Discussion and questions related to the course Boost Control
Hi, I race on my 240sx running a sr20det with 400whp at the moment, I race at time attack events with street tires (ad08R), so it can get tricky to put the power down at some corner exit. To be available to control the power better I opted for open loop boost control with TPS vs RPM table, this work great to control the power output with my right foot but it thas sacrifice some boost response on the spoolup areas where with a higher wgdc I would get better response (I'm running a soft spring so I can have a very low minimum boost at any time so thats the less power that I can control, in my case is something like 270whp at 10psi... and 400whp at 18psi), I cant go much higher on the spoolup areas (3500-4200) cause on higher gear I get to much boost, I will like to have a additional map that can add control with load vs rpm or something like that.
It may be other way to do this that I dont know of to control boost for better traction control on the right foot, right now I'm very happy with the tune and performance of the engine but I will like to add some more advantage agains the competition.
Which ECU? Some have switchable boost maps which could potentially be triggered by gear (if that's the only problem), so different WG DC maps
and trims for different gears (or groups of gears - e.g. Link has up to 3 tables so you'd have to pair gears).
Slightly different and quite specific to an ECU, but when I am up and running with my FI project I plan to try using the Link G4+ "interpolate between" setting (intended for multi-fuel) between two WG tables using TPS as the "interpolate between" input. It might not work as intended (it does in a spreadsheet), but should allow:
- boost by TPS **and** using Andre's famous passive closed loop using open loop WG table boost control method (normally you have to choose one or the other)
- different trims of WG duty cycle at either end of the TPS range (eg different per gear trim depending on small vs large TPS opening), which may be what you are after
I am not sure how useful or effective the above is, but I am sure it is a fun exercise if I ever find time. If anyone reads the above and can say "that won't work" then feel free to tell me as it will save me wasting time!
(edit - thought I better open the software and remind myself and unfortunately the trims are global not specific to each boost table)
Sorry I forget to tell that I am running on a link g4+ storm ecu.
Ok that helps narrow it down. Note I changed my post above when I looked again, and discovered the software doesn't do what I thought it did.
Does using multiple WG DC tables switched by gear (using a Virtual Aux) do what you want?
Alternatively you could ask Link to have boost tables operate as an overlay (as well as switched or interpolating). That would allow all sorts of flexibility and you wouldn't be the first to ask (and apparently the more that ask, the more likely to be implemented).
Yes it that have a trim per gear but that wouldnt improve the response it would just lower all the wgdc on the higher gears, I dont have problems of over boost on higher gear the problem is that for example I tune the wgdc for max reponse on third, that mean I have to turnup the wgdc until 4250rpm, that is where I have (more or less) fullboost, but on fourth I have fullboost much earlier and I will get overboost right until the 4250rpm... in fifth it get much worst.
If I could have the tps vs rpm and the load vs rpm or something else would be great... or other way around this...
I think I have a idea... Maybe I can run 3 boost tables (the max tables that g4+ software allowe), that activate from a Virtual Aux (each boost table), and 1 Virtual Aux for, let say firts to third gear and then fourth and fifth from a Virtual Aux 2 and 3. That way I can have 3 independent maps for the 3 more importants gear that I use on the track. I havent try this yet and I will have to wire the Speed sensor on the gearbox but it may work...
Yep - that's what I thought might work after realizing my initial idea wouldn't.
- 2 boost tables, each set up with rpm and gear as the axes
- one table is for lower throttle, the other full throttle
- use "interpolate between" on TPS position
This should give a semi-custom (interpolated) WG duty cycle for any given rpm, gear and TPS.
Well I already tune the car on the dyno, and to improve response I interpolate 2 boost map.
1- TPS vs RPM (the one that I have always use in this engine)
2- TPS vs Boost (the new one)
Close to the boost target I'm running 100% the 1st map and anything lower than that 90% the second map, this way at any rpm if I am off the boost with high throttle position I have really high WGDC to improve the response and onces is very close to the boost target it switch to the first map to run target boost as usual. On the dyno and on the street feel very good, I have a race this weekend so I will try it and let you guys know the results.
Good luck and it would be good to know how it works for you on the track.
Well it performe as expected, it help the spool up like on the street and on the dyno without problems with the boost target, no overboost or weir thing, very happy with the result it take some work to setting properly but is not that hard and it worth the time... For me having the control of the power output on the right foot is essential, and without E-throttle is not that easy to have for how the turbo engine works. I'm gonna try with closeloop control now to see how it work on my setup and if is consistent for my application, where i can have 1 psi of error.
Rossobianconero, do you have a copy of your map with this boost control strategy configured that I could look at? This sounds ideal for my 1jz setup.
No problem I will upload my tune file, I did it the other they but apparently it did not work. I dont have a my rom file here I will post it when I get back home.
You can also setup closed loop boost control and zero all PID. This makes it openloop again, but whith the difference, that WG is hold at 90% (or wat ever you want) until a defined amount of boost before the target(point where PID's get active). I don't have the excact name of the setting parameter in my head yet, but can tell you when I'm at my laptop again.