RomRaider

First post updated with new file to work with the higher sampling rates of "fast poll" option in the logger.

Thanks, these tools are invaluable - I've manually adjusted the CL/OL spreadsheet as well, given the CAN logging volume of data.

two updates in a day...
new file has default graphing range extended

Feature request:

1. Plot max achieved boost with RPM vs throttle position, to populate a boost target table. The objective is to minimize Turbo Dynamics error with initial WGDC. I would use the plot to adjust WGDC and flatten boost vs throttle response.

2. Continuous dyno plot - max achieved torque with RPM vs input throttle position. The data can be used to flatten throttle pedal response vs torque output.


Trying to minimize the binary throttle pedal response into something useful..

You can use the Boost interpolation sheet to plot out the boost/WGDC response at part throttle. You don't really want Initial WGDC to be too close to the final required value anyways since turbo takes time to spool.

Take logs at 50%, 75% and 100% throttle and you have all you need for filling out the WGDC table.



Not sure what you mean by that. Below wastegate boost, throttle opening controls boost, not WGDC.

The throttle plate angle is missing from that worksheet, that would make it useful. You would tune it by disabling Turbo dynamics completely, populating a guess at a initial WGDC table, the worksheet would make a table of RPM vs throttle plate opening angle with axis markers of the boost target map. The user adjusts the initial WGDC to achieve a flat (or whatever shape they want) boost map, when this is done the result of the worksheet is pasted into the boost target map.

I'm defeating integral windup quite simply. I don't know why other don't do this, it's the only strategy I found that works for control large turbo's. All you need is an initial WGDC table in your ECU to do this. Anyway - IMHO the initial WGDC should be as close to the correct steady state as possible, and the boost target table should reflect that steady state value. This actually works for small turbo's with OEM boost control like a VF40 (that has a response time that's close to the integration interval of the ECU, ~100ms or less), but it's difficult coming up with accurate tables. Even if the opinion is to start lower, it's simple to subtract a percentage from the whole table once you have the table. You would arrive at the table by crunching large amount part-throttle logs.


I want to flatten torque response with regard to throttle input so the throttle pedal is linear/useful. Fixing part-throttle boost response would be step 1, and this step 2. If all positive acceleration put into a table which has RPM on one axis and the throttle input position on the other, you can use this to fixup the DBW throttle maps. Those maps even use some value called "torque". I think most of it will be fixed by getting the part-throttle boost response right though.

I got a little spoiled by a car with a linear throttle/torque response... The throttle in my subaru has basically had most of the torque output somewhere around midway, the rest of the pedal travel does relatively little. Some people like this (*ahem* sport sharp mode) but it's terrible.

Throttle Opening is in the spreadsheet; it is one of the "required" entries to run the macros.

WGDC required for boost changes from gear to gear. There are comp tables for air temp and altitude but not all ecus have per gear compensation.

Growing buttons Excel 2003? They start big and get bigger almost every time I click on them.

Also I can't seem to get any of the buttons to work in Excel 2011 for mac, even though they are formatted OK. Anyone else?

Saw that a few years ago...if you use a screen resolution that is not the "native" value for the LCD screen, the button sizes will get all messed up.

Just tried it on my laptop...if excel is running and you change the screen resolution, it gets messed up.

Don't know about Excel and Mac...MS took macros away from mac office and supposedly added it back..who knows. changed the security settings to allow macros?

Yeah, it's probably VMware, arbitrary resolution.



Macro's are allowed. They actually errored out on a few non-related lines and I commented them out. Cursor turns to 4 arrows over the buttons (well all embedded controls). Can't find any context or menu item for any more details. I can create new buttons in developer mode - they look like OSX buttons. The buttons in the spreadsheet are rendered like windows buttons. They almost appear to be just images that have no functionality, although I can't seem to modify them in any way.

Airboy...

I just want to commend you for your work here. It is nice to see you still updating (fast polling) and helping others. Congrats on just passing the 5 year anniversary of this thread btw!!

dm

Holy crap, five years, really?

If you get a chance at some point in the future, could you look into an issue on the manual plot graphing. On the data page, where you select Y1 and Y2, if you have too many parameters, you cannot scroll all the way down in the lists.

Also, it would be awesome to be able to plot afr error relative to final fueling base rather than target AFR.

Once again, thanks for this spreadsheet. Airboy ftw!

That bug with the controls has always annoyed me. I don't know of a real fix for it. You could make the list box longer so all items fits into the list. To do this you have to enable the "Control Toolbox" by right-click on the tool bar area (blank) to bring up the list. Then go into "Design mode" (should be first button). Once in design mode, you can click on the list box and resize it. Exit Design mode and you should be good to go. Resave the file to keep the changes.



I'll look into it this winter.

I am using open office. The import csv button does not work with macros enabled.

Is there an alternate way to load the log I have??

Please help.