The problem with that is when you change altitude without cycling the key. You have locked yourself at one value and you may reach a very different value before restarting the car.

Probably not a major issue if only limited to idle and maybe cruis conditions.
The FD Rx-7 supposedly did something similar to grabbing MAP on start up and I've heard it is part of the reason that engine failures were so common. I'm not an RX-7 guy though, so I don't know how true that is.

I can also say altitude cause a lot of other problems though. All the load values like what controls closed loop, knock, idle control etc is all based on baro+IAT compensated load. A load of 100 is atmospheric pressure, regardless of altitude stock. With the SD patch, at 4500', a load of 85 is atmospheric pressure now and you end up going into closed loop while under boost and knock control starts late, etc. It leaves you A LOT of maps to adjust to accommodate the change.

Also, it makes this idea of matching load before and after the SD patch completely unrealistic. If you match load with baro locked at 101 kPa, the car is EXTREMELY rich. It just doesn't work at all. You could lock the Baro to local ambient pressure and it may help a lot, but then going 1:1 on the MAPVE table messes everything up too.

 

Yes, this is one possible solution, but my recollection is that baro is used in the idle control routines, and thus MAP at startup may not handle situations where someone drives to higher altitude and idle the car. What we probably need is someone to log and document SD idle response at different altitudes to determine if baro has a big impact on idle performance. (hint, hint)

 

I'm not sure I'm following this...maybe I'm just not thinking well right now.

From what I remember, jcsbanks made the SD patch pretty simple, replacing maf volume with map, to get to a master load again. If you match the load before and after SD, why would your fueling be different, irregardless of the baro chosen?

The baro is incorporated into the map reading. The load is calculated from the map VE and then the RPM VE. So, as long as you end up at the same load before and after you should be in the same cells before and after, with identical timing and fueling.

I understand that there may be special cases, such as startup, etc, that may be affected by different locked baro values that weren't taken into consideration with what load is used, when, etc, etc, etc...but the overal fueling (cruising, WOT) shouldn't be affected.

Am I misunderstanding what you are saying or are you just talking purely about the startup issues that you saw and not WOT fueling?

 

From what I have seen on here, it seems like a car at altitude sees a higher load for a given manifold absolute pressure. My car on 18 psi (gage) was around 260-270 load. That gauge pressure would be similar to 15.5 psi at sea level, and that boost level at sea level seems to correspond to roughly 220 load?

My absolute pressure is lower, yet my load is higher.

If I try to match load up, my car runs ~20% rich because I'm no longer having the baro compensation pull out fuel to make up for the lower MAP.

Trying to pull that fuel out using the VE tables obviously doesn't work, because you drop your load.


Think about what MAF Hz is, as that is actually what has been replaced and not load. It's almost strictly based on airflow velocity. Lower atmospheric pressure means a given mass airflow rate will have a higher velocity because the air is less dense. MAF frequency goes up, which later becomes load.

Load is used to look everything up, but baro is actually used as a final IPW correction.

If I try to match up load, I get higher loads for a given airflow condition, but now I don't have baro compensation taking the extra fuel out at the end.

 

I will have to look at the whole subroutines that jcsbanks disassembled and described, but I don't think it's as you are explaining.

Baro is needed in a maf based car because the maf is only measuring volumetric airflow. The temp and pressure are needed and why the MAF has both sensors incorporated. The master load and mass airflow (temp+baro corrected when using maf) is then calculated accordingly.

However, when on SD, the baro is no longer needed in these calculations. The mass airflow is calculated directly from pressure (map) and temperature (MAT).

I don't know why you would see higher load at higher altitude, though. I would have thought it would have been the opposite. Which load are you talking about? Either way though, if you match MAF based load to SD load, I still don't see how the fueling can be different. Again, I don't want to go through the threads that jcsbanks has posted as well as mrfred advanced fueling again, but I think maybe something else may be going on or something in particular to your case.

You mentioned that baro is used in the final IPW calculation. If that's true and hasn't been taken into account with the SD patch, then that makes sense.

 

Why not, as an easy test, just lock the baro at your 'normal' value and do some before and after logs. Maybe some at different altitudes as well.

 

I'm pretty sure that single change is going to make the car undriveable and I really don't have the time to mess around with it this week.

Pretty sure changing it to 85 kPa is going to cause the load to increase 20% and the AFR will go that much more rich as well, just based on Mrfred's thread and some older posts from Mrfred and JCSBanks. The correction will try to account for the change in baro but the MAF frequency calculated will not change because it's based on the VE tables and the MAP voltage.

This issue was even discussed in the original SD thread.

 

My intention was to take a baro sample before cranking, i.e. in ECU configure routine, the sub before main loop (88590015 @ 116F8), where initial value could be taken.

 

Its definitely one of the possible solutions and would probably work well for driving around at typical day-to-day elevation changes, but it wouldn't be the best solution for someone who drives over a wide range of altitude without restarting the motor.

What worries me about having these kinds of options is the chance that many different SD variations are going to appear as people implement different improvements in different ways, and then its going to be a big hassle to provide any level of support. We really need to agree how its going to be done before coding anything and setting it loose in the wild.

 

Testing various methods before deciding on one particular method would be beneficial.

Creating a standardized version in the end that could be loaded up and make the car fairly driveable would definitely be desireable.

 

Do we have anyone here who competes in hillclimbs?

 

Says the man who lives in a state where there is only 1000 ft between the lowest and highest elevation.

 

Only when there is over 2' of powder to be had.

 

I increased the MAP VE around atmospheric pressure to pick up a larger VE change from closed throttle to open throttle. I went from 71 to 80 at the 88kpa load site.

I had also noticed the large change in VE from 88kpa to 120 kpa and the relatively small VE changes below 88kpa, but that was were A LOT of tuning got the best results.

Anyway, it was just meant to be a quick check to see if it helped down low. It did completely the opposite of what I would have thought.

It made the lean issue below 2000 RPM WORSE. Odd that richening up the problem area with MAPVE made the lean problem worse.

It also had some other interesting side effects.
Car idles so rich it won't stay in closed loop? (I didn't even mess with the area where the car idles???)
The ECU didn't cut fuel on decel above 3000 RPM.
Car was knocking all over the place in transitions.
The ECU was very slow to go into closed loop in cruise conditions.
Car wouldn't start without giving it throttle. (typical of my car when it's warm, but not hot or cold, it just didn't want to start at all without throttle)

I reset the ECU and put about 60 miles on the changes. That 60 miles pissed me off so damn bad as the car just ran like absolute ****every where.
Needless to say, back to the old map.

 

Just a quick question...I forgot if you had answered this before:

When you switched to SD, did you base your tune off of a maf based baro+temp compensated 3D chart, or did you just start tuning the VE tables?

Your findings above do seem very strange. I'm wondering if it has to do with you adjusting different cells than you are actually in and/or that your fixed baro of 101 instead of 85 is really causing a problem ith the baro and baro+temp corrected loads.

Remember from mrfred, that the loads used are as follows:

So, for fueling at least, baro+temp compensated load is almost always used (except at startup it appears...load <20). If you were trying to match loads pre and post SD to uncompensated, that may be where your issues began. I know you don't want to hear it, but I think completely retuning with your 'true' baro may help.

The only other obvious thing, which I doubt is wrong, but figured I would ask, is that you are using the correct map 16bit scaling for your particular map sensor. If not, your map VE table will make no sense and you will be adjusting the wrong cells.