I found this too with my RA. I forgot about it for awhile and never got around to posting about it.

One thing I want(ed) to experiment with is forcing the car to run higher loads with the MAP tables to see if it would increase SST clamp pressure as all my SST logs have shown increase in clamp pressure depending on load.

At some point I'll get around to trying that out. I actually hope it works since we RA owners do not have S-Sport so there becomes an issue with not enough clamp pressure at some point in making more power, which I am now at. (The modes also increase pressure from what I have logged).

 

It's gotta be RPM related. Looks like it flipped to the MAP table above 4500 rpm - pretty much exactly. I've seen a 4500rpm-mark jump on mine too.

Rich

 

Interstingly enough, what I "felt" seemed to be SST related, almost like the clutches were slipping.... I don't think they were... what I think is when the load shot up there was some adjustment going on between the requested torque and calculated torque... I did not, however, see any change in throttle plate angle so it may have just been the rich mixture slow down i was feeling.....

Bryan, if you send me your data item list for the sst logging stuff I'll be happy to get a some more logs and see whats going on.... might be safer on a lower power, stock turbo car like mine rather than your 5.0 killer

 

yup, I'm sure it's a litte more complicated but there definitely seems to be an rpm component to it. I guess based on what I saw you really want the two values (map load and maf load) to stay very near one another in order to keep smooth transitions.

I would imagine there is some logic such that if the error between the two values gets too great, it defaults to using one or the other... it would make sense to favor the higher load value in a fail-over scenario as the higher load dumps more fuel and less timing.

 

Indeed! I had a track test today, and was checking for knock after each run. Revised a few areas in terms of fuel and timing (very minor adjustments).

On the last run, just one instance of knock occurred (ks=7). It was during spool-up, exactly as load crossed from below 4500rpm to above. If the load dropped suddenly at that point, it would lean out slightly, and the timing chosen would be more advanced.

After these MAP/MAF conversations, I found myself leaving alone the timing table... and instead started thinking of smoothing the load, so it doesn't jump around. There's really no way to tune timing/fuel "perfectly" if load drops 7 or 8 points "instantly".

So I'm keen to hear of peoples' experiences and approaches to smoothifying the load calculations.

Rich

 

Well, from looking at my logs, I can figure it out up to the point where it jumps up to the map load... and even at that point you can make an educated guess... problem is the stock RA turbo is not stable under boost over 20psi or so.... some of your boost curves look just like mine peak around 22psi, drop to 20.5 or so then comes back to 22 then starts to taper... this alone can play havoc on the load especially if the maf and map loads are too far apart.....

I would imagine the process involves some of what you have already done.. set the boost to a low point (wastegate pressure for starters) increase the map tables at that boost to make sure it is the higher of the two.... do a few logs, adjust the map table to be very near the maf load that is logged..... lather rinse repeat... however, this does not address the jump that seems to occur, at least with the ralliart.

 

This is my first full revision of the map load tables. it's mostly just smoothing the stock table getting rid of the peaks and valleys.

 

I've always assumed that Mitsu made that map look that way for a reason.

 

I try not to get too distracted by the boost curve wobbles now - it's just the nature of the "error correction" style of boost control the ECU uses. I spent ages trying to get it "perfect"... but what you gain in one area, you give away in another. Maybe I'll get back to fiddling with WGDC and BTEL at a later date, when I want to get all obsessive again!

Fortunately, for the purposes of dialling in various "other" things, setting the solenoid duty cycle to X percent bypasses all that error correction stuff!

The boost "ceiling" approach does seem to be the best way to get a ruler-straight boost curve... serving as input to the MAP calc.

Once I've learned a little more about the 3xMAP tables, and the MAF table, I reckon I'll give this a shot too. Good luck!

Rich

 

Actually, it has nothing to do with error correction, i have had ec turned off for several weeks now just running straight duty cycle with a 3port. I think the issue is the gate port is too large for the turbo... at higher boost anyway... gate opens, turbo slows down so much that boost drops then the turbo has to spool up again... I've got two things left to try, upgraded bov and putting a turn or two on the actuator rod.

Anyway... made a few pulls on the updated MAP load tables this am and the car felt great, as smooth if not smoother than ever.... load seemed a little more stable, fueling was a little more stable than previous maps.... will do more testing over the weekend.

 

I smoothed out my VE/MAP tables as well awhile back and it makes a pretty substantial difference in smoothness and logging smoothness.

One thing I posted about awhile back is when I did my turbo swap from the stock RA turbo to the BBX lite, when I kept the RA VE/MAP values the car would run too low load values for the boost being used and would also trigger the ASC light every time at WOT, even with ASC off, copying over the Evo VE/MAP values cleared it up and it ran proper load values.

I've also changed these tables for the stroker builds. The cars would always run odd until I did that.

 

Yeah,
It's really strange, these are not true VE tables as you can tell from the shape of the graphed tables. They almost remind me of boost comp tuning with AEM. Way back in 2008 when i was working with ecutek and had issues post MX-1 cam install the Ecutek guys mentioned that the ecu used a combination of maf and speed density that made tuning for cams etc a PITA.... Now I see what they were talking about.

It seem the main thing is to make sure the maf load and the map load are in agreement with each other within some degree of acceptable error.

Once again this points to the need to be able to log both load values so that the required changes can be made.

To your point about upping the clutch clamping pressure Bryan... I would assume that if both the maf and map tables were adjusted accordingly, one could easily accomplish this.... So long as any preset load limits that may exist in the tcu are not exceeded.

I doubt I will come across the need to up the clamping pressure while on the X turbo.... but you never know.. If I want to push the limits and spool the turbo as early as possible.... I may play around with upping both sets of tables once i do the swap.

 

Hi Golden / Tephra, are you able to provide the 3XMAP table address for ROM ID 53050009? Thanks.

 

added

 

should be the same as 53040010