Elevation Correction

[suberimakuri]

Not done internally.
The baro curve you adjust is replaces the barocorr value in calculation.
There is no hidden baro internally.

The code needs to get current baro reading from somewhere and that is from the sensor you choose. Either a permanent baro sensor or an initial reading at engine startup from the included onboard sensor.

Based on the reading the ecu will look at the baro curve for a correction value.

[Yves]

That would mean my curve is right as it does account for the change in airdensity when baro changes.

[suberimakuri]

If you're fueling stays good with altitude changes then yes you're good to go man.

[jsmcortina]

So the correction for altitude is done internally ? Is this right.

Not it is not right.

If so the barocorr curve is a corr on top of this if needed to include exhaust effects and such...

That's what the barocorr curve is for and it isn't on top of any other baro calc, it IS THE CALC.

As whittlebeast said, measured MAP is just that - however it occurred. That's the whole point of "Multiply MAP". That should allow for the majority of the change. The barocor curve is to allow for other stuff not measured by the MAP sensor.

James

[kaeman]

I happen to live in an area where I can be in the mountians 5000 feet higher than I live and tune the car at that elevation in about 40 minutes. once I have my car running well at my altitude where I live, I then get the car loaded up on a trailer warm up the engine and take note of the idle afrs and the afr at about 2000 rpm, then I hall it up to a higher elevation, start the engine let it settle into a good idle and adjust the baro-correction curve to achieve the readings I had at my base altitude, then I move up higher and repeat the procedure, I do this for several points and then when I drive the vehicle on trips the baro correction makes the corrections needed to have the vehicle run well.
I have been using the single map sensor, so I get the new baro reading each time I start the engine. I am changing over to using a second map sensor and will then be using the two independent sensors setting, but my baro correction settings should still be the same and will work continuously rather than having to restart the vehicle when I make a large change in altitude.
Good luck with your baro correction strategy.

I have found that as I went up in altitude I have had to add fuel, not large amounts, but in reno I was running way lean first time I went there, I ended up needing about 17% baro correction to get back to my normal readings. I live at 200 feet elevation so am pretty much at the sea level area of the barometer....

[Yves]

So the correction for altitude is done internally ? Is this right.

Not it is not right.

If so the barocorr curve is a corr on top of this if needed to include exhaust effects and such...

That's what the barocorr curve is for and it isn't on top of any other baro calc, it IS THE CALC.

As whittlebeast said, measured MAP is just that - however it occurred. That's the whole point of "Multiply MAP". That should allow for the majority of the change. The barocor curve is to allow for other stuff not measured by the MAP sensor.

James

James,

I agree on what you say, when you are running a pure speed density setup. However, when using ITB load, the tps is factored in and then the whole game changes.

[Yves]

I happen to live in an area where I can be in the mountians 5000 feet higher than I live and tune the car at that elevation in about 40 minutes. once I have my car running well at my altitude where I live, I then get the car loaded up on a trailer warm up the engine and take note of the idle afrs and the afr at about 2000 rpm, then I hall it up to a higher elevation, start the engine let it settle into a good idle and adjust the baro-correction curve to achieve the readings I had at my base altitude, then I move up higher and repeat the procedure, I do this for several points and then when I drive the vehicle on trips the baro correction makes the corrections needed to have the vehicle run well.
I have been using the single map sensor, so I get the new baro reading each time I start the engine. I am changing over to using a second map sensor and will then be using the two independent sensors setting, but my baro correction settings should still be the same and will work continuously rather than having to restart the vehicle when I make a large change in altitude.
Good luck with your baro correction strategy.

I have found that as I went up in altitude I have had to add fuel, not large amounts, but in reno I was running way lean first time I went there, I ended up needing about 17% baro correction to get back to my normal readings. I live at 200 feet elevation so am pretty much at the sea level area of the barometer....

How high is Reno ?

[Dennis_Zx7r]

James, I agree on what you say, when you are running a pure speed density setup. However, when using ITB load, the tps is factored in and then the whole game changes.

MAP is still multiplied in when using MultiplyMap, as is recommended in the general settings ("99% of installs..."). Even if ITB load doesn't decrease, at 100%TPS a MAP reading of 80kpa (ambient pressure) would still reduce fueling by 20% compared to a reading of 100kpa MAP. 100kpa -> Calculated fuel multiplied with 1; 80kpa -> multiplied with 0.8

[Yves]

James, I agree on what you say, when you are running a pure speed density setup. However, when using ITB load, the tps is factored in and then the whole game changes.

MAP is still multiplied in when using MultiplyMap, as is recommended in the general settings ("99% of installs..."). Even if ITB load doesn't decrease, at 100%TPS a MAP reading of 80kpa (ambient pressure) would still reduce fueling by 20% compared to a reading of 100kpa MAP. 100kpa -> Calculated fuel multiplied with 1; 80kpa -> multiplied with 0.8

Ok, I see what you mean.

So in this case the curve accounts for effects of exhaust back pressure and possibly intake wave tuning. Since these are dependant on rpm, load and on the intake side TP position I don't think you can correct for that with the curve in question. I leave it to 100% then and let ego take care of any corrections if needed.

[TheSilverBuick]

How high is Reno ?

4500Ft. Drove by there last week going to Lake Tahoe. 700 miles round trip and around 4,000 vertical feet of change over the course of the drive. Baro correction worked like a champ.

[suberimakuri]

Yves.
Regardless of all that theory, I suggest you still consider a baro curve.
If your ego correction has enough that's fine but next time you do a big elevation change you should check and adjust baro if necessary to reduce ego correction.

[sedd]

I recently logged a trip from 450 feet elevation to 6400 feet elevation. My baro sensor went from around 99 to around 78. I have not been able to figure out all of the data yet. I am trying out the new Megalog HD viewer, getting used to it still. Mostly I want to make a post so I can find this set easier.......... I don't think I can add much to this yet.

I used a set of baro corrections that was similar to ones suggested in the posts. My afr error was pretty small with my baro correction that I used. I think my baro correction is pretty close.

I had set the limit for correction at 118.8% and 85 baro. I used 100% at 101.3 baro. My curve was almost linear.

A quick look shows that for idle at elevation and baro around 78, the Idle valve steps opened to 30, which is 15 more than when at around 94 baro. It appears there was not enough air flow past the throttle body so the steps had to open to flow more air? So I assume a slightly higher map due to the larger steps?

I didn't see a linear change in the map at idle that follows the change in elevation baro. I show 46 map at idle and 78 baro with 56 map at idle and 94 baro. I get a 10 map change with a 16 baro change. Possibly the Idle valve steps account for this, not certain of that.

I need to look at the temperature change yet, it was a lot cooler ambient.

[suberimakuri]

Good testing.
I don't have the need for idle valves but I'm not surprised about step difference in your setup, good thinking to check that.
We will see the opposite with boost duty valves I think. Less duty for same boost.

Your idle kpa does seem to be moving linearly as a percentage of baro.
46/78=0.58
56/94=~0.58

So that's all good I think.

[whittlebeast]

Sedd, can you post a log of the climb? I would love to see how the system responded as you went up the hill.

[sedd]

I have tried to create a word document with a screen shot included. The actual picture and the log files are too large to attach. I took several logs, I think 5 total over the 6 hour duration to get there and then longer to make the climb.

I purchased the megalog HD and it has the option to upload all the logs and combine them. that is what is shown by the picture. I would really like to find a way to limit the data by hits.

I want to also to look at only values with 100% ego correction. that way if afr is off it is only due to baro and my table errors.

I attached the msq that should have been in place at the time of those logs. I use speed density control only, on a natural aspirated engine.

let me know if you can view this and have some ideas on how to interpret that data.

based on this set of posts I am more confused about how the "multiply map" affects this. I have seen notes that say this is a multiplier of map/baro and I don't understand that. the set up manual formulas don't say anything about dividing by baro so I will dig more to see what is really going on.

interesting about how the ratio of values is the same for what I found.

[KenK]

10/06/16

Gents:
This is what I have done to date to setup the engine to run at different elevations and temperatures.

The temperature issue had to be dealt with first. Since I have no way to control the MAT at different elevations, I need to get the temp issue under control first. At this time of the year here in southern California, the ambient temperature can run from the low 50s (F) early in the morning to as high as 118 (F) in the afternoon. This worked for me as I an running an open car and have no plans on driving the car when the temps go below +30 (F) or above 188 (F).

To get around some of the cam related issues, I ran the motor at 2500 rpm under no load. I setup the VE table up such that all of the VE numbers around the 2500 rpm point were the same so I could the most repeatable and stable data as possible. During the testing I ran the motor as the ambient temperature warmed up every 5 degrees or so. Using the data collected, I made the necessary changes to the "Temp Comp" table and repeated the tests. The corrected temp table worked pretty good. I will wait another month or so when the ambient temps drop to the low 40s (F) and high 30s (F) and add these points. The curve at this point is quite flat. Once I have all of this done, I plan on going back to the desert in November where I will be able to do some testing between about 2000 feet to about 8600 feet ASL. In February, I plan on going to the Imperial Valley where the elevation runs about 200 feet below ASL.

During this testing, I will adjust only the elevation curve.

Once this is done, I hope to have a motor that is happy anywhere between -200 feet to +8600 feet and from 30 (F) to 120(F).

I do have a question for the MS developers though. How does the MS software handle conditions where the temp, Baro pressure or MAP pressure go above or below the highest or lowest point in the available table or curve??

KenK

[suberimakuri]

Sounds good Ken.
MS uses the last entry in table. This can come up with Ve and ignition tables too.

[sedd]

KenK from KenS:

Could you explain a bit more about the details on the temperature table corrections. Did you use the MAT/CLT table for the corrections? Or the MAT air density table? or some other? I have not tried to correct for temperature yet. I suspect my run to the mountain might have been affected by the much cooler temps at elevation but have no proof of that. I have noticed that may VE table seems to move around based on ambient. I run autotune and get the table set only to have it tune to another value on the next auto tune. So maybe I have the cart before the horse? I should have gotten temps corrected first and then move towards baro? So I had an opportunity to gather data and at worst case it may not be too useful.........

Did you look at the afr changes or correction and then lean or fatten the pulse based on that.

Where is your air temp sensor, just curious.

Hood up or hood down? I have noticed that my MAT varies each time my electronic fan turns on and off from the radiator with the hood down is much more variation. Similar increase in temps during a longer stop and just setting around, then it cools quite a bit while driving. It is clear the air flow thru the engine compartment purges trapped hot air that is under the hood. the radiator fan has an impact as well. I have some data somewhere from sensors I had placed next to my MAP sensor and another placed inside the engine bay at the inlet to the horn of the air cleaner intake. I felt that the only time I should tune off idle is when the car was moving for a while. Idle should be tuned with the engine hood up and not when the fan was running.

Any heat soak indications, did you wait a while and run the engine a bit before making the analysis?

[sedd]

Getting back to baro issues.

I did a search on multiply map and there was conflicting info about what this did to the fuel required calcs. I see the set up manual shows the pulse width calc for Speed Density is multiplied by MAP along with other corrections. Does the calculation take the MAP value and divide by 100 for the calculation? Some posts said MAP was divided by Baro for the calculation.

Assume a 50 map reading, would the req'd fuel would be multipled by 50/100 or 0.5 times and of course those other corrections?

I think that understanding this is important because my data shows a map shift that occurs. If the VE table varies enough there could be an effect of the table values due to the map shift. IE: if the engine requires 56 map to run closer to sea level it might require 46 map at elevation. if the ve values vary a lot at those two points at the same RPM there could be afr changes due to that.

[KenK]

10/07/16

KenS:
Good questions!!

In my case I am running an air cooled engine in an off-road car. As such, I have no hood. There is no doubt in my mind that varying conditions under a hood will cause significant changes in the ARF.

At this time, I am working with the MAT air density table.

As for the location of the temp sensor, it is in the air filter. This caused me some trouble as the temp sensor usually used for this does not like the presence of gasoline fumes. In the presence of fuel fumes, the temperature jumped around wildly (-250 to + 250). I had to go to the type of sensor usually used for CLT applications. I know there are some heat soaking issues when using this type of sensor, but in my case, the heat soaking issue was far less problematic than the fume problem. Again, my MAT sensor is out in free air and virtually unaffected by engine heat.

As for running the engine a bit before taking the data, yes I did. Since I am running an air cooled engine, the engine cools down quite quickly when not running and also warms up quickly. I allowed the engine to run until the head temp was up to 200 degrees before starting the test. At this point, the WUE is completely out of the picture. The car was also in my shop out of direct sun light during the testing so solar heating was not a problem.

As for tuning the Baro, in my humble opinion, you can't do anything with the Baro until you have the temp comp under control. Since we have no choice but to drive up a mountain to change Baro, and the temperature will change as we do so (about 5-10 degrees per 1000 feet) the temp comp must work correctly. I did the temp comp testing in my shop over several days and at different times of the day but in my shop where the Baro doesn't change much. I still need some colder temps as the lowest temp I have tested at to date is 55 degrees F. As for the high end, I do have several data points above 100 degree F. Since my engine is completely exposed to outside conditions as is the driver (me) 120 degrees is plenty hot enough. The low 30s will be cold enough for my application for the same reasons as stated above.

To get around the VE moving issue, I am not using the EGO to correct the AFM, just as an indicator and I also set the VE at the point where I was running the engine such that they were all the same number. To get around the AFR dilution problem as well as the cam dilution issues, I ran the engine right at 2500 rpm. On my engine, the engine is real unstable at 2000 rpm as this is the point where the engine starts to come on the pipe. Anywhere lower than 2000 rpm, I have a dilution issue and the AFR is wildly erratic. 3000 rpm would be better but I'm not trying to wake up the dead and at that rpm the engine really starts to get load.

To make corrections based on temp, I made a table in Excel logging RPM, MAP, B-Volts, AFR, Fuel Flow, Baro, and MAT. Once I had a good table I through out the outliers and I generated an air density table based on the AFR.

The table came out pretty flat but not necessarily a straight line. I did have a dip in the 60 degree area. The reason???

I hope that helps. After I get a couple of lower temps in the data base, we're off to the mountains.

KenK