Can't get smooth boost target with Boost Limit Multiplier per Gear

[bradsm87]

I'm having trouble getting a smooth 1st gear boost target with boost limit multiplier per gear.

I run 160 load target in 1st and the following boost limit multiplier per gear values. It is a very smooth increase slope:



Here is a log:

http://datazap.me/u/bradsm87/rr600-r...8&zoom=225-245

As you can see, there is a huge sudden step in boost target somewhere between 2700 and 3000rpm, then the boost target slowly drops off again.

The problem with the big and sudden step is that WGDC drops right down when the actual boost approaches the very low target before the step, making it laggier than it needs to be.

Load per gear was a lot easier to set up. It's a shame about the big power surge between 1st and 2nd when LPG is used with AT.

How are people getting a smooth boost target with boost limit multiplier per gear? Can anyone share their values that work well?

[bradsm87]

Bump.

I'm about to try these settings. It's snowing here so it might be a while until I can actually test it.

Has anyone else had to start high and suddenly drop the BLM to get boost target smooth?

[Jeffman]

Hi Brad,
I've been working on a similar problem with Martial over the last few weeks (he's now off a few weeks en vacance). I thought this may be a MHD logic glitch but now I'm not sure. Still a work in progress. In the meantime I'm happy to share my last settings and comments to Martial with you:

I changed the BLM Per Gear table as (Martial) suggested, starting with lower MAF breakpoints, and gradually increasing the BLM with MAF, as follows

140 170 200 220 230 260 290 330
1 1.45 1.50 1.55 1.60 1.65 1.70 1.75 1.80
2 1.60 1.65 1.70 1.75 1.80 1.85 1.90 1.95
3 1.75 1.80 1.85 1.90 1.95 2.00 2.05 2.10
4 1.90 1.95 2.00 2.05 2.10 2.15 2.20 2.25
5 1.90 1.95 2.00 2.05 2.10 2.15 2.20 2.25
6 1.90 1.95 2.00 2.05 2.10 2.15 2.20 2.25

Load (AT) values same as before.
Remember: Flash Only (no JB4)
Here is my new WOT log in 3rd gear:
http://datazap.me/u/jeffman/log-1468...-23&zoom=36-81

Notice how the Boost Target psi and Boost Setpoint track each other very well, and they increase in a step-wise fashion, with the step-ups occurring when the maf req wgdc gs values are near the break points (140, 170, 200 etc). So it looks like a possible value interpolation glitch? And you can see how my WGDC Base 1 is responding on-off accordingly...

I'm not sure if this is a bug, but it's something to look into -- the breakpoint definitions in the Boost Limit Mulitplier tables:
Cell units in the breakpoint table for BLM Per Gear MAF is "MAF Req. (WGDC)(gs)
Column Units in the BLM Per Gear table is MAF..

I hope this info helps!

[bradsm87]

Thanks. It's good to know I'm not the only one.

I think I'll use a combination of lower load target and less of a restriction on BLM in the hope for a smoother boost target. I formerly used load per gear only and it was great except for the surge in power from 1st to 2nd due to a difference of around 55 between the load targets of 1st and 2nd gear. I think a difference of around 30 won't be an issue (around 140 load target in 1st and 170 in 2nd). I'll then just use BLM per gear to get the boost down a bit further in 1st (rather than using BLM per gear as the only way to reduce boost target).

[Jeffman]

I thought the load per gear tables were mainly used by the manual transmission guys? Mine is zeroed out and I only use load per rpm.
Let me know how you make out with both tables in play.

[08_335i]

I thought the load per gear tables were mainly used by the manual transmission guys? Mine is zeroed out and I only use load per rpm.
Let me know how you make out with both tables in play.

I was trying to figure out the same, why are you using LPG tables on an AT?

[bradsm87]

I was trying to figure out the same, why are you using LPG tables on an AT?

Both LPG and BLMPG can be used together. The only downside of LPG with AT is a sudden increase in power when changing to the next gear when you have big load target changes between those gears. Smaller differences are not noticeable. The other advantage to using a combination of LPG and BLMPG instead of just BLMPG is better throttle control in 1st gear (BLMPG caps the boost so most of the second half of the throttle travel does nothing).

[RSL]

I haven't messed with either limiting table much, but noticed the same issue with LPG. I made a "U" shaped load pattern, so anything in normal driving RPM range was close/same gear to gear and it alleviated that unevenness between them for just driving. I messed with BLM by gear just once so far and had smooth targets, but was overshooting. I'll get around to adjusting the base that low at some point, but this is what my target looked like:

http://www.datazap.me/u/rsl/v9a-blm-...-31&zoom=50-86

[TwistedTuning]

Use one or the other guys. Manual transmission car can use Boost or Load limiter. AT/DCT need to use Boost limiter. You dont need to be using both at once. Im willing to bet your issue is that while using bother your load targets in gear are not lining up with you Boost targets in gear. So its causing an issue, kinda like bumping heads.

[RSL]

Just for the record, mine were always separate, so Twisted might be on to something. I messed with LPG once or twice, but since then, they've not been set for limiting, they match my standard load target tables. When I went to BLM by gear, I zeroed out the LPG breaks/z-cells altogether and used the standard load target tables.

[bradsm87]

Use one or the other guys. Manual transmission car can use Boost or Load limiter. AT/DCT need to use Boost limiter. You dont need to be using both at once. Im willing to bet your issue is that while using bother your load targets in gear are not lining up with you Boost targets in gear. So its causing an issue, kinda like bumping heads.

I only used boost limit multiplier per gear in the beginning and the issue was there. I have started using a combination of both which has helped.

There is no harm with using both. Martial has confirmed. BLM caps boost without scaling the throttle so after that boost has been reached, the accelerator pedal does nothing past that point. Load target scales the lower load throughout the throttle. I will be using both to get a happy medium between the two. More useable accelerator pedal travel in 1st than what BLMPG usually has will be handy when a little wheelspin does occur or any other reason to have a bit of finer control over power output in 1st. Using a little lower load target in 1st is no different to those that have that lower load target in all gears.

[bradsm87]

When I went to BLM by gear, I zeroed out the LPG breaks/z-cells altogether and used the standard load target tables.

MHD copies the standard load table into the LPG table at flash time if the LPG table is empty and the logic still uses the LPG table. Once LPG has been used once, that logic is always used unless you flash back to stock.

[RSL]

Gotcha. Thought I had read that somewhere and saw no long flashes when changing between LPG and BLM by gear.

[08_335i]

Use one or the other guys. Manual transmission car can use Boost or Load limiter. AT/DCT need to use Boost limiter. You dont need to be using both at once. Im willing to bet your issue is that while using bother your load targets in gear are not lining up with you Boost targets in gear. So its causing an issue, kinda like bumping heads.

See this is what I thought it was supposed to be it too.

[RSL]

Gotcha. Thought I had read that somewhere and saw no long flashes when changing between LPG and BLM by gear.

I'm going to take that back, I just added BLM by gear back in and it was a 20 minute flash. I don't remember that happening the other time I tried it. LPG is zeroed out anyway.

[Jeffman]

I only used boost limit multiplier per gear in the beginning and the issue was there. I have started using a combination of both which has helped.

There is no harm with using both. Martial has confirmed. BLM caps boost without scaling the throttle so after that boost has been reached, the accelerator pedal does nothing past that point. Load target scales the lower load throughout the throttle. I will be using both to get a happy medium between the two. More useable accelerator pedal travel in 1st than what BLMPG usually has will be handy when a little wheelspin does occur or any other reason to have a bit of finer control over power output in 1st. Using a little lower load target in 1st is no different to those that have that lower load target in all gears.

Brad, this makes so much sense. Lately I've barely used first gear because pressing the pedal more than just a tap makes the car unruly. I'll try messing with the LPG table this weekend. Thanks for the advice.

[bradsm87]

So I've been through about 5 more revisions and now I seriously think that the BLM per gear on my car is not working properly at all. I realised as I lowered and lowered my BLM with each revision, that the boost target was not changing AT ALL.

It's still targeting 9.3psi in a spot where I have the BLM set at 1.08. It seems most people run a BLM in 1st starting from 1.3 and increasing. Very confusing!

I'm testing my 1st gear targets in 2nd gear so that it actually gets traction for logging until I can get to the bottom of it.

I also have tried to counteract the dramatic step between around 112 and 120 MAF, also completely without success.

Any ideas @martial@mhd ?

http://datazap.me/u/bradsm87/rev17-1...20&zoom=89-141

[Jeffman]

Brad, I think we are experience similar issues. Why should the boost targets be increasing in a step-wise fashion, tracking to a certain degree the BLM v MAF table entries? I also observed this phenomenon (see my Datazap link in an earlier post above) and raised this with Martial, suggesting that the target boost interpolation logic was a little buggy.

Your Datazap
MAF Boost Target
116 8.5
125 9.3

Your BLM v MAF Table 2nd gear
MAF BLM
113 1.40
120 1.08
158 1.14

You are getting an over boost-throttle closure when your MAF(WGDC) is 139-148. Perhaps your overboost is occurring because you're telling your engine to now have BLM = 1.08 @ MAF=120 while your turbos are spooling up and you're building pressure, then suddenly you're telling it to slow down. I may be wrong but it could be that our system (control logic in view of the dynamic limitations of our turbos, wastegates, charge pipes, etc) doesn't like sudden reduction in boost request during spooling. I'd try changing your BLM value at MAF=113 from 1.40 to less than or equal to 1.08. In other words don't reduce BLM in the table as MAF increases.

[bradsm87]

Brad, I think we are experience similar issues. Why should the boost targets be increasing in a step-wise fashion, tracking to a certain degree the BLM v MAF table entries? I also observed this phenomenon (see my Datazap link in an earlier post above) and raised this with Martial, suggesting that the target boost interpolation logic was a little buggy.

Your Datazap
MAF Boost Target
116 8.5
125 9.3

Your BLM v MAF Table 2nd gear
MAF BLM
113 1.40
120 1.08
158 1.14

You are getting an over boost-throttle closure when your MAF(WGDC) is 139-148. Perhaps your overboost is occurring because you're telling your engine to now have BLM = 1.08 @ MAF=120 while your turbos are spooling up and you're building pressure, then suddenly you're telling it to slow down. I may be wrong but it could be that our system (control logic in view of the dynamic limitations of our turbos, wastegates, charge pipes, etc) doesn't like sudden reduction in boost request during spooling. I'd try changing your BLM value at MAF=113 from 1.40 to less than or equal to 1.08. In other words don't reduce BLM in the table as MAF increases.

The boost target does not drop, it increases. The boost control/throttle closure logic knows nothing of BLM, just boost target. I can fix that little overboost no problem with PID, but it's not the target boost I want to run. I want lower boost but no amount of BLM lowering is getting the boost target lower.

[TwistedTuning]

email me the file you are currently running, i'll look it over and see if i see something out of sorts.

[RSL]

I don't normally start from that low of an RPM, which is apparently why I've never seen the target "step" issue. It seemed the first "step" flat line occurs from some starting point prior to first cell value until the first column value, but may be entirely coincidental. The step may be an interpolation thing. I always start WOT logs at or over 2000RPM and never see that step, even with a 100g/s first cell.

I tried a huge spread on MAF, starting at 80g/s with the next cell not until 200g/s with the same, obscenely low, BLM and target still steps going WOT <2000. So, as far as this table goes, nothing is requesting anything different between 2000-4500RPM (80-200g/s) and the step is there...flat target from just below first break (58-79g/s) and then steps up when crossing 80 my first column break point. All the logs I tried starting <2000 have this relative to first column MAF, any logs starting after 2000RPM do not. Yours exhibits the exact same behavior, flat for 20-30g/s before first break point and then stepping up when hitting it. Try starting your pull at 2000RPM instead and see if it disappears.

All that said, I'm not entirely sure it's that big a deal though. Actual boost curve is pretty smooth, as mentioned before. As for not being able to reduce target, you may need to start looking at other tables because they may be pushing it up in that area. I clearly have no problem running a target of <5psi max LOL

My WGDC is in no way adjusted for running this little boost LOL



http://www.datazap.me/u/rsl/v11i-blm...=13&zoom=56-92

[bradsm87]

Yeah I don't really care so much about the step any more. I only have 1 step. The first one in my log is just from going WOT. I will try sneaking another breakpoint in so that there are two before the step and see if that fixes it. The thing I need now is to get the boost target down. I'm at a loss as to why the BLM per gear does not seem to have an effect. I'm tempted to to try a long write because it seems like the BLM per gear simply isn't working.

[jyamona]

This may not be that simple to solve until more of the VE calc tables are released. There is a ton that plays into boost target / setpoint, like residual gas pressure model, engine efficiency factor for cam overlap, modeled valve lift position, boost request offset, and a very very complex air model for what the engine is taking in based off air temp, pre / post throttle body map sensors, etc. It uses a multi-node "neural network" to get it's calculations and I haven't found a good way to "remove" this added nonsense.