Shift strategy idea
 

Truck is finally out of the shop, and seems for now that everything is completely fixed. Whew, only took... what, 10 months?

Anyway, I've been pondering a new shift strategy to try. For a while I kept playing around with the shift points, making my graphs ramp up smoothly... and not getting the results I wanted. It wouldn't upshift when I liked, it wouldn't downshift when I was expecting. Sooooooo......

I'm doing something different. My graphs will no longer be smooth. More like stairs. Yep, flat spots, abrupt rises.... my graphs will litterly look like cartoon drawn stairs.

Ok, not that many steps, but you guys get the point.

Reason I'm doing this is that since my truck has been broken down so much, I've been driving my wife's BMW constantly. I've gotten very spoiled to how it shifts, which IMO is as perfect as any vehicle can get. However, in driving, I started paying attention to the feel of the pedal vs the vehicle speed, and how it responded to upshifts/downshifts.

After really driving it and testing it, I noticed that shift points seem to be more in graduated "steps", rather than smooth linear transitions. The pedal had specific ranges before it downshifted, and again would have ranges before it would upshift. So instead of a slight adjustment on the pedal resulting in an instant upshift or downshift... a slight adjustment of the pedal would change fueling, but without the annoying and unwanted gear change. However, these ranges were small enough that a quick downshift, when warranted, happened when I wanted, and precisely at the pedal position I was expecting.

So yeah, that's it. I'll keep you guys updated with what I find. If it works, and I get it adjusted how I like it... I'll post some screenshots.

If it doesn't work, then this thread will self-destruct in 3.....2.....1....... :yikes2:

Interesting idea.


What kind of increments were you planning on experimenting with, 2mph, 5mph, etc?

I'm going to try several different ones. I'm finishing one right now that has 3 "tiers" to it. Depending on the gear, the MPH increments between them varies. Lower gears have smaller jumps, higher gears have bigger jumps.

I may change it so that higher gears have maybe an extra "tier" or two than the lower gears, since there's a much larger MPH range to play with on the higher gears.

I also don't have the jumps following the same pedal position. I learned that lesson a while back, when my trans was jumping back and forth constantly between gears as I was holding a steady speed on a level patch of road :doh:

Anyway, I haven't even road tested anything yet. Just been sitting down with the tables playing with them. I'll probably finish up a few more tomorrow morning, and burn a couple of tunes to the chip. Then head out and run some errands and see what happens. :thumbs up yellow:

I am excited to hear how it works.

Should make the downshift points much easier to fine tune, which (to me) seems to be way more of a challenge then upshifts.

Thomas made an Excel file here (http://forum.gopowerhungry.com/minot...27-ya-hoo.html) that I'm using to help set the shift points. This is a cool little spreadsheet. :evillol: Helps me to see on paper what my RPM's should be at a given speed, then I can look at my pedal position and estimate where my throttle position will be at a given point. Of course, the last step would be to hook up AE and track pedal position to verify that I'm on target.

Got a few tunes done. Time to burn the chip and test it out.

Well, the shifting works. I still have a few tweaks to make, but it shouldn't be much at all. Works great driving uphill at over 6000 ft elevation.

Bad part is that I've had my truck back for less than 24 hours, and the power steering decided to puke on me :throwup: I'm not sure if it's the pump, or if I blew a line. Gonna have to dive in and find out, but there's power steering fluid sprayed all over the wheel well, and dripping down the axle and driver's side suspension. This truck is really pissing me off with one thing after another breaking :cursin:

Anyway, here's a quick screenshot of what I'm doing. This is the 1-2 shift, and the others look similar:

http://forum.gopowerhungry.com/membe...iftpattern.jpg

I just need to move the first jump over to the right just a bit. I don't fuel much at light pedal, and I'm having to hit the pedal a little harder, so I'm sitting at the second tier just a tad earlier in the pedal than I would like. Also I'll probably raise the first tier just a smidge, and possibly lower the second tier very slightly. Same goes for the 2-3 shift. However, the 3-4 shift is nearly dead on, and the torque converter lockup is happening almost exactly where I want it. Downshifting is also very predictable now, and happens just about where I want it. :thumbs up yellow:

For me, this just seems to make the trans shift points so much easier to manipulate, and helps me to make the shifts happen when I want it on the pedal. Might not look like much, but it helps me tremendously.... especially being just a novice tuner.

Glad it worked out!

I might have to try and make something similar when I get some spare time. I like the idea of having predictable, consistent shifting, which recently seems to be my biggest gripe.

Sorry to hear about the PS leak..........at least it will be an easy fix, whatever it is.

Buy an OBS. :cool:

Your stepped shift schedule seems like a good idea and I'm glad it seems to be working out for you (besides your recent leak). Goes to show that a smooth shift schedule doesn't always mean smooth/desireable shifting. I might have to try with a file and see how it goes for me as well on my next reburn.

I'm glad you liked the spreadsheet!

I haven't gotten in and messed with any shifting yet but I need to as my programs hold out my 1st gear a bit to long at light throttle position. When Bill was doing the tuning on my shifting the graph looked stepped just like the one you showed.

What is the measurement it's using on the bottom for throttle position? I haven't had a chance to get into a lot of the tables yet but it seems some use wird measurements.

Shifting is determined by comparing speed and throttle position. Throttle position is measured using a/d counts. Cody had a nice little layout on a/d counts vs volts, I will have to see if I can find it.

Here it is......

Thanks for the info! So what are A/D counts? What does A/D stand for?

Maybe this is a question for a different thread being it's a little off topic but at what point of APP does 100% mass fuel desired come in at? And is there a reason other than the throttle pedal won't push the sensor far enough that it won't hit a full 5v?

The reason I ask is because I trimmed the stop down a bit awhile back to get the voltage close to 5V. The problem that I ran into was that if I pushed real hard it would shut the engine down so I assume once it hit that full 5V it was to much and the PCM cut fuel. I thought about just putting a 5V regulator on it so that it would come up to 5V and not go over. This way I could get full load fuel and not have to worry about it shutting down. Is my thought process on this off?

A/D counts is basically a signal that has been converted from 0-5 volts to a digital signal 0-1024 A/D.

Here is some good reading on the subject.....

http://forum.gopowerhungry.com/minot...0-scaling.html

http://forum.gopowerhungry.com/minot...-question.html

A/D stands for Analog to Digital. The means that the 0 to 5 volt "analog" value is converted to a "digital" number that the PCM can use for processing. The A/D Converter in the PCM is a 10 bit converter which means that it has 1024 (2 to the 10th power) steps on the scale. Since the analog voltage scale is 0 to 5 volts, 0 volts = 0 A/D Counts (or "ticks" as Ford refers to them) and 5 volts = 1023 A/D Counts.

As to the "shutoff" error with excessive APP, this is because the maximum allowable APP is 922 A/D Counts, or roughly 4.5 volts. Anything about that is considered a pedal fault and the ECM instantly drops the engine to idle to prevent an accident or a runaway engine. At that point, you MUST drop the APP voltage back to an idle value and reset the IVS before the throttle will function again. This value can be raised in the parameters, but I've never found any reason to unless the throttle pedal has been modified. ;)

Hope this helps.

Thanks for the info Bill. So if I'm not getting full MFD and only hitting 3.8V of throttle pedal input would it be better to modify the Inferred Throttle Position table to deliver full MFD or to mofify the throttle position sensor so that it will read a full 4.5v?

Whatever you do, you'll have to match your inferred table to the output voltage of your sensor. If the sensor only reads 3.8v at WOT, then the table should reflect that. If it's reading 4.4v at WOT, it should again reflect that. This way you can make the most use out of your pedal travel.

Thanks Bill. That's kind of what I was thinking.