BMW E46 Fanatics Forum banner

1 - 15 of 15 Posts

·
Registered
Joined
·
972 Posts
Discussion Starter #1 (Edited)
Hello,

I am in the beginning stages of tuning my 2003 ZHP. I swapped in the M54b30 engine in that I used to have in my 328i on the MS42. On that car I had it making approx 250WHP (has cams, M50 intake manifold, ported head, exhaust etc... all NA)

I am currently working on the base MS45.1 tune in Renovelo and one area I am very unsure of is whether or not the torque tables will need adjustment and how to go about doing so intelligently. Does anyone have any experience with this. At what point does the engine get limited by the factory values? Do I need to adjust the torque tables on an NA motor making maybe 60-70HP more than the factory motor did at the engine... or is this still within the range that the factory torque map will accept without limiting throttle due to overages?

I'll have to check it in ByteTuner/ByteLogger, I suppose I should be able to see if the car is hitting the max values in the map.

Figured I'd ask here first in-case anyone has some insight.

Thanks guys.
 

·
Registered
Joined
·
1,682 Posts
The only thing you'll probably notice is the car going WOT sooner in your throttle pedal travel, if you notice much of anything. In NA form, even with more power, I dont think you'll have to adjust the torque tables. And with MS45.1 being wideband, as long as you're using your stock fuel injectors, you probably don't have to do too much with fueling.

EDIT: Just some more info. MS45 has the concept of "full load". once the car sees it is pulling in enough air at the MAF to trigger it, it will move the throttle plate to full wide open. So in cars that are making significantly more power (and thus pulling in more air), this wide open throttle condition could tend to happen too soon and too abruptly in the throttle pedal angle. When this happens you have to adjust the intake manifold model to tell the DME to expect more air before hitting full load, and then you'll want to adjust your torque and engine load tables accordingly as well. That is because the MS45 throttle is based on torque demand, and delivered torque is based on MAF, converted to engine load, converted to torque. So if you do adjust your intake manifold model without adjusting your torque and engine load tables, then you'll probably find the car never actually hits WOT.
 

·
Registered
Joined
·
972 Posts
Discussion Starter #3
The only thing you'll probably notice is the car going WOT sooner in your throttle pedal travel, if you notice much of anything. In NA form, even with more power, I dont think you'll have to adjust the torque tables. And with MS45.1 being wideband, as long as you're using your stock fuel injectors, you probably don't have to do too much with fueling.

EDIT: Just some more info. MS45 has the concept of "full load". once the car sees it is pulling in enough air at the MAF to trigger it, it will move the throttle plate to full wide open. So in cars that are making significantly more power (and thus pulling in more air), this wide open throttle condition could tend to happen too soon and too abruptly in the throttle pedal angle. When this happens you have to adjust the intake manifold model to tell the DME to expect more air before hitting full load, and then you'll want to adjust your torque and engine load tables accordingly as well. That is because the MS45 throttle is based on torque demand, and delivered torque is based on MAF, converted to engine load, converted to torque. So if you do adjust your intake manifold model without adjusting your torque and engine load tables, then you'll probably find the car never actually hits WOT.

Wow thank you. Tuned a bunch over the weekend and I've got it running pretty well now and pulling fairly strong but I know there is still some tweaking to improve things still. Sounds like quite a complex interplay between these tables in the MS45.1. With what you describe with the full load condition it sounds like an issue would generally manifest as a touchy pedal that punches you hard when you don't quite expect it. I haven't run into that so far, so I think I'm still not making enough power to really mess with the full load parameters. I also looked at the torque tables in depth and ya, I don't think its making enough power to cause any issues there yet. Most of that stuff would likely come into play with any serious FI I would imagine.

I did change the injectors due to the M50 manifold (the factory injectors need to be installed backwards if used and therefore would squirt the wrong direction in the intake tract). So I have attempted to scale for them a bit with global settings... I did notice like you said though, that the factory wide band does a good job of keeping the fueling where it should be most of the time. Really just cold idle and cold low throttle fueling maps could use a bit of a tweak for before the wide-bands kick in.
 

·
Registered
Joined
·
1,682 Posts
You would notice the throttle issue when you roll into the throttle pedal and power will be increasing and then at some point, maybe 60% into the pedal travel, BOOM, full throttle. In data form it looks like this:



The more air you are taking in through the MAF over the stock values, the more pronounced the phenomenon is.

Really just cold idle and cold low throttle fueling maps could use a bit of a tweak for before the wide-bands kick in.
Yes, this is still something I'd like to improve upon with my tune as well. Its been a problem really I think ever since I installed headers years ago, and was worse with the supercharger and larger injectors. Despite sizing the injectors correctly and having the correct deatimes etc, cold starts are still a bit bumpy before those widebands kick in.

The best change I've been able to introduce was by modifying ip_fac_ti_tco_wup and taking the values from mostly <1 to a little >1. And for the most part its usually quite good, its way better than it used to be, but I still get an occasional stumble, and well, I'm quite picky and it still bugs me that its not super perfect yet.

I mean I could just increase the values in that table even more and see what happens, but this is still on my "to do" list at some point to tackle and I just haven't pulled this little effort down off the shelf yet.
 

·
Registered
Joined
·
557 Posts
I was having the same cold start problems with my old MS42. It turned out that the hot start was a little lean from the stock little rich. It just gets worse as the engine starts get colder. I started with increasing the pre-start by around 5% and then the cranking and post start tables by 5% initially. After that it was a bit of a round robin type of thing with increasing one map by 5%(usually too much) to get both the cold and hot start working well.

When the maps were wrong, hot or cold start would go rich at cranking or post start(4 scenarios) with extended cranking to get it to start and/or run.

Hope it helps
 

·
Registered
Joined
·
972 Posts
Discussion Starter #6
Thanks Blue, was noticing something like that going on today with my cold start. Started pretty rich but not too rich for a cold start and then goes silly lean very quickly. So based on seeing that I assume the startup enrichment is working fairly well but my cold injection maps are too lean.
 

·
Registered
Joined
·
1,682 Posts
Blu302 and jared, maybe we should list our mods. I'm curious what is making the MS45.1 do this on cold starts.

For me, I'm running:
ESS TS2 supercharger kit with larger injectors
catless headers with the rear O2 sensors unplugged and disabled in the tune
SAP removed, plugged off, and disabled in the tune

The only cold start fueling table I've changed is ip_fac_ti_tco_wup.

I've wondered if it was removing the cats, or removing the SAP, that is causing this lean condition during cold starts that causes the stumbles during the first 60-90 seconds of operation.

There are some tables in the DME that make me think the MS45.1 will operate a little lean on cold starts to heat up the cats faster.

Also, keep in mind in stock form with a working SAP, there is air being added into the exhaust before the pre-cat O2 sensors on cold startups. So in stock form, the DME expects or looks for the exhaust to be a little lean. Now that the SAP is removed, the exhaust will appear richer than the car would have expected in stock form, and so maybe it tries to lean it back out. So maybe I'm missing a table that relates to the SAP.

Do either of you have similar mods to the deleted cats or deleted SAP on your cars?
 

·
Registered
Joined
·
557 Posts
I have full exhaust, headers, high flow cats and Schrick cams.

I noticed it started with the exhaust(first mod) as it was flowing more air(less reversion for emissions). During the initial cat warm-up cycle I had nothing but problems. I disabled it and it almost disappeared totally.
Being in Australia, the cars here never came with SAP, MS42/MS43/MS45.

The WUP tables come into play with multiple starts of the engine, especially if close together like a failed start leading to a retry.
I changed the pre-start, cranking and post-start maps to get a clean, instant start every time.
 

·
Registered
Joined
·
7,939 Posts
Also, keep in mind in stock form with a working SAP, there is air being added into the exhaust before the pre-cat O2 sensors on cold startups. So in stock form, the DME expects or looks for the exhaust to be a little lean. Now that the SAP is removed, the exhaust will appear richer than the car would have expected in stock form, and so maybe it tries to lean it back out. So maybe I'm missing a table that relates to the SAP.
Keep in mind that the MS45.1 cars have a mini mass airflow sensor for the SAP. Between that and the wideband O2 sensors, the DME should be able to determine the true AFR regardless.
 

·
Registered
Joined
·
1,682 Posts
Ah, that's right. I had assumed that was only for detection of a failing SAP, but I guess the DME could use it to impact its AFR too....
 

·
Registered
Joined
·
1,682 Posts
I was browsing for some of the parameters that effect the SAP, and I'm currently running these:

ip_t_sa_off_1 - all set to 0, this should disable the SA function
lc_inh_sa_man_deac - disables diag of SAP?
lc_inh_safm_man_deac - disables diag of secondary MAF?
lc_sap_et_di - disables automatic learning of SAP

So with all those off, the DME should not be expecting any air flowing from the SAP, in theory anyway. And if that's the case then the rough cold starts might not be SAP related?

these are also interesting:

ip_saf_mdl - Seconary air flow model
lc_sa_swi_acq - logical constant if SAF model is used or not

I'll test with these two disabled or zerod out and see if it helps.
 

·
Registered
Joined
·
7,939 Posts
I'll double check what I did to my ZHP. From what I recall, I disabled the automatic learning, and then copied the SAF stuff from an MS45.0 calibration. I also adjusted C_OBD_STATE_2 from 6D to 65 to get the SAP OBDII monitor to show as "Not Available", but that's more cosmetic than anything.

Unfortunately my virtual machine got corrupted and my newest backup was from like a year ago (dumb mistake on my part), so I'll have to wait until I go back to my parents' house where my ZHP is being stored to actually redump my tune and see what I did.
 

·
Registered
Joined
·
1,682 Posts
Looks like ip_t_sa_off_1 might only control how long the SA function runs, but doesnt do much to interrupt the whole SA initialization processes.

lc_sa_swi_acq will force the DME to use the SAFM model, rather than try to read the airflow values from the secondary air MAF.
ip_saf_mdl is the secondary air model and we could set that to all 0 to indicate zero airflow.

That should feed into the initial lambda calculations during startup and prevent any rich/lean conditions based on secondary air that isn't there anymore.

Its also possible that simply changing lc_sa_swi_acq to 0 and forcing the model will stop the DME from "hunting around" during startup, in case its looking for secondary air MAF signal and not getting it and spazzing out.

I set lc_sa_swi_acq to 0 yesterday and had a good startup this morning, but I want to run it a few more cycles and see if it holds up. If its still not good I'm going to zero the kg/hr values in ip_saf_mdl next and see if that finally solves it.
 

·
Registered
Joined
·
557 Posts
I have a MS45.0 that doesn't have SAP or anything. I can post the tune file and the software number if it helps.
 

·
Registered
Joined
·
1,682 Posts
I tested the following changes:

lc_sa_swi_acq 0
ip_saf_mdl 0's
lc_sa_swi_cmb 0

And none of them made much of a difference. Might have had a little more stumbling with lc_sa_swi_cmb, but not super noticeable and could have just been an anomoly (when it comes to cold starts, I generally need to observe over the course of several cold starts, like where the car has sat for at least 6 hours).

I'm starting to think the cold start stumble might not be SAP related, but have something to do with the cat heating cycles when cats don't exist.
 
1 - 15 of 15 Posts
Top