i use 5 degrees of timing in mine in the low end of the table for firm shifts down low, However i have negative 6 degrees in the top end for shifts.
negative timing seems to work better for quicker shifts anyway.
Printable View
i use 5 degrees of timing in mine in the low end of the table for firm shifts down low, However i have negative 6 degrees in the top end for shifts.
negative timing seems to work better for quicker shifts anyway.
5 degrees won't get me near where I want! Currently the ECU is pulling 40 degrees down to zero of advance.... I would want no more than 20 to be pulled, so I would have to put my mins at 20 for that to work at light load...no good.Quote:
Originally Posted by Tre-Cool
u dont understand the table.
if you put in -6 degrees in the minimum timing table. the ecu can not go any lower than -6. so if your seeing -40 now. it will hit the "wall" at -6. it's not a -40 +20 = -20 degrees equation.
B5153 Power Reduction Minimum timing in my e38 is set to -10 degrees for the whole lot
B5157 -Minimum Final Timing=
Goes from +6 to -10 after 0.72g/cyl
I have -15 in B5153, and -25 as the largest in B5157. So if that were true...I would never get more than 25 degrees pulled (because that is the largest negative number in B5157). Yet, I am getting 40+ pulled...so what am I missing?Quote:
Originally Posted by Tre-Cool
what b5156 or alternativily post your tune files and i'll take a look and see if anything stands out.
Attachment 14273 Here you go.
Quote:
Originally Posted by Tre-Cool
well it all looks good. i have put the settings that i use in from my e38, so see how it goes.
GM certainly likes to run some timing in those motors. knock sensors are seem pretty relaxed too. i run forged pistons in my ls3 block and my knock threshold numbers are at 6 for all mg.
Attachment 14274
What all did you change, so I know what to look for?Quote:
Originally Posted by Tre-Cool
do a compare, but it's mostly the reduced power spark and minimum timing tables.
O.k. will do! Thanks! I'll let you know how it does.Quote:
Originally Posted by Tre-Cool
Well...no good. Still getting the same timing numbers being pulled, and now, after it shows SMRET being reinstated back to zero...it is still holding timing and ramping it in at a timed rate. Feels HORRIBLE! Surely someone knows why it is doing this.
Take a look in your :Torque Control Parameters: you might find your answer in there!!!Quote:
Originally Posted by Jason S.
Cheers Scotty
Care to be more specific? Table numbers? We both have looked through those and don't see anything.
I have come to the conclusion that there is no way to reduce the amount of timing pulled during the shift for torque reduction. Several of the tables listed as being related to this clearly have the wrong descriptions. TCS must = Traction Control System and not Torque Control System, as none of those tables when adjusted affect regular torque reduction.
TCS is traction control. If you put minimum timing in that rpm/load table on the minimum final timing table to say....10*, it will not go below 10* in that area, ever.
TCS is always been Traction Control. It is in the Torque Control section because it is a type of request for reduction and there are also different methods that it uses. This always takes a bit of experimentation for me as well.
I'm trying to sort out this thread and have a question about scaled ECM tunes. I normally modify the TCM pressure tables using Torque and offset them in accordance with the percentage of scaling. The shifts are sometimes good and sometimes OK...kinda of inconsistant from setup to setup. Would you apply this ECM Torque model update in conjunction with the scaling in the TCM tune or leave the TCM tune stock and change the ECM model? Anyone have any experience with this?Quote:
Thanks,
Ed M
i still run higher pressure in my tcm tune and large multiplier update to increase the torque output readings.
After much testing and many 6l80s tuned I have found the trucks do very well with just the scaled torque model tables or tables on vvt setups. They also tend to like a little wider shift time vs the car setups as well.
On the cars I run a combo of tq model and on the low side of the table I will still bump pressure to get the desired feel. On cars I run a tighter shift time, at wot anywhere from .15 to .25 will bump pretty well.
Quote:
i still run higher pressure in my tcm tune and large multiplier update to increase the torque output readings.
So if I understand, you will increase the PCM Torque Model Airflow coeficient table (B3033) to "realign" the calculated torque model which has been reduced as a result of the Tune scaling process? This would then make it not necessary to scale the Torque related TCM Torque axis values (offset table typically as you do with the timing table)?Quote:
On the cars I run a combo of tq model and on the low side of the table I will still bump pressure to get the desired feel
Is it as simple as saying if my tune is scaled by 30% then I multiply the PCM Torque Model airflow coefficient values by 30%?
Thanks for your input. Will experiment when I get a chance but any expereince from the forum would be appreciated.
Ed M
Correct, however if your running mafless, I would add another 5%.
Ed,Quote:
Originally Posted by mowton
Assuming GM has proportionally aligned the reported torque from the engine controller to match the values on the TCM; any changes to the airflow that are scaled will effect the final torque value. If the MAF is cut by 50% then the reported torque is 50% less and tables in the TCM will off by 50%. EFIL needs a scaling hot key that can do this in one key stroke!! hmmm......V8 idea!
This torque table is unfortunately difficult to reason through as its a coefficient table much like the airflow tables. What we need is this math reverse engineered to look like a VE surface which we could plot over airflow and really correct this table.......I have noticed that when you raise this table the reported actual torque value will go up.
Howard
I just started playing with the TM table(s) today on my 08 TBSS, Im running SD (failed the MAF) so I can tune the VVE. Lets just say even with the MAFless patches applied, the shifts were horrible. Completely different than they were before I installed my injectors (ID850s :D) and scaled my tune 50%.
I started with a 10% bump over stock and it made a difference, but still not what I was hoping. I then flashed the same tune with a 50% over stock, and HOLY CRAP it shifts just like normal again. So anyone wondering if you need to adjust these in a scaled tune, YOU DO! I bumped it 10% more over that and its pretty dang good for my taste.
Howard, I like that idea. Have the ability to click a button and have all the VVE tables halved, Torque Model table doubled, etc.
Definitely sounds like we are on to something good :) Will make A4/A6 setups much easier to tune. The other tuning software lets you change the axis values which helps a bit. Thanks Howard (that table would be cool for sure) and Eric for your inputs. Last tune I tweaked the coefficient by 10% (also scaled the TCM Torque related pressure values) and it got better. Didn't want to go overboard until I had a chance to log and experiement.
Thanks again guys,
Ed M
Yes it works great.
Some questions about this:
Will it alter the ECM's torque PID? it has been hailed as so accurate and I myself found that it is within about 4 ft-lbs of factory rating. So after we make the tune shift better, is the accuracy of this PID invalid?
Also, is it my imagination or does the car also "respond better" even when not shifting?
I have increased all 4 degree positions on my VVT by 5%. The part throttle shifts are nice, almost a little harsh, the full throttle shifts have improved, though on the soft side. So should I increase just in the high RPM sections as stated before?
Thanks!
You always want to leave pressures low enough for the tcm to firm up during learning, not soften. I run wot pressures on the base tables at 1200-1400 kpa. Max line pressure limits up to match that. Wot shift times of .2 to .27. The tighter the converter the looser you can run shift timing for a good feel.
Yes, in fact this PID is your feedback to track your changes.Quote:
Originally Posted by Boost
People have ridiculed this method (usually because they aren't using EFILive), the conversation usually goes something like this.
Is the car MAF-Less ?
Yes, and since doing that the trans shifts bad.
Put the MAF back on, log the Torque PID, observe the max value (lets assume 450Nm).
Now fail the MAF so it run SD and repeat the Torque PID test, observe max value, oh look, it only hit 390Nm that time.
So now the TCM figures the engine is making less power than before.
Mess with the coefficients to get the reported torque back up to where it was and what do you know, the trans is shifting ok in SD mode now.
Most people run in to this problem when going SD, it's too variable to 'auto scale' for some they need 10% more, others 40%. Just monitor the torque PID so see which way you are heading.Quote:
Originally Posted by Redline Motorsports
Again, this is the whole idea of messing with the tables, not so much to fool everything else in to thinking the engine is making 1000Nm of torque, rather to have some control to fix the torque calculation errors the ECM might be dong.Quote:
Originally Posted by Redline Motorsports
To represent it as a VVE type table unfortunately isn't really possible, the VVE is plotted from numerous 2D tables all combined and manipulated at the same time. The tables the are with the Torque calculation are all 3D and are not calculated all in one shot. GM must have a killer bit of software to plot the numbers, it's over our head.
Remember guys, the Torque PID is your friend, consider it the WBO2 of torque tuning.
i just swapped out some 63lb injectors for some id1000 sized ones which with an allready maxxed out ifr have to run an even more scaled tune. this has resulted in a loss of resolution but no biggie. i've also switched to e85 by the drum so thought i'd show the difference in the tunes to get the reported torque figures back to "normal".
Attachment 14472Attachment 14473Attachment 14474Attachment 14475
the grey screen is percentage difference between the old 63lb vs 100lbs.
Thanks for the very good info and responses! So as I take it - yes - of couse by altering these tables the reported torque PID changes (higher) and it may no longer be so accurate and close to the factory rating (I am with you Ross about the WBO2 comparison, I've come to trust this PID more than some dynos).
However, adjusting it makes a significant improvement to shift quality and almost feels like the final missing piece to the TCM dance.
So, I am working on an E39 / T43. I did 10% on the four tables to start, and there was a noticeable difference, though still too sloppy. Then I tried 15% and I almost feel like it got worse?Quote:
Originally Posted by Rhino79
I ended at 10% but needed to just decrease the shift times a bit and increase the pressure on the 1-2 and 2-3... I had to experiment, but in this application which was a truck, just increasing the tables was not enough to provide the desired results.Quote:
Originally Posted by Boost
Seems you need to work a few things in tandem... Nothing is easy and direct with these things...
Hope this helps... Again, just my experience..
John
I hear that. Thank you John!
On the V6 it might be a little tricky because you have the cam phasing to deal with too, each column is very different, unfortunately just adding percentages on the E39 doesn't always work. The other thing is these tables on the E39 will change the throttle response.
I should let you all know that there is a pretty major change to these torque table configurations coming which will mean if you have been scripting changes you will have to start again, the table configurations are changing.
I will shortly be in NZ working along side Paul on a few projects, one of them will be to plot the torque maps similar to how we do the Virtual VE's. However, at least for this trip there is no possibility of adding in the reverse calculations.
But the plan is what we can offer is a 3D plot of what the calculations will produce. We did some logging on Paul's car today (6.0L L98) and based on the numbers it was amazingly accurate.
So as another test I set up the calculation in Excel and punched in the numbers to match the L98's peak torque figure at 4,400 RPM from those in the torque model tables. I assumed a couple of things, timing (23 deg) airflow (860mg).
The L98's rated torque at 4,400RPM is 530Nm, the calculation produced 532Nm! If I dropped the timing back to 10deg it calculated out 436, upped it to a knock producing 30deg it only went up to 548Nm.
It's really fascinating playing around with the numbers and seeing the results, should be a lot of fun once it's implemented, but for the new ECM's (E39, E78 etc) pretty important.
It will be very tricky with the variable cam engines, it may take a few software revisions to get the operational side of this ideal based on everyone's feedback.
http://download.efilive.com/Staff/GMPX/TrqMdlCalc.png
Oh man this is great news!
Any body know the answer for this question ?Quote:
Originally Posted by Protec
Do we need to change all 4 tables with VVT cars ?
Yes change them all, my cam is vvt in my car and it helped changing all 4 airflow tables
Hmmm - may explain the high trans failures on blown cars running around 12deg ign timingQuote:
Originally Posted by GMPX
Good point Luis, all the more reason for us to get them displayed so you can change the spark coeffs to match.