HOWTO: Open Loop Boost Control (FG)

[Roland@pcmtec]

When tuning with a different wastegate actuator or boost control solenoid it is best to tune the vehicle in open loop first to get your initial open loop duty cycle correct. Once this is correct you would enable closed loop again

There are 4 modes the boost control works in in the FG.

• Idle/Cruising mode (0% duty cycle)
• Overboost (0% duty cyle)
• Underboost (100% duty cycle)
• Closed Loop (Variable %)

In closed loop the wastegate duty cycle is determined via the following (this ignores the boost adder functionality at WOT and assumes you have zero'd them out)

duty_cycle = auF16463 (predicted open loop duty cycle) * auF0318 (injector duty cycle) * auF16634 (boost multiplier ECT) * auF0315 (Barometric multipler) * auF16475 (desired boost multiplier TPS) + KP + KI

Where KP and KI are the PID outputs (closed loop control).

So knowing the above we can force permanent open loop control by the following:

Setting the PID constants to all zero.

We can then set the underboost value to a very high number so that we never get true underboost (100% duty cycle)

Set the overboost value to a very low number so we never get true overboost (0% duty cycle)

Set the "cruising mode" TP Count to a low number so you never get 0% duty cycle unless you really are at cruise or idle.

 

If you would like a dummys guide change the following items:

auF2645 P1228 Timer - Set this to 1000 for the entire table

auF2150 P1228 - Wastegate Failed Open (Under Pressure) - Set to "No Error"

auF2279 P1227 - Wastegate Failed Closed (Over Pressure) - Set to "No Error"

auF16459 Desired Boost (inHG) set this out of range to a high number like 35

auF16475 TPS Boost multipler - Set this all to 1 other than idle/cruise.

auF3184 Underboost Required For Open Loop - Set this to a high number like 40-100

auF3204 Intake Cam Adder for boost error - Set the degree adder to 0. This is important otherwise your VCT system will command incorrect angles.

auF3264 Exhaust Cam Adder for boost error - set the degree adder to 0. This is important otherwise your VCT system will command incorrect angles.

auF3002 Spark Adder/Delta Change in Wastegate - set to 0

auF0286 Integral gain - Set to 0

auF0287 Proportional gain - Set to 0 (try 1 as well, however most likely will be ignored in open loop)

auF0296 Overboost to trigger open loop - set to -50 to -100

auF16555 Integral Gain - Set to 0

auF16519 Integral max - Set to 0

auF16453 Integral min - Set to 0

auF0289 Proportional Max - Set to 0

auF0288 Proportional Min - Set to 0

auF1193 P1227 Fault Filter - Recommended to leave unchanged.

auF1328 P1227 Fault Filter - Recommended to leave unchanged.

auF1898 - Open loop duty cycle for underboost - set to 40-50%

auF1098 - P1227 overboost error - Recommended to leave unchanged.

auF2625 - Slope due to pre turbing temperature - This function accounts for heat soak into the wastegate actuator. It will modify the final wastegate duty cycle to try and keep a more consistent boost pressure when the exhaust is hot or cold. If you want a completely open loop boost control, set the unitless column to 0. If you leave it as stock, the wastegate duty cycle will change with exhaust temp, however it should give you a flatter boost curve when then exhaust is extremely hot vs cold.

auF16680 Boost Clip based on Inferred Turbo Temp - This is not in all FGs, only the later model vehicles. This will clip the desired boost. If you are using proper closed loop boost control, or you are utilising this table, make sure you raise this table to be higher than what you are requesting in your desired boost table.

auF0286 - Integral Clip (this is the "cruising mode" table which sets duty cycle to 0% below these TPS values. Set the TPS column to ~150 counts at higher rpm to ensure you always get the commanded duty cycle.

 

[Roland@pcmtec]

Will I notice from your uploaded file you have auF2776 set to -100, this may be causing you issues as well. You also have P1228 and P1227 set to "No MIL" which will still trigger them I believe, auF1098 (boost error) is set to 18 as well which may not be high enough.

[will15]

just did all that's recommended in the starting point list, plus changed auf2776 back to 100 and still the same.

alex fg ute pcmtech.tec 4.tec

[Roland@pcmtec]

I will have a look at your file and see if any of the other workshops would be willing to give advice as I've never personally done an FG, though I know lots of people do what you are trying to do.

[Roland@pcmtec]

Can you try setting auF0287 to 1.0 instead of 0.0 ? It may use this as a multiplier still.

Everything else looks good.

edit: Can you use a scanner to see if P1228 or P1227 are being set?

[will15]

I don't know whats going on now I just tried 0 desired boost and still commands 100% dc and it's all ways at the 4000rpm

[will15]

tried 1 for auf0287 still the same, 4000rpm 100dc, no codes

[Puffwagon]

Corrupted file? Does that happen with these cars?

I've had that happen with an aftermarket unit and it wouldn't change the fuelling no matter what I did. I reloaded the file and it fixed it.

Just a thought.

[Roland@pcmtec]

Just now, Puffwagon said:

Corrupted file? Does that happen with these cars?

I've had that happen with an aftermarket unit and it wouldn't change the fuelling no matter what I did. I reloaded the file and it fixed it.

Just a thought.

Unlikely. You would get a checksum error, and if any of the code gets corrupted the car wouldn't even start as it would crash and just constant restart the CPU.

[Puffwagon]

Ah ok.

[Roland@pcmtec]

Will it looks like you have whole number in the duty cycle. Eg 42 will cause 100% duty cycle. 1.00 is 100% duty so scale these numbers back by a factor of 100

[will15]

• I totally over looked that, thanks man that's fixed it. you guys are doing a awesome job thanks for the quick replies.

[Roland@pcmtec]

I didn't even think to look either!

Glad to see it is working now. We will make up some parameter files (a file with presets) for enabling/disabling open loop at some point soon.

[will15]

I have boost control ?

[Puffwagon]

Winning!

[Shaunyg88]

Hi Guys,

Sorry bit of a newbie question. With applying all the above settings, how do you then set your targeted boost setting??

Say i want to run 15 psi?

[Roland@pcmtec]

Depends.

If you are running everything open loop you need to adjust your wastegate duty cycle until you reach your desired boost. If you are running closed loop you need to adjust your desired boost (uses inHG units) and also adjust your open loop wastegate duty cycle otherwise you will get oscillations when going in and out of closed loop/open loop.

The recommended procedure is to force open loop. Get your 15psi via the duty cycle table, then set your desired boost table to match what you are measuring boost wise. Then if you run your car on a freezing cold night it won't over boost.

[Shaunyg88]

Thanks for the reply roland. 

Could you add the tables that need to be modified? Or is it all apart of what's in your original post?

[Roland@pcmtec]

In an FG the wastegate duty cycle is "auF16463 Predicted Open Loop Wastegate Duty Cycle" and Desired Boost is auF16459 "Base Desired Boost Pressure for Turbo Control"

[Shaunyg88]

Thanks for that mate. Do you also have to disable closed loop o2 like in your how to guide?

[Roland@pcmtec]

The closed loop fueling controls the afr to 1.0 lambda, it has nothing to do with closed loop boost control which controls boost pressure to that which is in the desired boost table. 

There are probably 20 closed loop PID controllers in the PCM. 

[Roland@pcmtec]

Bump. Just updated this guide with some more information, an equation and some extra tables.

[Superb]

Hi, need some help.

Followed the guide to enable open loop.

I'm finding that something is not set right, Initially i'm getting full boost like 15-18psi on 1 run then after only 3 psi.

Any help would be appreciated. Looking to have a valet and low boost profile

Also not sure which parameter is to set boost by gear.

Using HAER1UB sprint as base with custom OS

 

F6 Sprint_CustomOS-boost open loop.tec

[Roland@pcmtec]

Sprint is complicated as it has overboost and boost by gear from factory. You are best off disabling all of the temporary overboost multiplier tables and boost by gear tables then enabling our custom os boost by gear tables.

The factory ones work off timers and will only run for a few seconds.

[Superb]

Ok thanks, I will try that.

Already had all WOT waste gate and boost adders set to 0

Set all WOT boost multipliers to 0 now as well

[Puffwagon]

1 hour ago, Superb said:

Initially i'm getting full boost like 15-18psi on 1 run then after only 3 psi.

Is the throttle blade staying completely open while this is happening? The only way I can think of making 3 pounds at WOT via the tune is for the throttle to be closed for torque reduction.

[Superb]

1 minute ago, Puffwagon said:

Is the throttle blade staying completely open while this is happening? The only way I can think of making 3 pounds at WOT via the tune is for the throttle to be closed for torque reduction.

Unsure? I will log that tomorrow and see if that is happening.

Thanks

[Roland@pcmtec]

Always log torque source. It is the most important thing to log. 

Also log map pressure you could have an intake pipe sucking shut. 

[Superb]

19 hours ago, Roland@pcmtec said:

Always log torque source. It is the most important thing to log. 

Also log map pressure you could have an intake pipe sucking shut. 

Is that TQ_ETC_SRC ?

No its not a intake pipe problems, too wet today to test much but I think the car is going into limp mode after a WOT run.

As @Puffwagon said I noticed when this happens I'm only getting 20% throttle. 

If I don't push it, I have boost but i'm still not getting boost control to where I set it. 

Need to spend more time seeing whats going on

[Puffwagon]

As an example of torque reduction, I tried loading the car up on the brake in 3rd gear at low rpm, before logging the run. The car felt very sluggish compared to just planting it with no brake boosting.

When I reviewed the log, the throttle wasn't fully open until just after 4000rpm, which showed as a massive dip in the virtual dyno graph I made.

I haven't gotten around to removing any torque reductions yet and I'm ok with that, as it rarely affects how the car drives, but in your case 20% throttle clearly needs sorting out.

[Roland@pcmtec]

14 hours ago, Superb said:

Is that TQ_ETC_SRC ?

Yes this is the one.

[Superb]

I know this sounds like a dumb question but just for confirmation

0% duty cycle is the wastegate 100% open & 100% dc fully closed ?

which would be in auF16463 0 for 100% open & 1 for fully closed?

 

 

 

[Roland@pcmtec]

It's the bleed valve that is open /closed and indirectly the waste gate. 

1.0 is 100% duty and hence maximum bypass or keeping the wastegate as closed as possible. 

[Superb]

1 minute ago, Roland@pcmtec said:

It's the bleed valve that is open /closed and indirectly the waste gate. 

1.0 is 100% duty and hence maximum bypass or keeping the wastegate as closed as possible. 

Perfect, thanks

[Superb]

So recently I've had a new turbo fitted and trying to enable closed loop boost control.

My tuner only tuned it in open loop and i would like to have some engine protection if anything were to go wrong.

For some reason the over boost flag is being triggered after about 5500rpm

The desired boost table auF16459 is set correctly

If you look at the logs "desired boost target" drops to 24inHg at 5500rpm, like its defaulting to the stock setting. (TestB1)

As a test and still in open loop I changed auF16459 to 25inHg across the table and it still dropped to 24inHg (TestB2)

I'm kind of lost to what is going on.

Any input or advice would be appreciated 

 

 

TestB1.teclog TestB2.teclog F6 Sprint - Open Loop Setup.tec

Edited March 17, 2022 by Superb

[hjtrbo]

How I have my open loop setup is my desired boost is setup about 3psi higher than what I actually see. This keeps it away from any flags etc but if my wastegate fails it will take action. MK1 FG XR6T

Edit:

I think you are hitting the 5 sec timer in auF11827. Upping your desired boost like I do should get you out of that timer.

 

 

85_In Car.tec

Edited March 17, 2022 by hjtrbo

[Puffwagon]

25 minutes ago, Superb said:

The desired boost table auF16459 is set correctly

 

No it isn't, it is actually set up kinda arse about face. That lower line (98.889C) is when the intercooler is hot enough to cook an egg and the boost is basically 20psi. This is a safety row where you pull all of the boost out to mitigate the heat in the engine, not increase it! As unlikely as it is to get to 100c iat, if it did you'd kiss your motor goodbye unless it's on e85.

The wg duty table could be ok in open loop and also might be fine for closed loop. It depends on things and stuff.

Timing wise it's fine for 12psi on 98 so there is a positive thing.

This is as far as I'm gonna touch this one as imho it needs a full retune from stock due to the sheer amount of things that have been touched. Stuff has been switched off or out of range so you won't even know there's an issue until it's too late. It's not all bad but yeah, I would start from stock if I were to fix it.

 

Anyhow I could go on but I've gotta watch netflix with the wife. Good luck with it.

[Superb]

14 minutes ago, Puffwagon said:

 

No it isn't, it is actually set up kinda arse about face. That lower line (98.889C) is when the intercooler is hot enough to cook an egg and the boost is basically 20psi. This is a safety row where you pull all of the boost out to mitigate the heat in the engine, not increase it! As unlikely as it is to get to 100c iat, if it did you'd kiss your motor goodbye unless it's on e85.

The wg duty table could be ok in open loop and also might be fine for closed loop. It depends on things and stuff.

Timing wise it's fine for 12psi on 98 so there is a positive thing.

This is as far as I'm gonna touch this one as imho it needs a full retune from stock due to the sheer amount of things that have been touched. Stuff has been switched off or out of range so you won't even know there's an issue until it's too late. It's not all bad but yeah, I would start from stock if I were to fix it.

 

Anyhow I could go on but I've gotta watch netflix with the wife. Good luck with it.

Hey thanks, yeah I know about the lower line not being correct. I did have a custom os with boost by gear enabled and didn't bother to change it. The lower line was actually the boost curve that was set by my tuner. 18-19psi and I left it there as a reference.

I've got a closed loop config which I didn't upload because I'm getting boost cut after 1 hit through the redline.

Thinking I need to find another tuner 

 

 

 

 

 

 

F6 Sprint_CustomOS - ZF Mk1 - Closed Loop Boost.tec

Edited March 17, 2022 by Superb

[Superb]

1 hour ago, hjtrbo said:

I think you are hitting the 5 sec timer in auF11827. Upping your desired boost like I do should get you out of that timer.

Thanks but it seems I'm pegged at 35inHg/ 17.2 psi in the mid and still dropping to 24inHG/ 11.8 psi up top.

Any change doesn't seem to reflect any increase beyond that.

Something weird going on with the FGX sprint calibration.

[Superb]

@hjtrbo you are right though the timer in auF11827 is coming into play.

Last run down the highway... quick change into 2nd with my foot flat and wheelspin through redline didn't trigger it until I lifted and got back on it

 

Test C.teclog

[hjtrbo]

Dump this in and see how it goes. It's based off your open loop file you posted earlier. Have put the majority of turbo DTC's back to stock and raised up your desired boost table a couple of pounds above where you are running it. Also used the wastegate multiplier on spark delta to force the wastegate open during a gear shift to help reduce the boost spike. Just do a simple 3rd gear pull to 6000rpm so you can see if it works or not.

Disclaimer: I didn't check out the rest of your tune, you're responsible for that. Knock ears and a wideband are mandatory when learning to tune.

13551451_F6Sprint-OpenLoopSetup_Rev1.tec

[Superb]

4 hours ago, hjtrbo said:

Disclaimer: I didn't check out the rest of your tune, you're responsible for that. Knock ears and a wideband are mandatory when learning to tune.

13551451_F6Sprint-OpenLoopSetup_Rev1.tec 1.82 MB · 1 download

Yes 100%

Thanks

[Puffwagon]

@Superb Don't run it to redline until you have verified that the boost isn't gonna skyrocket. Just bring it up slowly in a high gear like 5th so you can easily catch it if it runs away.

[Roland@pcmtec]

We replied to your support ticket. When you were running your low boost tune with 0% duty cycle, you were seeing boost creep to 18psi by redline. Either the gate is not working properly or the spring preload/base pressure is too high.

If you get boost creep to 18psi on 0% duty cycle, you'll never be able to control a flat desired boost at this level, it will always spike.

Then your high boost run, the extra DC brings the turbo to 18psi sooner, but it cannot drop it below 18psi at redline. You've also disabled the overboost action (WG OL OBOOST) as its reporting false. However the DTC warning (no action associated to it) is showing true WG OBOOST, hence even though overboost is occurring its not reducing the duty cycle to 0, however even if you did fix this as per the first log it would unlikely drop the boost by any measurable amount. This is also why your desired boost drops dramatically.

So there are several issues, one which is mechanical and then another which is tune related (eg the overboost not dropping duty cycle to 0%).

 

15 hours ago, Puffwagon said:

 

No it isn't, it is actually set up kinda arse about face. That lower line (98.889C) is when the intercooler is hot enough to cook an egg and the boost is basically 20psi. This is a safety row where you pull all of the boost out to mitigate the heat in the engine, not increase it! As unlikely as it is to get to 100c iat, if it did you'd kiss your motor goodbye unless it's on e85.

It might not be obvious if you are using the compare function (as it will show the axis of your file, not his) but he has boost by gear enabled. So whilst what you said is true with the standard boost tables, with boost by gear its not.

[Puffwagon]

I opened his file and saw that it was filled with 12psi except the lower line. It is all in degrees C, not boost by gear.

[Puffwagon]

Pic from his file in the post where he said, "The desired boost table auF16459 is set correctly".

 

[Superb]

I'd first like to thank everyone for their input.

@Puffwagon I uploaded the wrong file as I was testing reverting back from custom-os as I was trying to rule out software issues.

The changed you pointed out were set incorrectly.

@Roland@pcmtec Yes I received the email from support, thank you.

I wasn't aware I had an mechanical issue with the current tune, I just assumed the wastegate not being large enough to reduce boost at higher RPM in low boost.

It is a modified housing but it's looking like I'm going external gate.

 

 

 

[Puffwagon]

26 minutes ago, Superb said:

It is a modified housing but it's looking like I'm going external gate.

Are you any good with a die grinder? You can use a high speed drill and a long carbide bit too.

Before you go external gate you should check your turbine housing and make sure there is a good port job on it. You want a nice radius going into the wg hole and you can actually raise the roof slightly (wg hole facing up) so there is less height to the angle into it. I leave the back side of the hole square to help with wg priority. That combined with minimal preload should give you a much lower minimum boost. It takes about an hours worth of grinding to go from stock to perfect. Suffice to say, don't touch the seat or it will probably leak.

I recently did this to a pulsar 3584rs with a TS 12psi actuator, decent preload and straight through 3.5 inch exhaust. I could run as low as 13psi the whole way to redline. Same turbo with untouched turbine housing on my territory with a very free flowing custom 4 inch exhaust and it would boost spike to 28psi with 0 duty.

Anyhow there's some info for you and other passers by. It's what I would be doing before spending money and if it doesn't help it's only some wasted time.

Edited March 18, 2022 by Puffwagon
clarification

[Roland@pcmtec]

On 3/18/2022 at 3:55 PM, Superb said:

I wasn't aware I had an mechanical issue with the current tune, I just assumed the wastegate not being large enough to reduce boost at higher RPM in low boost.

It is a modified housing but it's looking like I'm going external gate.

The internal gate should be sufficient for 18psi if its modified correctly. I don't see any need to go external.

[Superb]

On 7/27/2018 at 10:57 AM, Roland@pcmtec said:

auF16680 Boost Clip based on Inferred Turbo Temp - This is not in all FGs, only the later model vehicles. This will clip the desired boost. If you are using proper closed loop boost control, or you are utilising this table, make sure you raise this table to be higher than what you are requesting in your desired boost table.

Found the missing key 😎

[overlap]

Evening guys.

If I had carried out the above open FG loop procedure but had left some small numbers in the "boost integral gain" auF0286 , would this be why im seeing 5-10% duty cycle even with zero DC in the auF16463 wastegate duty cycle table? Im yet to get a chance to test it.

 

Cheers 

[Roland@pcmtec]

 This is good information worth reading in relation to boost control

 

 

[overlap]

Can we still effectively run an over boost shutdown in open loop boost control ?

[Roland@pcmtec]

4 hours ago, overlap said:

Can we still effectively run an over boost shutdown in open loop boost control ?

Yes

[overlap]

8 hours ago, Roland@pcmtec said:

Yes

would that be wastegate failed closed? or a combination ? thanks

[Roland@pcmtec]

3 hours ago, overlap said:

would that be wastegate failed closed? or a combination ? thanks

Test it and find out. Set it to some low number and make sure it goes into limp mode. You can also use tmap and boost sensor max voltage. 

[overlap]

Which part of the open loop procedure causes applied dc to the solenoid at idle and stalled? 

[overlap]

8 minutes ago, overlap said:

Which part of the open loop procedure causes applied dc to the solenoid at idle and stalled? 

Found it 👍