Boost Control discussion thread | AEM
58 posts / 0 new
Last post
admin
Offline
Last seen: 5 years 5 months ago
Joined: 07/21/2011 - 02:08
Boost Control discussion thread

This thread is dedicated to the topic of boost control in the AEM Infinity EMS system. Feel free to post anything and everything related in this thread, including questions, comments, suggestions, examples, screenshots, etc.

Because Infinity's boost control is so configurable and powerful, we'd love to hear about all the creative strategies being utilised by our customers and tuners!

To start this off, below is an example of boost control on an AWD time attack vehicle (not traction limited) that doesn't need boost-by-gear (or speed). However, there is an AEM 12-position switch (p/n 30-2056) installed so the boost can be adjusted on the fly.

 Notice that there are also several failsafe strategies that will limit boost if the charge air temperature is too high (heatsoak) or if the coolant temp is out of range (engine too cold or overheating). In addition, the boost target mapping for the 12-position switch only allows certain (higher) boost levels if the Ethanol content is high enough to support the boost. The boost target is also varied based on throttle position, to allow the driver to use the throttle for linear power delivery, which makes the car much easier to drive on a road course where part-throttle is common. To aid in spool, the duty-cycle is increased at lower RPM where the engine cannot attain full boost, and at lower boost pressures, again to keep the wastegate closed as long as possible to improve response.

The PID system on the Infinity is extremely fast and you'll notice no Derivative Gain (BoostGain_Kd) was needed.

[hr]

The example below is a boost strategy for a high horsepower, severely traction limited drag car that runs on varying mixes of Ethanol. Boost compensated for changes in baro pressure (weather, altitude, etc) and changes in exhaust backpressure. Typically, higher duty cycle is needed to retain boost level as baro pressure is reduced and EBP is increased. However, due to the Infinity's advanced PID system, you could theoretically get away without having these sensors and still have consistent boost control by relying on boost feedback.

Failsafe strategies include injected pressure (this is the pressure delta between the fuel pressure and manifold pressure). On a manifold-pressure-referenced fuel pressure regulator, injected pressure should theoretically remain constant at all times, as the pressure regulator directly combats changes to manifold pressure by changing fuel pressure in a 1:1 ratio. However, if there is a pressure regulator failure, or a fuel flow issue (weak pump, restricted lines, etc) injected pressure can drop. The boost control is set up to automatically reduce boost in this condition. Side note: Due to the Infinity's model-based fuel calculations, any variation in injected pressure will also be compensated for in fueling.

You'll notice that the boost target increases with each gear, as it should. Also, with higher Ethanol contents, more boost can be run safely, but of course we only want this in the higher gears as we're traction limited. However, it was found on this car that higher ethanol contents (at the same boost level) caused an increase in torque due to the fuel/tuning, and this caused additional wheelspin. Therefore, boost target is actually [i]reduced[/i]slightly in lower gears and higher ethanol contents to combat this tendency.

Beau

outlawtuning
Offline
Last seen: 7 months 1 week ago
Joined: 05/31/2014 - 04:30
It looks like that infinity

It looks like that infinity tuner is currently locking the X and Y-axis to default sensor/channel types (eg for BoostTargetTable2 Y is InjPressure and X is FlexContent), without the ability to change them to a type of our choice inorder to implement a strategy.

I would very much like to experiment on a Gear/WheelSlip BoostTarget Table, how did you manage to change the X-Y axis channels?

 

AEM_NS
Offline
Last seen: 17 hours 30 min ago
Joined: 06/06/2014 - 09:31
You can assign the axes for

You can assign the axes for both the base duty and target tables in the Setup Wizard under Boost Control.

outlawtuning
Offline
Last seen: 7 months 1 week ago
Joined: 05/31/2014 - 04:30
ya, i noticed after i posted

ya, i noticed after i posted the question blush

However, only prediefined channels can be used, it's better than any other ECU out there but still limited [at least for the capabilities of the framework you are using]

Give the masses and tuners the ability to use thier imagination, abstraction is the key [while you're developing the next firmware version] smiley

AEM_NS
Offline
Last seen: 17 hours 30 min ago
Joined: 06/06/2014 - 09:31
Is there something specific

Is there something specific that you'd like to reference but can't in the current configuration?

outlawtuning
Offline
Last seen: 7 months 1 week ago
Joined: 05/31/2014 - 04:30
[quote=AEM_NS]

AEM_NS wrote:

Is there something specific that you'd like to reference but can't in the current configuration?

Ya, boost target table, X -> gear, Y -> Slip%

PST
Offline
Last seen: 4 years 6 months ago
Joined: 08/04/2014 - 18:44
How about using CO2 in the

How about using CO2 in the wastegate dome with 2 solenoids in a drag racing application? This link shows the common way to plumb in the co2:http://www.google.com/imgres?imgurl=http%3A%2F%2Fi103.photobucket.com%2F...

Precision Speed Technology

AEM_NS
Offline
Last seen: 17 hours 30 min ago
Joined: 06/06/2014 - 09:31
[quote=outlawtuning]

outlawtuning wrote:

Ya, boost target table, X -> gear, Y -> Slip%

Currently, this isn't supported as our current traction control strategy trims fuel and/or ignition which gives a more immediate response over targeting a lower boost target if slipage occurs.  There has been some discussion about implementing a function that will trim boost based on slip but this is more likely to be rolled into traction control than boost control.  That way you still maintain the full flexibility of the current boost targeting setup versus having to lose a target reference to a TC function.

Sick SRT4
Offline
Last seen: 5 years 11 months ago
Joined: 08/13/2014 - 05:16
[quote=PST]How about using

PST wrote:
How about using CO2 in the wastegate dome with 2 solenoids in a drag racing application? This link shows the common way to plumb in the co2:http://www.google.com/imgres?imgurl=http%3A%2F%2Fi103.photobucket.com%2F...

 I agree, co2 boost control is becoming more and more popular 

GGM
Offline
Last seen: 5 years 9 months ago
Joined: 10/30/2014 - 04:29
Hi all. I will soon be on a

Hi all. I will soon be on a Dyno with my first Infinity. Looking forward to using this ecu. I am hoping its as good as It seems to be.

I work with lots of different ecu's and Pcm's with most of the units being Cosworth.

I would where pos use turbo speed as a target not pressure. This is more feed forward and in my experiance gives much faster boost control, especialy when running high compression boosted engines.

Anyone ever used G as a boost control axis? Another take on previously mentioned slip.

 

PST
Offline
Last seen: 4 years 6 months ago
Joined: 08/04/2014 - 18:44
Another good one would be

Another good one would be waste gate position targets. Wish AEM would be more proactive in this discussion, as lacking response about co2 use is the only thing preventing me from pulling the trigger on an infinity system for our current NMRA true street build.

Precision Speed Technology

sportbikeryder
Offline
Last seen: 2 weeks 3 days ago
Joined: 05/23/2014 - 05:39
No way to control 2 solenoids

No way to control 2 solenoids and use them as a pressure control in the software currently, onlt a bleed style control.

AEM_BB
AEM_BB's picture
Offline
Last seen: 1 year 5 months ago
Joined: 06/04/2014 - 11:53
The next firmware release

The next firmware release will have expanded boost control function, and we're looking into boost contol based on wastegate "can" pressure as opposed to simply MAP. Please list specific details/requirements on this and I'll have the engineers look into it.

blacksaleen95
Offline
Last seen: 1 year 2 months ago
Joined: 06/03/2014 - 19:32
+1 on the ability to use a

+1 on the ability to use a dual solenoid boost control setup with C02. Boost Control setups like the Holley Dominator & AMS1000 are standard in the domestic Drag scene and if the infinity could over similar control I think you would gain more interest from that crowd.

VitViper
Offline
Last seen: 3 years 7 months ago
Joined: 11/12/2014 - 02:04
I wish we had at

I wish we had at BoostTargetTable3 and BoostTargetTable4 with configurable axis. I would like to be able to apply extra protection strategies (currently doing flexcontent/gear & throttle/oiltemp)

AEM_BB
AEM_BB's picture
Offline
Last seen: 1 year 5 months ago
Joined: 06/04/2014 - 11:53
The next firmware update will

The next firmware update will have more selectable inputs into the boost tables.

 

Also, don't forget that you can always set up any available LS output as "General Purpose" and run a secondary solenoid. It will run as open loop but you could easily add inputs like RPM vs MAP.

joe@kpracing
Offline
Last seen: 2 years 5 months ago
Joined: 01/05/2015 - 06:37
any update on when the new

any update on when the new firmware will be aviable? I need c02 boost control.

kb58
Offline
Last seen: 1 year 11 months ago
Joined: 10/29/2014 - 09:21
+1 on adding more factors to

+1 on adding more factors to configure boost control. In your two examples at the top, both setups have inputs that would be good on my setup: fuel pressure, MAP, throttle, ethanol content, engine speed, coolant, air temp, gear... that's 8 inputs!

jetnbyu93
Offline
Last seen: 4 years 7 months ago
Joined: 03/17/2015 - 04:40
+1 I am also interested in

+1 I am also interested in looking at Co2 on my AEM series 2, I have had nothing but positive results with Co2 on my previous nlr ams 1000 boost controller. If theirs any help/details on how to get the Co2 to work can someone please respond to this thread or PM me?? 

 

kern303
Offline
Last seen: 5 years 3 months ago
Joined: 04/23/2015 - 08:38
2001 s2000 how does the mac

2001 s2000 how does the mac solenoid for boost control need to be wired into the ems? can i wire it through the purge valve like if i had a V2 ems? (ive searched and searched the infinity system is still so new with s2000's i cant find any info on it)

 

 

thanks for your help!

AEM_BB
AEM_BB's picture
Offline
Last seen: 1 year 5 months ago
Joined: 06/04/2014 - 11:53
Pin 7 in the AUX connector

Pin 7 in the AUX connector included in the 2000-2005 S2000 PnP jumper harness is hooked to pin LS1 in the ECU, which is typically used for boost control. Attach one side of the solenoid to pin 7, and the other side to pin 8 (+12v) of the AUX connector. More info on the AUX connectors and pinout can be found in the instruction manual included with the harness, which is also available on our website under Support --> Instructions.

PST
Offline
Last seen: 4 years 6 months ago
Joined: 08/04/2014 - 18:44
With co2 boost control, the

With co2 boost control, the co2 source is regulated to 50-60psi.
The regulated co2 is fed to the top can of the waste gate via the boost control solenoid. The EMS needs to monitor and actively try to control this top can pressure via the bcs.
The bottom can of the waste gate is fed manifold pressure.

The EMS needs to allow the user to manipulate boost pressure by either targeting specific "dome" or top of waste gate pressure, or by MAP. The latter being most desireable.

With Holley, the preferred solenoid setup uses to solenoids, one to increase dome pressure, one to decrease. Doing it this way uses less co2.

I'd imagine that in AEM fashion, you would have a boost target table, and a duty cycle table, then use PID control to offer closed loop duty cycle regulation.

Holley does a great job with their setup, but it does lack the refinement that AEM has always seemed to offer when it tackles a specific tasks.

Burnout boost target, launch boost target (when on the Trans brake), then as for boost when going down track.. The drag racers like to target boost by time, or boost by gear (with gear recognition from engine rpm/vehicle speed), or my favorite is just a simple boost by rpm and tps.

Precision Speed Technology

SCHNELLENGINEERING
Offline
Last seen: 10 months 2 days ago
Joined: 03/31/2015 - 11:37
So I have an odd situation

So I have an odd situation currently, was wondering if anyone had any ideas on how to do this.  IMO the only real way to do this would be to have a separate Duty table but maybe there is another compromise.

Turbo RSX making 850+, I programmed base duty using 4th gear (it's a close ratio box 1-4, 5th and final is stock).  

My issue is this, even with feedback enabled I haven't been able to dial in the controller to compensate for the very large load difference between 4th and 5th, and I constantly overboost.

I feel that simply the load difference between the two is so large that the base duty needs to come down to keep the boost down when in 5th gear, but then I don't have enough duty for the lower, shorter gears.  Again trying to mess with the feedback control, I haven't really been able to get to a happy medium.

Does anyone have suggestions?  Is there a way to trigger a 2nd duty table for 5th gear only maybe where I can offset the main base duty table?

It's a 4 port solenoid BTW, target boost approx 33-34 psi.

Thanks!

Gpro
Offline
Last seen: 3 months 2 weeks ago
Joined: 08/31/2014 - 04:33
Can't you set a 2nd boost

You could set a 2nd boost duty table by gear and remove duty in 5th gear

i.e. -8 duty in 5ths gear

SCHNELLENGINEERING
Offline
Last seen: 10 months 2 days ago
Joined: 03/31/2015 - 11:37
You're right actually, just

You're right actually, just seen that.

Now for a less obvious question... When is the 2nd table active?  When all the conditions are true?

ie. 2nd duty table only 5th gear row is filled out and when in 5th the table is automatically factored in?  Or is there a switch elsewhere to enable that table?

Thanks

AEM_NS
Offline
Last seen: 17 hours 30 min ago
Joined: 06/06/2014 - 09:31
It's always active.  The two

It's always active.  The two base duty tables and the two target tables are always adding together.

AEM_BB
AEM_BB's picture
Offline
Last seen: 1 year 5 months ago
Joined: 06/04/2014 - 11:53
Side note, if you are having

Side note, if you are having issues tuning the PID boost feedback system, logging the following channels is very useful:

BC_P

BC_I

BC_D

These three channels will report the feedback contribution from each term of the PID loop, which can give you some insight as to which gains to adjust. These channels will add up to the final feedback value channel, BoostFB_PID [%]. 

SCHNELLENGINEERING
Offline
Last seen: 10 months 2 days ago
Joined: 03/31/2015 - 11:37
Yes that part is understood

Yes that part is understood and I use them to help dial in the feedback, but in this case the main issue was simply too much duty due to the large difference in gearing.

Cheers!

bry195
Offline
Last seen: 2 years 10 months ago
Joined: 05/21/2015 - 20:01
could you create a table for

could you create a table for changing "I" based on Vehicle speed? If "I" is not enough you could add more variables.

AEM_BB
AEM_BB's picture
Offline
Last seen: 1 year 5 months ago
Joined: 06/04/2014 - 11:53
Are you referring to "I" as

Are you referring to "I" as the Integral term of the PID feedback for boost control?

bry195
Offline
Last seen: 2 years 10 months ago
Joined: 05/21/2015 - 20:01
Yes. if you want a more or

Yes. if you want a more or less responsive system with dynamic changes in loading the way we do it in the industrial world is to load multiple tuning parameters that shift response. or use a formula. seems like an array based on vehicle and speed could change the integral oe others on the fly. 

Pages