Upshifts Too Late, Downshifts Too Soon In One Truck But Works Great In Another

[Supercharged111]

I've had an absolute motherfucker of a time trying to get this stupid big block to act like a big block. For the longest time I blamed the stock tune, this truck drives unlike anything I've driven GM or otherwise. With a 4000 RPM big block, this thing sure seems to love its redline. Part throttle it REFUSES to upshift and the thing also will NOT hold a damn gear past what feels like 40% without reaching for the next so it's never at atmospheric at torque peak, rather revving its face off at 4000 RPM for no reason. I based a lot of my trans stuff on my 1500 as it moves with much less throttle input, upshifts far more aggressively, doesn't take 10 seconds to lock the converter, and doesn't downshift at the sight of a damn hill. One day I had an epiphany as I suddenly realized this truck was not hitting atmospheric prior to a downshift. Made sense to me that WOT seemed on par with what I expected and part throttle sucked so damn bad. Both 350 and 454 came stock with a 75mm TB, so I bought an 85mm for the 454. Slightly better response, but nowhere near what I was expecting. I figured it'd artificially fix the upshifts and downshifts but it had very little effect on those whatsoever. I calculated that the 85mm would have around a 27% increase in airflow, maybe I should have sprung for the 90. It just seems like the speed values I program in, no matter how I skew them, are simply ignored. My logs indicate that the speed EFILive displays on its dash is accurate within a couple and the speedometer in the truck is within 1 mph so I really don't know what to make of this. The 1 ton will, at this point, get out towed by a half ton simply because the half ton will lug a damn gear. It's pretty sad that I'm about to change plans this weekend to tow with the otherwise inferior truck simply because it won't piss me off by making me listen to it at 4000 RPM for 5 hours. Attached is the current dually tune and the current 1500 tune.

21932
21933

[statesman]

I've had an absolute motherfucker of a time trying to get this stupid big block to act like a big block.

That "stupid big block" is only doing what you're telling it to do. Have a look at the WHOLE tune, not just two or three parameters.

Start off with this...

{D0960}... change ALL values to 100
{D0961}... change ALL values to 100
{D0962}... change ALL values to 100

See if that helps at all.

I'll have a good look at your tune after YOU'VE had a good look at your tune.

[Supercharged111]

If those are the kickdown tables I have modified them. They're similar to the tables that work well on my 350.

[Supercharged111]

According to D0962, at 103kph (~63mph), I should not downshift from 4-3 until 70%. According to D0908, at 105kph, I should not downshift until 75%. This is more than I want, and it does not deliver in accordance with what I am commanding. Less aggressive downshift tables on my 350 work much better, this is what's confusing the hell out of me and pissing me off. When I look at what's commanded in the tune, it's not what I'm getting when I drive. I must be missing something critical that's skewing everything and for the life of me I just don't know what it is. I'll get a log later after I fix all the issues with the new throttle body.

[Supercharged111]

Well I figured it out, wasn't the tune after all. While response was improved with the larger throttle body I installed, it was nowhere near what I wanted. New TB had progressive linkage installed which didn't strike me as the stock TB was also progressive. When I compared it to a small block TB though I noticed that that had linear linkage so I changed the arm out and just like that the engine responds, upshifts sooner, downshifts later, locks sooner, EVERYTHING this thing absolutely sucked at stock (Plus TPS read in the teens at idle so I modded the TPS to clock it properly). As happy as I am I figured out what the holdup was, I'm kind of pissed at the same time that they'd intentionally make this thing suck so hard from the get go.

The problem I'm left with is the slam-bang shifts at mid/higher throttle angles. It must pertain to the added airflow at lower than stock throttle angles, something that was precluded in the factory tune, but I'm not too sure what to do about it since the blue truck never did this even after adding a blower to it. If I grab a handful to get up to speed and then let off it hits the next gear hard enough to break stuff. If I just stay in it it just hits more like a shift kit. Definitely didn't hit this hard with the stock throttle body and I got it to do it on the 1-2 shift so that eliminates locked TCC on upshift. For tomorrow I think I'll just turn shift times back to stock to soften things up, but I'm not sure even that's gonna fix it. IIRC I left TM intact on the upshifts.

[joecar]

Which table has priority, D0901 or D0960...?
What about D0902 and D0961...?
D0903 and D0962...?

It turns out that in different OS's the answer is different.

Look at attached graphs drawn from your file 411 Whipple V18 Stomp + 20%.tun (https://forum.efilive.com/attachment.php?attachmentid=21933&d=1521085352)

If you want predictable shifts, then set the throttle kickdown tables to 100%.



219342193521936

[joecar]

( and should also do same with TCC Throttle Release tables D1007,8,9 )

[joecar]

To soften the harsh shifts, add some Torque Reduction back in, right at/around the places where it shifts harsh...

[joecar]

Set the shift times to 0.2 or 0.3 seconds.

[Supercharged111]

Set TCC release to 100%? Here I've been setting a curve via trial and error to get it to jive with the downshifts. Statesman did get me to thinking when he posted about the kickdown, I even did a search and that seems to be the common thing to do. Now that the throttle body has completely transformed how this truck acts, I can work on tweaking up and downshifts. Before the TB and linkage no tune in the world was going to give me what I was after. Have any new products popped up other than that handy Excel spreadsheet?

[joecar]

The other problem is that your PT shift tables cross (in four places)... they should not cross anywhere, see attached graph:

21938

At any TPS % value, the order of the shift table should be as follows
(from top to bottom as you look at them stacked onto the same graph):
3->4
4->3
2->3
3->2
1->2
2->1

Do the same with WOT shift points.



PT shift table relationships:

basic relationships:
1->2 curve should be above 2->1 curve;
2->3 curve should be above 3->2 curve;
3->4 curve should be above 4->3 curve;

additional relationships:
3->2 curve should be above 1->2 curve;
4->3 curve should be above 2->3 curve;

this can be summarized as: the curves cannot cross anywhere;

if they cross anywhere, then you will not be able to predict what gear the PCM selects when the operating point hits the cross.



A4 shift table spreadsheet is here (https://forum.efilive.com/attachment.php?attachmentid=1942&d=1185999343).

[Supercharged111]

The shifts were not this harsh before the TB swap and I'm commanding some quick shift times in the 1500 and it doesn't slam the way this one is. It really feels like something has faulted and driven the line pressure up, it even hits when shifting from park/neutral to gear. Now I did modify my TPS to be adjustable and set it so it'd read 0 at idle and figured that was good enough, but the throttle keeps on going a decent ways past 100%. I ASSume this is normal as I hooked up other throttle bodies and witnessed them do the same thing. The truck does have a P0122 since the TB swap which is TPS circuit low but every time I'm looking at TPS while logging it seems normal? That said, it is now being weird about shifting out of 1st and 2nd at low speed, give it some lovin and it upshifts. I'm not sure what to make of it because my log from last night doesn't seem to show any breaks in TPS data?

21940

You'll need to forgive last night's optimism in naming my log, it was clearly unwarranted.

[joecar]

No worries, stuff happens.


It sounds like you need to reset/relearn the TPS voltage.


Log the pid GM.TFMPRS and see what pressure the PCM is commanding.

Take a look at the current DTC's in your log... then in your tune under the Transmission Diagnostics section, try to find those DTC's in each of the fault tables see if they are enabled... if they are, disable them and test drive.


Post your current tune file.

[statesman]

The shifts were not this harsh before the TB swap and I'm commanding some quick shift times in the 1500 and it doesn't slam the way this one is.

Shift times does not cause the hit... lack of torque reduction is what does it. Put some torque reduction back in and make the shift times a bit longer... 0.35 for 1-2 and 0.4 for 2-3 and 3-4. Don't be tempted to reduce line pressures... you'll just burn up your clutch packs.

[Supercharged111]

I'm scared shitless to touch line pressure, even to bump it to lessen the adaptive correction factor when shorter than stock shifts are commanded (I ASSume that's how you're supposed to do it?). I am thinking I ought to rip off some of the trans stuff from an 8.1 4L80E Avalanche or Suburban though, all this trans stuff right now is based on the 5.7 4L80E in the Express van. Shift times have all been put back to stock for a 454 4L80E 1 ton pickup and they're not much if any different feeling than when I was commanding the shorter times. The log for GM.TFMPRS was parked at 90 psi for the entire drive. My DTCs are as follows:

P0122 - Throttle Position Sensor 1 Circuit Low (C)
P0405 - EGR Position Sensor A Low Voltage (C)
P0463 - Fuel Level Sensor 1 Circuit High Voltage (C)
P1122 - Throttle Position Sensor Intermittent Low Voltage (H X S)
P1637 - Generator 2 L Terminal Circuit (C), Not Used
P1638 - Generator F Terminal Circuit (C), Not Used
P1665 - Engine Oil Pressure Output Circuit (C)

I did a search and it looks like I need TPS voltage to be .5-.6V at an idle, then reflash to get it to learn? I imagine that ought to fix the P0122 which I was so sure would be a trans pressure enabler/modifier, but it was already set to no in both. P1122 wasn't in there at all (though since it isn't current, I assume it simply wouldn't be a factor). EGR is disabled in the tune, is it then appropriate to not only disable MIL but also processing enablers? Would that prevent it from coming up in the scanner when I pull codes? Also, since my truck is a 97, the PCM does not have its mitts in the fuel gauge or the alternator output. P1665 comes as a surprise, going to have to investigate that. P1122 really implies I went too far to get the TPS to read zero, it seems a relearn is in order before I rip my hair out over nothing. Is the relearn as simple as I found above?

21947
21948

I didn't deconflict the shift points yet, trying to get this right first as its affecting low speed shifting. If feels like there's a dead zone where the PCM doesn't see throttle input, but the logs seem to imply otherwise.

[joecar]

Supercharged111,

make sure to understand what statesman said:


Shift times does not cause the hit... lack of torque reduction is what does it.


Don't be tempted to reduce line pressures... you'll just burn up your clutch packs.

you can add Torque Reduction back in to soften the shifts (TR injects momentary spark timing reduction during each shift).

[joecar]

...
The log for GM.TFMPRS was parked at 90 psi for the entire drive.

This would explain hard shifts at low/moderate part throttle... we have to find out what is causing TFMPRS to be 90 psi always (max is 96 psi).




My DTCs are as follows:

P0122 - Throttle Position Sensor 1 Circuit Low (C)
P0405 - EGR Position Sensor A Low Voltage (C)
P0463 - Fuel Level Sensor 1 Circuit High Voltage (C)
P1122 - Throttle Position Sensor Intermittent Low Voltage (H X S)
P1637 - Generator 2 L Terminal Circuit (C), Not Used
P1638 - Generator F Terminal Circuit (C), Not Used
P1665 - Engine Oil Pressure Output Circuit (C)

Those DTC's are enabled in the Trans Diag tables E010x... we have to see which of those DTC's are present, and fix the base cause.



I did a search and it looks like I need TPS voltage to be .5-.6V at an idle, then reflash to get it to learn?

You can get it to relearn idle voltage by disconnecting TPS, powering up PCM for say 10 seconds, then powering off, reconnecting TPS, and powering up again.



I imagine that ought to fix the P0122 which I was so sure would be a trans pressure enabler/modifier, but it was already set to no in both. P1122 wasn't in there at all (though since it isn't current, I assume it simply wouldn't be a factor). EGR is disabled in the tune, is it then appropriate to not only disable MIL but also processing enablers? Would that prevent it from coming up in the scanner when I pull codes? TPS problem has to be fixed regardless.

If the DTC is disabled, you can leave the MIL on (since the DTC will not trigger).



Also, since my truck is a 97, the PCM does not have its mitts in the fuel gauge or the alternator output. P1665 comes as a surprise, going to have to investigate that. P1122 really implies I went too far to get the TPS to read zero, it seems a relearn is in order before I rip my hair out over nothing. Is the relearn as simple as I found above?

21947
21948

I didn't deconflict the shift points yet, trying to get this right first as its affecting low speed shifting. If feels like there's a dead zone where the PCM doesn't see throttle input, but the logs seem to imply otherwise.
Do you have a wiring diagram for your specific vehicle, to see if PCM is involved in alternator and/or fuel gauge...?

[Supercharged111]

This would explain hard shifts at low/moderate part throttle... we have to find out what is causing TFMPRS to be 90 psi always (max is 96 psi).



Those DTC's are enabled in the Trans Diag tables E010x... we have to see which of those DTC's are present, and fix the base cause.


You can get it to relearn idle voltage by disconnecting TPS, powering up PCM for say 10 seconds, then powering off, reconnecting TPS, and powering up again.

TPS problem has to be fixed regardless.

If the DTC is disabled, you can leave the MIL on (since the DTC will not trigger).


Do you have a wiring diagram for your specific vehicle, to see if PCM is involved in alternator and/or fuel gauge...?

Alright, looks like a TPS relearn with the initial TB install would have been the ticket. I was getting around .24V at idle with the reclocked TPS and could only muster .4X V spun back to stock, so I installed an unmolested sensor and it was sitting at .55V with the idle stop as I set it. Since there's so much to be tweaked, I killed the MIL and some enablers for stuff I don't care about, bumped TCC lock to 100% min/max, and went for a spin (not harsh at all, shouldn't come as a surprise since the 700R4 was on/off as well). TFMPRS was much lower initially and reacted with throttle, shifts felt stock as they should.

To be clear, I wasn't blowing off what Statesman had said about TR, but the fact is I had left all the stock TR intact on the upshifts and haven't touched it on any iteration of tune on this truck. I'm pretty sure I left it all in the 1500 as well, just seems to make sense to me. I can go back to those shortened shift times I'd posted before as they were anything but harsh.

I have a nasty habit of only posting for help on here when I'm nuclear which usually happens when I've tried fixing something for the 10th failed time or I'm about to bust a deadline of some sort for needing a function I seek (like trailer pulling ability in OD to pick up a car from paint). So the cause of crappy throttle response was a progressive linkage and the cause of the bang shifts was me not doing a TPS relearn. Since this all came about during the TB install, my suspicion was something physical rather than tune related, but man did that response thing piss me off. Now I can focus on deconflicting shift points, correcting idle airflow, and seeing why my timing seems capped. I really didn't want to end up trying to tune around something and then have to turn around and undo it all. I did that with the blue truck when I was running speed density and didn't get very far. Only code to come back on the drive was the 1665 oil pressure thing, going to have to have a peek at the pinout and see what's going on with that. No need for a wiring diagram, 97 was the last year the PCM was not involved with the fuel gauge or alternator. 98-00 black box PCMs like that on my 1500 were involved. The fuel pump and sending unit/basket were changed, fuel gauge began to receive a PWM signal from PCM, and the PCM began to command alternator current. They also got the next gen airbag and a different stereo among other things I'm sure. My dually is nice and dumb.

[joecar]

Do the upshifts have a firm feel (firm meaning not hard, but positive and definite)...?

[Supercharged111]

When shift times are shortened yes, they were more firm and not harsh at all. The 1500 has a shift kit and hits hard enough to break loose in the rain on 1-2, but again nobang like I had with the dually. Just a huge shove in the backside, feels like a slingshot more than anyhing. Dually doesn't get that spicy without the shift kit.

[statesman]

the cause of the bang shifts was me not doing a TPS relearn.

Glad to hear you found the cause. I've never come across that before... so I've learned something new today. :cheers:

[Supercharged111]

Correction: went for a rip earlier and the dually was softer than I remembered so I shortened the min shift time to .3. My guess is that it needs to learn to the quicker times, but I suspect there's a way to land the base pressure so that less correction is required for shorter shift times. What might I need to log for that and how would I plot it? It's something I've been pondering for a bit. If it's the way ahead, then I have 3 trucks to apply it to.

[statesman]

That's something which I can't help you with... hopefully someone else will be able to help. What I do is leave the line pressures stock and give it some more room to learn by changing {D1101}, {D1102}, {D1103}, {D1107}. There probably is a way to adjust base line pressures, but I would never be comfortable doing that.

[Supercharged111]

Glad to hear you found the cause. I've never come across that before... so I've learned something new today. :cheers:

If I had corrected some of the known wrongs in my tune in conjunction with the throttle body swap and done a flash at the same time I would have not known it was a factor either (kinda odd that I didn't as I tend to flash something new into the truck most days). Made for a slower haul as I was afraid to give the thing any throttle with the trailer to try and keep the shifts as soft as possible but now that it's healthy again I can concentrate on some of the things you and Joecar have highlighted. The real prize was taking this back into posession.

21972
21971


That's something which I can't help you with... hopefully someone else will be able to help. What I do is leave the line pressures stock and give it some more room to learn by changing {D1101}, {D1102}, {D1103}, {D1107}. There probably is a way to adjust base line pressures, but I would never be comfortable doing that.

Well there's another reason to fear base pressures as I currently do. Seems these tables would also need to be smartly coordinated with a shift in base pressure. Definitely going to poke around an 8.1 4L80E tune to scout for significant differences.

[Supercharged111]

Hooked up the trailer again and I must say it is much better than last weekend even if you ignore the smoother shifting. Deleting the kickdown has the transmission downshifting in a far more intuitive manner and having the TPS read properly I'm sure is also helping. I didn't find myself cursing the truck, just needs more power now. Shifts aren't even really that firm, but they are quicker now. I feel I now have the 90% solution as far as drivability, I can't imagine being stuck with a stock tune on one of these things with an automatic.