This is about how PID's, which I know net delayed data, or wrong data, or both.
As mentioned in another post; https://forum.hptuners.com/showthread.php?96990
I was hitting my rev limiter and the eng was cutting fuel, but the scanner was showing I was not hitting the rev limiter, not even close. Apparently two things were happening, one is the Scanner data is slow so it doesn't always catch a quick rpm spike to the rev limiter. Second, which I recently discovered, is I had tried two different RPM PID's since I've been using HP Tuners; PID 12.56 (Engine RPM SAE) and 2135 (Engine RPM).
For most things that have more than one option, SAE seems like the one to select? Now I'm not so sure. Whatever the case, I had randomly used both 12.56 SAE and 2135 for rpm. It wasn't until recently that I noticed that my shift points were suddenly much lower, as low as 5Krpm. So I made two rpm gauges, one for 12.56 and one for 2135, and 12.56 was much slower, which you can see in the attached pix.
I had assumed they both fed from the Crank Sensor, and they probably do, but for whatever reason 12.56 had a huge delay. Plus note the PID's on the far left of the bottom pix, which typically give me delayed data compared to the gauges but here it's more accurate.
There are also these two PID's, which I've yet to try:
7223 (Engine RPM From Crank Sensor)
7175 (Engine RPM - High Resolution)
So I'll make four gauges to find the best, and I guess I'll be forced to make multiple gauges of everything else that has more than one PID. Uggg...
Since rpm is used in my horsepower math, 12.56 lowered my horsepower reading. It also upset things like my est of how much boost I have at xx rpm, which I use to est how well a tune is working. Since I switched those PID's a few times, I have no idea which boost values were on 12.56. And of course I still have no idea how delayed 2135 is.
Another is Load vs Load SAE. I don't know if these are meant to be different but Load SAE is always higher than Load. Since Load SAE seems to match my Torque to Load settings I've been assuming Load SAE is the correct one? I don't really know, but these and other PID's that seem the same, may not be.
Manifold pressure (MAP), I noticed was different as well.
I think the PID with the most delay is speed? Hard to tell since I have no reference for most things, but I do notice when I come to a stop and the Scanner never dropped below 2-3mph. Or maybe I roll through a stop at maybe 3mph and it never drops below 12. It seems it reads acceleration speed much better because those "seem" to be accurate, but hard to tell. Really I just know the speed goes up right away when I gas it, but it certainly doesn't read right when slowing or stopped.
So I have since made three speed gauges. The one normal gauge we probably all use, plus one for front axle, one for rear axle. After doing this I discovered the main speedo is an average of front/rear. Makes sense I suppose... I did this to help troubleshot the aforementioned issue with speed, and the newly discovered rpm issue making me wonder if maybe the front or rear would be better than the normal speedo. Plus I wanted to see how much wheel spin I get (RWD). This enlightened me to another problem I have; when doing a 0-60 run, the speedo would simply stop altogether ~30-40mph. When and for how long it stops varies but it would often stop for about 1 second. Two things were happening; one is it was shifting so I'm losing power with the throttle closing and cutting boost. The second is the rear tires where spinning, but during that power cut they slowed and stopped spinning.
So I was getting a false higher speed before the 1-2 shift (avg of front & rear), then when shifting the rears were slowing but the front still accelerating which netted an avg of zero acceleration. It varies but it takes about the same one second for the wheels to stop spinning as it takes to shift, so this explains my one second of zero acceleration. Watching the front axle speed I can see acceleration has slowed in that moment, and I can feel it, it's just not nearly as bad as the Scanner data shows.
Also, rear wheel spin is minimal, and with the new tires I bought 10 months ago I had assumed it was zero. As the rears wore down and started spinning more easily, my speedo delay at the 1-2 shift started to return.
Now, 10 months and 10K miles later, my tires spin easily enough that it often spins through the 1-2 shift and hooks up at the 2-3 shift.
So now I have my main speedo, plus two small ones for front/rear, plus a bar that shows rear speed minus front to give me wheel spin relative to ground speed. That last one is pretty neat to have regardless. See pix
The point of all this is your data, like mine, maybe skewed a little or a lot, and you may not know it. I guess you just have to make multiple gauges to figure out which has the least delay and/or reads closest to what you think it should. My MAP for example, I have to assume the higher reading is faster and more correct.
Other things like spark retard show up on the gauge long before they do in the PID list. Boost psi shows a psi increase before it shows the throttle closing. Etc etc.
I can set the polling time on any of these to read as fast as it can, but it doesn't actually update that fast. Say it's set to read 100x a second, but if in reality there is only data twice a second, and rpm or psi peaked and dropped back down between data points, I don't see it. Which, imo, explains why some data doesn't make sense. Eg; my MAP can be ~4 psi higher than my boost pressure, which is not possible but it happens a lot and I can only assume the MAP data is faster than the Boost data.
The Speed data, when wrong, like at a stop, reads the same for the front and rear axles. So I can only assume whatever delays that data has is affecting both equally. My guess is the ECU sits on the data for a while before spitting it out, like an after thought. Or maybe for acceleration it provides the data as it should, but during decel, when it's less important, it's programmed to give it out less often to save bandwidth? I'm guessing here, but fyi.
RPM Diff problem.jpg
Wheel spin.jpg
Reply With Quote