PDA

View Full Version : Fuel Pressure Tables



FUBAR
February 20th, 2012, 03:41 PM
Can someone please tell me what the difference in the paremters and the tables that follow below, how they differ? They look the same to me, save for the different units. So if I want to calm down my pump where it erraticly overshoots and undershoots fuel pressure when I mash on the pedal, which should I look to. Units wise. I understand the P.I.D. meanings.

Thanks,

12709

2007 5.9
February 20th, 2012, 04:42 PM
I'd look to CompD for a meaningful explanation on how PID loops work...lol...

Sorry Rich...just had to!!!

FUBAR
February 20th, 2012, 05:12 PM
Lmao!

But seriously...with the parameters having % units and psi, and the tables below the parameters using a gain factor of up to 4, I'm not sure whether to adjust the tables of gain (really makes me pause when I see minuscule numbers compared to the range of 0-4) and / or the tables in the parameters section to obtain my goals of smoothing out my modded FCA or other various intents. If I adjust my Base Duty cycle table closer to what the truck wants, it sounds like that will rely less on the P.I.D. loop system for adjustments???

DoghouseDiesel
February 20th, 2012, 10:14 PM
I'd look to CompD for a meaningful explanation on how PID loops work...lol...

Sorry Rich...just had to!!!

DON'T YOU "F'n" START HERE TOO!

http://images.cheezburger.com/completestore/2009/6/22/128901670087269133.jpg

:muahaha:

DoghouseDiesel
February 20th, 2012, 10:24 PM
Here's the thing with the P.I.D loop....(and this is good, because this is one of the things that no one likes to mess with)

All it is error correction, plain and simple.

It's a game of reacting fast enough to correct the errors, but not so arressive that you continually overshoot the intended target value.

The best example of is during cruise control; this is where you'll really feel it. If you get a little too aggressive in the error correction, you'll feel the buck as it tries to maintain a steady speed. Usually you'll feel it under light throttle.

Ideally, you want the values slower, but smoother, in the lower RPM ranges and faster and more aggressive in the upper RPM ranges. Keep it smooth for daily driving, but yet fast and aggressive enough under high RPM / high loads.

Just getting the duty cycle close, while it'll help, for the most part you won't notice. All that is is the ECM's guideline for how much it can open the FCA. Yes, you want it close, you certainly don't want 100% duty cycle, but as long as the duty cycle allows you to maintain the target rail pressure, you're not going to notice the duty cycle all the much. Now, if the duty cycle isn't high enough, you'll get a reduced power condition, low rail pressure and very likely you'll hear the popping as you cavitate the rail.

triton
February 21st, 2012, 12:27 AM
Lmao!

But seriously...with the parameters having % units and psi, and the tables below the parameters using a gain factor of up to 4, I'm not sure whether to adjust the tables of gain (really makes me pause when I see minuscule numbers compared to the range of 0-4) and / or the tables in the parameters section to obtain my goals of smoothing out my modded FCA or other various intents. If I adjust my Base Duty cycle table closer to what the truck wants, it sounds like that will rely less on the P.I.D. loop system for adjustments???

I'm pretty sure he is serious lol.... sorry Rich :-) You have nothing to hide man ! This would probably be a better place to learn more about it

FUBAR
February 21st, 2012, 02:02 AM
Ok, Cool. So to help me better understand, let's put this to use in a real world situation. For example in the below picture. This is where I have been at a cruise on flat ground going around 60mph. Then I matt the skinny pedal and the fuel pressure overshoots and undershoots initally and then smooths out till the truck builds boost and my Boost Fuel Limiter table allows full fueling. Then actual rail pressure dips down for a moment but recovers. At the time the rail is recovering, I see a trend in the command and actual dutycycle %. Just not sure if my rail is really draing or something is funky in my P.I.D. or base dutycycle causing that now.

Would the erratic saw tooth pressure trend be a situation where dutycycle is being respond to too fast, or not fast enough. And for the situation when boost builds and is fueling heavy, would this be a case where the 3D dutycycle table would play into?

Thanks,

12711

triton
February 21st, 2012, 02:14 AM
Ok, Cool. So to help me better understand, let's put this to use in a real world situation. For example in the below picture. This is where I have been at a cruise on flat ground going around 60mph. Then I matt the skinny pedal and the fuel pressure overshoots and undershoots initally and then smooths out till the truck builds boost and my Boost Fuel Limiter table allows full fueling. Then actual rail pressure dips down for a moment but recovers. At the time the rail is recovering, I see a trend in the command and actual dutycycle %. Just not sure if my rail is really draing or something is funky in my P.I.D. or base dutycycle causing that now.

Would the erratic saw tooth pressure trend be a situation where dutycycle is being respond to too fast, or not fast enough. And for the situation when boost builds and is fueling heavy, would this be a case where the 3D dutycycle table would play into?

Thanks,

12711

I'm sure one of the guys here will help you and this may not answer this direct question but, really.... if you didn't read this thread at CompD, you should... from start to finish. I think you'll find a lot of questions you have about PID's, duty cycles and others things getting answered...

http://www.competitiondiesel.com/forums/showthread.php?t=129607

FUBAR
February 21st, 2012, 03:09 AM
I did. I admit-tingly skipped thru some stuff just because of the nature of the conversation. My question to purse a better understanding is where the fuel pressure spikes and drops erratically. Is that a case where the P component is being too aggressive and the D component is reacting to fast causing it to over and undershoot by so much? That's just my thinking and am not sure. I understand what Rich is saying you don't want to have the PID loop making a butt load of changes and quick when cruising or DD'ing because that would make for one heck of a roller coaster ride! But in the instance tuning a modded FCA when it is being command to open and supplies too much pressure or volume....(please do not start that debate in this thread) any how, it seems to me the ECM is taking hay maker swings to get it back into control via how the PID is setup aggressive. Thoughts?

Thanks,

FUBAR
February 21st, 2012, 03:12 AM
Kinda sounds to me like a rule of thumb anytime we ugrade components on our trucks...usually what the stock was calling for, was too much, giving that we usually upgrade everything..BIGGER, lol. Kinda like tuning uS for bigger injectors. The same 2000uS pulse is a lot more with a 60% over injector vs stock.

2007 5.9
February 21st, 2012, 03:36 AM
I like where your heads at Andrew!!!

killerbee
February 21st, 2012, 05:53 AM
If you want a good intro lesson on process controls, and the significant of P, I and D, here is a good one.

http://en.wikipedia.org/wiki/PID_controller

Joesixpack
February 25th, 2012, 03:29 PM
Ok, Cool. So to help me better understand, let's put this to use in a real world situation. For example in the below picture. This is where I have been at a cruise on flat ground going around 60mph. Then I matt the skinny pedal and the fuel pressure overshoots and undershoots initally and then smooths out till the truck builds boost and my Boost Fuel Limiter table allows full fueling. Then actual rail pressure dips down for a moment but recovers. At the time the rail is recovering, I see a trend in the command and actual dutycycle %. Just not sure if my rail is really draing or something is funky in my P.I.D. or base dutycycle causing that now.

Would the erratic saw tooth pressure trend be a situation where dutycycle is being respond to too fast, or not fast enough. And for the situation when boost builds and is fueling heavy, would this be a case where the 3D dutycycle table would play into?

Thanks,

12711


Your example would have been great a couple days ago. IMO you have a lot to gain working on the FCA base duty cycle. The PID looks to be a little jumpy, if you could trim it at different pressure's and commanded mm3's I would trim it down some there where it jumps. However the response when your rail drops is not hard at all so likely wouldn't fool with the PID portion of the pressure regulator. The base duty cycle portion of the pressure regulator, it would be a slow pain full process when you can't do it in real time, but its worth it for the system response you can get back.

A Hint on your pump capacity. If you log FCA current, 1.8 amps is near no flow. .4 amps is all out full stroke. So if you can pull your pump down to .4 amps and you can still maintain rail your golden. If your down to .4 amps and your pressure is still dropping your looking for more flow.