new $EE tuning thing!

[steveo]

So what do you imagine this OS Code hack is for high stall OBDI?

ed says some odd stuff sometimes.

thing is, that thread talks about disabling the VSS error code, not moving the 'RPM threshold', they might be different. i'm not even sure that 'rpm threshold for vss error' or whatever is an accurate label.

The fix is NOT the VSS error code. Moving the WOT mph threshold for the 1-2 shift down to 1 mph and letting it shift on the rpm trigger was the fix for mine.

this seems reasonable too.

[lionelhutz]

if i said make me a set of shift tables that'd be not great but just passable for a midsize cam with a medium and large stall, could you rough them out? how about just more aggressive for a stock-ish car, a bit more line pressure and higher shift points?

Well, the transmission operation is rather complex. Gear ratio and tire size changes also change the input shaft torque in each gear which then affects the converter slip. So, even just changing the shift curves 21% because the rear gear changed from 3.08 to 3.73 is not likely to produce the desired results.

This is how I made my shift curves. I went for a drive with each gear manually selected and recorded the speed at 10%, 20%, 30%, 40% throttle. Then, I created a shift curve that follows these just below those values blending it to the WOT shift speed. I logged WOT to determine the top speed reached in each gear. You have to set the downshift mph points in the kickdown mode table to the same upper values as the main shift table. Otherwise, funny things happen as the throttle transitions back and forth past the kickdown mode scalar.

So, I suppose you could calculate the approximate WOT mph for each shift by using gear ratios and allowing some rpm drop for the stall. Then, make curves that start at a minimum speed for each gear (maybe 1/4 for 1st, 1/3 for 2nd etc) and curves up into this maximum mph. Make the curve whatever shape you think would work best. I don't know how well this would work but it'd be a good start.

I have not experimented with line pressure enough to say just do this to make it better. But, you could pull-up the low and middle parts of the main line pressure table and still be safe. Messing with the force motor current isn't a very good idea. But then, if you make a tune for an unknown transmission build then who knows how well the force motor current table matches with the transmission. I mostly just find lots of comments about too much line pressure blowing the pump or low pressure burning out clutches from people who have blindly messed with line pressure changes (via tuning or shift kits or shims or turning the screw in the force motor or ???) when trying to find useful info about this. Which basically tells me you actually need to put a pressure gauge on your transmission to do it right. So, once again a way to manually set the line pressure to do testing would be beneficial. I would basically linearly vary the line pressure from 0 to 96 and ensure the actual pressure also linearly increases from the minimum to maximum. Then, if I didn't like the minimum or maximum I would adjust the force motor table and test again.

[lionelhutz]

ed says some odd stuff sometimes.

thing is, that thread talks about disabling the VSS error code, not moving the 'RPM threshold', they might be different. i'm not even sure that 'rpm threshold for vss error' or whatever is an accurate label.

this seems reasonable too.

I already described this one. Setting the upshift mph in the kickdown table to 1mph and then using rpm only fixes the kickdown mph being too high. Obviously, you don't have to use 1mph, but the speed the car reaches at WOT has to go above the shift mph before hitting the rev limiter or it hits the limiter instead of shifting. This IS NOT the same as the stall wall where the transmission goes into limp mode.

Another stall wall "fix" is to put the 100% TPS 1-2 upshift mph in the main shift table to 1mph. I don't really understand why this one would work. At WOT the PCM is using the kickdown table values, not the main shift table so the 100% TPS mph setting in the main shift table should mean nothing.

[fbody_Brian]

Mine is slipping mostly on the 1-2 shift and in 2, it'll hold a decent amount of torque in 3 and 4,

That's probably the band going out. You may be able to get more life by getting a longer pin for band engagement, but you may want to do some searching. I don't think that any tuning is going to help your issue.

[fbody_Brian]

... but the speed the car reaches at WOT has to go above the shift mph before hitting the rev limiter or it hits the limiter instead of shifting.

I always found it odd the way it's set up by default and just lowered all my kickdown shift mph settings really low in all gears so I can just specify an rpm for shift.
good to know about the TPS 1-2 upshift mph in the main shift table!

[steveo]

this is all great info, keep it coming. im learning a lot.

i dont really want to tune automatic transmissions but understanding how they work is really useful for my tools.

how about 'adaptive pressure' thing? is that slip measurement that increases line pressure to compensate? or...

how does that work?

[fbody_Brian]

I found where I read about the v4p towing package trans tables...
it was on this forum. LOL
http://www.gearhead-efi.com/Fuel-Inj...ission-VIA-PCM

Also buddrow who is active on here seems to have some knowledge about how transmissions work so maybe has some insight.
http://www.gearhead-efi.com/Fuel-Inj...mission-tuning

[fbody_Brian]

You have to set the downshift mph points in the kickdown mode table to the same upper values as the main shift table. Otherwise, funny things happen as the throttle transitions back and forth past the kickdown mode scalar.

I find this very interesting. I wonder if I haven't been experiencing these funny things.

[steveo]

this entire line of tuning conversation makes me really glad i have a stick. the human mind is a perfectly good transmission controller.

seems there's a lot of opportunity for a bad transmission tune to cause long term damage too.

[fbody_Brian]

this entire line of tuning conversation makes me really glad i have a stick. the human mind is a perfectly good transmission controller.

seems there's a lot of opportunity for a bad transmission tune to cause long term damage too.

Yes, and on top of that the 4l60e is notoriously difficult to build. I'll be investing in a new trans soon and after the cost of this rebuild plus the previous rebuild I could have had a t56 in my car and been happier, until I grenaded the ten bolt.

[steveo]

4.5 stable: http://fbodytech.com/eehack-2/download-eehack/

[lionelhutz]

I find this very interesting. I wonder if I haven't been experiencing these funny things.

It's not just the same points, but the curve as well. Otherwise, what basically happens is that when the normal downshift is lower between the 2 tables then you can get this downshift-upshift thing happening if you move the throttle into and out of kickdown. Also, you get to the speed where you have to be at WOT to downshift when it should happen at a lower throttle. Here is an example where the normal 2->1 shift point is 10mph while the kickdown mode is 28mph. So, one you are over 10mph you have to go WOT to get a kickdown.

Attachment 10378

Now, if you modified the normal curve so it came up to 28mph as well then you could get part throttle kickdown between 10mph and 28mph. For example, at 17mph 50% TPS might cause a kickdown. To me, it just makes more sense to build a kickdown curve so that it will kickdown into a lower gear once I give it enough throttle to be accelerating in the lower gear, rather than requiring WOT for a kickdown.

[steveo]

some people reported issues with log playback

apparently if you go back and forth a bit it was possible for stale data to be displayed.

this was due to some edge cases in logic i was struggling with earlier, related to the fact that i maintain a single linear log for all datastream messages, making truly time-linear display more expensive.

released 4.5.1 to address these issues

[fbody_Brian]

It's not just the same points, but the curve as well. Otherwise, what basically happens is that when the normal downshift is lower between the 2 tables then you can get this downshift-upshift thing happening if you move the throttle into and out of kickdown. Also, you get to the speed where you have to be at WOT to downshift when it should happen at a lower throttle. Here is an example where the normal 2->1 shift point is 10mph while the kickdown mode is 28mph. So, one you are over 10mph you have to go WOT to get a kickdown.

Ok this is not the kind of issue I've been having. Mine sometimes doesn't want to shift into 3rd gear until way high up in rpm, and sometimes not at all. When it does this I put it in neutral and let the rpms come down then I shift it back into gear and it shifts. Always has shifted firm and no slipping at wot. I had some bad shutter above part throttle recently and after that 3rd and 4th slip above around 50% throttle, but hold fine otherwise. Time to hook up a pressure gauge.

Attachment 10378

Now, if you modified the normal curve so it came up to 28mph as well then you could get part throttle kickdown between 10mph and 28mph. For example, at 17mph 50% TPS might cause a kickdown. To me, it just makes more sense to build a kickdown curve so that it will kickdown into a lower gear once I give it enough throttle to be accelerating in the lower gear, rather than requiring WOT for a kickdown.

I agree, seems logical.

[steveo]

now packaging an installer-free eehack for 64 bit linux:

http://fbodytech.com/now-packaging-linux64-build/