Mysterious power flatline from 5k rpm - tuner is stumped

[jdniss]

On 2/14/2024 at 1:22 AM, CowsWithGuns said:

Is the boost solenoid plumbed correctly?

 

Tuner plumbed it to his preferred way after I agreed to the recommendation to ditch my 4 port. It is done like so, except the tee is using the top port of the wastegate actuator (it has the option of two fittings on top of the membrane).

I've heard several stories where 4-ports were ditched due to poor boost control resolution, both external gate and twin-port internal gate (including a number of tuners) - I bought one myself to replace a 3-port but never installed it.

Stephan Papadakis mentioned they use dual solenoids per wastegate on their Supra drift car as opposed to a single 4-port ~ around the 11min40 mark, should be timestamped:

 

[Lithium]

8 minutes ago, jdniss said:

I've heard several stories where 4-ports were ditched due to poor boost control resolution, both external gate and twin-port internal gate (including a number of tuners) - I bought one myself to replace a 3-port but never installed it.

Stephan Papadakis mentioned they use dual solenoids per wastegate on their Supra drift car as opposed to a single 4-port ~ around the 11min40 mark, should be timestamped:

I've only tuned one 4-port setup and I've essentially ended up finding it increasingly difficult to control nicely as I go further from gate pressure and haven't been tooooo certain that it's resolution alone that's causing the issues for me.

It's an RB30 in a RWD R32 and owner went with a ~.4bar spring in a 66mm Precision wastegate and 4 port BCS, the aim to be able to get the turbo (75lb/min Borg Warner) to it's limit with it while also being able to hold back power in the earlier gears to make it usable.  If I use "open loop" boost control boost follows a nice curve at high boost levels buuuut it will deviate from that curve easily in different atmospheric conditions. 

If I use closed loop to manage it the thing is pretty stable up to around 20psi, but anything past then and it starts getting a little erratic - any input from closed loop boost control is too aggressive and if I dial the PID back then its "not enough" either, I guess you could blame the tuner but this is far from the first car I've done PID with with using the same engine management (G4+ Link).  I've put it down to being a perfect storm of soft spring, huge bypass area with the gate, the aforementioned resolution with the 4-port and also limitations with the closed loop tuneability on the Link which isnt NORMALLY a problem.

What we're aiming at trying soon is as you mentioned, two 3-port gates.  I'm going to try running one of them as a standard PWM with a fixed value for each boost target just to "bring up the base pressure", then use the other one as the target of the actual boost control logic and see if that helps things or makes them messier.   

[CowsWithGuns]

Saga complete, coming back with a summary before bed

Installed the smallest kelford cams, replaced the turbo, added an NZ wiring trigger kit for good measure. No more problems, only good times.

 

Interesting bonus feedback from the tuner is the knock threshold has improved, and there is room for more power with more boost, but he advises against it due to the stock rods + reused bolts in my motor.

Back to back boost curves show no change, implying that the turbo failure had no relation to the root issue. They had no trigger issues in the first place so no reason to suspect the NZ wiring kit to be the solution. Hence, we conclude my stock cams were faulty somehow as indicated by measurements of duration (exhaust side), and the symptoms presented.

Cam centrelines at 125 and 114 per kelford's recommendations, no changes made on the dyno.

Car is fun as all hell, I'm traction limited in 1st and 2nd which is where most of my fun time is anyway. Major gains from 2000 rpm onwards, subjectively no change at 1500, significantly slower sub 1500. Probably due to the cams (?), but the improvement is violent at exactly 1500 so I think that's the currect VCT-on threshold and could be lowered. Regardless - car drive, car good, car fun 

Time to take a break and enjoy, then get some ABS and aircon back in the beast!

Many thanks to everyone who shared their advice and moral support during the trouble period

 

Edited April 15 by CowsWithGuns

[joshuaho96]

Never a dull moment with these engines is there?

[Murray_Calavera]

Very nice

On the plus side, it isn't too hard to hook up the power you have if your prepared to use semi slicks on the road... But then you'll be back to chasing more power so maybe don't get sticky tyres lol.

[Duncan]

well....glad to hear that worked out and thanks for coming back with the info.

No idea how stock cams themselves could be an issue but at least it is sorted

[Lithium]

6 hours ago, CowsWithGuns said:

Saga complete, coming back with a summary before bed

Installed the smallest kelford cams, replaced the turbo, added an NZ wiring trigger kit for good measure. No more problems, only good times.

Interesting bonus feedback from the tuner is the knock threshold has improved, and there is room for more power with more boost, but he advises against it due to the stock rods + reused bolts in my motor.

Cam centrelines at 125 and 114 per kelford's recommendations, no changes made on the dyno.

Car is fun as all hell, I'm traction limited in 1st and 2nd which is where most of my fun time is anyway. Major gains from 2000 rpm onwards, subjectively no change at 1500, significantly slower sub 1500. Probably due to the cams (?), but the improvement is violent at exactly 1500 so I think that's the currect VCT-on threshold and could be lowered. Regardless - car drive, car good, car fun 

 

Awesome outcome, thanks for the update.

One thing that leaps out to me here, and I (and others) had been asking about this back at the time you were working through issues... it seemed like timing was held back on and your tuner said it was normal, yet here it's mentioned that the knock threshold has improved and suddenly it could be pushed further if the engine was stronger - which to me very much implies that knock was an issue where it wouldn't normally be so much of an issue.

I don't know what testing was done for triggering issues but I can say first hand that just relying on your typical ECU trigger problem detection to pick up on the issues you can get with the stock CAS setup isn't the way.  I know of plenty of people who have had a stock CAS and had issues that have not resulted in an error logged but have 100% got timing drift which adversely effects the knock threshold and even general driveability.

Here's example of a car that had trigger issues (knocky and just not making the power it should) and they changed to an NZ Wiring kit, 0 mechanical OR tune changes at this point from when it was tuned with the dodgy triggering... so would have made more with further adjustments, this is just to show the effects of a stock CAS not playing nice.


Doesn't look wildly unlike where yours woke up.... but either way, highlights the importance of a good trigger setup and why I always insist that at the very least someone gets an NZ Wiring kit if they're modding their RB at all and still have the stock CAS.

Edited April 15 by Lithium

[Dose Pipe Sutututu]

9 hours ago, CowsWithGuns said:

Car is fun as all hell, I'm traction limited in 1st and 2nd

you would be with the boost coming in hard & fast, do you run a proper LSD in the back? If not, time to invest in a proper LSD. The viscous one is terrible for putting power down or doing anything motorsport related.