| View previous topic :: View next topic |
| Author |
Message |
Aurélien
Joined: 30 Jun 2009
Posts: 137
|
 Posted: Wed May 30, 2012 5:56 am Post subject: How much breakpoints ? |
 |
|
Hello !
I'm back with a question that keep me awake...
How to know how much breakpoint should we use for a particular application ?
On a basic reflection :
- Less breakpoint -> Less cell to tune -> quicker.
- More breakpoint -> More cell to tune -> more resolution/ ability to fine tune.
But if we go further, everyone will admit that you will have to make compromise if you have too little breakpoints, and so you will LOOSE time trying to get the best compromise.
On the other hand, too much breakpoints will be overkill and you will LOOSE time too.
So how are you doing ? starting with a relatively average number like 16*16 then adding breakpoints in some aera if you need ?
I've been working a little in order to reference how much breakpoints are using OEM ecu calibrator, those are just above for curious people !
Car : Porsche Cayenne 956 4.8T
Ignition
RPM breakpoints : 16
Load breakpoints : 16
Load type : Relative load
Fuel ( Aim lambda )
RPM breakpoints : 16
Load breakpoints : 12
Load type : Relative load
------------------------------------------
Car : Porsche 997 3.6T
Ignition
RPM breakpoints : 24
Load breakpoints : 16
Load type : Relative load
Fuel ( Aim lambda )
RPM breakpoints : 12
Load breakpoints : 16
Load type : Relative load
------------------------------------------
Car : Ferrari 458 Italia
Ignition
RPM breakpoints : 18
Load breakpoints : 12
Load type : Relative load
Fuel ( Aim lambda )
RPM breakpoints : 16
Load breakpoints : 12
Load type : Relative load
------------------------------------------
Car : Audi RS4 B5
Ignition
RPM breakpoints : 16
Load breakpoints : 12
Load type : Relative load
Fuel ( Aim lambda )
RPM breakpoints : 16
Load breakpoints : 12
Load type : Relative load
------------------------------------------
Car : BMW 135i
Ignition
RPM breakpoints : 20
Load breakpoints : 16
Load type : Relative load
Fuel ( Aim lambda )
RPM breakpoints : 16
Load breakpoints : 12
Load type : Relative load
------------------------------------------
Car : Mitsubishi lancer evolution VIII
Ignition
RPM breakpoints : 20
Load breakpoints : 19
Load type : Relative load
Fuel ( "Aim" lambda )
RPM breakpoints : 15
Load breakpoints : 18
Load type : Relative load
------------------------------------------
Car : Megane 3 RS
Ignition
RPM breakpoints : 20
Load breakpoints : 16
Load type : Relative load
Fuel ( Aim lambda )
RPM breakpoints : 20
Load breakpoints : 11
Load type : Relative load
------------------------------------------
Car : Citroen DS3-R
Ignition
RPM breakpoints : 16
Load breakpoints : 12
Load type : Relative load
Fuel ( Aim lambda )
RPM breakpoints : 16
Load breakpoints : 12
Load type : Relative load |
|
| Back to top |
|
 |
StantonWarrior
Joined: 10 Dec 2010
Posts: 188
Location: UK
|
| Posted: Wed May 30, 2012 7:01 am Post subject: |
|
|
It depends on that particular application.
If its a motorsport type vehicle, where drive-ability isn't of particular importance is, then you will typically want higher resolution under full load areas, specifically where you encounter large changes in airflow e.g due to cams/cam timing, boost pressure etc.
If its solely a road car, then particular attention will be paid to areas which concern shunting etc, to make the drive very smooth.
RL type calibrations will really be a reflection of the torque model, and as such its not a fair comparison to a relatively simple after market system.
i.e on OEM stuff I very rarely alter axis span, unless targeting a higher IQ/Boost/whatever elsewhere int he cal.
But on aftermarket stuff I will regularly alter axis points to accommodate a specific area of interest. |
|
| Back to top |
|
|
Warpspeed
Joined: 05 Jun 2006
Posts: 483
Location: Melbourne
|
| Posted: Wed May 30, 2012 4:30 pm Post subject: |
|
|
I suppose it depends on how rapidly Ve and combustion conditions change with rpm.
If you have some very steep changes due to peaky induction/exhaust tuning, variable valve timing, or some massive boost pressure variations, then higher resolution is going to be required in certain critical areas than a much more mild state of tune. |
|
| Back to top |
|
|
MrDomino
Joined: 01 Oct 2009
Posts: 188
Location: Indianapolis, IN
|
| Posted: Thu May 31, 2012 6:41 am Post subject: |
|
|
One
_________________
I am a Mechanical Engineer. |
|
| Back to top |
|
|
Roberto Arano
Joined: 22 Nov 2005
Posts: 358
Location: colorado
|
| Posted: Sat Jun 02, 2012 8:21 am Post subject: |
|
|
Also depends on what MODE you are using . Forexample with a turbo car MAP x RPM will need more resolution that MAP/EBP x rpm. Since the extra EBP input will allow a more comprehensive algorithm , the fueling table can be "flatter"/less resolution.
_________________
www.circuitse7en.net
http://stores.ebay.com/Circuit-Se7en |
|
| Back to top |
|
|
MrDomino
Joined: 01 Oct 2009
Posts: 188
Location: Indianapolis, IN
|
| Posted: Sun Jun 03, 2012 9:06 am Post subject: |
|
|
Is EBP turbine in pressure? I agree that MAP/EMAP and RPM is the right way to do things although I wonder how a Delta-P and RPM algorithm would work since the pressure rise or drop across the engine will largely dictate airflow.
_________________
I am a Mechanical Engineer. |
|
| Back to top |
|
|
Roberto Arano
Joined: 22 Nov 2005
Posts: 358
Location: colorado
|
| Posted: Sun Jun 03, 2012 6:44 pm Post subject: |
|
|
| MrDomino wrote: | | Is EBP turbine in pressure? I agree that MAP/EMAP and RPM is the right way to do things although I wonder how a Delta-P and RPM algorithm would work since the pressure rise or drop across the engine will largely dictate airflow. |
Yes EBP=EMap=TIP etc
EBP is just another variable that can be used in the mass flow calculation that can help flatten the VE table. So with it you may have very similar "~VE"numbers for all MAP ranges in any RPM column, but still have "~VE" numbers that are specific to each rpm column (Because MAP/EMAP doesn't account for breathing efficiency due to rpm,cam,intake tuning ranges)
Extreme example for a VE table : one row of Map cells x 12 rpm columns.
Only theoretical, still have injectors that aren't linear vs PW etc etc
_________________
www.circuitse7en.net
http://stores.ebay.com/Circuit-Se7en |
|
| Back to top |
|
|
Tomak
Joined: 20 Apr 2008
Posts: 427
Location: •Calgary •Alberta
|
| Posted: Wed Jun 06, 2012 11:27 am Post subject: |
|
|
use as few points as you can to get the job done.
_________________
www.Dynomotive.ca
AEM Factory Trained, Accel EMIC, Haltech Trained, Advanced GM EFI, Diablo CMR Mopar Dealer, SCT TUner - Viper/SRT10, Delta Force Dealer.
Analog DD 450
Mustang MD250 - gone but not forgotten. |
|
| Back to top |
|
|
giskard
Joined: 24 Oct 2005
Posts: 251
|
| Posted: Wed Jun 06, 2012 2:07 pm Post subject: |
|
|
By MAP/EMAP, do you mean, MAP divided by EMAP, or MAP minus EMAP or some other formula?
On my motor, there are areas where EMAP is less than MAP, and there are som areas where EMAP is greater than MAP.
I don't see how MAP-EMAP will help flatten the VE table. Maybe MAP - (EMAP divided by factor)... |
|
| Back to top |
|
|
Roberto Arano
Joined: 22 Nov 2005
Posts: 358
Location: colorado
|
| Posted: Wed Jun 06, 2012 3:11 pm Post subject: |
|
|
| giskard wrote: | By MAP/EMAP, do you mean, MAP divided by EMAP, or MAP minus EMAP or some other formula?
On my motor, there are areas where EMAP is less than MAP, and there are som areas where EMAP is greater than MAP.
I don't see how MAP-EMAP will help flatten the VE table. Maybe MAP - (EMAP divided by factor)... |
Yes , I meant the inteligent use of MAP with respect to EMap in order to get the net pressure differential across the Cylinder Head ant any boost vs backpressure combination.
_________________
www.circuitse7en.net
http://stores.ebay.com/Circuit-Se7en |
|
| Back to top |
|
|
Tomak
Joined: 20 Apr 2008
Posts: 427
Location: •Calgary •Alberta
|
| Posted: Wed Jun 06, 2012 3:42 pm Post subject: |
|
|
| Roberto Arano wrote: | | giskard wrote: | By MAP/EMAP, do you mean, MAP divided by EMAP, or MAP minus EMAP or some other formula?
On my motor, there are areas where EMAP is less than MAP, and there are som areas where EMAP is greater than MAP.
I don't see how MAP-EMAP will help flatten the VE table. Maybe MAP - (EMAP divided by factor)... |
Yes , I meant the inteligent use of MAP with respect to EMap in order to get the net pressure differential across the Cylinder Head ant any boost vs backpressure combination. |
Never done it this way.
So you have a MAP versus rpm table, then some sort of Map/Emap muliplier?
So say at 100 kpa map, and 100kpa emap, ratio is 1:1, no modification to base VE.
Get boost to say 200kpa map, and efficient turbo, so 200kpa emap, still 1:1 so again no modification to base VE.
But if we get 200kpa map, and 300kpa emap, we would decrease fueling in the modifier table to pull fuel ? What is the point of this? Seems complicated, and for any particular engine combo. emap will always be the same at a particualr MAP and rpm.... so its already "baked into" the main VE table.
I guess if you had Cosworths facility where you could simulate ANY map, and any emap, you could then produce a tune that would not need to be changed if the customer changed turbos etc....
What am I missing?
_________________
www.Dynomotive.ca
AEM Factory Trained, Accel EMIC, Haltech Trained, Advanced GM EFI, Diablo CMR Mopar Dealer, SCT TUner - Viper/SRT10, Delta Force Dealer.
Analog DD 450
Mustang MD250 - gone but not forgotten. |
|
| Back to top |
|
|
giskard
Joined: 24 Oct 2005
Posts: 251
|
| Posted: Wed Jun 06, 2012 4:11 pm Post subject: |
|
|
| Tomak wrote: |
What am I missing? |
Transients.
When the wastegate is shut and the turbine is accelerating, and close to target boost, TIP (aka EMAP), will be higher, than steady state. |
|
| Back to top |
|
|
Roberto Arano
Joined: 22 Nov 2005
Posts: 358
Location: colorado
|
| Posted: Wed Jun 06, 2012 5:35 pm Post subject: |
|
|
Transients, like giskard said. And also changes in elevation create higher EMAP to maintain the same target boost (Absolute Pressure not load%)
And if the ECU does it in the background, it's LESS complicated to calibrate the VE table, because it's "flatter" (cells more similiar to each other). Autronic offers the option.
_________________
www.circuitse7en.net
http://stores.ebay.com/Circuit-Se7en |
|
| Back to top |
|
|
Tomak
Joined: 20 Apr 2008
Posts: 427
Location: •Calgary •Alberta
|
| Posted: Wed Jun 06, 2012 5:46 pm Post subject: |
|
|
Transients as in quick "pump shot",
or are you calling a car coming onto target boost a "Transient"?
_________________
www.Dynomotive.ca
AEM Factory Trained, Accel EMIC, Haltech Trained, Advanced GM EFI, Diablo CMR Mopar Dealer, SCT TUner - Viper/SRT10, Delta Force Dealer.
Analog DD 450
Mustang MD250 - gone but not forgotten. |
|
| Back to top |
|
|
Tomak
Joined: 20 Apr 2008
Posts: 427
Location: •Calgary •Alberta
|
| Posted: Wed Jun 06, 2012 5:48 pm Post subject: |
|
|
| Roberto Arano wrote: | Transients, like giskard said. And also changes in elevation create higher EMAP to maintain the same target boost (Absolute Pressure not load%)
And if the ECU does it in the background, it's LESS complicated to calibrate the VE table, because it's "flatter" (cells more similiar to each other). Autronic offers the option. |
And how can this be done in the "background" without a large order of fudge factor?
Every particular cam timing would affect how the engine responds to the pressure differential, so this "map/emap" mulitpler would be different for each particular combo.
Good god, boggles my mind how complicated this can get, for seemingly small results.
_________________
www.Dynomotive.ca
AEM Factory Trained, Accel EMIC, Haltech Trained, Advanced GM EFI, Diablo CMR Mopar Dealer, SCT TUner - Viper/SRT10, Delta Force Dealer.
Analog DD 450
Mustang MD250 - gone but not forgotten. |
|
| Back to top |
|
|
Roberto Arano
Joined: 22 Nov 2005
Posts: 358
Location: colorado
|
| Posted: Wed Jun 06, 2012 9:37 pm Post subject: |
|
|
I don't know how else to explain it. You are correctthe columns (rpm) sites are still going to be different as I think I mentioned earlier that is the effect of the cam(and other rpm dependent VE factors),really. The EMAP factor isn't going to account for that.
It's not more complicated it's less complicated (for the tuner), just like speed density is simpler and more accurate than ALpha-N this is just one more step in a more comprehensive direction.
I think if I could explain it better you would see what I mean....maybe think about how when tuning Speed density the tuner doesn't have to do any math to use the MAP value, it's all done in the software.
hmm maybe I just can't express it....you'll probably figure it out before I find a way to explain it.
_________________
www.circuitse7en.net
http://stores.ebay.com/Circuit-Se7en |
|
| Back to top |
|
|
StantonWarrior
Joined: 10 Dec 2010
Posts: 188
Location: UK
|
| Posted: Thu Jun 07, 2012 5:41 am Post subject: |
|
|
Ive never used this kind of function on any ecu before; can you explain what you have to input to make it engine specific?
As Tomak says, there are MANY combinations of cam angle, lift, etc etc etc that will effect the delta pressure over the cylinder head, and unless the TIP is a real measured value, it is going to be nigh on impossible to quantify a usable theoretical representation of this.
Call me a cynic, but it sounds like more of a selling point than a purposeful functional feature. |
|
| Back to top |
|
|
giskard
Joined: 24 Oct 2005
Posts: 251
|
| Posted: Thu Jun 07, 2012 7:09 am Post subject: |
|
|
| Roberto Arano wrote: | | Autronic offers the option. |
Do you set the formula of MAP and EMAP, or does it have a set formula? Do you know what the formula is?
Does it have a failsafe mode that detects loss of EMAP signal? |
|
| Back to top |
|
|
baldur
Joined: 12 Oct 2007
Posts: 430
|
| Posted: Thu Jun 07, 2012 7:45 am Post subject: |
|
|
| Even just logging EMAP on one of my own machines I've had the 4mm brakeline that connected the sensor clog up twice with carbon and lead deposits after not many hours of running. I think any ECU offering this as a part of the control strategy should have the option of running redundant extra sensors. |
|
| Back to top |
|
|
giskard
Joined: 24 Oct 2005
Posts: 251
|
| Posted: Thu Jun 07, 2012 8:41 am Post subject: |
|
|
| Tomak wrote: | Every particular cam timing would affect how the engine responds to the pressure differential, so this "map/emap" mulitpler would be different for each particular combo.
|
True, but if you map the cam timing to the main fuel table (e.g. MAPxRPM or MAP/EMAPxRPM), then there is a fixed cam timing for a given table entry. This ignores transient in cam timing, but in my experience cam timing settles much more quickly than boost. (thus my comment on transients wrt turbine backpressure aka EMAP) |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
