MAP conversion for high boosted application

Discuss Bosch (Porsche, BMW, Volvo, etc) tuning topics here. Request definitions, discuss parameters, etc.

Moderators: robertisaar, dex

Post Reply
Evil
Posts: 133
Joined: Tue Jul 18, 2017 12:53 pm
Location: France

MAP conversion for high boosted application

Post by Evil » Sun Jan 14, 2018 5:04 pm

I was reflecting about using a MAP sensor with our 3.X(.X) motronics.
For low boost applications the simplest solution is a MAF swap for a bigger one, and seems to work well.
But for high boost, using a MAP sensor would be a better idea than a MAF which could saturate quickly.

So here is my idea: remove the MAF and replace the TPS by a MAP sensor.
-TPS and MAP sensors are both 0-5V, so electrically speaking it will work, but maybe it needs to modify a transfer function, if it exists?
-TPS is used for Idle and WOT switching: we know the adresses for these switches values, so we can set the Idle/PT to 0 and the PT/WOT to 255 and then use only PT fuel tables. We could also imagine to use the WOT switch to use 2 tables in 16x16, a kind of low boost/high boost tables (the TPS value for switching becoming a MAP value). That would mean to tune other things than fuel and ignition tables.
-With MAF removed, the DME will use "alpha/n limp mode table" as a VE table. But instead of throttle/RPM it becomes boost/RPM, no problem about that but this table must be absolutely extended to 16x16.
-If I understood correctly what we talked about accel enrichments recently, TPS has nothing to do with it, so there should be no problem ?

Thats just some assumptions for the moment, it will obviously ask a lot of work. It seems achievable for me, but maybe have I forget anything crucial ?
Idle could be a problem, I dont know if Idle switch can work correctly with a MAP value... If not, Idle speed regulation will be obviously not working.

What do you guys think about it?

Mykk
Posts: 99
Joined: Sat Sep 05, 2015 8:28 am

Post by Mykk » Mon Jan 15, 2018 4:38 am

On my 404DME M3.3 I have replaced the TPS with a MAP sensor.

I've tried to run my car solely in P/T tables, much like how a standalone EFI you only get one VE & one Timing map for entire operation (not including temp modifiers).

The Motronic doesn't like to idle in the P/T maps. I would need to keep my engine idling in the 900+ RPM range to prevent from stalling. During cold starts I'd need to manually keep RPM's up with gas pedal until engine was warm enough to maintain idle on it's own.

Unfortunately there are too many maps and too many unknown variables left in the Motronic code that come into play during the Idle maps that make things run much smoother.

Yes, you do need to change the Idle/Part Throttle threshold map. Without the RPM header for those thresholds we're stuck with the switching happening at 1800 RPM. Which can make any kind of low RPM cruising, low speed (parking lots, driveways..etc) driving bucky & hesitant and kind of tricky as the DME switches to Idle maps at 1800RPM. Remember, you'll have a lower KPA while driving and cruising (higher manifold vacuum, less MAP voltage signal) then while idling. So while driving and cruising you'll be under the Idle threshold in the table you have used for the MAP to idle.

2nd observation: The idle never really cleans up and sometimes stalls, AFR's are inconsistent and unpredicable. My theory is that because manifold vacuum is always changing even slightly and the fact the engine will have more manifold vacuum while warm then cold. The MAP sensor signal is constantly moving/changing, usually the TPS signal at idle is static. The DME is always in the acceleration enrichment table and is adding fuel to some degree.

Not that all of this couldn't be cleaned up with tuning. But man I wish we could find the Idle/PT threshold RPM header.

Also, since Motronic uses TPS for load calculations. Replacing it with a MAP doesn't give an accurate enough variable for tuning ignition tables in boost.

Sure boost will read higher loads in the maps, but there is no way to get specific like "I want 20 degrees timing at .5 bar and 15 degrees of timing at 1 bar boost" It would be a guessing game.

By going MAP vs RPM based VE you're converting Alpha-N to Speed Density. Which is cool, I encourage that. Unfortunately Speed Density works much better with a TPS sensor too. If the engine is solely dependent on MAP values the computer will always be Re-active to throttle transients making adjustments as the changes in manifold pressure happen instead of Pro-active; making changes as the throttle changes before manifold vacuum reacts...if even by only a couple milliseconds.

Evil
Posts: 133
Joined: Tue Jul 18, 2017 12:53 pm
Location: France

Post by Evil » Mon Jan 15, 2018 11:49 am

good to have a real life feedback about that!

For ignition tables, on MP3.2 they are RPM/Load related, like us. I dont see what could be a problem, for a given RPM load is directly related to boost value.
Anyway I think that those tables can be converted to RPM/Boost instead of RPM/Load. For a better definition the PT/WOT switch could be used, to allow using extended WOT tables

About Idle, the main problem I see is that when idling on PT tables, idle speed regulation is'nt active any more :?
A solution could be an electical switch activated off-throttle with a resistor, to deviate signal when throttle is off, then have a constant signal and allow use of idle tables?

Mykk
Posts: 99
Joined: Sat Sep 05, 2015 8:28 am

Post by Mykk » Thu Jan 18, 2018 1:18 am

I'm experimenting this morning; keeping the Idle to PT & PT to WOT threshold values but changing the Idle, PT, WOT Ignition and Fueling tables to all reflect the PT table in the Table of Tables to see if I can get the car to run on just PT tables all of the time. In theory this will still allow any other tables that come alive at Idle condition to still function. I'll report back later to verify function and practicality.

Worst case scenario the M3.3 won't like sharing the same table address for different operating conditions. Reducing the amount of control tables available for tuning isn't exactly desirable as more control over different operating conditions is a good thing. But this is an experiment and I'll be curious to see how the vehicle operates.

Evil
Posts: 133
Joined: Tue Jul 18, 2017 12:53 pm
Location: France

Post by Evil » Thu Jan 18, 2018 8:47 am

On M4.4 (volvo turbo) I have seen that the same tables is used at several points of the "table of tables", something like 4/5 times, so maybe you can be optimistic about that :D I don't remember what table it was and what was is function, I was just making comparisons with MP3.2 so didnt make very deep researchs about that . Good luck with your tests :D

Mykk
Posts: 99
Joined: Sat Sep 05, 2015 8:28 am

Post by Mykk » Fri Jan 19, 2018 2:31 am

Everything worked out great until the system went into WOT values.

At WOT the DME still used ignition values from part throttle map as planned, however the fueling stopped cell hit tracing from PT fuel table during WOT protocol and with the throttle open the engine did indeed act like a fuel cut and rev'd down.

At first I thought perhaps I goofed at the address change in the Table of Tables. But further inspection showed that the addresses were correct.

I've currently got it set up for all Ign & Fuel values are taken from P.T. tables and the PT/WOT threshold is 254 to prevent WOT switching. It actually is working out pretty good. The car doesn't feel faster but the transitions & cruising feel smoother and Idle has cleaned up.

I'm going to leave it like this for the time being.

Evil
Posts: 133
Joined: Tue Jul 18, 2017 12:53 pm
Location: France

Post by Evil » Fri Jan 19, 2018 7:54 am

Great news!

For the wot tables, maybe is there an unknown thresold anywhere that cause these wot fuel problems. I think both ignition and fuel wot tables could be extended, with complementary scaling of the load scale between PT and wot. Asks some work for "programming" but I believe it should work.
Anyway thank you for the tests, that shows there ls good solutions To work on for boosted applications :D

Evil
Posts: 133
Joined: Tue Jul 18, 2017 12:53 pm
Location: France

Re: MAP conversion for high boosted application

Post by Evil » Wed Jan 30, 2019 2:40 pm

I've boosted my m50b25 with a gt3076. I've done a "housing swap" on my MAF (not an easy thing with the hot wire maf from non-vanos engines ) but maf signal is very unstable at idle and low loads.
The only maf swap for 402/403 dme is the M5 3.8, so I think I will do the TPS/MAP swap.

Mykk, is your setup still working good?

Evil
Posts: 133
Joined: Tue Jul 18, 2017 12:53 pm
Location: France

Re: MAP conversion for high boosted application

Post by Evil » Sat Feb 02, 2019 11:11 am

Did the swap today and a pre-tune (No more idle and wot maps, changed the adresses in the table index, PT/wot switch To 254, all fueling and ignition will work with PT tables).
Good thing To know: the MAP sensor I use is a 2bar from a sierra cosworth, the electrical plug is absolutely the same :D I just have To verify if wiring is ok.
Unfortunately a lot of Snow today so couldn't do some tests ...

Post Reply