Advanced theory in excel
#1
Elite Member
Thread Starter
iTrader: (1)
Join Date: Jun 2012
Location: NH
Posts: 9,483
Total Cats: 104
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
I'm stuck and its time for advanced excel functions time.
So I have data coming in columns thats related, IE I have RPM, Wideband, MAP. How do I make a table that has RPM on one axis, MAP on the other, and the average wideband value in each cell displayed. Like if a cell overs 800-1200 rpm and 20-32 kpa it will display the average of all the wideband values that fall in that range. Size of the input data will always be different but assume it'll be in the 10,000 data points area. Does this require a macro? And what would that look like? Making macros are my excel weak-point.
This is the more simple version. The end goal is to have my target AFR table from the tune also in this spreadsheet and make a table the same size as my fueling table in the tune that displays AFR % Error, so I know where and how much to change the tune. I know how to do this part though, its the easy part.
Unneeded but will probably be asked. AEM EMS4 is the ecu in question.
So I have data coming in columns thats related, IE I have RPM, Wideband, MAP. How do I make a table that has RPM on one axis, MAP on the other, and the average wideband value in each cell displayed. Like if a cell overs 800-1200 rpm and 20-32 kpa it will display the average of all the wideband values that fall in that range. Size of the input data will always be different but assume it'll be in the 10,000 data points area. Does this require a macro? And what would that look like? Making macros are my excel weak-point.
This is the more simple version. The end goal is to have my target AFR table from the tune also in this spreadsheet and make a table the same size as my fueling table in the tune that displays AFR % Error, so I know where and how much to change the tune. I know how to do this part though, its the easy part.
Unneeded but will probably be asked. AEM EMS4 is the ecu in question.
#2
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
BTDT, but forgotten how, sorry.
One thing you gotta watch out for, is that you have some kind of marker that the data is "stable", i.e., it's been in that cell for x seconds, so you weed out transients. You need to weed out the effect of throttle enrichment, and the time lag of the O2 sensor (especially at low airflows).
For such large datasets, I would use Matlab, Scilab, or R instead of Excel.
R has the advantage of being designed for statistical analysis (which is what you're doing). You can get fancy and produce a distribution curve (or median and standard deviation), per cell in your map.
One thing you gotta watch out for, is that you have some kind of marker that the data is "stable", i.e., it's been in that cell for x seconds, so you weed out transients. You need to weed out the effect of throttle enrichment, and the time lag of the O2 sensor (especially at low airflows).
For such large datasets, I would use Matlab, Scilab, or R instead of Excel.
R has the advantage of being designed for statistical analysis (which is what you're doing). You can get fancy and produce a distribution curve (or median and standard deviation), per cell in your map.
Thread
Thread Starter
Forum
Replies
Last Post
StratoBlue1109
Miata parts for sale/trade
21
09-30-2018 01:09 PM