Closed loop manifold pressure using DBW throttle

[bradsm87]

I'm still undecided what would be the simplest way to implement or present this but long story short, I'd really like to have closed loop manifold pressure for targeting boost below the wastegate actuator's setpoint for some zones of the accelerator pedal's travel.

An example scenario. 10psi wastegate actuator setpoint. 15psi WOT boost target.

If I hit the accelerator pedal to 50%, I'd like the throttle to snap wide open but have it pull back to maintain 100kpa absolute manifold pressure once boost starts to build.

If I hit the accelerator pedal to 60%, throttle to snap wide open and pull back once boost reaches 5psi to maintain 5psi.

If I hit the accelerator pedal to 75%, throttle stays wide open and normal boost control is used to maintain 11psi.

It would of course need a table for default accelerator pedal to throttle angle mapping which it defaults to after sudden pedal changes and go open loop with sudden pedal changes. Maybe have the closed loop control inactive until manifold pressure goes above a bit above atmosphere pressure and only start applying closed loop above that point so that everything below that is still very predictable.

I get that there are a shitload of variables involved but damn it would be nice to have the car effectively go full throttle then pull back as boost builds for me!

Edit: Actually closed loop probably isn't needed. Using current manifold pressure as an axis or influence by another means for throttle angle would be enough to get that basic control which I'm guessing is already available. I'll play with it in some base maps while I eagerly await the G4X Fury's release.

[bradsm87]

So I had a play with a base map file.

Am I missing something with the definition of general purpose table? I want to create a new separate table somewhere with say X axis as RPM and Y axis as MAP then fill in table values myself manually (say between 1 and 100) then use the output from that table as an axis on another table (in this case, the throttle 1 table) so that I can still use both RPM and MAP as influences on the throttle table rather than just being able to choose one or the other.

Here is an example of roughly how I'd want to use MAP to influence throttle position, subject to in-car testing and making sure it's not aggressive to the point of oscillating.

[dx4picco]

I mean with a 10psi spring and wide open throttle blade, if you are in the Region where your Turbo can spool you will not be able to target less than these 10psi.

How ever you may be able to target 10 or above. 

You may use a 4 port solenoid and very soft spring to come closer to what you want. Even though I'm not sure about why you would want to do that

[bradsm87]

30 minutes ago, dx4picco said:

I mean with a 10psi spring and wide open throttle blade, if you are in the Region where your Turbo can spool you will not be able to target less than these 10psi.

How ever you may be able to target 10 or above. 

I'm familiar with that. What I am talking about is using the throttle blade to target pressures in the intake manifold after the throttle blade, using the throttle blade, which is very possible.

 

30 minutes ago, dx4picco said:

Even though I'm not sure about why you would want to do that

Why not get predictable output from your accelarator pedal all the way from closed throttle up to your WOT target boost like all modern factory DBW cars? Use throttle blade to regulate manifold pressure below wastegate actuator setpoint and normal boost control above wastegate actuator setpoint. I already use accelerator pedal position between around 70% and 90% to dictate target boost between actuator boost and full throttle boost targets. It'd be nice to better use the zone between around 55% and 70% to use the throttle blade to regulate boost below wastegate actuator boost. That way if you put your accelerator to say 65%, throttle will open wide open until say 5psi is reached then gradually back the throttle off to maintain that 5pi. This is better than the old school driveability issue of going to 60% accelerator pedal and having not enough torque then boost kicks in and suddenly you have more than you want and need to back off a bit. This would effectively automate that process.

[dx4picco]

I totally agree with your statement and did not understand your first request that way. 

Boost by pedal + closed loop boost control will for sure bring you your result. 

I have a similar result that what you looking for with linear WG duty according to tps. Output is pretty linear and repeatable. Even though it may not bring the same manifold pressure at different rpms (you may need different pedal to tps tables) 

[bradsm87]

Does moving the APS % runtime out to another table to feed values into the final e-throttle target table like this have any impact on any safety features? I'm definitely wanting more control than just one 3d table.

[Lotussuper7]

I have set this up on my car, however haven't had a chance to test it yet,  as I am still working on the fueling in the boosted parts of my map.

It's a open loop strategy, so not exactly like you wanted. But I recon it will work well to give you a more linear feel at boost pressures below wastgate.

In general it just uses a gp pwm table with Map on the x axis and Aps on the y axis, then your main e throttle table y axis is changed from Aps to the gp target table.

The gp target table is then set up to reduce the sudo Aps value as you come into positive pressure. I have no change in the vacuum part of the map.

Some testing will be needed to find the right reduction in Aps per psi of boost. Hopefully it will not induce instability, only testing will tell.

My gp table has an approx 10% reduction in Aps value from 0 psi Mgp to around 15 psi Mgp.

Have you done any further testing?