the injector and spray pattern in there, but this is what it will look like....and yes, I'll take a real quick video of it in motion.

 

[Cummin-a-long]

Sorry, I started getting a little sexier with it than I intended.....

I wasn't really happy with the stick and block model. lol

 

[Cummin-a-long]

Just waiting on the video to load.

Here is just the quick data export with it turning 600 RPM's

The piston velocity ranges from 0 in/sec at +/- 90*'s (straight up, straight down) to approx 150 in/sec (12.5 ft/sec) when the crank is horizontal. That's just turning 600 RPM's

 

[Cummin-a-long]

Okay, here's what it looks like actually in motion. Keep in mind, this is just a quick throw together and not a real sophisticated model or analysis.

I really should have done it showing the pistons acceleration vs the cranks angular displacement and you can really see how fast it pulls away as the crank angle gets away from vertical.

You can obviously see the crank, rod, and piston. The outline at the top is the injector and the 143* spray pattern.

If you move that manually, or if I set it up to show interference, you can see when it's within the piston bowl, how deep it's in, and how long it's in there for.

[ame]http://www.youtube.com/watch?v=ojkYywdhFGg[/ame]

 

[Cummin-a-long]

Just for reference, I had it recording at 200 frames per second....those 6 seconds of analysis produced 1200 data points. That excel sheet is just where I cut it down to 2 rotations.

 

[Cummin-a-long]

I've got a little time to kill before I head over and start measuring for the bracket for the 3rd pump, so I'm gonna see if I can make that simulation a little more realistic....

 

[Cummin-a-long]

Shouldn't take TOO long to knock this out....

 

[Cummin-a-long]

It's gettin' there....

 

[Cummin-a-long]

Originally from Smokem...

Maybe this will help.

 

[Cummin-a-long]

lol.....little late in the process right now, but I'll make a more accurate one with that later.

This should get us close enough for 'ish and giggles....

 

[Cummin-a-long]

I'm not gonna go nuts on the rod or crank, I just need them there for the actual motion.

 

[Cummin-a-long]

This is all I'm gonna do for the crank.

 

[Cummin-a-long]

Almost there....just gotta get the alignments and rotations set up.

 

[Cummin-a-long]

Okay, now we've got the rotation and alignments set, I just need to add the motor and build an injector setup and we'll be good.

 

[Cummin-a-long]

Here's the initial animation without analysis or the injector spray pattern.

This is spinning at 200 RPM's with 200 frames per second resolution.

[ame]http://www.youtube.com/watch?v=1UxH-INZcaQ[/ame]

 

[Cummin-a-long]

Here's the injector nozzle and a simple 4 hole spray pattern shown. Keep in mind that pray pattern would only represent the liquid and not the vapor cloud surrounding it.

The way I set this up I made the nozzle and the center of the piston concentric so they would stay in alignment, but it allows me to move the nozzle up and down for various installed heights and also rotate the spray pattern without disturbing that alignment.

 

[Cummin-a-long]

I'm loading the last video now.

What I did was slow it down to 5 RPM's and zoomed in on that area of the nozzle and piston. I just put the nozzle and spray pattern in there as a static element.

If I wanted to set this up for a really accurate analysis, I could take the known density of diesel, the total quantity injected, the orifice size, the number of orifices and the pressure being applied and calculate the velocity the fuel is actually leaving the nozzle. That combined with the relative piston speed would let you figure out just how much timing before and after TDC you can go without being outside the bowl.