Boostpill Base Map users - check your boost levels
Dec 2, 2011, 08:41 AM
#1
Evolved Member
Thread Starter
Boostpill Base Map users - check your boost levels
This thread is for Ralliart owners who have fitted a boost pill, and have flashed their ROM with one of razorlab's fine Base Maps.
Introductory Blurb
As you've heard, when it comes to tuning, every vehicle is different. When you flash your Ralliart's ECU with a base map, you should:
Note: There are other resources on EvoM about boost tuning, ranging from detailed threads about the wonders of ECU-controlled boost, how to patch for different styles of boost control. There's Merlin's Ralliart Tuning Guide. The amount of information can be rather headache-inducing. But I'm going to keep this all really simple, and focus purely on Ralliarts running a razorlab Base Map for boost pill.
And I'll focus on Base Map v27, as that's the version razorlab is rolling out right now.
So... let's begin.
Log Your Boost - Accurately
NOTE! Much of the "logging" advice below has been made redundant by Rax Fast Logging. As of Base Maps v31, RAX Fast Logging is built in. You should no longer need to muck about with the following instructions!
------------------------
With the default EvoScan setup, you'll probably have a boost scaling formula like this:
Boost Eval: (x/4*0.19347)-14.5
The "14.5" constant is adjusting absolute boost pressure by your supposed atmospheric pressure. If you're not driving where atmospheric pressure is 14.5psi, your logged boost will not be accurate.
You either need to set up your "constant" so it is correct, or log an additional "Baro" parameter, and base your boost calc on that. ie.
Baro Eval: x*0.07251887
Boost Eval: (x/4*0.1953125)-[Baro]
Here's how the two items look in my Evoscan Edit Data Settings window...
Note: You should ensure "Baro" is logged before "Boost".
Log Your Boost Target
The v27 Base Maps have switched to Direct (psi-based) Boost Control. You can therefore log the dynamic boost target along with your actual boost, and see how close you are.
For each Ralliart ROM ID, here is the"Boost Target" variable address to use:
53600009: RequestID 23808b63
53610010: RequestID 23808b5b
53610013: RequestID 23808b69
55590006: RequestID 23808b75
56940007: RequestID 23808b99
56950006: RequestID 23808b99
58030005: RequestID 23808bb9
Boost Target Eval: x*0.19347
ResponseBytes: 1
When you put all this together, you'll see something like this:
Checking Boost Levels - The Easy Way
There are many different ways of setting, managing and fine-tuning boost. The v27 Base Maps work use one particular method, and the notes below work well for that specific method. It may not translate effectively to other boost tuning approaches... but I'm keeping it tightly focused here! So here goes...
The initial WOT boost peak is the key to fine-tuning your overall boost.
Check it like this:
You don't have to do all of them in one drive - it's not an endurance event.
In fact, when you're first starting out it would be better to try just one test, to confirm you're in the ballpark with reference to your logged boost target. If that test is good, try different rpm starting points.
The initial "peak" boost level is predominantly determined by your BWGDC table. Sure, there is error correction processing active in the ECU, but your spool-up occurs so fast there's no chance of it "catching" an overboost condition and getting the wastegate to somehow open on-target. Won't happen.
On initial spool-up, your wastegate-open point will be driven by your BWGDC Table, pure and simple.
So perform WOT spool-up tests, and check your initial peak.
Adjusting Boost Levels
Say you've performed a few WOT spool-up tests, and have found your initial boost peaks are consistently 1.0psi above target.
This means you should reduce the values in your "Reactive Solenoid Base WGDC Map" tables.
Open your EcuFlash, load up your ROM.
Scroll down until you find the "Turbo" section.
Look for the three "Reactive Solenoid Base WGDC Map" tables. Check the box to open them.
With tables like this, hit Ctrl-G to open up a 3D graph. You can spin it around with the mouse or arrow keys.
One Editing Approach - Applying a MULTIPLIER
Say our hypothetical case, we need to broadly reduce BWGDC a bit. Choose one of the maps, and select all of the cells, ie. from top left to bottom right. Highlight them all.
Next, hit *, which is the "multiply" hotkey. Enter a value... by which all of the map cells will be multiplied. If we want BWGDC to drop by 5% across the board, enter 0.95 - new values will be 95% of old values.
Make the same style change to all three tables. Note that the "Low Gear" table differs from the pair of "High Gear" tables.
Another Editing Approach - SUBTRACTING Some BWGDC %
It is just as easy to lower the entire BWGDC map by a fixed amount. This may yield better results than applying a "multiplier".
Choose one of the maps and select everything except the topmost row and leftmost column (we want these left as zero).
Next, hit the minus key (-), which is the "incremental subtract" hotkey. All the selected BWGDC cells will drop by 0.5%. Repeat until your map cells have dropped to where you want them.
Make the same style change to all three tables. Note that the "Low Gear" table differs from the pair of "High Gear" tables.
Save the ROM.
Then reload and double-check all your work! There is a good "Compare ROMs" option under "File". Load up old and new ROMs, use the "Compare", and ensure the changes are exactly and only the ones you expect.
And that's it in a nutshell.
The above examples adjust the entire BWGDC table - obviously quite a broad-brush approach... but you get the idea!
It's not a big step going from "whole of table" adjustment to making more refined modifications, eg. selecting different RPM "rows" and raising/lowering BWGDC as required.
Good luck!
Rich
Introductory Blurb
As you've heard, when it comes to tuning, every vehicle is different. When you flash your Ralliart's ECU with a base map, you should:
- Log to start with... and then keep on logging!
- Review your logs carefully (and regularly).
- If you're not sure what to look for in your logs, learn.
Note: There are other resources on EvoM about boost tuning, ranging from detailed threads about the wonders of ECU-controlled boost, how to patch for different styles of boost control. There's Merlin's Ralliart Tuning Guide. The amount of information can be rather headache-inducing. But I'm going to keep this all really simple, and focus purely on Ralliarts running a razorlab Base Map for boost pill.
And I'll focus on Base Map v27, as that's the version razorlab is rolling out right now.
So... let's begin.
Log Your Boost - Accurately
NOTE! Much of the "logging" advice below has been made redundant by Rax Fast Logging. As of Base Maps v31, RAX Fast Logging is built in. You should no longer need to muck about with the following instructions!
------------------------
With the default EvoScan setup, you'll probably have a boost scaling formula like this:
Boost Eval: (x/4*0.19347)-14.5
The "14.5" constant is adjusting absolute boost pressure by your supposed atmospheric pressure. If you're not driving where atmospheric pressure is 14.5psi, your logged boost will not be accurate.
You either need to set up your "constant" so it is correct, or log an additional "Baro" parameter, and base your boost calc on that. ie.
Baro Eval: x*0.07251887
Boost Eval: (x/4*0.1953125)-[Baro]
Here's how the two items look in my Evoscan Edit Data Settings window...
Note: You should ensure "Baro" is logged before "Boost".
Log Your Boost Target
The v27 Base Maps have switched to Direct (psi-based) Boost Control. You can therefore log the dynamic boost target along with your actual boost, and see how close you are.
For each Ralliart ROM ID, here is the"Boost Target" variable address to use:
53600009: RequestID 23808b63
53610010: RequestID 23808b5b
53610013: RequestID 23808b69
55590006: RequestID 23808b75
56940007: RequestID 23808b99
56950006: RequestID 23808b99
58030005: RequestID 23808bb9
Boost Target Eval: x*0.19347
ResponseBytes: 1
When you put all this together, you'll see something like this:
Checking Boost Levels - The Easy Way
There are many different ways of setting, managing and fine-tuning boost. The v27 Base Maps work use one particular method, and the notes below work well for that specific method. It may not translate effectively to other boost tuning approaches... but I'm keeping it tightly focused here! So here goes...
The initial WOT boost peak is the key to fine-tuning your overall boost.
Check it like this:
- In 3rd gear, SPORT MANUAL mode, coast down with no throttle.
- When you reach a set rpm, apply full throttle.
- Let boost build, peak and level out.
- Get off the throttle.
- Review.
You don't have to do all of them in one drive - it's not an endurance event.
In fact, when you're first starting out it would be better to try just one test, to confirm you're in the ballpark with reference to your logged boost target. If that test is good, try different rpm starting points.The initial "peak" boost level is predominantly determined by your BWGDC table. Sure, there is error correction processing active in the ECU, but your spool-up occurs so fast there's no chance of it "catching" an overboost condition and getting the wastegate to somehow open on-target. Won't happen.
On initial spool-up, your wastegate-open point will be driven by your BWGDC Table, pure and simple.
So perform WOT spool-up tests, and check your initial peak.
Adjusting Boost Levels
Say you've performed a few WOT spool-up tests, and have found your initial boost peaks are consistently 1.0psi above target.
This means you should reduce the values in your "Reactive Solenoid Base WGDC Map" tables.
Open your EcuFlash, load up your ROM.
Scroll down until you find the "Turbo" section.
Look for the three "Reactive Solenoid Base WGDC Map" tables. Check the box to open them.
With tables like this, hit Ctrl-G to open up a 3D graph. You can spin it around with the mouse or arrow keys.
One Editing Approach - Applying a MULTIPLIER
Say our hypothetical case, we need to broadly reduce BWGDC a bit. Choose one of the maps, and select all of the cells, ie. from top left to bottom right. Highlight them all.
Next, hit *, which is the "multiply" hotkey. Enter a value... by which all of the map cells will be multiplied. If we want BWGDC to drop by 5% across the board, enter 0.95 - new values will be 95% of old values.
Make the same style change to all three tables. Note that the "Low Gear" table differs from the pair of "High Gear" tables.
Another Editing Approach - SUBTRACTING Some BWGDC %
It is just as easy to lower the entire BWGDC map by a fixed amount. This may yield better results than applying a "multiplier".
Choose one of the maps and select everything except the topmost row and leftmost column (we want these left as zero).
Next, hit the minus key (-), which is the "incremental subtract" hotkey. All the selected BWGDC cells will drop by 0.5%. Repeat until your map cells have dropped to where you want them.
Make the same style change to all three tables. Note that the "Low Gear" table differs from the pair of "High Gear" tables.
Save the ROM.
Then reload and double-check all your work! There is a good "Compare ROMs" option under "File". Load up old and new ROMs, use the "Compare", and ensure the changes are exactly and only the ones you expect.
And that's it in a nutshell.
The above examples adjust the entire BWGDC table - obviously quite a broad-brush approach... but you get the idea!
It's not a big step going from "whole of table" adjustment to making more refined modifications, eg. selecting different RPM "rows" and raising/lowering BWGDC as required.
Good luck!
Rich
Last edited by richardjh; Mar 14, 2013 at 04:41 PM. Reason: Revised instructions on table edits.
Dec 2, 2011, 08:44 AM
#3
Evolved Member
Thread Starter
Guys and gals,
Please let me know if you think there is anything more I should add, or anything unnecessary I should take out. And point out any speeling mistaks (as it's very late now!).
Hope it's useful and readable...
Rich
Please let me know if you think there is anything more I should add, or anything unnecessary I should take out. And point out any speeling mistaks (as it's very late now!).
Hope it's useful and readable...
Rich
Dec 4, 2011, 03:05 PM
#5
Evolved Member
Thread Starter
Yes, you certainly can... like this...
edit: Revised the above formula to use value "0.048828", to align the maths with the earlier Evoscan example. It's a wee bit more accurate.
You can probably speed up logging a lot by using "sampgroup=..." declarations for items that don't change very often. Baro is one of these.
Give things like Baro, IAT, MAT, Speed, Octane, LTFT_Idle, LTFT_Cruise a "sampgroup=1" declaration... and make sure Baro is first in the list and comes before the Boost item.
Rich
Code:
paramname=Boost
paramid=0x??????
scalingrpn=x,0.048828,*,Baro,-
You can probably speed up logging a lot by using "sampgroup=..." declarations for items that don't change very often. Baro is one of these.
Give things like Baro, IAT, MAT, Speed, Octane, LTFT_Idle, LTFT_Cruise a "sampgroup=1" declaration... and make sure Baro is first in the list and comes before the Boost item.
Rich
Last edited by richardjh; Jan 31, 2012 at 12:58 PM. Reason: Revised scaling constant in the "Boost" example.
Dec 15, 2011, 10:50 PM
#6
Evolved Member
Thread Starter
I've added a bit, explaining how to select a chunk of the BWGDC table and use "-" to reduce values. In hindsight, this is more natural than applying a "multiplier".
Rich
Rich
Jan 31, 2012, 01:03 PM
#7
Evolved Member
Thread Starter
I've revised a couple of things...
Post #1 now has example images taken from Evoscan Edit Data Settings, showing how my "Baro" and "Boost" items are configured.
Post #5 has an example of standalone logging of "Boost". Its scaling multiplier wasn't ideal. Now fixed.
If anyone wants further clarification on this stuff, just shout.
Rich
Post #1 now has example images taken from Evoscan Edit Data Settings, showing how my "Baro" and "Boost" items are configured.
Post #5 has an example of standalone logging of "Boost". Its scaling multiplier wasn't ideal. Now fixed.
If anyone wants further clarification on this stuff, just shout.
Rich
Trending Topics
]
Apr 10, 2012, 07:15 PM
#12
Newbie
Join Date: Oct 2011
Location: New Jersey
Posts: 13
Likes: 0
Received 0 Likes
on
0 Posts
I am running about 2-3psi below target boost, my max boost seems to be about 20.3 psi. Does it seem likely I am leaking boost since its so low from target? Or will changing some values to increase boost solve the problem?
Thanks!
Derik
Thanks!
Derik
Apr 11, 2012, 02:52 AM
#13
Evolved Member
Thread Starter
Well, every car is different... and the Base Map simply shoots for "near the middle of the bell curve". From the logs we saw while dialling it all in, I'd say more vehicles fall slightly below target boost - but that's better than the other way round.
Yes, it's likely some fine-tuning of your BWGDC tables will yield excellent results. That's what this thread is all about. Good luck!
Rich
Yes, it's likely some fine-tuning of your BWGDC tables will yield excellent results. That's what this thread is all about. Good luck!
Rich
Apr 17, 2012, 05:20 PM
#14
Newbie
Join Date: Oct 2011
Location: New Jersey
Posts: 13
Likes: 0
Received 0 Likes
on
0 Posts
I have made adjustments to those tables and attached a picture of the RPM vs Boost vs Boost Target graph and the horsepower torque graph. Does this look good? It drops off later than the boost target, should I be concerned about that?
Are there any other tables in the ROM that I could begin editing to increase the performance?
Thanks!
Derik
Are there any other tables in the ROM that I could begin editing to increase the performance?
Thanks!
Derik
Apr 17, 2012, 06:51 PM
#15
Evolved Member
Richard mentions that the ECU struggles to initially get boost under control as the boost level is rising very rapidly.
This can be helped to some extent by reducing the waste-gate correction interval - there are two 1D parameters here, high range and low range.
As scaling may vary with different deffinitions, I would suggest setting these to the second smalest value.
Depending on your xml, this may translate to either 2 or 200mS.
This can be helped to some extent by reducing the waste-gate correction interval - there are two 1D parameters here, high range and low range.
As scaling may vary with different deffinitions, I would suggest setting these to the second smalest value.
Depending on your xml, this may translate to either 2 or 200mS.